{"collection":"philosophy-ethics","full_text_available":true,"graph_url":"https://unchartedknowledge.5gfusion.net/v2/api/page/gutenberg_62856/graph","html_url":"https://unchartedknowledge.5gfusion.net/v2/page/gutenberg_62856","id":"gutenberg_62856","license":"Public Domain (US)","rights_basis":"gutenberg_metadata_copyright_false","source":"gutenberg","source_url":"https://www.gutenberg.org/ebooks/62856","summary":null,"tags":["Category: Philosophy & Ethics","Emotions","Ethics","Knowledge, Theory of","Philosophy, English"],"text":"Introductions to Books I and II of\nDavid Hume\u2019s Treatise of Human Nature\n\nThomas Hill Green\n\nA Treatise of Human Nature, being an attempt to introduce the\nExperimental Method of Reasoning into Moral Subjects and Dialogues\nConcerning Natural Religion\n\nby David Hume,\n\nEdited, with preliminary dissertation and Notes, by T.H. Green\nand T.H. Grose\n\nLondon, Longmans Green & Co, 1874\n\n\nTranscriber's Note:\n\nThe Introduction to Book I is taken from an 1898 reprint; that to\nBook II from an 1882 reprint, both by Longmans.\n\nThe tables of contents have been changed to refer to paragraphs\ninstead of pages, as was done by R.L. Nettleship in his edition of\nGreen's _Philosophical Works_. The paragraph numbers are the same as\nin the originals, and as in Nettleship's edition.\n\nThe Notes which were printed in the margins of the originals have\nbeen placed as captions above the relevant paragraphs.\n\nGreen\u2019s footnotes have been placed below the paragraphs to which they\nrelate. Because this book does not contain Hume\u2019s text, where Green\ncites Hume by page number, a reference to the relevant section has\nbeen added in square brackets. Greek phrases are translated in\nfootnotes marked \"Tr.\"\n\n\n\n\nPREFACE.\n\nIn this edition we have sought to avoid the inconveniences which are\napt to attend commentaries on philosophical writers, by the plan\nof putting together, in the form of continuous introductions, such\nexplanation and criticism as we had to offer, and confining the\nfootnotes almost entirely to references, which have been carefully\ndistinguished from Hume\u2019s own notes. For the introductions to\nthe first and second volumes Mr. Green alone is responsible. The\nintroduction to the third is the work of Mr. Grose, who also has\nundertaken the revision of Hume\u2019s text.\n\nThroughout the introductions to Volumes I. and II., except where\nthe contrary is stated, \u2018Hume\u2019 must be understood to mean Hume as\nrepresented by the \u2018Treatise on Human Nature.\u2019 In taking this as\nintrinsically the best representation of his philosophy, we may\nbe thought to have overlooked the well-known advertisement which\n(in an edition posthumously published) he prefixed to the volume\ncontaining his \u2018Inquiries concerning the Human Understanding and\nthe Principles of Morals.\u2019 In it, after stating that the volume,\nis mainly a reproduction of what he had previously published in\nthe \u2018Treatise,\u2019 he expresses a hope that \u2018some negligences in his\nformer reasoning, and more in the expression,\u2019 have been corrected,\nand desires \u2018that the following Pieces may alone be regarded as\ncontaining his philosophical sentiments and principles.\u2019 Was not Hume\nhimself then, it may be asked, the best judge of what was an adequate\nexpression of his thoughts, and is there not an unbecoming assurance\nin disregarding such a voice from his tomb?\n\nOur answer is that if we had been treating of Hume as a great\nliterary character, or exhibiting the history of his individual\nmind, due account must have been taken of it. Such, however, has not\nbeen the object which, in the Introductions to Volumes I. and II.,\nwe have presented to ourselves, (See Introd. to Vol. I. \u00a7 4.) Our\nconcern has been with him as the exponent of a philosophical system,\nand therefore specially with that statement of his system which\nalone purports to be complete, and which was written when philosophy\nwas still his chief interest, without alloy from the disappointment\nof literary ambition. Anyone who will be at the pains to read the\n\u2018Inquiries\u2019 alongside of the original \u2018Treatise\u2019 will find that their\nonly essential difference from it is in the way of omission. They\nconsist in the main of excerpts from the \u2018Treatise,\u2019 re-written in\na lighter style, and with the more difficult parts of it left out.\nIt is not that the difficulties which logically arise out of Hume\u2019s\nsystem are met, but that the passages which most obviously suggest\nthem have disappeared without anything to take their place. Thus in\nthe \u2018Inquiry concerning the Human Understanding\u2019 there is nothing\nwhatever corresponding to Parts II. and IV. of the first Book of the\n\u2018Treatise.\u2019 The effect of this omission on a hasty reader is, no\ndoubt, a feeling of great relief. Common-sense is no longer actively\nrepelled by a doctrine which seems to undermine the real world, and\ncan more easily put a construction on the account of the law of\ncausation, which remains, compatible with the \u2018objective validity\u2019 of\nthe law--such a construction as in fact forms the basis of Mr. Mill\u2019s\nLogic. How inconsistent this construction is with the principles\nfrom which Hume started, and which he never gave up; how impossible\nit would be to anyone who had assimilated his system as a whole; how\nclose is the organic connection between all the parts of this as he\noriginally conceived it--we must trust to the following introductions\nto show. (See, in particular, Introd. to Vol. I. \u00a7\u00a7 301 and 321.)\n\nThe only discussion in the \u2018Inquiry concerning Human Understanding,\u2019\nto which nothing in his earlier publication corresponds, is that\non Miracles. On the relation in which this stands to his general\ntheory some remarks will be found in the Introduction to Vol. I. (\u00a7\n324, note). The chief variations, other than in the way of omission,\nbetween the later redaction of his ethical doctrine and the earlier,\nare noticed in the Introduction to Vol. II. (\u00a7\u00a7 31, 43, and 46, and\nnotes).\n\n\nSUMMARY OF THE CONTENTS\n\nof the\n\nGENERAL INTRODUCTION TO VOL. I\n\n1. How the history of philosophy should be studied.\n\n2. Hume the last great English philosopher.\n\n3. Kant his true successor.\n\n4. Distinction between literary history and the history of\nphilosophical systems.\n\n5. Object of the present enquiry.\n\n6. Locke\u2019s problem and method.\n\n7. His notion of the \u2018thinking thing\u2019.\n\n8. This he will passively observe.\n\n9. Is such observation possible?\n\n10. Why it seems so.\n\n11. Locke\u2019s account of origin of ideas.\n\n12. Its ambiguities _(a)_ In regard to sensation.\n\n13. _(b)_ In regard to ideas of reflection.\n\n14. What is the \u2018tablet\u2019 impressed?\n\n15. Does the mind make impressions on itself?\n\n16. Source of these difficulties. The \u2018simple\u2019 idea, as Locke\ndescribes it, is a complex idea of substance and relation.\n\n17. How this contradiction is disguised.\n\n18. Locke\u2019s way of interchanging \u2018idea\u2019 and \u2018quality\u2019 and its effects.\n\n19. Primary and secondary qualities of bodies.\n\n20. \u2018Simple idea\u2019 represented as involving a theory of its own cause.\n\n21. Phrases in which this is implied.\n\n22. Feeling and felt thing confused.\n\n23. The simple idea as \u2018ectype\u2019 other than mere sensation.\n\n24. It involves a judgment in which mind and thing are distinguished.\n\n25. And is equivalent to what he afterwards calls \u2018knowledge of\nidentity\u2019. Only as such can it be named.\n\n26. The same implied in calling it an idea of an object.\n\n27. made _for_, not _by_, us, and therefore according to Locke really\nexistent.\n\n28. What did he mean by this?\n\n30. Existence as the mere presence of a feeling.\n\n31. Existence as reality.\n\n32. By confusion of these two meanings, reality and its conditions\nare represented as given in simple feeling.\n\n33. Yet reality involves complex ideas which are made by the mind.\n\n34. Such are substance and relation which must be found in every\nobject of knowledge.\n\n35. Abstract idea of substance and complex ideas of particular sorts\nof substance.\n\n36. The abstract idea according to Locke at once precedes and follows\nthe complex.\n\n37. Reference of ideas to nature or God, the same as reference to\nsubstance.\n\n38. But it is explicitly to substance that Locke makes them refer\nthemselves.\n\n39. In the process by which we are supposed to arrive at complex\nideas of substances the beginning is the same as the end.\n\n40. Doctrine of abstraction inconsistent with doctrine of complex\nideas.\n\n41. The confusion covered by use of \u2018particulars\u2019.\n\n42. Locke\u2019s account of abstract general ideas.\n\n43. \u2018Things not general.\u2019\n\n44. Generality an invention of the mind.\n\n45. The result is, that the feeling of each moment is alone real.\n\n46. How Locke avoids this result.\n\n47. The \u2018particular\u2019 was to him the individual qualified by general\nrelations.\n\n48. This is the real thing from which abstraction is supposed to\nstart.\n\n49. Yet, according to the doctrine of relation, a creation of thought.\n\n50. Summary of the above contradictions.\n\n51. They cannot be overcome without violence to Locke\u2019s fundamental\nprinciples.\n\n52. As real existence, the simple idea carries with it \u2018invented\u2019\nrelation of cause.\n\n53. Correlativity of cause and substance.\n\n54. How do we know that ideas correspond to reality of things?\nLocke\u2019s answer.\n\n55. It assumes that simple ideas are consciously referred to things\nthat cause them.\n\n56. Lively ideas real, because they must be effects of things.\n\n57. Present sensation gives knowledge of existence.\n\n58. Reasons why its testimony must be trusted.\n\n59. How does this account fit Locke\u2019s definition of knowledge?\n\n60. Locke\u2019s account of the testimony of sense renders his question as\nto its veracity superfluous.\n\n61. Confirmations of the testimony turn upon the distinction between\n\u2018impression\u2019 and \u2018idea\u2019.\n\n62. They depend on language which pre-supposes the ascription of\nsensation to an outward cause.\n\n63. This ascription means the clothing of sensation with invented\nrelations.\n\n64. What is meant by restricting the testimony of sense to _present_\nexistence?\n\n65. Such restriction, if maintained, would render the testimony\nunmeaning.\n\n66. But it is not maintained: the testimony is to operation of\npermanent identical things.\n\n67. Locke\u2019s treatment of relations of cause and identity.\n\n68. That from which he derives idea of cause pre-supposes it.\n\n69. Rationale of this \u2018petitio principii\u2019.\n\n70. Relation of cause has to be put into sensitive experience in\norder to be got from it.\n\n71. Origin of the idea of identity according to Locke.\n\n72. Relation of identity not to be distinguished from idea of it.\n\n73. This \u2018invented\u2019 relation forms the \u2018very being of things\u2019.\n\n74. Locke fails to distinguish between identity and mere unity.\n\n75. Feelings are the real, and do not admit of identity. How then can\nidentity be real?\n\n76. Yet it is from reality that the idea of it is derived.\n\n77. Transition to Locke\u2019s doctrine of essence.\n\n78. This repeats the inconsistency found in his doctrine of substance.\n\n79. Plan to be followed.\n\n80. What Locke understood by essence.\n\n81. Only to nominal essences that general propositions relate, _i.e._\nonly to abstract ideas having no real existence.\n\n82. An abstract idea may be a simple one.\n\n83. How then is science of nature possible?\n\n84. No \u2018uniformities of phenomena\u2019 can be known.\n\n85. Locke not aware of the full effect of his own doctrine ...\n\n86. ... which is to make the real an abstract residuum of\nconsciousness.\n\n87. Ground of distinction between actual sensation and ideas in the\nmind is itself a thing of the mind.\n\n88. Two meanings of real essence.\n\n89. According to one, it is a collection of ideas as qualities of a\nthing:\n\n90. ... about real essence in this sense there may be general\nknowledge.\n\n91. But such real essence a creature of thought.\n\n92. Hence another view of real essence as unknown qualities of\nunknown body.\n\n93. How Locke mixes up these two meanings in ambiguity about body.\n\n94. Body as \u2018parcel of matter\u2019 without essence.\n\n95. In this sense body is the mere individuum.\n\n96. Body as qualified by circumstances of time and place.\n\n97. Such body Locke held to be subject of \u2018primary qualities\u2019: but\nare these compatible with particularity in time?\n\n98. How Locke avoids this question.\n\n99. Body and its qualities supposed to be outside consciousness.\n\n100. How can primary qualities be outside consciousness, and yet\nknowable?\n\n101. Locke answers that they copy themselves in ideas--Berkeley\u2019s\nrejoinder. Locke gets out of the difficulty by his doctrine of\nsolidity.\n\n102. In which he equivocates between body as unknown opposite of mind\nand body as a \u2018nominal essence\u2019.\n\n103. Rationale of these contradictions.\n\n104. What knowledge can feeling, even as referred to a \u2018solid\u2019 body,\nconvey?\n\n105. Only the knowledge that something is, not _what_ it is.\n\n106. How it is that the real essence of things, according to Locke,\nperishes with them, yet is immutable.\n\n107. Only about qualities of matter, as distinct from matter itself,\nthat Locke feels any difficulty.\n\n108. These, as knowable, must be our ideas, and therefore not a \u2018real\nessence\u2019.\n\n109. Are the \u2018primary qualities\u2019 then, a \u2018nominal essence\u2019?\n\n110. According to Locke\u2019s account they are relations, and thus\ninventions of the mind.\n\n111. Body is the complex in which they are found. Do we derive the\nidea of body from primary qualities, or the primary qualities from\nidea of body?\n\n112. Mathematical ideas, though ideas of \u2018primary qualities of body,\u2019\nhave \u2018barely an ideal existence\u2019.\n\n113. Summary view of Locke\u2019s difficulties in regard to the real.\n\n114. Why they do not trouble him more.\n\n115. They re-appear in his doctrine of propositions.\n\n116. The knowledge expressed by a proposition, though certain, may\nnot be real ...\n\n117. ... when the knowledge concerns substances. In this case general\ntruth must be merely verbal. Mathematical truths, since they concern\nnot substances, may be both general and real.\n\n118. Significance of this doctrine.\n\n119. Fatal to the notion that mathematical truths, though general,\nare got from experience:\n\n120. ... and to received views of natural science: but Locke not so\nclear about this.\n\n121. Ambiguity as to real essence causes like ambiguity as to science\nof nature. Particular experiment cannot afford general knowledge.\n\n122. What knowledge it can afford, according to Locke.\n\n123. Not the knowledge which is now supposed to be got by induction.\nYet more than Locke was entitled to suppose it could give.\n\n124. With Locke mathematical truths, though ideal, true also of\nnature.\n\n125. Two lines of thought in Locke, between which a follower would\nhave to choose.\n\n126. Transition to doctrine of God and the soul.\n\n127. Thinking substance--source of the same ideas as outer substance.\n\n128. Of which substance is perception the effect?\n\n129. That which is the source of substantiation cannot be itself a\nsubstance.\n\n130. To get rid of the inner source of ideas in favour of the outer\nwould be false to Locke.\n\n131. The mind, which Locke opposes to matter, perpetually shifting.\n\n132. Two ways out of such difficulties. \u2018Matter\u2019 and \u2018mind\u2019 have the\nsame source in self-consciousness.\n\n133. Difficulties in the way of ascribing reality to substance as\nmatter, re-appear in regard to substance as mind.\n\n134. We think not always, yet thought constitutes the self.\n\n135. Locke neither disguises these contradictions, nor attempts to\novercome them.\n\n136. Is the idea of God possible to a consciousness given in time?\n\n137. Locke\u2019s account of this idea.\n\n138. \u2018Infinity,\u2019 according to Locke\u2019s account of it, only applicable\nto God, if God has parts.\n\n139. Can it be applied to him \u2018figuratively\u2019 in virtue of the\nindefinite number of His acts?\n\n140. An act, finite in its nature, remains so, however often repeated.\n\n141. God only infinite in a sense in which time is _not_ infinite,\nand which Locke could not recognize\n\n142. --the same sense in which the self is infinite.\n\n143. How do I know my own real existence?--Locke\u2019s answer.\n\n144. It cannot be known consistently with Locke\u2019s doctrine of real\nexistence.\n\n145. But he ignores this in treating of the self.\n\n146. Sense in which the self is truly real.\n\n147. Locke\u2019s proof of the real existence of God. There must have been\nsomething from eternity to cause what now is.\n\n148. How \u2018eternity\u2019 must be understood if this argument is to be\nvalid:\n\n149. ... and how \u2018cause\u2019.\n\n150. The world which is to prove an eternal God must be itself\neternal.\n\n151. But will the God, whose existence is so proven, be a thinking\nbeing?\n\n152. Yes, according to the true notion of the relation between\nthought and matter.\n\n153. Locke\u2019s antinomies--Hume takes one side of them as true.\n\n154. Hume\u2019s scepticism fatal to his own premises. This derived from\nBerkeley.\n\n155. Berkeley\u2019s religious interest in making Locke consistent.\n\n156. What is meant by relation of mind and matter?\n\n157. Confusions involved in Locke\u2019s materialism.\n\n158. Two ways of dealing with it. Berkeley chooses the most obvious.\n\n159. His account of the relation between visible and tangible\nextension. We do not see bodies without the mind ...\n\n160. ... nor yet feel them. The \u2018esse\u2019 of body is the \u2018percipi\u2019.\n\n170. What then becomes of distinction between reality and fancy?\n\n171. The real = ideas that God causes.\n\n172. Is it then a succession of feelings?\n\n173. Berkeley goes wrong from confusion between thought and feeling.\n\n174. Which, if idea = feeling, does away with space and body.\n\n175. He does not even retain them as \u2018abstract ideas\u2019.\n\n176. On the same principle all permanent relations should disappear.\n\n177. By making colour = relations of coloured points, Berkeley\nrepresents relation as seen.\n\n178. Still he admits that space is constituted by a succession of\nfeelings.\n\n179. If so, it is not space at all; but Berkeley thinks it is only\nnot \u2018pure\u2019 space. _Space_ and _pure_ space stand or fall together.\n\n180. Berkeley disposes of space for fear of limiting God.\n\n181. How he deals with possibility of general knowledge.\n\n182. His theory of universals ...\n\n183. ... of value, as implying that universality of ideas lies in\nrelation.\n\n184. But he fancies that each idea has a positive nature apart from\nrelation.\n\n185. Traces of progress in his idealism.\n\n186. His way of dealing with physical truths.\n\n187. If they imply permanent relations, his theory properly excludes\nthem. He supposes a divine decree that one feeling shall follow\nanother.\n\n188. Locke had explained reality by relation of ideas to outward\nbody. Liveliness in the idea evidence of this relation.\n\n189. Berkeley retains this notion, only substituting \u2018God\u2019 for \u2018body\u2019.\n\n190. Not regarding the world as a system of intelligible relations,\nhe could not regard God as the subject of it.\n\n191. His view of the soul as \u2018naturally immortal\u2019.\n\n192. Endless succession of feelings is not immortality in true sense.\nBerkeley\u2019s doctrine of matter fatal to a true spiritualism:\n\n193. ... as well as to a true Theism. His inference to God from\nnecessity of a power to produce ideas;\n\n194. ... a necessity which Hume does not see. A different turn should\nhave been given to his idealism, if it was to serve his purpose.\n\n195. Hume\u2019s mission. His account of impressions and ideas. Ideas are\nfainter impressions.\n\n196. \u2018Ideas\u2019 that cannot be so represented must be explained as mere\nwords.\n\n197. Hume, taken strictly, leaves no distinction between impressions\nof reflection and of sensation.\n\n198. Locke\u2019s theory of sensation disappears. Physiology won\u2019t answer\nthe question that Locke asked.\n\n199. Those who think it will don\u2019t understand the question.\n\n200. Hume\u2019s psychology will not answer it either.\n\n201. It only seems to do so by assuming the \u2018fiction\u2019 it has to\naccount for; by assuming that impression represents a real world.\n\n202. So the \u2018Positivist\u2019 juggles with \u2018phenomena\u2019.\n\n203. Essential difference, however, between Hume and the \u2018Positivist\u2019.\n\n204. He adopts Berkeley\u2019s doctrine of ideas, but without Berkeley\u2019s\nsaving suppositions,\n\n205. ... in regard to \u2018spirit\u2019,\n\n206. ... in regard to relations. His account of these.\n\n207. It corresponds to Locke\u2019s account of the sorts of agreement\nbetween ideas.\n\n208. Could Hume consistently admit idea of relation at all?\n\n210. Only in regard to identity and causation that he sees any\ndifficulty. These he treats as fictions resulting from \u2018natural\nrelations\u2019 of ideas: _i.e._ from resemblance and contiguity.\n\n212. Is resemblance then an impression?\n\n213. Distinction between resembling feelings and idea of resemblance.\n\n214. Substances = collections of ideas.\n\n215. How can ideas \u2018in flux\u2019 be collected?\n\n216. Are there general ideas? Berkeley said, \u2018yes and no\u2019.\n\n217. Hume \u2018no\u2019 simply. How he accounts for the appearance of there\nbeing such.\n\n219. His account implies that \u2018ideas\u2019 are conceptions, not feelings.\n\n220. He virtually yields the point in regard to the _predicate_ of\npropositions.\n\n221. As to the subject, he equivocates between singleness of feeling\nand individuality of conception.\n\n222. Result is a theory which admits predication, but only as\nsingular.\n\n223. All propositions restricted in same way as Locke\u2019s propositions\nabout real existence.\n\n224. The question, how the _singular_ proposition is possible, the\nvital one.\n\n225. Not relations of resemblance only, but those of quantity also,\ntreated by Hume as feelings.\n\n226. He draws the line between certainty and probability at the same\npoint as Locke; but is more definite as to probability,\n\n227. ... and does not admit opposition of mathematical to physical\ncertainty--here following Berkeley.\n\n228. His criticisms of the doctrine of primary qualities.\n\n229. It will not do to oppose bodies to our feeling when only feeling\ncan give idea of body.\n\n230. Locke\u2019s shuffle of \u2018body,\u2019 \u2018solidity,\u2019 and \u2018touch,\u2019 fairly\nexposed.\n\n231. True rationale of Locke\u2019s doctrine.\n\n232. With Hume \u2018body\u2019 logically disappears. What then?\n\n233. Can space survive body? Hume derives idea of it from sight and\nfeeling. Significance with him of such derivation.\n\n234. It means, in effect, that colour and space are the same, and\nthat feeling may be extended.\n\n235. The parts of space are parts of a perception.\n\n236. Yet the parts of space are co-existent not successive.\n\n237. Hume cannot make space a \u2018perception\u2019 without being false to his\nown account of perception;\n\n238. ... as appears if we put \u2018feeling\u2019 for \u2018perception\u2019 in the\npassages in question.\n\n239. To make sense of them, we must take perception to mean perceived\nthing,\n\n240. ... which it can only mean as the result of certain \u2018fictions\u2019.\n\n241. If felt thing is no more than feeling, how can it have qualities?\n\n242. The thing will have ceased before the quality begins to be.\n\n243. Hume equivocates by putting \u2018coloured points\u2019 for colour.\n\n245. Can a \u2018disposition of coloured points\u2019 be an impression?\n\n246. The points must be themselves impressions, and therefore not\nco-existent.\n\n247. A \u2018compound impression\u2019 excluded by Hume\u2019s doctrine of time.\n\n248. The fact that colours mix, not to the purpose.\n\n249. How Hume avoids appearance of identifying space with colour, and\naccounts for the abstraction of space.\n\n250. In so doing, he implies that space is a relation, and a relation\nwhich is not a possible impression.\n\n251. No logical alternative between identifying space with colour,\nand admitting an idea not copied from an impression.\n\n252. In his account of the idea as _abstract_, Hume really introduces\ndistinction between feeling and conception;\n\n253. ... yet avoids appearance of doing so, by treating\n\u2018consideration\u2019 of the relations of a felt thing as if it were itself\nthe feeling.\n\n254. Summary of contradictions in his account of extension.\n\n255. He gives no account of quantity as such.\n\n256. His account of the relation between Time and Number.\n\n257. What does it come to?\n\n258. Unites alone really exist: number a \u2018fictitious denomination\u2019.\nYet \u2018unites\u2019 and \u2018number\u2019 are correlative; and the supposed fiction\nunaccountable.\n\n259. Idea of time even more unaccountable on Hume\u2019s principles.\n\n260. His ostensible explanation of it.\n\n261. It turns upon equivocation between feeling and conception of\nrelations between felt things.\n\n262. He fails to assign any impression or compound of impressions\nfrom which idea of time is copied.\n\n263. How can he adjust the exact sciences to his theory of space and\ntime?\n\n264. In order to seem to do so, he must get rid of \u2018Infinite\nDivisibility\u2019.\n\n265. Quantity made up of impressions, and there must be a least\npossible impression.\n\n266. Yet it is admitted that there is an idea of number not made up\nof impressions. A finite division into impressions no more possible\nthan an infinite one.\n\n267. In Hume\u2019s instances it is not really a feeling, but a conceived\nthing, that appears as finitely divisible.\n\n268. Upon true notion of quantity infinite divisibility follows of\ncourse.\n\n269. What are the ultimate elements of extension? If not extended,\nwhat are they?\n\n270. Colours or coloured points? What is the difference?\n\n271. True way of dealing with the question.\n\n272. \u2018If the point were divisible, it would be no termination of a\nline.\u2019 Answer to this.\n\n273. What becomes of the exactness of mathematics according to Hume?\n\n274. The universal propositions of geometry either untrue or\nunmeaning.\n\n275. Distinction between Hume\u2019s doctrine and that of the hypothetical\nnature of mathematics.\n\n276. The admission that no relations of quantity are data of sense\nremoves difficulty as to general propositions about them.\n\n277. Hume does virtually admit this in regard to numbers.\n\n278. With Hume idea of vacuum impossible, but logically not more so\nthan that of space.\n\n279. How it is that we talk as if we had idea of vacuum according to\nHume.\n\n280. His explanation implies that we have an idea virtually the same.\n\n281. By a like device that he is able to explain the appearance of\nour having such ideas as Causation and Identity.\n\n283. Knowledge of relation in way of Identity and Causation excluded\nby Locke\u2019s definition of knowledge.\n\n284. Inference a transition from an object perceived or remembered to\none that is not so.\n\n285. Relation of cause and effect the same as this transition.\n\n286. Yet seems other than this. How this appearance is to be\nexplained.\n\n287. Inference, resting on supposition of necessary connection, to be\nexplained before that connection.\n\n288. Account of the inference given by Locke and Clarke rejected.\n\n289. Three points to be explained in the inference according to Hume.\n\n290. _a_. The original impression from which the transition is made\n_b_. The transition to inferred idea\n\n291. _c_. The qualities of this idea.\n\n292. It results that necessary connection is an impression of\nreflection, _i.e._, a propensity to the transition described.\n\n293. The transition not to anything beyond sense.\n\n294. Nor determined by any objective relation.\n\n295. Definitions of cause: _a_. As a \u2018philosophical\u2019 relation.\n\n296. Is Hume entitled to retain \u2018philosophical\u2019 relations as distinct\nfrom \u2018natural\u2019?\n\n297. Examination of Hume\u2019s language about them.\n\n298. Philosophical relation consists in a comparison, but no\ncomparison between cause and effect.\n\n299. The comparison is between present and past experience of\nsuccession of objects.\n\n300. Observation of succession already goes beyond sense.\n\n301. As also does the \u2018observation concerning identity,\u2019 which the\ncomparison involves.\n\n302. Identity of objects an unavoidable crux for Hume. His account of\nit.\n\n303. Properly with him it is a fiction, in the sense that we have no\nsuch idea. Yet he implies that we have such idea, in saying that we\nmistake something else for it.\n\n304. Succession of like feelings mistaken for an identical object:\nbut the feelings, as described, are already such objects.\n\n305. Fiction of identity thus implied as source of the propensity\nwhich is to account for it.\n\n306. With Hume continued existence of perceptions a fiction different\nfrom their identity. Can perceptions exist when not perceived?\n\n307. Existence of objects, distinct from perceptions, a further\nfiction still.\n\n308. Are these several \u2018fictions\u2019 really different from each other?\n\n309. Are they not all involved in the simplest perception?\n\n310. Yet they are not possible ideas, because copied from no\nimpressions.\n\n311. Comparison of present experience with past, which yields\nrelation of cause and effect, pre-supposes judgment of identity;\n\n312. ... without which there could be no recognition of an object as\none observed before.\n\n313. Hume makes conceptions of identity and cause each come before\nthe other. Their true correlativity.\n\n314. Hume quite right in saying that we do not go _more_ beyond sense\nin reasoning than in perception.\n\n315. How his doctrine might have been developed. Its actual outcome.\n\n316. No philosophical relation admissible with Hume that is not\nderived from a natural one.\n\n317. Examination of his account of cause and effect as \u2018natural\nrelation\u2019.\n\n318. Double meaning of natural relation. How Hume turns it to account.\n\n319. If an effect is merely a constantly observed sequence, how can\nan event be an effect the first time it is observed? Hume evades this\nquestion;\n\n320. Still, he is a long way off the Inductive Logic, which supposes\nan objective sequence.\n\n321. Can the principle of uniformity of nature be derived from\nsequence of feelings?\n\n322. With Hume the only uniformity is in expectation, as determined\nby habit; but strength of such expectation must vary indefinitely.\n\n323. It could not serve the same purpose as the conception of\nuniformity of nature.\n\n324. Hume changes the meaning of this expectation by his account\nof the \u2018remembrance\u2019 which determines it. Bearing of his doctrine\nof necessary connexion upon his argument against miracles. This\nremembrance, as he describes it, supposes conception of a system of\nnature.\n\n326. This explains his occasional inconsistent ascription of an\nobjective character to causation.\n\n327. Reality of remembered \u2018system\u2019 transferred to \u2018system of\njudgment\u2019.\n\n328. Reality of the former \u2018system\u2019 other than vivacity of\nimpressions.\n\n329. It is constituted by relations, which are not impressions at\nall; and in this lies explanation of the inference from it to \u2018system\nof judgment\u2019.\n\n330. Not seeing this, Hume has to explain inference to latter system\nas something forced upon us by habit.\n\n331. But if so, \u2018system of judgment\u2019 must consist of feelings\nconstantly experienced;\n\n332. ... which only differ from remembered feelings inasmuch as their\nliveliness has faded. But how can it have faded, if they have been\nconstantly repeated?\n\n333. Inference then can give no new knowledge.\n\n334. Nor does this merely mean that it cannot constitute new\nphenomena, while it can prove relations, previously unknown, between\nphenomena. Such a distinction inadmissible with Hume.\n\n335. His distinction of probability of causes from that of chances\nmight seem to imply conception of nature, as determining inference.\n\n336. But this distinction he only professes to adopt in order to\nexplain it away.\n\n337. Laws of nature are unqualified habits of expectation.\n\n338. Experience, according to his account of it, cannot be a parent\nof knowledge.\n\n339. His attitude towards doctrine of thinking substance.\n\n340. As to Immateriality of the Soul, he plays off Locke and Berkeley\nagainst each other, and proves Berkeley a Spinozist.\n\n341. Causality of spirit treated in the same way.\n\n342. Disposes of \u2018personal\u2019 identity by showing contradictions in\nLocke\u2019s account of it.\n\n343. Yet can only account for it as a \u2018fiction\u2019 by supposing ideas\nwhich with him are impossible.\n\n344. In origin this \u2018fiction\u2019 the same as that of \u2018Body\u2019.\n\n345. Possibility of such fictitious ideas implies refutation of\nHume\u2019s doctrine.\n\n\n\n\nSUMMARY OF THE CONTENTS\n\nof the\n\nGENERAL INTRODUCTION TO VOL. II.\n\n1. Hume\u2019s doctrine of morals parallel to his doctrine of nature.\n\n2. Its relation to Locke: Locke\u2019s account of freedom, will, and\ndesire.\n\n3. Two questions: Does man always act from the strongest motive?\nand, What constitutes his motive? The latter the important question.\nDistinction between desires that are, and those that are not,\ndetermined by the conception of self.\n\n4. Effect of this conception on the objects of human desire.\n\n5. Objects so constituted Locke should consistently exclude: But he\nfinds room for them by treating every desire for an object, of which\nthe attainment gives pleasure, as a desire for pleasure.\n\n6. Confusion covered by calling \u2018happiness\u2019 the general object of\ndesire.\n\n7. \u2018Greatest sum of pleasure\u2019 and \u2018Pleasure in general\u2019 unmeaning\nexpressions.\n\n8. In what sense of happiness is it true that it \u2018is really just as\nit appears?\u2019 In what sense that it is every one\u2019s object?\n\n9. No real object of human desire can ever be just as it appears.\n\n10. Can Locke consistently allow the distinction between true\nhappiness and false? Or responsibility?\n\n11. Objections to the Utilitarian answer to these questions.\n\n12. According to Locke present pleasures may be compared with future,\nand desire suspended till comparison has been made.\n\n13. What is meant by \u2018present\u2019 and \u2018future\u2019 pleasure? By the supposed\ncomparison Locke ought only to have meant the competition of\npleasures equally present in imagination:\n\n14.... and this could give no ground for responsibility. In order to\ndo so, it must be understood as implying determination by conception\nof self.\n\n15. Locke finds moral freedom in necessity of pursuing happiness.\n\n16. If an action is moved by desire for an object, Locke asks no\nquestions about origin of the object. But what is to be said of\nactions, which we only do because we ought?\n\n17. Their object is pleasure, but pleasure given not by nature but by\nlaw.\n\n18. Conformity to law not the moral good, but a means to it.\n\n19. Hume has to derive from \u2018impressions\u2019 the objects which Locke\ntook for granted.\n\n20. Questions which he found at issue, _a_. Is virtue interested?\n_b_. What is conscience?\n\n21. Hobbes\u2019 answer to first question.\n\n22. Counter-doctrine of Shaftesbury. Vice is selfishness; but no\nclear account of selfishness.\n\n23. Confusion in his notions of self-good and public good; Is all\nliving for pleasure, or only too much of it, selfish?\n\n24. What have Butler and Hutcheson to say about it? Chiefly that\naffections terminate upon their objects; but this does not exclude\nthe view that all desire is for pleasure.\n\n25. Of moral goodness Butler\u2019s account circular. Hutcheson\u2019s\ninconsistent with his doctrine that reason gives no end.\n\n26. Source of the moral judgment: received notion of reason\nincompatible with true view. Shaftesbury\u2019s doctrine of rational\naffection; spoilt by doctrine of \u2018moral sense\u2019.\n\n27. Consequences of the latter.\n\n28. Is an act done for \u2018virtue\u2019s sake\u2019 done for pleasure of moral\nsense?\n\n29. Hume excludes every object of desire but pleasure.\n\n30. His account of \u2018direct passions\u2019: all desire is for pleasure.\n\n31. Yet he admits \u2018passions\u2019 which produce pleasure, but proceed not\nfrom it.\n\n32. Desire for objects, as he understands it, excluded by his theory\nof impressions and ideas.\n\n33. Pride determined by reference to self.\n\n34. This means that it takes its character from that which is not a\npossible \u2018impression\u2019.\n\n35. Hume\u2019s attempt to represent idea of self as derived from\nimpression.\n\n36. Another device is to suggest a physiological account of pride.\n\n37. Fallacy of this: it does not tell us what pride is to the subject\nof it.\n\n38. Account of love involves the same difficulties; and a further one\nas to nature of sympathy.\n\n39. Hume\u2019s account of sympathy.\n\n40. It implies a self-consciousness not reducible to impressions.\n\n41. Ambiguity in his account of benevolence: it is a desire and\ntherefore has pleasure for its object. What pleasure?\n\n42. Pleasure of sympathy with the pleasure of another.\n\n43. All \u2018passions\u2019 equally interested or disinterested. Confusion\narises from use of \u2018passion\u2019 alike for desire and emotion. Of this\nHume avails himself in his account of active pity.\n\n44. Explanation of apparent conflict between reason and passion.\n\n45. A \u2018reasonable\u2019 desire means one that excites little emotion.\nEnumeration of possible motives.\n\n46. If pleasure sole motive, what is the distinction of self-love?\nIts opposition to disinterested desires, as commonly understood,\ndisappears. It is desire for pleasure in general.\n\n47. How Hume gives meaning to this otherwise unmeaning definition.\n\u2018Interest,\u2019 like other motives described, implies determination by\nreason.\n\n48. Thus Hume, having degraded morality for the sake of consistency,\nafter all is not consistent.\n\n49. If all good is pleasure, what is moral good? Ambiguity in Locke\u2019s\nview.\n\n50. Development of it by Clarke, which breaks down for want of true\nview of reason.\n\n51. With Hume, moral good is pleasure excited in a particular way,\nviz.; in the spectator of the \u2018good\u2019 act and by the view of its\ntendency to produce pleasure.\n\n53. Moral sense is thus sympathy with pleasure qualified by\nconsideration of general tendencies.\n\n54. In order to account for the facts it has to become sympathy with\nunfelt feelings.\n\n55. Can the distinction between the \u2018moral\u2019 and \u2018natural\u2019 be\nmaintained by Hume? What is \u2018artificial virtue\u2019?\n\n56. No ground for such distinction in relation between motive and act.\n\n57. Motive to artificial virtues.\n\n58. How artificial virtues become moral.\n\n59. Interest and sympathy account for all obligations civil and moral.\n\n60. What is meant by an action which ought to be done.\n\n61. Sense of morality no motive: when it seems so the motive is\nreally pride.\n\n62. Distinction between virtuous and vicious motive does not exist\nfor person moved.\n\n63. \u2018Consciousness of sin\u2019 disappears.\n\n64. Only respectability remains: and even this not consistently\naccounted for.\n\n\n\n\nGENERAL INTRODUCTION TO VOL. I.\n\nHow the history of philosophy should be studied.\n\n1. There is a view of the history of mankind, by this time\nfamiliarised to Englishmen, which detaches from the chaos of events\na connected series of ruling actions and beliefs--the achievement of\ngreat men and great epochs, and assigns to these in a special sense\nthe term \u2018historical.\u2019 According to this theory--which indeed, if\nthere is to be a theory of History at all, alone gives the needful\nsimplification--the mass of nations must be regarded as left in\nswamps and shallows outside the main stream of human development.\nThey have either never come within the reach of the hopes and\ninstitutions which make history a progress instead of a cycle, or\nthey have stiffened these into a dead body of ceremony and caste, or\nat some great epoch they have failed to discern the sign of the times\nand rejected the counsel of God against themselves. Thus permanently\nor for generations, with no principle of motion but unsatisfied want,\nwithout the assimilative ideas which from the strife of passions\nelicit moral results, they have trodden the old round of war, trade,\nand faction, adding nothing to the spiritual heritage of man. It\nwould seem that the historian need not trouble himself with them,\nexcept so far as relation to them determines the activity of the\nprogressive nations.\n\nHume the last great English philosopher.\n\n2. A corresponding theory may with some confidence be applied to\nsimplify the history of philosophical opinion. The common plan of\nseeking this history in compendia of the systems of philosophical\nwriters, taken in the gross or with no discrimination except in\nregard to time and popularity, is mainly to blame for the common\nnotion that metaphysical enquiry is an endless process of threshing\nold straw. Such enquiry is really progressive, and has a real\nhistory, but it is a history represented by a few great names.\nAt rare epochs there appear men, or sets of men, with the true\nspeculative impulse to begin at the beginning and go to the end, and\nwith the faculty of discerning the true point of departure which\nprevious speculation has fixed for them. The intervals are occupied\nby commentators and exponents of the last true philosopher, if it has\nbeen his mission to construct; if it has been sceptical, by writers\nwho cannot understand the fatal question that he has asked, and\nthus still dig in the old vein which he had exhausted, and of which\nhis final dilemma had shown the bottom. Such an interval was that\nwhich in the growth of continental philosophy followed on the epoch\nof Leibnitz; an interval of academic exposition or formulation, in\nwhich the system, that had been to the master an incomplete enquiry,\nbecame in the hands of his disciples a one-sided dogmatism. In the\nline of speculation more distinctively English, a like _r\u00e9gime_\nof \u2018strenua inertia\u2019 has prevailed since the time of Hume. In the\nmanner of its unprofitableness, indeed, it has differed from the\nWolfian period in Germany, just as the disinterested scepticism of\nHume differed from the system-making for purposes of edification to\nwhich Leibnitz applied himself. It has been unprofitable, because\nits representatives have persisted in philosophising upon principles\nwhich Hume had pursued to their legitimate issue and had shown,\nnot as their enemy but as their advocate, to render all philosophy\nfutile. Adopting the premises and method of Locke, he cleared them of\nall illogical adaptations to popular belief, and experimented with\nthem on the body of professed knowledge, as one only could do who\nhad neither any twist of vice nor any bias for doing good, but was a\nphilosopher because he could not help it.\n\nKant his true successor.\n\n3. As the result of the experiment, the method, which began with\nprofessing to explain knowledge, showed knowledge to be impossible.\nHume himself was perfectly cognisant of this result, but his\nsuccessors in England and Scotland would seem so far to have been\nunable to look it in the face. They have either thrust their\nheads again into the bush of uncriticised belief, or they have\ngone on elaborating Hume\u2019s doctrine of association, in apparent\nforgetfulness of Hume\u2019s own proof of its insufficiency to account\nfor an intelligent, as opposed to a merely instinctive or habitual,\nexperience. An enquiry, however, so thorough and passionless as the\n\u2018Treatise of Human Nature,\u2019 could not be in vain; and if no English\nathlete had strength to carry on the torch, it was transferred to a\nmore vigorous line in Germany. It awoke Kant, as he used to say, from\nhis \u2018dogmatic slumber,\u2019 to put him into that state of mind by some\ncalled wonder, by others doubt, in which all true philosophy begins.\nThis state, with less ambiguity of terms, may be described as that of\nfreedom from presuppositions. It was because Kant, reading Hume with\nthe eyes of Leibnitz and Leibnitz with the eyes of Hume, was able to\na great extent to rid himself of the presuppositions of both, that he\nstarted that new method of philosophy which, as elaborated by Hegel,\nclaims to set man free from the artificial impotence of his own false\nlogic, and thus qualify him for a complete interpretation of his own\nachievement in knowledge and morality. Thus the \u2018Treatise of Human\nNature\u2019 and the \u2018Critic of Pure Reason,\u2019 taken together, form the\nreal bridge between the old world of philosophy and the new. They are\nthe essential \u2018Propaedeutik,\u2019 without which no one is a qualified\nstudent of modern philosophy. The close correspondence between the\ntwo works becomes more apparent the more each is studied. It is\nsuch as to give a strong presumption that Kant had studied Hume\u2019s\ndoctrine in its original and complete expression, and not merely as\nit was made easy in the \u2018Essays.\u2019 The one with full and reasoned\narticulation asks the question, which the other with equal fulness\nseeks to answer. It is probably because the question in its complete\nstatement has been so little studied among us, that the intellectual\nnecessity of the Kantian answer has been so little appreciated.\nTo trace the origin and bring out the points of the question, in\norder to the exhibition of that necessity, will be the object of\nthe following treatise. To do this thoroughly, indeed, would carry\nus back through Hobbes to Bacon. But as present limits do not allow\nof so long a journey, we must be content with showing Hume\u2019s direct\nfiliation to Locke, who, indeed, sufficiently gathered up the results\nof the \u2018empirical\u2019 philosophy of his predecessors.\n\nDistinction between literary history and the history of philosophical\nsystems.\n\n4. Such a task is very different from an ordinary undertaking in\nliterary history, and requires different treatment. To the historian\nof literature a philosopher is interesting, if at all, on account\nof the personal qualities which make a great writer, and have a\npermanent effect on letters and general culture. Locke and Hume\nundoubtedly had these qualities and produced such an effect--an\neffect in Locke\u2019s case more intense upon the immediately following\ngenerations, but in Hume\u2019s more remarkable as having reappeared\nafter near a century of apparent forgetfulness. Each, indeed, like\nevery true philosopher, was the mouth-piece of a certain system\nof thought determined for him by the stage at which he found the\ndialectic movement that constitutes the progress of philosophy, but\neach gave to this system the stamp of that personal power which\npersuades men. Their mode of expression had none of that academic\nor \u2018ex cathedra\u2019 character, which has made German philosophy almost\na foreign literature in the country of its birth. They wrote as\ncitizens and men of the world, anxious (in no bad sense) for effect;\nand even when their conclusions were remote from popular belief,\nstill presented them in the flesh and blood of current terms used in\nthe current senses. It is not, however, in their human individuality\nand its effects upon literature, but as the vehicles of a system of\nthought, that it is proposed here to treat them; and this purpose\nwill best be fulfilled if we follow the line of their speculation\nwithout divergence into literary criticism or history, without\nremarks either on the peculiarities of their genius or on any of the\nsecondary influences which affected their writings or arose out of\nthem. For a method of this sort, it would seem, there is some need\namong us. We have been learning of late to know much more about\nphilosophers, but it is possible for knowledge about philosophers\nto flourish inversely as the knowledge of philosophy. The revived\ninterest which is noticeable in the history of philosophy may be\nan indication either of philosophical vigour or of philosophical\ndecay. In those whom intellectual indolence, or a misunderstood and\ndisavowed metaphysic, has landed in scepticism there often survives a\ncuriosity about the literary history of philosophy, and the writings\nwhich this curiosity produces tend further to spread the notion that\nphilosophy is a matter about which there has been much guessing by\ngreat intellects, but no definite truth is to be attained. It is\notherwise with those who see in philosophy a progressive effort\ntowards a fully-articulated conception of the world as rational. To\nthem its past history is of interest as representing steps in this\nprogress which have already been taken for us, and which, if we will\nmake them our own, carry us so far on our way towards the freedom of\nperfect understanding; while to ignore them is not to return to the\nsimplicity of a pre-philosopic age, but to condemn ourselves to grope\nin the maze of \u2018cultivated opinion,\u2019 itself the confused result of\nthose past systems of thought which we will not trouble ourselves to\nthink out.\n\nObject of the present enquiry.\n\n5. The value of that system of thought, which found its clearest\nexpression in Hume, lies in its being an effort to think to\ntheir logical issue certain notions which since then have become\ncommonplaces with educated Englishmen, but which, for that reason, we\nmust detach ourselves from popular controversy to appreciate rightly.\nWe are familiar enough with these in the form to which adaptation to\nthe needs of plausibility has gradually reduced them, but because\nwe do not think them out with the consistency of their original\nexponents, we miss their true value. They do not carry us, as they\nwill do if we restore their original significance, by an intellectual\nnecessity to those truer notions which, in fact, have been their\nsequel in the development of philosophy, but have not yet found their\nway into the \u2018culture\u2019 of our time. An attempt to restore their\nvalue, however, if this be the right view of its nature, cannot but\nseem at first sight invidious. It will seem as if, while we talk of\ntheir value, we were impertinently trying to \u2018pull them to pieces.\u2019\nBut those who understand the difference between philosophical\nfailures, which are so because they are anachronisms, and those which\nin their failure have brought out a new truth and compelled a step\nforward in the progress of thought, will understand that a process,\nwhich looks like pulling a great philosopher to pieces, may be the\ntrue way of showing reverence for his greatness. It is a Pharisaical\nway of building the sepulchres of philosophers to profess their\ndoctrine or extol their genius without making their spirit our own.\nThe genius of Locke and Hume was their readiness to follow the lead\nof Ideas: their spirit was the spirit of Rationalism--the spirit\nwhich, however baffled and forced into inconsistent admissions,\nis still governed by the faith that all things may ultimately be\nunderstood. We best do reverence to their genius, we most truly\nappropriate their spirit, in so exploring the difficulties to which\ntheir enquiry led, as to find in them the suggestion of a theory\nwhich may help us to walk firmly where they stumbled and fell.\n\nLocke\u2019s problem and method.\n\n6. About Locke, as about every other philosopher, the essential\nquestions are, What was his problem, and what was his method?\nLocke, as a man of business, gives us the answers at starting. His\nproblem was the origin of \u2018ideas\u2019 in the individual man, and their\nconnection as constituting knowledge: his method that of simply\n\u2018looking into his own understanding and seeing how it wrought.\u2019\nThese answers commend themselves to common sense, and still form\nthe text of popular psychology. If its confidence in their value,\nas explained by Locke, is at all beginning to be shaken, this is\nnot because, according to a strict logical development, they issued\nin Hume\u2019s unanswered scepticism, which was too subtle for popular\neffect, but because they are now open to a rougher battery from the\nphysiologists. Our concern at present is merely to show their precise\nmeaning, and the difficulties which according to this meaning they\ninvolve.\n\nHis notion of the \u2018thinking thing\u2019.\n\n7. There are two propositions on which Locke is constantly insisting:\none, that the object of his investigation is _his own_ mind; the\nother, that his attitude towards this object is that of mere\nobservation. He speaks of his own mind, it is to be noticed, just\nas he might of his own body. It meant something born with, and\ndependent on, the particular animal organism that first saw the\nlight at Wrington on a particular day in 1632. It was as exclusive\nof other minds as his body of other bodies, and he could only infer\na resemblance between them and it. With all his animosity to the\ncoarse spiritualism of the doctrine of innate ideas, he was the\nvictim of the same notion which gave that doctrine its falsehood and\ngrotesqueness. He, just as much as the untutored Cartesian, regarded\nthe \u2018minds\u2019 of different men as so many different things; and his\nrefutation of the objectionable hypothesis proceeds wholly from this\nview. Whether the mind is put complete into the body, or is born and\ngrows with it; whether it has certain characters stamped upon it to\nbegin with, or receives all its ideas through the senses; whether it\nis simple and therefore indiscerptible, or compound and therefore\nperishable--all these questions to Locke, as to his opponents,\nconcern a multitude of \u2018thinking things\u2019 in him and them, merely\nindividual, but happening to be pretty much alike.\n\nThis he will passively observe.\n\n8. This \u2018thinking thing,\u2019 then, as he finds it in himself, the\nphilosopher, according to Locke, has merely and passively to observe,\nin order to understand the nature of knowledge. \u2018I could look into\nnobody\u2019s understanding but my own to see how it wrought,\u2019 he says,\nbut \u2018I think the intellectual faculties are made and operate alike\nin most men. But if it should happen not to be so, I can only make\nit my humble request, in my own name and in the name of those that\nare of my size, who find their minds work, reason, and know in the\nsame low way that mine does, that the men of a more happy genius will\nshow us the way of their nobler flights.\u2019 (Second Letter to Bishop of\nWorcester.) As will appear in the sequel, it is from this imaginary\nmethod of ascertaining the origin and nature of knowledge by passive\nobservation of what goes on in one\u2019s own mind that the embarrassments\nof Locke\u2019s system flow. It was the function of Hume to exhibit the\nradical flaw in his master\u2019s method by following it with more than\nhis master\u2019s rigour.\n\nIs such observation possible?\n\n9. As an observation of the \u2018thinking thing,\u2019 the \u2018philosophy of\nmind\u2019 seems to assume the character of a natural science, and\nthus at once acquires definiteness, and if not certainty, at\nleast plausibility. To deny the possibility of such observation,\nin any proper sense of the word, is for most men to tamper with\nthe unquestioned heritage of all educated intelligence. Hence the\nunpalatability of a consistent Positivism; hence, too, on the\nother side, the general conviction that the Hegelian reduction of\nPsychology to Metaphysics is either an intellectual juggle, or a\nwilful return of the philosophy, which psychologists had washed, to\nthe mire of scholasticism. It is the more important to ascertain what\nthe observation in question precisely means. What observes, and what\nis observed? According to Locke (and empirical psychology has never\nsubstantially varied the answer) the matter to be observed consists\nfor each man firstly in certain impressions of his own individual\nmind, by which this mind from being a mere blank has become\nfurnished--by which, in other words, his mind has become actually\na mind; and, secondly, in certain operations, which the mind, thus\nconstituted, performs upon the materials which constitute it. The\nobserver, all the while, is the constituted mind itself. The question\nat once arises, how the developed man can observe in himself (and\nit is only to himself, according to Locke, that he can look) that\nprimitive state in which his mind was a \u2018tabula rasa.\u2019 In the first\nplace, that only can be observed which is present; and the state in\nquestion to the supposed observer is past. If it be replied that it\nis recalled by memory, there is the farther objection that memory\nonly recalls what has been previously known, and how is a man\u2019s own\nprimitive consciousness, as yet void of the content which is supposed\nto come to it through impressions, originally known to him? How can\nthe \u2018tabula rasa\u2019 be cognisant of itself?\n\nWhy it seems so.\n\n10. The cover under which this difficulty was hidden from Locke,\nas from popular psychologists ever since, consists in the implicit\nassumption of certain ideas, either as possessed by or acting upon\nthe mind in the supposed primitive state, which are yet held to be\narrived at by a gradual process of comparison, abstraction, and\ngeneralisation. This assumption, which renders the whole system\nresting upon the interrogation of consciousness a paralogism, is yet\nthe condition of its apparent possibility. It is only as already\ncharged with a content which is yet (and for the individual, truly)\nmaintained to be the gradual acquisition of experience, that the\nprimitive consciousness has any answer to give to its interrogator.\n\nLocke\u2019s account of origin of ideas.\n\n11. Let us consider the passage where Locke sums up his theory of\nthe \u2018original of our ideas.\u2019 (Book II. chap. i. sec. 23, 24.) \u2018Since\nthere appear not to be any ideas in the mind, before the senses have\nconveyed any in, I conceive that ideas in the understanding are\ncoeval with sensation; which is such an impression, made in some part\nof the body, as produces some perception in the understanding. It is\nabout these impressions made on our senses by outward objects, that\nthe mind seems first to employ itself in such operations as we call\nperception, remembering, consideration, reasoning, &c. In time the\nmind comes to reflect on its own operations about the ideas got by\nsensation, and thereby stores itself with a new set of ideas, which\nI call ideas of reflection. These impressions that are made on our\nsenses by outward objects, that are extrinsical to the mind; and\nits own operations, proceeding from powers intrinsical and proper\nto itself, which, when reflected on by itself, become also objects\nof its contemplation, are, as I have said, the original of all\nknowledge.\u2019\n\nIts ambiguities _(a)_ In regard to sensation.\n\n12. Can we from this passage elicit a distinct account of the\nbeginning of intelligence? In the first place it consists in an\n\u2018idea,\u2019 and an idea is elsewhere (Introduction, sec. 8) stated\nto be \u2018whatsoever is the object of the understanding, when a man\nthinks.\u2019 But the primary idea is an \u2018idea of sensation.\u2019 Does this\nmean that the primary idea _is_ a sensation, or is a distinction\nto be made between the sensation and the idea thereof? The passage\nbefore us would seem to imply such a distinction. Looking merely\nto it, we should probably say that by _sensation_ Locke meant \u2018an\nimpression or motion in some part of the body;\u2019 by the _idea of\nsensation_ \u2018a perception in the understanding,\u2019 which this impression\nproduces. The account of perception itself gives a different result.\n(Book II. chap. ix. sec. 3.) \u2018Whatever impressions are made on the\noutward parts, if they are not taken notice of within, there is no\nperception. Fire may burn our bodies with no other effect than it\ndoes a billet, unless the motion be continued to the brain, and\nthere the _sense_ of heat or _idea_ of pain be produced in the mind,\nwherein consists actual _perception_.\u2019 Here sensation is identified\nat once with the idea and with perception, as opposed to the\nimpression on the bodily organs. [1] To confound the confusion still\nfarther, in a passage immediately preceding the above, \u2018Perception,\u2019\nhere identified with the idea of sensation, has been distinguished\nfrom it, as \u2018exercised about it.\u2019 \u2018Perception, as it is the first\nfaculty of the mind exercised about our ideas, so it is the first and\nsimplest idea we have from reflection.\u2019 Taking Locke at his word,\nthen, we find the beginning of intelligence to consist in having an\nidea of sensation. This idea, however, we perceive, and to perceive\nis to have an idea; _i.e._ to have an idea of an idea of sensation.\nBut of perception again we have a simple or primitive idea. Therefore\nthe beginning of intelligence consists in having an idea of an idea\nof an idea of sensation.\n\n[1] Cf. Book II. chap. xix. sec. 1. \u2018The _perception_, which actually\naccompanies and is annexed to any impression on the body, made by\nan external object, being distinct from all other modifications of\nthinking, furnishes the mind with a distinct idea which we call\n_sensation_; which is, as it were, the actual entrance of any idea\ninto the understanding by the senses.\u2019\n\n_(b)_ In regard to ideas of reflection.\n\n13. By insisting on Locke\u2019s account of the relation between the\nideas of sensation and those of reflection we might be brought to a\ndifferent but not more luminous conclusion. In the passages quoted\nabove, where this relation is most fully spoken of, it appears that\nthe latter are essentially sequent to those of sensation. \u2018_In\ntime_ the mind comes to reflect on its own operations, about the\nideas got by sensation, and thereby stores itself with a new set\nof ideas, which I call ideas of reflection.\u2019 Of these only two are\nprimary and original (Book II. c. xxi. sec. 73), viz. motivity or\npower of moving, with which we are not at present concerned, and\nperceptivity or power of perception. But according to Locke, as we\nhave seen, there cannot be any, the simplest, idea of sensation\nwithout perception. If, then, the _idea_ of perception is only\ngiven later and upon reflection, we must suppose perception to take\nplace without any idea of it. But with Locke to have an idea and\nto perceive are equivalent terms. We must thus conclude that the\nbeginning of knowledge is an unperceived perception, which is against\nhis express statement elsewhere (Book II. c. xxvii. sec. 9), that it\nis \u2018impossible for any one to perceive without perceiving that he\ndoes perceive.\u2019\n\nWhat is the \u2018tablet\u2019 impressed?\n\n14. Meanwhile a perpetual equivocation is kept up between a supposed\nimpression on the \u2018outward parts,\u2019 and a supposed impression on the\n\u2018tablet of the mind.\u2019 It is not the impression upon, or a motion\nin, the outward parts, as Locke admits, that constitutes the idea\nof sensation. It is not an agitation in the tympanum of the ear,\nor a picture on the retina of the eye, that we are conscious of\nwhen we see a sight or hear a sound. [1] The motion or impression,\nhowever, has only, as he seems to suppose, to be \u2018continued to the\nbrain,\u2019 and it becomes an idea of sensation. Notwithstanding the\nrough line of distinction between soul and body, which he draws\nelsewhere, his theory was practically governed by the supposition\nof a cerebral something, in which, as in a third equivocal tablet,\nthe imaginary mental and bodily tablets are blended. If, however,\nthe idea of sensation, as an object of the understanding when a man\nthinks, differs absolutely from \u2018a motion of the outward parts,\u2019\nit does so no less absolutely, however language and metaphor may\ndisguise the difference, from such motion as \u2018continued to the\nbrain.\u2019 An instructed man, doubtless, may come to think about a\nmotion in his brain, as about a motion of the earth round the sun,\nbut to speak of such motion as an idea of sensation or an immediate\nobject of intelligent sense, is to confuse between the object of\nconsciousness and a possible physical theory of the conditions of\nthat consciousness. It is only, however, by such an equivocation that\nany idea, according to Locke\u2019s account of the idea, can be described\nas an \u2018impression\u2019 at all, or that the representation of the mind as\na tablet, whether born blank or with characters stamped on it, has\neven an apparent meaning. A metaphor, interpreted as a fact, becomes\nthe basis of his philosophical system.\n\n[1] Cf. Locke\u2019s own statement (Book III. c. iv. sec. 10). \u2018The cause\nof any sensation, and the sensation itself, in all the simple ideas\nof one sense, are two ideas; and two ideas so different and distant\none from another, that no two can be more so.\u2019\n\nDoes the mind make impressions on itself?\n\n15. As applied to the ideas of reflection, indeed, the metaphor loses\neven its plausibility. In its application to the ideas of sensation\nit gains popular acceptance from the ready confusion of thought\nand matter in the imaginary cerebral tablet, and the supposition\nof actual impact upon this by \u2018outward things.\u2019 But in the case of\nideas of reflection, it is the mind that at once gives and takes\nthe impression. It must be supposed, that is, to make impressions\non itself. There is the further difficulty that as perception is\nnecessary in order to give an _idea_ of sensation, the impress of\nperception must be taken by the mind in its earliest receptivity; or,\nin other words, it must impress itself while still a blank, still\nvoid of any \u2018furniture\u2019 wherewith to make the impression. There is no\nescape from this result unless we suppose perception to precede the\nidea of it by some interval of time, which lands us, as we have seen,\nin the counter difficulty of supposing an unperceived perception.\nLocke disguises the difficulty from himself and his reader by\nconstantly shifting both the receptive subject and the impressive\nmatter. We find the \u2018tablet\u2019 perpetually receding. First it is the\n\u2018outward part\u2019 or bodily organ. Then it is the brain, to which the\nimpression received by the outward part must somehow be continued,\nin order to produce sensation. Then it is the perceptive mind, which\ntakes an impression of the sensation or has an idea of it. Finally,\nit is the reflective mind, upon which in turn the perceptive mind\nmakes impressions. But the hasty reader, when he is told that the\nmind is passively impressed with ideas of reflection, is apt to\nforget that the matter which thus impresses it is, according to\nLocke\u2019s showing, simply its perceptive, _i.e._ its passive, self.\n\nSource of these difficulties. The \u2018simple\u2019 idea, as Locke describes\nit, is a complex idea of substance and relation.\n\n16. The real source of these embarrassments in Locke\u2019s theory,\nit must be noted, lies in the attempt to make the individual\nconsciousness give an answer to its interrogator as to the beginning\nof knowledge. The individual looking back on an imaginary earliest\nexperience pronounces himself in that experience to have been simply\nsensitive and passive. But by this he means consciously sensitive\n_of something_ and consciously passive _in relation to something_.\nThat is, he supposes the primitive experience to have involved\nconsciousness of a self on the one hand and of a thing on the other,\nas well as of a relation between the two. In the \u2018idea of sensation\u2019\nas Locke conceived it, such a consciousness is clearly implied,\nnotwithstanding his confusion of terms. The idea is a perception,\nor consciousness _of a thing_, as opposed to a sensation proper or\naffection of the bodily organs. Of the perception, again, there is\nan idea, _i.e._ a consciousness by the man, in the perception, of\nhimself in negative relation to the thing that is his object, and\nthis consciousness (if we would make Locke consistent in excluding an\nunperceived perception) must be taken to go along with the perceptive\nact itself. No less than this indeed can be involved in any act that\nis to be the beginning of knowledge at all. It is the minimum of\npossible thought or intelligence, and the thinking man, looking for\nthis beginning in the earliest experience of the individual human\nanimal, must needs find it there. But this means no less than that he\nis finding there already the conceptions of substance and relation.\nHence a double contradiction: firstly, a contradiction between the\nprimariness of self-conscious cognisance of a thing, as the beginning\nof possible knowledge, on the one hand, and the primariness of\nanimal sensation in the history of the individual man on the other;\nsecondly, a contradiction between the primariness in knowledge of\nthe ideas of substance and relation, and the seemingly gradual\nattainment of those \u2018abstractions\u2019 by the individual intellect. The\nformer of these contradictions is blurred by Locke in the two main\nconfusions which we have so far noticed: _(a)_ the confusion between\nsensation proper and perception, which is covered under the phrase\n\u2018idea of sensation;\u2019 a phrase which, if sensation means the first act\nof intelligence, is pleonastic, and if it means the \u2018motion of the\noutward parts continued to the brain,\u2019 is unmeaning; and _(b)_ the\nconfusion between the physical affection of the brain and the act of\nthe self-conscious subject, covered under the equivocal metaphor of\nimpression. The latter contradiction, that concerning the ideas of\nsubstance and relation, has to be further considered.\n\nHow this contradiction is disguised.\n\n17. It is not difficult to show that to have a simple idea, according\nto Locke\u2019s account of it, means to have already the conception of\nsubstance and relation, which are yet according to him \u2018complex\nand derived ideas,\u2019 \u2018the workmanship of the mind\u2019 in opposition to\nits original material, the result of its action in opposition to\nwhat is given it as passive. The equivocation in terms under which\nthis contradiction is generally covered is that between \u2018idea\u2019\nand \u2018quality.\u2019 \u2018Whatever the mind perceives in itself, or is the\nimmediate object of perception, thought, or understanding, that\nI call _idea_; and the power to produce that idea I call quality\nof the subject wherein that power is. Thus a snowball having\npower to produce in us the ideas of white, cold, and round, the\npowers to produce these ideas in us, as they are in the snowball,\nI call qualities; and as they are sensations or perceptions in\nour understandings, I call them ideas; which ideas, if I speak of\nsometimes as in the things themselves, I would be understood to mean\nthose qualities in the object which produce them in us.\u2019 (Book II.\nchap. viii. sec. 8.)\n\nLocke\u2019s way of interchanging \u2018idea\u2019 and \u2018quality\u2019 and its effects.\n\n18. An equivocation is not the less so because it is announced.\nIt is just because Locke allows himself at his convenience to\ninterchange the terms \u2018idea\u2019 and \u2018quality\u2019 that his doctrine is at\nonce so plausible and so hollow. The essential question is whether\nthe \u2018simple idea,\u2019 as the original of knowledge, is on the one hand\na mere feeling, or on the other a thing or quality of a thing.\nThis question is the crux of empirical psychology. Adopting the\none alternative, we have to face the difficulty of the genesis of\nknowledge, as an apprehension of the real, out of mere feeling;\nadopting the other, we virtually endow the nascent intelligence with\nthe conception of substance. By playing fast and loose with \u2018idea\u2019\nand \u2018quality,\u2019 Locke disguised the dilemma from himself. Here again\nthe metaphor of Impression did him yeoman\u2019s service. The idea, or\n\u2018immediate object of thought,\u2019 being confused with the affection\nof the sensitive organs, and this again being accounted for as the\nresult of actual impact, it was easy to represent the idea itself as\ncaused by the action of an outward body on the \u2018mental tablet.\u2019 Thus\nLocke speaks of the \u2018objects of our senses obtruding their particular\nideas on our minds, whether we will or no.\u2019 (Book II. chap. i. sec.\n25.) This sentence holds in solution an assumption and two fallacies.\nThe assumption (with which we have no further concern here) is the\nphysical theory that matter affects the sensitive organs in the way\nof actual impact. Of the fallacies, one is the confusion between this\naffection and the idea of which it is the occasion to the individual;\nthe other is the implication that this idea, as such, in its prime\nsimplicity, recognises itself as the result of, and refers itself as\na quality to, the matter supposed to cause it. This recognition and\nreference, it is clearly implied, are involved in the idea itself,\nnot merely made by the philosopher theorising it. Otherwise the\n\u2018obtrusion\u2019 would be described as of a property or effect, not of an\nidea, which means, it must be remembered, the object of consciousness\njust as the object of consciousness. Of the same purport is the\nstatement that \u2018the mind is furnished with simple ideas as they are\nfound in exterior things.\u2019 (Book II. chap, xxiii. sec. 1.) It only\nrequires a moment\u2019s consideration, indeed, to see that the beginning\nof consciousness cannot be a physical theory, which, however true it\nmay be and however natural it may have become to us, involves not\nonly the complex conception of material impact, but the application\nof this to a case having no palpable likeness to it. But the\n\u2018interrogator of consciousness\u2019 finds in its primitive state just\nwhat he puts there, and thus Locke, with all his pains \u2018to set his\nmind at a distance from itself,\u2019 involuntarily supposes it, in the\nfirst element of intelligence, to \u2018report\u2019 that action of matter upon\nitself, which, as the result of a familiar theory--involving not\nmerely the conceptions of substance, power, and relation, but special\nqualifications of these--it reports to the educated man.\n\nPrimary and secondary qualities of bodies.\n\n19. This will appear more clearly upon an examination of his doctrine\nof \u2018the ideas of primary and secondary qualities of bodies.\u2019 The\ndistinction between them he states as follows. The primary qualities\nof bodies are \u2018the bulk, figure, number, situation, motion, and rest\nof their solid parts; these are in them, whether we perceive them\nor no; and when they are of that size that we can discover them, we\nhave by these an idea of the thing as it is in itself.\u2019 ... Thus\n\u2018the ideas of primary qualities are resemblances of them, and their\npatterns do really exist in the bodies themselves. But the ideas\nproduced in us by the secondary qualities have no resemblance of them\nat all. There is nothing like them existing in the bodies themselves.\nThey are in the bodies, we denominate from them, only a power to\nproduce these sensations in us; and what is sweet, blue, or warm in\nidea is but the certain bulk, figure, and motion of the insensible\nparts in the bodies themselves which we call so.\u2019 This power is then\nexplained to be of two sorts: _(a)_ \u2018The power that is in any body,\nby reason of its insensible primary qualities, to operate after a\npeculiar manner on any of our senses, and thereby produce in us the\ndifferent ideas of several colours, sounds, smells, tastes, &c. These\nare usually called sensible qualities, _(b)_ The power that is in\nany body, by reason of the particular constitution of its primary\nqualities, to make such a change in the bulk, figure, texture, and\nmotion of another body, as to make it operate differently on our\nsenses from what it did before. Thus the sun has a power to make wax\nwhite, and fire to make lead fluid. These are usually called powers.\u2019\n(Book II. chap. viii. sec. 15, 23.)\n\n\u2018Simple idea\u2019 represented as involving a theory of its own cause.\n\n20. What we have here is a theory of the causes of simple ideas; but\nwe shall find Locke constantly representing this theory as a simple\nidea itself, or the simple idea as involving this theory. By this\nunconscious device he is enabled readily to exhibit the genesis of\nknowledge out of \u2018simple ideas,\u2019 but it is at the cost of converting\nthese into \u2018creations of the mind,\u2019 which with him are the antitheses\nof \u2018facts\u2019 or \u2018reality.\u2019 The process of conversion takes a different\nform as applied respectively to the ideas of primary and to those\nof secondary qualities. We propose to follow it in the latter\napplication first.\n\nPhrases in which this is implied.\n\n21. The simple idea caused by a quality he calls the idea _of_ that\nquality. Under cover of this phrase, he not only identifies the idea\nof a primary quality with the quality itself of which he supposes it\nto be a copy, but he also habitually regards the idea of a secondary\nquality as the consciousness of a quality _of a thing_, though under\nwarning that the quality as it is to consciousness is not as it is in\nthe thing. This reservation rather adds to the confusion. There are\nin fact, according to Locke, as appears from his distinction between\nthe \u2018nominal\u2019 and \u2018real essence,\u2019 two different things denoted by\nevery common noun; the thing as it is in itself or in nature, and\nthe thing as it is for consciousness. The former is the thing as\nconstituted by a certain configuration of particles, which is only\nan object for the physical philosopher, and never fully cognisable\neven by him; [1] the latter is the thing as we see and hear and\nsmell it. Now to a thing in this latter sense, according to Locke,\nsuch a simple idea as to the philosopher is one of a secondary\nquality (_i.e._ not a copy, but an effect, of something in a body),\nis already in the origin of knowledge referred as a quality, though\nwithout distinction of primary and secondary. He does not indeed\nstate this in so many words. To have done so might have forced him to\nreconsider his doctrine of the mere passivity of the mind in respect\nof simple ideas. But it is implied in his constant use of such\nphrases as \u2018reports of the senses,\u2019 \u2018inlet through the senses\u2019--which\nhave no meaning unless something is reported, something let in--and\nin the familiar comparison of the understanding to a \u2018closet, wholly\nshut from light, with only some little opening left, to let in\nexternal visible resemblances, or ideas, of things without.\u2019 (Book\nII. chap. xi. sec. 17.)\n\n[1] This distinction is more fully treated below, paragraphs 88, &c.\n\nFeeling and felt thing confused.\n\n22. Phraseology of this kind, the standing heritage of the philosophy\nwhich seeks the origin of knowledge in sensation, assumes that the\nindividual sensation is from the first consciously representative;\nthat it is more than what it is simply in itself--fleeting,\nmomentary, unnameable (because, while we name it, it has become\nanother), and for the same reason unknowable, the very negation of\nknowability; that it shows the presence of something, whether this\nbe a \u2018body\u2019 to which it is referred as a quality, or a mind of which\nit is a modification, or be ultimately reduced to the permanent\nconditions of its own possibility. This assumption for the present\nhas merely to be pointed out; its legitimacy need not be discussed.\nNor need we now discuss the attempts that have been made since Locke\nto show that mere sensations, dumb to begin with, may yet become\narticulate upon repetition and combination; which in fact endow them\nwith a faculty of inference, and suppose that though primarily they\nreport nothing beyond themselves, they yet somehow come to do so\nas an explanation of their own recurrence. The sensational theory\nin Locke is still, so to speak, unsophisticated. It is true that,\nin concert with that \u2018thinking gentleman,\u2019 Mr. Molyneux, he had\nsatisfied himself that what we reckon simple ideas are often really\ninferences from such ideas which by habit have become instinctive;\nbut his account of this habitual process presupposes the reference of\nsensation to a thing. \u2018When we set before our eyes a round globe of\nany uniform colour, it is certain that the idea thereby imprinted in\nour mind is of a flat circle, variously shadowed with several degrees\nof light and brightness coming to our eyes. But we having by use been\naccustomed to perceive what kind of appearance convex bodies are wont\nto make in us, what alterations are made in the reflections of light\nby the difference of the sensible figures of bodies; the judgment\npresently, by an habitual custom, alters the appearances into their\ncauses. So that from that which truly is variety of colour or shadow,\ncollecting the figure, it makes it pass for a mark of figure, and\nframes to itself the perception of a convex figure, and an uniform\ncolour.\u2019 (Book II. chap. ix. sec. 8.) The theory here stated involves\ntwo assumptions, each inconsistent with the simplicity of the simple\nidea. _(a)_ The actual impression of the \u2018plane variously coloured\u2019\nis supposed to pronounce itself to be of something outward. Once\ncall the sensation an \u2018impression,\u2019 indeed, or call it anything, and\nthis or an analogous substantiation of it is implied. It is only\nas thus reporting something \u2018objective\u2019 that the simple idea of\nthe plane variously coloured gives anything to be corrected by the\n\u2018perception of the kind of appearance convex bodies are wont to make\nin us,\u2019 _i.e._ \u2018of the alterations made in the reflections of light\nby the difference of the sensible figure of bodies.\u2019 This perception,\nindeed, as described, is already itself just the instinctive judgment\nwhich has to be accounted for, and though this objection might be\nmet by a better statement, yet no statement could serve Locke\u2019s\npurpose which did not make assumption _(b)_ that sensations of light\nand colour--\u2018simple ideas of secondary qualities\u2019--are in the very\nbeginning of knowledge _appearances_, if not of _convex_ bodies, yet\nof bodies; if not of bodies, yet of something which they reveal,\nwhich remains there while they pass away.\n\nThe simple idea as \u2018ectype\u2019 other than mere sensation.\n\n23. The same assumption is patent in Locke\u2019s account of the\ndistinction between \u2018real and fantastic,\u2019 \u2018adequate and inadequate,\u2019\nideas. This distinction rests upon that between the thing as\narchetype, and the idea as the corresponding ectype. Simple ideas he\nholds to be necessarily \u2018real\u2019 and \u2018adequate,\u2019 because necessarily\nanswering to their archetypes. \u2018Not that they are all of them images\nor representations of what does exist: ... whiteness and coldness\nare no more in snow than pain is: ... yet are they real ideas in\nus, whereby we distinguish the qualities that are really in things\nthemselves. For these several appearances being designed to be the\nmarks whereby we are to know and distinguish things which we have\nto do with, our ideas do as well serve us to that purpose, and are\nas real distinguishing characters, whether they be only constant\neffects, or else exact resemblances of something in the things\nthemselves.\u2019 (Book II. chap. xxx. sec. 2.) The simple idea, then, is\na \u2018mark\u2019 or \u2018distinguishing character,\u2019 either as a copy or as an\neffect, of something other than itself. Only as thus regarded, does\nthe distinction between real and fantastic possibly apply to it.\nSo too with the distinction between true and false ideas. As Locke\nhimself points out, the simple idea in itself is neither true nor\nfalse. It can become so only as \u2018referred to something extraneous\nto it.\u2019 (Book II. chap, xxxii. sec. 4.) For all that, he speaks of\nsimple ideas as true and necessarily true, because \u2018being barely\nsuch perceptions as God has fitted us to receive, and given power to\nexternal objects to produce in us by established laws and ways ...\ntheir truth consists in nothing else but in such appearances as are\nproduced in us, and must be suitable to those powers He has placed in\nexternal objects, or else they could not be produced in us.\u2019 (Book\nII. chap, xxxii. sec. 14.) Here again we are brought to the same\npoint. The idea is an \u2018appearance\u2019 of something, necessarily true\nwhen it cannot seem to be the appearance of anything else than that\nof which it is the appearance. We thus come to the following dilemma.\nEither the simple idea is referred to a thing, as its pattern or its\ncause, or it cannot be regarded as either real or true. If it is\nstill objected that it need not be so referred in the beginning of\nknowledge, though it comes to be so in the developed intelligence,\nthe answer is the further question, how can that be knowledge even in\nits most elementary phase--the phase of the reception of simple ideas\n--which is not a capacity of distinction between real and apparent,\nbetween true and false? If its beginning is a mode of consciousness,\nsuch as mere sensation would be--which, because excluding all\nreference, excludes that reference of itself to something else\nwithout which there could be no consciousness of a distinction\nbetween an \u2018is\u2019 and an \u2018is not,\u2019 and therefore no true judgment at\nall--how can any repetition of such modes give such a judgment? [1]\n\n[1] Cf. the ground of distinction between clearness and obscurity of\nideas; (Book II. chap. xxix. sec. 2) \u2018Our simple ideas are clear when\nthey are such as the objects themselves, whence they are taken, did\nor might in a well-ordered sensation or perception, present them.\u2019 As\nLocke always assumes that immediate consciousness can tell whether an\nidea is clear or not, it follows that immediate consciousness must\ntell of \u2018the object itself, whence the idea is taken.\u2019\n\nIt involves a judgment in which mind and thing are distinguished.\n\n24. The fact is that the \u2018simple idea\u2019 with Locke, as the beginning\nof knowledge, is already, at its minimum, the judgment, \u2018I have\nan idea different from other ideas, which I did not make for\nmyself.\u2019 His confusion of this judgment with sensation is merely\nthe fundamental confusion, on which all empirical psychology rests,\nbetween two essentially distinct questions--one metaphysical, What is\nthe simplest element of knowledge? the other physiological, What are\nthe conditions in the individual human organism in virtue of which\nit becomes a vehicle of knowledge? Though he failed, however, to\ndistinguish these questions, their difference made itself appear in a\ncertain divergence between the second and fourth books of his Essay.\nSo far we have limited our consideration to passages in the second\nbook, in which he treats _eo nomine_ of ideas; of simple ideas as\nthe original of knowledge, of complex ones as formed in its process.\nHere the physical theory is predominant. The beginning of knowledge\nis that without which the animal is incapable of it, viz. sensation\nregarded as an impression through \u2018animal spirits\u2019 on the brain. But\nit can only be so represented because sensation is identified with\nthat which later psychology distinguished from it as Perception, and\nfor which no physical theory can account. As we have seen, the whole\ntheory of this (the second) book turns upon the supposition that\nthe simple idea of sensation is in every case an idea of a sensible\nquality, and that it is so, not merely for us, considering it _ex\nparte post_, but consciously for the individual subject, which can\nmean nothing else than that it distinguishes itself from, and refers\nitself to, a thing. Locke himself, indeed, according to his plan of\nbringing in a \u2018faculty of the mind\u2019 whenever it is convenient, would\nperhaps rather have said that it is so distinguished and referred \u2018by\nthe mind.\u2019 He considers the simple idea not, as it truly is, the mind\nitself in a certain relation, but a datum or material of the mind,\nupon which it performs certain operations as upon something other\nthan itself, though all the while it is constituted, at least in its\nactuality, by this material. Between the reference of the simple\nidea to the thing, however, by itself and \u2018by the mind,\u2019 there is no\nessential difference. In either case the reference is inconsistent\nwith the simplicity of the simple idea; and if the latter expression\navoids the seeming awkwardness of ascribing activity to the idea,\nit yet ascribes it to the mind in that elementary stage in which,\naccording to Locke, it is merely receptive.\n\nAnd is equivalent to what he afterwards calls \u2018knowledge of\nidentity\u2019. Only as such can it be named.\n\n25. So much for the theory \u2018of ideas.\u2019 As if, however, in treating\nof ideas he had been treating of anything else than knowledge, he\nafterwards considers \u2018knowledge\u2019 in a book by itself (the fourth)\nunder that title, and here the question as to the relation between\nidea and thing comes before him in a somewhat different shape.\nAccording to his well-known definition, knowledge is the perception\nof the agreement or disagreement of any of our ideas. The agreement\nor disagreement may be of four sorts. It may be in the way (1)\nof identity, (2) of relation, (3) of co-existence, (4) of real\nexistence. In his account of the last sort of agreement, it may\nbe remarked by the way, he departs at once and openly from his\ndefinition, making it an agreement, not of idea with idea, but of an\nidea with \u2018actual real existence.\u2019 The fatal but connatural wound in\nhis system, which this inconsistency marks, will appear more fully\nbelow. For the present, our concern is for the adjustment of the\ndefinition of knowledge to the doctrine of the simple idea as the\nbeginning of knowledge. According to the definition, it cannot be\nthe simple idea, as such, that constitutes this beginning, but only\nthe perception of agreement or disagreement between simple ideas.\n\u2018There could be no room,\u2019 says Locke distinctly, \u2018for any positive\nknowledge at all, if we could not distinguish any relation between\nour ideas.\u2019 (Book IV. chap. i. sec. 5.) Yet in the very context\nwhere he makes this statement, the perception of relation is put\nas a distinct kind of knowledge apart from others. In his account\nof the other kinds, however, he is faithful to his definition, and\ntreats each as a perception (_i.e._ a judgment) of a relation in the\nway of agreement or disagreement. The primary knowledge is that of\nidentity--the knowledge of an idea as identical with itself. \u2018A man\ninfallibly knows, as soon as ever he has them in his mind, that the\nideas he calls _white_ and _round_, are the very ideas they are,\nand not other ideas which he calls _red_ and _square_.\u2019 (Book IV.\nchap. i. sec. 4.) Now, as Hume afterwards pointed out, identity is\nnot simple unity. It cannot be predicated of the \u2018idea\u2019 as merely\nsingle, but only as a manifold in singleness. To speak of an idea\nas the \u2018same with itself\u2019 is unmeaning unless it mean \u2018same with\nitself _in its manifold appearances_\u2019 _i.e._ unless the idea is\ndistinguished, as an object existing continuously, from its present\nappearance. Thus \u2018the infallible knowledge,\u2019 which Locke describes\nin the above passage, consists in this, that on the occurrence of a\ncertain \u2018idea\u2019 the man _recognises_ it as one, which at other times\nof its occurrence he has called \u2018_white_.\u2019 Such a \u2018synthesis of\nrecognition,\u2019 however, expressed by the application of a common term,\nimplies the reference of a present sensation to a permanent object of\nthought, in this case the object thought under the term \u2018white,\u2019 so\nthat the sensation becomes an idea of that object. Were there no such\nobjects, there would be no significant names, but only noises; and\nwere the present sensation not so referred, it would not be named. It\nmay be said indeed that the \u2018permanent object of thought\u2019 is merely\nthe instinctive result of a series of past resembling sensations,\nand that the common name is merely the register of this result. But\nthe question is thus merely thrown further back. Unless the single\nfleeting sensation was, to begin with, fixed and defined by relation\nto and distinction from something permanent--in other words, unless\nit ceased to be a mere sensation--how did it happen that other\nsensations were referred to it, as different cases of an identical\nphenomenon, to which the noise suggested by it might be applied as a\nsign?\n\nThe same implied in calling it an idea of an object.\n\n26. This primary distinction and relation of the simple idea Locke\nimplicitly acknowledges when he substitutes for the simple idea,\nas in the passage last quoted, the man\u2019s knowledge that he has the\nidea; for such knowledge implies the distinction of the idea from its\npermanent conscious subject, and its determination by that negative\nrelation. [1] Thus determined, it becomes itself a permanent object,\nor (which comes to the same) an idea _of an object_; a phrase which\nLocke at his convenience substitutes for the mere idea, whenever it\nis wanted for making his theory of knowledge square with knowledge\nitself. Once become such an object, it is a basis to which other\nsensations, like and unlike, may be referred as differentiating\nattributes. Its identity becomes a definite identity.\n\n[1] Cf. the passage in Book II. chap. vii. sec. 7. \u2018When ideas are\nin our minds, we consider them as being actually there.\u2019 The mere\n\u2018idea\u2019 is in fact essentially different from the \u2018consideration\nof it as actually there,\u2019 as sensation is different from thought.\nThe \u2018consideration, &c.,\u2019 really means the thought of the \u2018idea\u2019\n(sensation) as determined by relation to the conscious subject.\n\nMade _for_, not _by_, us, and therefore according to Locke really\nexistent.\n\n27. Upon analysis, then, of Locke\u2019s account of the most elementary\nknowledge, the perception of identity or agreement of an idea with\nitself, we find that like the \u2018simple idea,\u2019 which he elsewhere\nmakes the beginning of knowledge, it really means the reference\nof a sensation to a conception of a permanent object or subject,\n[1] either in such a judgment as \u2018this is white\u2019 (_sc._ a white\nthing), or in the more elementary one, \u2018this is an object to me.\u2019\nIn the latter form the judgment represents what Locke puts as the\nconsciousness, \u2018I have an idea,\u2019 or as the \u2018consideration that the\nidea is actually there;\u2019 in the former it represents what he calls\n\u2018the knowledge that the idea which I have in my mind and which I call\nwhite is the very idea it is, and not the idea which I call red.\u2019\nIt is only because _referred_, as above, that the sensation is in\nLocke\u2019s phraseology \u2018a testimony\u2019 or \u2018report\u2019 of something. As we\nsaid above, his notion of the beginning of knowledge is expressed\nnot merely in the formula \u2018I have an idea different from other\nideas,\u2019 but with the addition, \u2018which I did not make for myself.\u2019\n[2] The simple idea is supposed to testify to something without that\ncaused it, and it is this interpretation of it which makes it with\nhim the ultimate criterion of reality. But unless it were at once\ndistinguished from and referred to both a thing of which it is an\neffect and a subject of which it is an experience, it could not in\nthe first place testify to anything, nor secondly to a thing as made\nfor, not by, the subject. This brings us, however, upon Locke\u2019s whole\ntheory of \u2018real existence,\u2019 which requires fuller consideration.\n\n[1] For a recognition by Locke of the correlativity of these (of\nwhich more will have to be said below) cf. Book II. chap. xxiii. sec.\n15. \u2018Whilst I know by seeing or hearing, &c., that there is some\ncorporeal being without me, the object of that sensation, I do more\ncertainly know that there is some spiritual being within me that sees\nand hears.\u2019\n\n[2] Cf. Book II. chap. xii. sec. 1.\n\nWhat did he mean by this?\n\n28. It is a theory, we must premise, which is nowhere explicitly\nstated. It has to be gathered chiefly from those passages of the\nsecond book in which he treats of \u2018complex\u2019 or \u2018artificial\u2019 ideas in\ndistinction from simple ones, which are necessarily real, and from\nthe discussion in the fourth book of the \u2018extent\u2019 and \u2018reality\u2019 of\nknowledge. We have, however, to begin with, in the enumeration of\nsimple ideas, a mention of \u2018existence,\u2019 as one of those \u2018received\nalike through all the ways of sensation and reflection.\u2019 It is an\nidea \u2018suggested to the understanding by every object without and\nevery idea within. When ideas are in our minds, we consider them as\nbeing actually there, as well as we consider things to be actually\nwithout us; which is, that they exist, or have existence.\u2019 (Book II.\nchap. vii. sec. 7.)\n\n29. The two considerations here mentioned, of \u2018ideas as actually\nin our minds,\u2019 of \u2018things as actually without us,\u2019 are meant\nseverally to represent the two ways of reflection and sensation,\nby which the idea of existence is supposed to be suggested. But\nsensation, according to Locke, is an organ of \u2018ideas,\u2019 just as much\nas reflection. Taking his doctrine strictly, there are no \u2018objects\u2019\nbut \u2018ideas\u2019 to suggest the idea of existence, whether by the way of\nsensation or by that of reflection, and no ideas that are not \u2018in the\nmind.\u2019 (Book II. chap. ix. sec. 3, &c.)\n\nExistence as the mere presence of a feeling.\n\n30. The designation of the idea of existence, then, as \u2018suggested\nby every idea within,\u2019 covers every possible suggestion. It can\nmean nothing else than that it is given in every act and mode of\nconsciousness; that it is inseparable from feeling as such, being\nitself at the same time a distinct simple idea. This, we may remark\nby the way, involves the conclusion that every idea is composite,\nmade up of whatever distinguishes it from other ideas together with\nthe idea of existence. Of this idea of existence itself, however,\nit will be impossible to say anything distinctive; for, as it\naccompanies all possible objects of consciousness, there will be no\ncases where it is absent to be distinguished from those where it is\npresent. Not merely will it be undefinable, as every simple idea is;\nit will be impossible \u2018to send a man to his senses\u2019 (according to\nLocke\u2019s favourite subterfuge) in order to know what it is, since it\nis neither given in one sense as distinct from another, nor in all\nsenses as distinct from any other modification of consciousness. Thus\nregarded, to treat it as a simple idea alongside of other simple\nideas is a palpable contradiction. It is the mere \u2018It is felt,\u2019 the\nabstraction of consciousness, no more to be reckoned as one among\nother ideas than colour in general is to be co-ordinated with red,\nwhite, and blue. Whether I smell a rose in the summer or recall the\nsmell in winter; whether I see a horse or a ghost, or imagine a\ncentaur or think of gravitation or the philosopher\u2019s stone--in every\ncase alike the idea or \u2018immediate object of the mind\u2019 _exists_. Yet\nwe find Locke distinguishing between real ideas, as those that \u2018have\na conformity with the existence of things,\u2019 and fantastic ideas, as\nthose which have no such conformity (Book II. chap. xxx. sec. 1); and\nagain in the fourth book (chap. i. sec. 7, chap. iii. sec. 21, &c.)\nhe makes the perception of the agreement of an idea with existence a\nspecial kind of knowledge, different from that of agreement of idea\nwith idea; and having done so, raises the question whether we have\nsuch a knowledge of existence at all, and decides that our knowledge\nof it is very narrow.\n\nExistence as reality.\n\n31. How are such a distinction and such a question to be reconciled\nwith the attribution of existence to every idea? The answer of\ncourse will be, that when he speaks of ideas as not conforming to\nexistence, and makes knowledge or the agreement of ideas with each\nother something different from their agreement with existence, he\nmeans and generally says \u2018real actual existence,\u2019 or the \u2018existence\nof _things_,\u2019 _i.e._ an existence, whatever it be, which is opposed\nto mere existence in consciousness. Doubtless he so means, but this\nimplies that upon mere consciousness, or the simple presence of\nideas, there has supervened a distinction, which has to be accounted\nfor, of ideas from things which they represent on the one hand, and\nfrom a mind of which they are affections on the other. Even in the\npassage first quoted (Book II. chap. vii. sec. 7), where existence is\nascribed to every idea, on looking closely we find this distinction\nobtruding itself, though without explicit acknowledgment. In the\nvery same breath, so to speak, in which the idea of existence is\nsaid to be suggested by every idea, it is further described as being\neither of two considerations--either the consideration of an idea\nas actually in our mind, or of a thing as actually without us. Such\nconsiderations at once imply the supervention of that distinction\nbetween \u2018mind\u2019 and \u2018thing,\u2019 which gives a wholly new meaning to\n\u2018existence.\u2019 They are not, in truth, as Locke supposed, two separate\nconsiderations, one or other of which, as the case may be, is\ninterchangeable with the \u2018idea of existence.\u2019 One is correlative with\nthe other, and neither is the same as simple feeling. Considered as\nactually in the mind, the feeling is distinguished from the mind as\nan affection from the subject thereof, and just in virtue of this\ndistinction is referred to a thing as the cause of the affection, or\nbecomes representative of a thing. But for such consideration there\nwould for us, if the doctrine of ideas means anything, be no \u2018thing\nwithout us\u2019 at all. To \u2018consider things as actually without us\u2019 is to\nconsider them as causes of the ideas in our mind, and this is to have\nan idea of existence quite different from mere consciousness. It is\nto have an idea of it which at once suggests the question whether the\nexistence is real or apparent; in other words, whether the thing, to\nwhich an affection of the mind is referred as its cause, is really\nits cause or no.\n\nBy confusion of these two meanings, reality and its conditions are\nrepresented as given in simple feeling.\n\n32. Between these two meanings of existence--its meaning as\ninterchangeable with simple consciousness, and its meaning as\nreality--Locke failed to distinguish. Just as, having announced\n\u2018ideas\u2019 to be the sole \u2018materials of knowledge,\u2019 he allows himself\nat his convenience to put \u2018things\u2019 in the place of ideas; so having\nidentified existence with momentary consciousness or the simple\nidea, he substitutes for existence in this sense _reality_, and in\nconsequence finds reality given solely in the simple idea. Thus when\nthe conceptions of cause or substance, or relations of any kind, come\nunder view, since these cannot be represented as given in momentary\nconsciousness, they have to be pronounced not to exist, and since\nexistence is reality, to be unreal or \u2018fictions of the mind.\u2019 But\nwithout these unreal relations there could be no knowledge, and if\nthey are not given in the elements of knowledge, it is difficult\nto see how they are introduced, or to avoid the appearance of\nconstructing knowledge out of the unknown. Given in the elements of\nknowledge, however, they cannot be, if these are simple ideas or\nmomentary recurrences of the \u2018it is felt.\u2019 But by help of Locke\u2019s\nequivocation between the two meanings of existence, they can be\ncovertly introduced as the real. Existence is given in the simple\nidea, existence equals the real, therefore the real is given in the\nsimple idea. But think or speak of the real as we will, we find that\nit exhibits itself as substance, as cause, and as related; _i.e._\naccording to Locke as a \u2018complex\u2019 or \u2018invented\u2019 or \u2018superinduced\u2019\nidea.\n\nYet reality involves complex ideas which are made by the mind.\n\n33. In the second book of his Essay, which treats of ideas, he\nmakes the grand distinction between \u2018the simple ideas which are all\nfrom things themselves, and of which the mind can have no more or\nother than what are suggested to it,\u2019 and the \u2018complex ideas which\nare the workmanship of the mind.\u2019 (Book II. chap. xii.) In his\naccount of the latter there are some curious cross-divisions, but he\nfinally enumerates them as ideas either of _modes_, _substances_,\nor _relations_. The character of these ideas he then proceeds to\nexplain in the order given, one after the other, and as if each were\nindependent of the rest; though according to his own statement the\nidea of mode presupposes that of substance, and the idea of substance\ninvolves that of relation. \u2018Modes I call such complex ideas, which,\nhowever compounded, contain not in them the supposition of subsisting\nby themselves, but are considered as dependencies on, or affections\nof, substances; such are the ideas signified by the words \u2018triangle,\u2019\n\u2018gratitude,\u2019 \u2018murder,\u2019 &c. Of these there are two sorts. First, there\nare some which are only variations or different combinations of the\nsame simple idea without the mixture of any other--as a dozen, or\nscore--which are nothing but the ideas of so many distinct units\nadded together; and these I call simple modes, as being contained\nwithin the bounds of one simple idea. Secondly, there are others\ncompounded of simple ideas of several kinds, put together to make\none complex one; e. g. beauty, ... and these I call _mixed modes_.\u2019\n(Book II. chap. xii. secs. 4, 5.) So soon as he comes to speak more\nin detail of simple modes, he falls into apparent contradiction with\nhis doctrine that, as complex ideas, they are the mere workmanship\nof the mind. All particular sounds and colours are simple modes of\nthe simple ideas of sound and colour. (Book II. chap, xviii. sees. 3,\n4.) Again, the ideas of figure, place, distance, as of all particular\nfigures, places, and distances, are simple modes of the simple idea\nof space. (Book II. chap, xiii.) To maintain, however, that the ideas\nof space, sound, or colour _in general_ (as simple ideas) were taken\nfrom things themselves, while those of _particular_ spaces, sounds,\nand colours (as complex ideas) were \u2018made by the mind,\u2019 was for Locke\nimpossible. Thus in the very next chapter after that in which he\nhas opposed all complex ideas, those of simple modes included, as\nmade by the mind to all simple ones as taken from things themselves,\nhe speaks of simple modes \u2018either _as found in things existing_,\nor as made by the mind within itself.\u2019 (Book II. chap. xiii. sec.\n1.) It was not for Locke to get over this confusion by denying the\nantithesis between that which the mind \u2018makes\u2019 and that which it\n\u2018takes from existing things\u2019 and for the present we must leave it as\nit stands. We must further note that a mode being considered \u2018as an\naffection of a substance,\u2019 space must be to the particular spaces\nwhich are its simple modes, as a substance to its modifications.\nSo too colour to particular colours, &c., &c. But the idea of a\nsubstance is a complex idea \u2018framed by the mind.\u2019 Therefore the idea\nof space--at any rate such an idea as we have of it when we think of\ndistances, places, or figures, and when else do we think of it at\nall?--must be a complex and artificial idea. But according to Locke\nthe idea of space is emphatically a simple idea, given immediately\n_both_ by sight and touch, concerning which if a man enquire, he\n\u2018sends him to his senses.\u2019 (Book II. chap, v.)\n\nSuch are substance and relation which must be found in every object\nof knowledge.\n\n34. These contradictions are not avoidable blunders, due to\ncarelessness or want of a clear head in the individual writer, \u2018The\ncomplex idea of substance\u2019 will not be exorcised; the mind will show\nits workmanship in the very elements of knowledge towards which its\nrelation seems most passive--in the \u2018existing things\u2019 which are\nthe conditions of its experience no less than in the individual\u2019s\nconscious reaction upon them. The interrogator of the individual\nconsciousness seeks to know that consciousness, and just for that\nreason must find in it at every stage those formal conceptions,\nsuch as substance and cause, without which there can be no object\nof knowledge at all. He thus substantiates sensation, while he\nthinks that he merely observes it, and calls it a sensible thing.\nSensations, thus unconsciously transformed, are for him the real, the\nactually existent. Whatever is not given by immediate sense, outer or\ninner, he reckons a mere \u2018thing of the mind.\u2019 The ideas of substance\nand relation, then, not being given by sense, must in his eyes be\nthings of the mind, in distinction from, really existent things. But\nspeech bewrayeth him. He cannot state anything that he knows save\nin terms which imply that substance and relation are in the things\nknown; and hence an inevitable obtrusion of \u2018things of the mind\u2019 in\nthe place of real existence, just where the opposition between them\nis being insisted on. Again, as a man seems to observe consciousness\nin himself and others, it has nothing that it has not received. It\nis a blank to begin with, but passive of that which is without, and\nthrough its passivity it becomes informed. If the \u2018mind,\u2019 then, means\nthis or that individual consciousness, the things of the mind must be\ngradually developed from an original passivity. On the other hand,\nlet anyone try to know this original passive consciousness, and in\nit, as in every other known object-matter, he must find these things\nof the mind, substance and relations. If nature is the object, he\nmust find them in nature; if his own self-consciousness, he must\nfind them in that consciousness. But while nature knows not what is\nin herself, self-consciousness, it would seem, _ex vi termini_, does\nknow. Therefore not merely substance and relation must be found in\nthe original consciousness, but the knowledge, the ideas, of them.\n\nAbstract idea of substance and complex ideas of particular sorts of\nsubstance.\n\n35. As we follow Locke\u2019s treatment of these ideas more in detail, we\nshall find the logical see-saw, here accounted for, appearing with\nscarcely a disguise. His account of the origin of the \u2018complex ideas\nof substances\u2019 is as follows. \u2018The mind being furnished with a great\nnumber of the simple ideas, conveyed in by the senses, as they are\nfound in exterior things, or by reflection on its own operations,\ntakes notice also that a certain number of these simple ideas go\nconstantly together; which being presumed to belong to one thing,\nand words being suited to common apprehensions and made use of for\nquick despatch, are called, so united in one subject, by one name;\nwhich by inadvertency we are apt afterwards to talk of and consider\nas one simple idea, which indeed is a complication of many ideas\ntogether; because, as I have said, not imagining how these simple\nideas can subsist by themselves, we accustom ourselves to suppose\nsome _substratum_, wherein they do subsist, and from which they do\nresult; which therefore we call _substance_.\u2019 (Book II. chap, xxiii.\nsec. 1.) In the controversy with Stillingfleet, which arose out of\nthis chapter, Locke was constrained further to distinguish (as he\ncertainly did not do in the original text) between the \u2018ideas of\ndistinct substances, such as man, horse,\u2019 and the \u2018general idea of\nsubstance.\u2019 It is to ideas of the former sort that he must be taken\nto refer in the above passage, when he speaks of them as formed by\n\u2018complication of many ideas together,\u2019 and these alone are _complex_\nin the strict sense. The _general_ idea of substance on the other\nhand, which like all general ideas (according to Locke) is made\nby abstraction, means the idea of a \u2018substratum which we accustom\nourselves to suppose\u2019 as that wherein the complicated ideas \u2018do\nsubsist, and from which they do result.\u2019 This, however, he regards as\nitself one, \u2018the first and chief,\u2019 among the ideas which make up any\nof the \u2018distinct substances.\u2019 (Book II. chap. xii. sec. 6.) Nor is\nhe faithful to the distinction between the general and the complex.\nIn one passage of the first letter to Stillingfleet, he distinctly\nspeaks of the _general_ idea of substance as a \u2018_complex_ idea made\nup of the idea of something plus that of relation to qualities.\u2019 [1]\nNotwithstanding this confusion of terms, however, he no doubt had\nbefore him what seemed a clear distinction between the \u2018abstract\ngeneral idea\u2019 of substance, as such, _i.e._ of \u2018something related\nas a support to accidents,\u2019 but which does not include ideas of any\nparticular accidents, and the composite idea of a substance, made up\nof a multitude of simple ideas plus that of the something related to\nthem as a support. We shall find each of these ideas, according to\nLocke\u2019s statement, presupposing the other.\n\n[1] Upon a reference to the chapter on \u2018complex ideas\u2019 (Book II.\nchap, xii.), it will appear that the term is used in a stricter and\na looser sense. In the looser sense it is not confined to _compound_\nideas, but in opposition to simple ones includes those of relation\nand even \u2018abstract general ideas.\u2019 When Locke thinks of the _general_\nidea of substance apart from the complication of accidents referred\nto it, he opposes it to the complex idea, according to the stricter\nsense of that term. On the other hand, when he thinks of it as \u2018made\nup\u2019 of the idea of _something_ plus that of relation to qualities (as\nif there could be an idea of something apart from such relation), it\nseems to him to have two elements, and therefore to be complex.\n\nThe abstract idea according to Locke at once precedes and follows the\ncomplex.\n\n36. In the passage above quoted, our aptness to consider a\ncomplication of simple ideas, which we notice to go constantly\ntogether, as one simple idea, is accounted for as the result of a\npresumption that they belong to one thing. This presumption is again\ndescribed in the words that \u2018we accustom ourselves to suppose some\nsubstratum, wherein they do subsist, and from which they do result;\nwhich therefore we call substance.\u2019 Here it is implied that the idea\nof substance, _i.e._ \u2018the general idea of something related as a\nsupport to accidents,\u2019 is one gradually formed upon observation of\nthe regular coincidence of certain simple ideas. In the sequel (sec.\n3 of the same chapter I. xxiii.) we are told that such an idea--\u2018an\nobscure and relative idea of substance in general--being thus made,\nwe come to have the ideas of particular sorts of substances by\ncollecting such combinations of simple ideas as are, by experience\nand observation of men\u2019s senses, taken notice of to exist together.\u2019\nThus a _general_ idea of substance having been formed by one\ngradual process, ideas of particular sorts of substances are formed\nby another and later one. But then the very same \u2018collection of\nsuch combinations of simple ideas as are taken notice of to exist\ntogether,\u2019 which (according to sec. 3) constitutes the later process\nand follows upon the formation of the _general_ idea of substance,\nhas been previously described as preceding and conditioning that\nformation. It is the complication of simple ideas, noticed to go\nconstantly together, that (according to sec. 1) leads to the \u2018idea\nof substance in general.\u2019 To this see-saw between the process\npreceding and that following the formation of the idea in question\nmust be added the difficulty, that Locke\u2019s account makes the general\nidea precede the particular, which is against the whole tenor of\nhis doctrine of abstraction as an operation whereby \u2018the mind makes\nthe particular ideas, received from particular objects, to become\ngeneral.\u2019 (Book II. chap. xi. sec. 9.)\n\nReference of ideas to nature or God, the same as reference to\nsubstance.\n\n37. It may be said perhaps that Locke\u2019s self-contradiction in\nthis regard is more apparent than real; that the two processes of\ncombining simple ideas are essentially different, just because\nin the later process they are combined by a conscious act of the\nmind as accidents of a \u2018something,\u2019 of which the _general_ idea\nhas been previously formed, whereas in the earlier one they are\nmerely presented together \u2018by nature,\u2019 and, _ex hypothesi_, though\nthey gradually suggest, do not carry with them any reference to a\n\u2018substratum.\u2019 But upon this we must remark that the presentation of\nideas \u2018by nature\u2019 or \u2018by God,\u2019 though a mode of speech of which Locke\nin his account of the origin of knowledge freely avails himself,\nmeans nothing else than their relation to a \u2018substratum,\u2019 if not\n\u2018wherein they do subsist,\u2019 yet \u2018from which they do result.\u2019 If then\nit is for consciousness that ideas are presented together by nature,\nthey already carry with them that reference to a substratum which is\nsupposed gradually to result from their concurrence. If it is not for\nconsciousness that they are so presented, if they do not _severally_\ncarry with them a reference to \u2018something,\u2019 how is it they come to do\nso in the gross? If a single sensation of heat is not referred to a\nhot thing, why should it be so referred on the thousandth recurrence?\nBecause perhaps, recurring constantly in the same relations, it\ncompels the inference of permanent antecedents? But the \u2018same\nrelations\u2019 mean relations to the same things, and the observation of\nthese relations presupposes just that conception of _the thing_ which\nit is sought to account for,\n\nBut it is explicitly to substance that Locke makes them refer\nthemselves.\n\n38. We are estopped, however, from any such explanation of Locke as\nwould suggest these ulterior questions by his explicit statement\nthat \u2018all simple ideas, all sensible qualities, carry with them a\nsupposition of a substratum to exist in, and of a substance wherein\nthey inhere.\u2019 The vindication of himself against the pathetic\ncomplaint of Stillingfleet, that he had \u2018almost discarded substance\nout of the reasonable part of the world,\u2019 in which this statement\noccurs, was certainly not needed. Already in the original text\nthe simple ideas, of which the association suggests the idea of\nsubstance, are such as \u2018the mind finds in exterior things or by\nreflection on its own operations.\u2019 But to find them in an exterior\nthing is to find them in a substance, a \u2018something it knows not\nwhat,\u2019 regarded as outward, just as to find them by reflection on\nits own operations, as its own, is to find them in such a substance\nregarded as inward. The process then by which, according to Locke,\nthe general idea of substance is arrived at, presupposes this idea\njust as much as the process, by which ideas of particular sorts of\nsubstances are got, presupposes it, and the distinction between the\ntwo processes, as he puts it, disappears.\n\nIn the process by which we are supposed to arrive at complex ideas of\nsubstances the beginning is the same as the end.\n\n39. The same paralogism appears under a slightly altered form when\nit is stated (in the first letter to Stillingfleet) that the idea\nof substance as the \u2018general indetermined idea of _something_ is\nby the abstraction of the mind derived from the simple ideas of\nsensation and reflection.\u2019 Now \u2018abstraction\u2019 with Locke means the\n\u2018separation of an idea from all other ideas that accompany it in its\nreal existence.\u2019 (Book II. chap. xii. sec. 1.) It is clear then that\nit is impossible to abstract an idea which is not _there_, in real\nexistence, to be abstracted. Accordingly, if the \u2018general idea of\nsomething\u2019 is derived by abstraction from simple ideas of sensation\nand reflection, it must be originally given with these ideas, or it\nwould not afterwards be separated from them. Conversely they must\ncarry this idea with them, and cannot be simple ideas at all, but\ncompound ones, each made up of \u2018the general idea of something or\nbeing,\u2019 and of an accident which this something supports. How then\ndoes the general idea of substance or \u2018something,\u2019 _as derived_,\ndiffer from the idea of \u2018something,\u2019 as given in the original ideas\nof sensation and reflection from which the supposed process of\nabstraction starts? What can be said of the one that cannot be said\nof the other? If the derived general idea is of something related\nto qualities, what, according to Locke, are the original ideas\nbut those of qualities related to something? It is true that the\ngeneral idea is of something, of which nothing further is known,\nrelated to qualities in general, not to any particular qualities.\nBut the \u2018simple idea\u2019 in like manner can only be of an indeterminate\nquality, for in order to any determination of it, the idea must be\nput together with another idea, and so cease to be simple; and the\n\u2018something,\u2019 to which it is referred, must for the same reason be\na purely indeterminate something. If, in order to avoid concluding\nthat Locke thus unwittingly identified the abstract general idea of\nsubstance with any simple idea, we say that the simple idea, because\nnot abstract, is not indeterminate but of a real quality, defined\nby manifold relations, we fall upon the new difficulty that, if so,\nnot only does the simple idea become manifoldly complex, but just\nsuch an \u2018idea of a particular sort of substance\u2019 as, according to\nLocke, is derived from the derived idea of substance in general. As\nan idea of a quality, it is also necessarily an idea of a correlative\n\u2018something;\u2019 and if it is an idea of a quality in its reality, _i.e._\nas determined by various relations, it must be an idea of a variously\nqualified something, _i.e._ of a particular substance. Then not\nmerely the middle of the twofold process by which we are supposed\nto get at \u2018complex ideas of substances\u2019--_i.e._ the _abstract_\nsomething; but its end--_i.e._ the _particular_ something--turns out\nto be the same as its beginning.\n\nDoctrine of abstraction inconsistent with doctrine of complex ideas.\n\n40. The fact is, that in making the general idea of substance precede\nparticular ideas of sorts of substances (as he certainly however\nconfusedly does, in the 23rd chapter of the Second Book, [1] as well\nas by implication in his doctrine of modes. Book II. chap. xii.\nsec. 4), Locke stumbled upon a truth which he was not aware of, and\nwhich will not fit into his ordinary doctrine of general ideas: the\ntruth that knowledge is a process from the more abstract to the more\nconcrete, not the reverse, as is commonly supposed, and as Locke\u2019s\ndefinition of abstraction implies. Throughout his prolix discussion\nof \u2018substance\u2019 and \u2018essence\u2019 we find two opposite notions perpetually\ncross each other: one that knowledge begins with the simple idea,\nthe other that it begins with the real thing as particularized by\nmanifold relations. According to the former notion, simple ideas\nbeing given, void of relation, as the real, the mind of its own act\nproceeds to bring them into relation and compound them: according\nto the latter, a thing of various properties (_i.e._ relations [2])\nbeing given as the real, the mind proceeds to separate these from\neach other. According to the one notion the intellectual process, as\none of complication, ends just where, according to the other notion,\nas one of abstraction, it began.\n\n[1] See above, paragraph 35.\n\n[2] Cf. Book II. chap, xxiii. sec. 37. Most of the simple ideas that\nmake up our complex ideas of substances are only powers ... _e.g._\nthe greater part of the ideas which make up our complex idea of gold\n... are nothing else _but so many relations to other substances_.\u2019\n\nThe confusion covered by use of \u2018particulars\u2019.\n\n41. The chief verbal equivocation, under which Locke disguises the\nconfusion of these two notions, is to be found in the use of the\nword \u2018particular,\u2019 which is sometimes used for the mere individual\nhaving no community with anything else, sometimes for the thing\nqualified by relation to a multitude of other things. The simple idea\nor sensation; the \u2018something\u2019 which the simple idea is supposed to\n\u2018report,\u2019 and which Locke at his pleasure identifies with it; the\ncomplex idea; and the thing as the collection of the properties which\nthe simple idea \u2018reports,\u2019 all are merged by Locke under the one term\n\u2018particulars.\u2019 As the only consistency in his use of the term seems\nto lie in its opposition to \u2018generals,\u2019 we naturally turn to the\npassage where this opposition is spoken of most at large.\n\nLocke\u2019s account of abstract general ideas.\n\n42. \u2018General and universal belong not to the real existence of\nthings, but are the inventions and creatures of the understanding,\nmade by it for its own use, and concern only signs, whether words or\nideas. Words are general when used for signs of general ideas, and\nso are applicable indifferently to many particular things; and ideas\nare general, when they are set up as the representatives of many\nparticular things; but universality belongs not to things themselves,\nwhich are all of them particular in their existence, even those words\nand ideas which in their signification are general. When therefore\nwe quit particulars, the generals that rest are only creatures of\nour own making, their general nature being nothing but the capacity\nthey are put into by the understanding, of signifying or representing\nmany particulars. For the signification they have is nothing but a\nrelation that by the mind of man is added to them. ... The sorting of\nthings under names is the workmanship of the understanding, taking\noccasion from the similitude it observes among them to make abstract\ngeneral ideas, and set them up in the mind, with names annexed to\nthem, as patterns or forms (for in that sense the word form has a\nvery proper signification), to which as particular things are found\nto agree, so they come to be of that species, have that denomination,\nor are put into that classis. For when we say this is a man, that a\nhorse; this justice, that cruelty, what do we else but rank things\nunder different specific names, as agreeing to those abstract ideas,\nof which we have made those names the signs? And what are the\nessences of those species, set out and marked by names, but those\nabstract ideas in the mind; which are, as it were, the bonds between\nparticular things that exist, and the names they are to be ranked\nunder?\u2019 (Book III. chap. iii. secs. 11 and 13.)\n\n\u2018Things not general.\u2019\n\n43. In the first of these remarkable passages we begin with the\nfamiliar opposition between ideas as \u2018the creatures of the mind\u2019\nand real things. Ideas, and the words which express them, may be\ngeneral, but things cannot. \u2018They are all of them particular in their\nexistence.\u2019 Then the ideas and words themselves appear as things,\nand as such \u2018in their existence\u2019 can only be particular. It is only\nin its signification, _i.e._ in its relation to other ideas which\nit represents, that an idea, particular itself, becomes general,\nand this relation does not belong to the \u2018existence\u2019 of the idea or\nto the idea in itself, but \u2018by the mind of man is added to it.\u2019 The\nrelation being thus a fictitious addition to reality, \u2018general and\nuniversal are mere inventions and creatures of the understanding.\u2019\nThe next passage, in spite of the warning that all ideas are\nparticular in their existence, still speaks of general ideas, but\nonly as \u2018set up in the mind.\u2019 To these \u2018particular things existing\nare found to agree,\u2019 and the agreement is expressed in such judgments\nas \u2018this is a man, that a horse; this is justice, that cruelty;\u2019 the\n\u2018this\u2019 and \u2018that\u2019 representing \u2018particular existing things,\u2019 \u2018horse\u2019\nand \u2018cruelty\u2019 abstract general ideas to which these are found to\nagree.\n\nGenerality an invention of the mind.\n\n44. One antithesis is certainly maintained throughout these\npassages--that between \u2018real existence which is always particular,\nand the workmanship of the mind,\u2019 which \u2018invents\u2019 generality. Real\nexistence, however, is ascribed _(a)_ to things themselves, _(b)_ to\nwords and ideas, even those which become of general signification,\n_(c)_ to mixed modes, for in the proposition \u2018this is justice,\u2019\nthe \u2018this\u2019 must represent a mixed mode. (Cf. II. xii. 5.) The\ncharacteristic of the \u2018really existent,\u2019 which distinguishes it from\nthe workmanship of the mind, would seem to be mere individuality,\nexclusive of all relation. The simple \u2018this\u2019 and \u2018that,\u2019 apart from\nthe relation expressed in the judgment, being mere individuals,\nare really existent; and conversely, ideas, which in themselves\nhave real existence, when a relation, in virtue of which they\nbecome significant, has been \u2018added to them by the mind,\u2019 become\n\u2018inventions of the understanding.\u2019 This consists with the express\nstatement in the chapter on \u2018relation\u2019 (II. xxv. 8), that it is \u2018not\ncontained in the existence of things, but is something extraneous\nand superinduced.\u2019 Thus generality, as a relation between any one\nof a multitude of _single_ (not necessarily _simple_) ideas, _e.g._\nsingle ideas of horses, and all the rest--a relation which belongs\nnot to any one of them singly--is superinduced by the understanding\nupon their _real_, _i.e._ their _single_ existence. Apart from this\nrelation, it would seem, or in their mere singleness, even ideas of\nmixed modes, _e.g._ _this act_ of justice, may have real existence.\n\nThe result is, that the feeling of each moment is alone real.\n\n45. The result of Locke\u2019s statement, thus examined, clearly is that\nreal existence belongs to the present momentary act of consciousness,\nand to that alone. Ascribed as it is to the \u2018thing itself,\u2019 to the\nidea which, _as general_, has it not, and to the mixed mode, it is in\neach case the momentary presence to consciousness that constitutes\nit. To a thing itself, as distinct from the presentation to\nconsciousness, it cannot belong, for such a \u2018thing\u2019 means that which\nremains identical with itself under manifold appearances, and both\nidentity and appearance imply relation, _i.e._ \u2018an invention of the\nmind.\u2019 As little can it belong to the _content_ of any idea, since\nthis is in all cases constituted by relation to other ideas. Thus\nif I judge \u2018this is sweet,\u2019 the real existence lies in the simple\n\u2018this,\u2019 in the mere form of presentation at an individual _now_,\nnot in the relation of this to other flavours which constitutes the\ndeterminate sweetness, or to a sweetness at other times tasted.\nIf I judge \u2018this is a horse,\u2019 a present vision really exists, but\nnot so its relation to other sensations of sight or touch, closely\nprecedent or sequent, which make up the \u2018total impression;\u2019 much less\nits relation to other like impressions thought of, in consideration\nof which a common name is applied to it. If, again, I judge \u2018this\nis an act of justice,\u2019 the present thought of the act, as present,\nreally exists; not so those relations of the act which either make\nit just, or make me apply the name to it. It is true that according\nto this doctrine the \u2018really existent\u2019 is the unmeaning, and that\nany statement about it is impossible. We cannot judge of it without\nbringing it into relation, in which it ceases to be what in its\nmere singleness it is, and thus loses its reality, overlaid by the\n\u2018invention of the understanding.\u2019 Nay, if we say that it is the mere\n\u2018this\u2019 or \u2018that,\u2019 as such--the simple \u2018here\u2019 and \u2018now\u2019--the very\n\u2018this,\u2019 in being mentioned or judged of, becomes related to other\nthings which we have called \u2018this,\u2019 and the \u2018now\u2019 to other \u2018nows.\u2019\nThus each acquires a generality, and with it becomes fictitious.\nAs Plato long ago taught--though the lesson seems to require to\nbe taught anew to each generation of philosophers--a consistent\nsensationalism must be speechless. Locke, himself, in one of the\npassages quoted, implicitly admits this by indicating that only\nthrough relations or in their generality are ideas \u2018significant.\u2019\n\nHow Locke avoids this result.\n\n46. He was not the man, however, to become speechless out of sheer\nconsistency. He has a redundancy of terms and tropes for disguising\nfrom himself and his reader the real import of his doctrine. In\nthe latter part of the passage quoted we find that the relation\nor community between ideas, which the understanding invents, is\noccasioned by a \u2018similitude which it observes among things.\u2019 The\ngeneral idea having been thus invented, \u2018things are found to agree\nwith it\u2019--as is natural since they suggested it. Hereupon we are\nforced to ask how, if all relation is superinduced upon real\nexistence by the understanding, an _observed_ relation of similitude\namong things can occasion the superinduction; and again how it\nhappens, if all generality of ideas is a fiction of the mind, that\n\u2018things are found to agree with general ideas.\u2019 How can the real\nexistence called \u2018this\u2019 or \u2018that,\u2019 which only really exists so far as\nnothing can be said of it but that it is \u2018this\u2019 or \u2018that,\u2019 agree with\nanything whatever? Agreement implies some content, some determination\nby properties, _i.e._ by relations, in the things agreeing, whereas\nthe really existent excludes relation. How then can it agree with the\nabstract general idea, the import of which, according to Locke\u2019s own\nshowing, depends solely on relation?\n\nThe \u2018particular\u2019 was to him the individual qualified by general\nrelations.\n\n47. Such questions did not occur to Locke, because while asserting\nthe mere individuality of things existent, and the simplicity of\nall ideas as _given_, _i.e._ as real, he never fully recognised the\nmeaning of his own assertion. Under the shelter of the ambiguous\n\u2018particular\u2019 he could at any time substitute for the _mere_\nindividual the _determinate_ individual, or individual qualified\nby community with other things; just as, again, under covering\nof the \u2018simple idea\u2019 he could substitute for the mere momentary\nconsciousness the perception of a definite thing. Thus when he speaks\nof the judgment \u2018this is gold\u2019 as expressing the agreement of a real\n(_i.e._ individual) thing with a general idea, he thinks of \u2018this\u2019\na& already having, apart from the judgment, the determination which\nit first receives in the judgment. He thinks of it, in other words,\nnot as the mere \u2018perishing\u2019 sensation [1] or individual void of\nrelation, but as a sensation symbolical of other possibilities of\nsensation which, as so many relations of a _thing_ to us or to other\nthings, are connoted by the common noun \u2018gold.\u2019 It thus \u2018agrees\u2019\nwith the abstract idea or conception of qualities, _i.e._ because\nit is already the \u2018creature of the understanding,\u2019 determined by\nrelations which constitute a generality and community between it and\nother things. Such a notion of the really existent thing--wholly\ninconsistent with his doctrine of relation and of the general--Locke\nhas before him when he speaks of general ideas as formed by\nabstraction of certain qualities from real things, or of certain\nideas from other ideas that accompany them in real existence. \u2018When\nsome one first lit on a parcel of that sort of substance we denote by\nthe word _gold_, ... its peculiar colour, perhaps, and weight were\nthe first he abstracted from it, to make the complex idea of that\nspecies ... another perhaps added to these the ideas of fusibility\nand fixedness ... another its ductility and solubility in aqua regia.\nThese, or part of these, put together, usually make the complex idea\nin men\u2019s minds of that sort of body we call _gold_.\u2019 (Book II. chap.\nxxxi. sec. 9.) Here the supposition is that a thing, multitudinously\nqualified, is given apart from any action of the understanding,\nwhich then proceeds to act in the way of successively detaching\n(\u2018abstracting\u2019) these qualities and recombining them as the idea of a\nspecies. Such a recombination, indeed, would seem but wasted labour.\nThe qualities are assumed to be already found by the understanding\nand found as in a thing; otherwise the understanding could not\nabstract them from it. Why should it then painfully put together in\nimperfect combination what has been previously given to it complete?\nOf the complex idea which results from the work of abstraction,\nnothing can be said but a small part of what is predicable of the\nknown thing which the possibility of such abstraction presupposes.\n\n[1] \u2018All impressions are perishing existences.\u2019--Hume. See below,\nparagraph 208.\n\nThis is the real thing from which abstraction is supposed to start.\n\n48. \u2018The complex idea of a species,\u2019 spoken of in the passage last\nquoted, corresponds to what, in Locke\u2019s theory of substance, is\ncalled the \u2018idea of a particular sort of substance.\u2019 In considering\nthat theory we saw that, according to his account, the beginning of\nthe process by which the \u2018abstract idea of substance\u2019 was formed,\nwas either that abstract idea itself, the mere \u2018something,\u2019 or by\na double contradiction the \u2018complex idea of a particular sort of\nsubstance\u2019 which yet we only come to have _after_ the abstract idea\nhas been formed. In the passage now before us there is no direct\nmention of the abstraction of the \u2018substratum,\u2019 as such, but only of\nthe quality, and hence there is no ambiguity about the paralogism. It\nis not a mere \u2018something\u2019 that the man \u2018lights upon,\u2019 and thus it is\nnot this that holds the place at once of the given and the derived\nbut a something having manifold qualities to be abstracted. In other\nwords, it is the \u2018idea of a particular sort of substance\u2019 that he\nstarts from, and it is just this again to which as a \u2018complex idea\nof a species,\u2019 his understanding is supposed gradually to lead him.\nThe understanding, indeed, according to Locke, is never adequate\nto nature, and accordingly the qualities abstracted and recombined\nin the complex idea always fall vastly short of the fulness of\nthose given in the real thing; or as he states it in terms of the\nmultiplication table (Book II. chap. xxxi. sec. 10), \u2018some who have\nexamined this species more accurately could, I believe, enumerate ten\ntimes as many properties in gold, all of them as inseparable from its\ninternal constitution, as its colour or weight; and it is probable\nif any one knew all the properties that are by divers men known of\nthis metal, there would an hundred times as many ideas go to the\ncomplex idea of gold, as any one man has yet in his; and yet perhaps\nthat would not be the thousandth part of what is to be discovered\nin it.\u2019 These two million properties, and upwards, which await\nabstraction in gold, are all, it must be noted, according to Locke\u2019s\nstatement elsewhere (Book II. chap xxiii. sec. 37), \u2018nothing but so\nmany relations to other substances.\u2019 It is just on account of these\nmultitudinous relations of the real thing that the understanding is\ninadequate to its comprehension. Yet according to Locke\u2019s doctrine of\nrelation these must all be themselves \u2018superinductions of the mind,\u2019\nand the greater the fulness which they constitute, the further is the\ndistance from the _mere_ individuality which elsewhere, in contrast\nwith the fictitiousness of \u2018generals,\u2019 appears as the equivalent of\nreal existence.\n\nYet, according to the doctrine of relation, a creation of thought.\n\n49. The real thing and the creation of the understanding thus change\nplaces. That which is given to the understanding as the real, which\nit finds and does not make, is not now the bare atom upon which\nrelations have to be artificially superinduced. Nor is it the\nmere present feeling, which has \u2018by the mind of man\u2019 to be made\n\u2018significant,\u2019 or representative of past experience. It is itself an\ninexhaustible complex of relations, whether they are considered as\nsubsisting between it and other things, or between the sensations\nwhich it is \u2018fitted to produce in us.\u2019 These are the real, which is\nthus a system, a community; and if the \u2018general,\u2019 as Locke says,\nis that which \u2018has the capacity of representing many particulars,\u2019\nthe real thing itself is general, for it represents--nay, is\nconstituted by--the manifold particular feelings which, mediately or\nimmediately, it excites in us. On the other hand, the invention of\nthe understanding, instead of giving \u2018significance\u2019 or content to\nthe mere individuality of the real, as it does according to Locke\u2019s\ntheory of \u2018generals,\u2019 now appears as detaching fragments from the\nfulness of the real to recombine them in an \u2018abstract essence\u2019 of its\nown. Instead of adding complexity to the simple, it subtracts from\nthe complex.\n\nSummary of the above contradictions.\n\n50. To gather up, then, the lines of contradiction which traverse\nLocke\u2019s doctrine of real existence as it appears in his account of\ngeneral and complex ideas:--The idea of substance is an abstract\ngeneral idea, not given directly in sensation or reflection, but\n\u2018invented by the understanding,\u2019 as by consequence must be ideas of\nparticular sorts of substances which presuppose the abstract idea.\nOn the other hand, the ideas of sensation and reflection, from which\nthe idea of substance is abstracted, and to which as _real_ it as an\n_invention_ is opposed, are ideas of \u2018something,\u2019 and are only real\nas representative of something. But this idea of something = the idea\nof substance. Therefore the idea of substance is the presupposition,\nand the condition of the reality, of the very ideas from which it\nis said to be derived. Again, if the general idea of substance is\ngot by abstraction, it must be originally given in conjunction with\nthe ideas of sensation or reflection from which it is afterwards\nabstracted, _i.e._ separated. But in such conjunction it constitutes\nthe ideas of particular sorts of substances. Therefore these latter\nideas, which yet we \u2018come to have\u2019 after the general idea of\nsubstance, form the prior experience from which this general idea is\nabstracted. Further, this original experience, from which abstraction\nstarts, being of \u2018sorts of substances,\u2019 and these sorts being\nconstituted by relations, it follows that relation is given in the\noriginal experience. But that which is so given is \u2018real existence\u2019\nin opposition to the invention of the understanding. Therefore these\nrelations, and the community which they constitute, really exist. On\nthe other hand, mere individuals alone really exist, while relations\nbetween them are superinduced by the mind. Once more, the simple idea\ngiven in sensation or reflection, as it is made _for_ not _by_ us,\nhas or results from real existence, whereas general and complex ideas\nare the workmanship of the mind. But this workmanship consists in the\nabstraction of ideas from each other, and from that to which they are\nrelated as qualities. It thus presupposes at once the general idea\nof \u2018something\u2019 or substance, and the complex idea of qualities of\nthe something. Therefore it must be general and complex ideas that\nare real, as made for and not by us, and that afford the inventive\nunderstanding its material. Yet if so--if they are _given_--why make\nthem over again by abstraction and recomplication?\n\nThey cannot be overcome without violence to Locke\u2019s fundamental\nprinciples.\n\n51. We may get over the last difficulty, indeed, by distinguishing\nbetween the complex and confused, between abstraction and analysis.\nWe may say that what is originally given in experience is the\nconfused, which to us is simple, or in other words has no definite\ncontent, because, till it has been analysed, nothing can be said of\nit, though in itself it is infinitely complex; that thus the process,\nwhich Locke roughly calls abstraction, and which, as he describes\nit, consists merely in taking grains from the big heap that is\ngiven in order to make a little heap of one\u2019s own, is yet, rightly\nunderstood, the true process of knowledge--a process which may be\nsaid at once to begin with the complex and to end with it, to take\nfrom the concrete and to constitute it, because it begins with that\nwhich is in itself the fulness of reality, but which only becomes so\nfor us as it is gradually spelt out by our analysis. To put the case\nthus, however, is not to correct Locke\u2019s statement, but wholly to\nchange his doctrine. It renders futile his easy method of \u2018sending\na man to his senses\u2019 for the discovery of reality, and destroys the\nsupposition that the elements of knowledge can be ascertained by the\ninterrogation of the individual consciousness. Such consciousness can\ntell nothing of its own beginning, if of this beginning, as of the\npurely indefinite, nothing can be said; if it only becomes defined\nthrough relations, which in its state of primitive potentiality are\nnot actually in it. The senses again, so far from being, in that\nmere passivity which Locke ascribes to them, organs of ready-made\nreality, can have nothing to tell, if it is only through the active\nprocesses of \u2018discerning, comparing, and compounding,\u2019 that they\nacquire a definite content. But to admit this is nothing else than,\nin order to avoid a contradiction of which Locke was not aware, to\nefface just that characteristic of his doctrine which commends it to\n\u2018common sense\u2019--the supposition, namely, that the simple datum of\nsense, as it is for sense or in its mere individuality, is the real,\nin opposition to the \u2018invention of the mind.\u2019 That this supposition\nis to make the real the unmeaning, the empty, of which nothing can be\nsaid, he did not see because, under an unconscious delusion of words,\neven while asserting that the names of simple ideas are undefinable\n(Book III. chap. iv. sec. 4), which means that nothing can be said\nof such ideas, and while admitting that the processes of discerning,\ncomparing, and compounding ideas, which mean nothing else than the\nbringing them into relation [1] or the superinduction upon them of\nfictions of the mind, are necessary to constitute even the beginnings\nof knowledge, he yet allows himself to invest the simple idea, as\nthe real, with those definite qualities which can only accrue to\nit, according to his showing, from the \u2018inventive\u2019 action of the\nunderstanding.\n\n[1] Locke only states this explicitly of comparison, \u2018an operation\nof the mind about its ideas, upon which depends all that large tribe\nof ideas, comprehended under relation.\u2019 (Book II. chap. xi. sec.\n4.) It is clear, however, that the same remark must apply to the\n\u2018discernment of ideas,\u2019 which is strictly correlative to comparison,\nand to their composition, which means that they are brought into\nrelation as constituents of a whole.\n\nThat these three processes are necessary to constitute the beginnings\nof knowledge, according to Locke, appears from Book II. chap. xi.\nsec. 15, taken in connection with what precedes in that chapter.\n\nAs real existence, the simple idea carries with it \u2018invented\u2019\nrelation of cause.\n\n52. Thus invested, it is already substance or symbolical of\nsubstance, not a mere feeling but a felt thing, recognised either\nunder that minimum of qualification which enables us merely to say\nthat it is \u2018something,\u2019 or (in Locke\u2019s language) abstract substance,\nor under the greater complication of qualities which constitutes\na \u2018particular sort of substance\u2019--gold, horse, water, &c. Real\nexistence thus means substance. It is not the simple idea or\nsensation by itself that is real, but this idea as caused by a thing.\nIt is the thing that is primarily the real; the idea only secondarily\nso, because it results from a power in the thing. As we have seen,\nLocke\u2019s doctrine of the necessary adequacy, reality, and truth of\nthe simple idea turns upon the supposition that it is, and announces\nitself as, an \u2018ectype\u2019 of an \u2018archetype.\u2019 But there is not a\ndifferent archetype to each sensation; if there were, in \u2018reporting\u2019\nit the sensation would do no more than report itself. It is the\nsupposed single cause of manifold different sensations or simple\nideas, to which a single name is applied. \u2018If sugar produce in us the\nideas which we call whiteness and sweetness, we are sure there is a\npower in sugar to produce those ideas in our minds. ... And so each\nsensation answering the power that operates on any of our senses, the\nidea so produced is a real idea (and not a fiction of the mind, which\nhas no power to produce any single idea), and cannot but be adequate\n... and so all simple ideas are adequate.\u2019 (Book II. chap. xxxi.\nsec. 2.) The sugar, which is here the \u2018archetype\u2019 and the source of\nreality in the idea, is just what Locke elsewhere calls \u2018a particular\nsort of substance,\u2019 as the \u2018something\u2019 from which a certain set of\nsensations result, and in which, as sensible qualities, they inhere.\nStrictly speaking, however, according to Locke, that which inheres in\nthe thing is not the quality, as it is to us, but a power to produce\nit. (Book II. chap. viii. sec. 28, and c. xxiii. 37.)\n\nCorrelativity of cause and substance.\n\n53. In calling a sensation or idea the product of a power, substance\nis presupposed just as much as in calling it a sensible quality;\nonly that with Locke \u2018quality\u2019 conveyed the notion of inherence in\nthe substance, power that of relation to an effect not _in_ the\nsubstance itself. \u2018Secondary qualities are nothing but the powers\nwhich _substances_ have to produce several ideas in us by our senses,\nwhich ideas are not in the things themselves, otherwise than as\nanything is in its cause.\u2019 (Book II. chap, xxiii. sec. 9.) \u2018Most of\nthe simple ideas, that make up our complex ideas of substances, are\nonly powers ... or relations to other substances (or, as he explains\nelsewhere, \u2018relations to our perceptions,\u2019 [1]), and are not really\nin the substance considered barely in itself.\u2019 (Book II. chap,\nxxiii. sec. 37, and xxxi. 8.) That this implies the inclusion of the\nidea of cause in that of substance, appears from Locke\u2019s statement\nthat \u2018whatever is considered by us to operate to the producing any\nparticular simple idea which did not before exist, hath thereby in\nour minds the relation of a cause.\u2019 (Book II. chap. xxvi. sec. 1.)\nThus to be conscious of the reality of a simple idea, as that which\nis not made by the subject of the idea, but results from a power\nin a thing, is to have the idea of substance as cause. This latter\nidea must be the condition of the consciousness of reality. If the\nconsciousness of reality is implied in the beginning of knowledge, so\nmust the correlative ideas be of cause and substance.\n\n[1] Book II. chap. xxi. sec. 3.\n\nHow do we know that ideas correspond to reality of things? Locke\u2019s\nanswer.\n\n54. On examining Locke\u2019s second rehearsal of his theory in the fourth\nbook of the Essay--that \u2018On Knowledge\u2019--we are led to this result\nquite as inevitably as in the book \u2018On Ideas.\u2019 He has a special\nchapter on the \u2018reality of human knowledge,\u2019 where he puts the\nproblem thus:--\u2018It is evident the mind knows not things immediately,\nbut only by the intervention of the ideas it has of them. Our\nknowledge therefore is real only so far as there is a conformity\nbetween our ideas and the reality of things. But what shall be here\nthe criterion? How shall the mind, when it perceives nothing but its\nown ideas, know that they agree with things themselves?\u2019 (Book IV.\nchap. iv. sec. 3.) It knows this, he proceeds to show, in the case\nof simple ideas, because \u2018since the mind can by no means make them\nto itself, they must be the product of things operating on the mind\nin a natural way. ... Simple ideas are not fictions of our fancies,\nbut the natural and regular productions of things without us, really\noperating upon us; and so carry with them all the conformity which\nis intended, or which our state requires, for they represent to us\nthings under those appearances which they are fitted to produce in\nus; whereby we are enabled to distinguish the sorts of particular\nsubstances,\u2019 &c. &c. (Book IV. chap. iv. sec. 4.) The whole force\nof this passage depends on the notion that simple ideas are already\nto the subject of them not his own making, but the product of a\nthing, which in its relation to these ideas is a \u2018particular sort of\nsubstance.\u2019 It is the reception of such ideas, so related, that Locke\ncalls \u2018sensitive knowledge of particular existence,\u2019 or a \u2018perception\nof the mind, employed about the particular existence of finite\nbeings without us.\u2019 (Book IV. chap. ii. sec. 14.) This, however, he\ndistinguishes from two other \u2018degrees of knowledge or certainty,\u2019\n\u2018intuition\u2019 and \u2018demonstration,\u2019 of which the former is attained when\nthe agreement or disagreement of two ideas is perceived immediately,\nthe latter when it is perceived mediately through the intervention\nof certain other agreements or disagreements (less or more), each of\nwhich must in turn be perceived immediately. Demonstration, being\nthus really but a series of intuitions, carries the same certainty\nas intuition, only it is a certainty which it requires more or less\npains and attention to apprehend. (Book IV. chap. ii. sec. 4.) Of\nthe \u2018other perception of the mind, employed about the particular\nexistence of finite beings without us,\u2019 which \u2018passes under the\nname of knowledge,\u2019 he explains that although \u2018going beyond bare\nprobability, it reaches not perfectly to either of the foregoing\ndegrees of certainty.\u2019 \u2018There can be nothing more certain,\u2019 he\nproceeds, \u2018than that the idea we receive from an external object\nis in our minds; this is intuitive knowledge. But whether there be\nanything more than barely that idea in our minds, whether we can\nthence certainly infer the existence of anything without us which\ncorresponds to that idea, is that whereof some men think there may be\na question made; because men may have such ideas in their minds, when\nno such thing exists, no such object affects their senses.\u2019 (Book IV.\nchap. ii. sec. 14.)\n\nIt assumes that simple ideas are consciously referred to things that\ncause them.\n\n55. It is clear that here in his very statement of the question\nLocke begs the answer. If the intuitive certainty is that \u2018the idea\nwe _receive from an external object_ is in our minds,\u2019 [1] how is\nit possible to doubt whether such an object exists and affects\nour senses? This impossibility of speaking of the simple idea,\nexcept as received from an object, may account for Locke\u2019s apparent\ninconsistency in finding the assurance of the reality of knowledge\n(under the phrase \u2018evidence of the senses\u2019) just in that \u2018perception\u2019\nwhich reaches not to intuitive or demonstrative certainty, and only\n\u2018passes under the name of knowledge.\u2019 In the passage just quoted he\nshows that he is cognizant of the distinction between the simple\nidea and the perception of an existence corresponding to it, and in\nconsequence distinguishes this perception from proper intuition, but\nin the very statement of the distinction it eludes him. The simple\nidea, as he speaks of it, becomes itself, as consciously \u2018received\nfrom an external object,\u2019 the perception of existence; just as we\nhave previously seen it become the judgment of identity or perception\nof the \u2018agreement of an idea with itself,\u2019 which is his first kind of\nknowledge.\n\n[1] I do not now raise the question, What are here the ideas, which\nmust be immediately perceived to agree or disagree in order to make\nit a case of \u2018intuitive certainty\u2019 or knowledge according to Locke\u2019s\ndefinition. See below, paragraphs 59, 101, and 147.\n\nLively ideas real, because they must be effects of things.\n\n56. In short, with Locke the simple idea, the perception of existence\ncorresponding to the idea, and the judgment of identity, are\nabsolutely merged, and in mutual involution, sometimes under one\ndesignation, sometimes under another, are alike presented as the\nbeginning of knowledge. As occasion requires, each does duty for the\nother. Thus, if the \u2018reality of knowledge\u2019 be in question, the simple\nidea, which is given, is treated as involving the perception of\nexistence, and the reality is established. If in turn this perception\nis distinguished from the simple idea, and it is asked whether the\ncorrespondence between idea and existence is properly matter of\nknowledge, the simple idea has only to be treated as involving the\njudgment of identity, which again involves that of existence, and\nthe question is answered. So in the context under consideration\n(Book IV. chap. ii. sec. 14), after raising the question as to the\nexistence of a thing corresponding to the idea, he answers it by\nthe counter question, \u2018whether anyone is not invincibly conscious\nto himself of a different perception, when he looks on the sun by\nday, and thinks on it by night; when he actually tastes wormwood, or\nsmells a rose, or only thinks on that savour or odour? We as plainly\nfind the difference there is between any idea revived in our minds\nby our own memory, and actually coming into our minds by our senses,\nas we do between any two distinct ideas.\u2019 The force of the above\nlies in its appeal to the perception of identity, or--to apply the\nlanguage in which Locke describes this perception--the knowledge that\nthe idea which a man calls the smell of a rose is the very idea it\nis. [1] The mere difference in liveliness between the present and\nthe recalled idea, which, as Berkeley and Hume rightly maintained,\nis the only difference between them as mere ideas, cannot by itself\nconstitute the difference between the knowledge of the presence of a\nthing answering to the idea and the knowledge of its absence. It can\nonly do this if the more lively idea is _identified_ with past lively\nideas as a representation of one and the same thing which \u2018agrees\nwith itself\u2019 in contrast to the multiplicity of the sensations,\nits signs. Only in virtue of this identification can either the\nliveliness of the idea show that the thing--the sun or the rose--is\nthere, or the want of liveliness that it is not, for without it there\nwould be no thing to be there or not to be there. It is because\nthis identification is what Locke understands by the first sort of\nperception of agreement between ideas, and because he virtually finds\nthis perception again in the simple idea, that the simple idea is\nto him the index of reality. But if so, the idea in its primitive\nsimplicity is the sign of a thing that is ever the same in the same\nrelations, and we find the \u2018workmanship of the mind,\u2019 its inventions\nof substance, cause, and relation, in the very rudiments of knowledge.\n\n[1] See above, paragraph 25.\n\nPresent sensation gives knowledge of existence.\n\n57. With that curious tendency to reduplication, which is one of his\ncharacteristics, Locke, after devoting a chapter to the \u2018reality of\nhuman knowledge,\u2019 of which the salient passage as to simple ideas has\nbeen already quoted, has another upon our \u2018knowledge of existence.\u2019\nHere again it is the sensitive knowledge of things actually present\nto our senses, which with him is merely a synonym for the simple\nidea, that is the prime criterion. (Book IV. chap. iii. secs. 5 and\n2, and chap. ii. sec. 2.) After speaking of the knowledge of our own\nbeing and of the existence of a God (about which more will be said\nbelow), he proceeds, \u2018No particular man can know the existence of any\nother being, but only when, by actually operating upon him, it makes\nitself perceived by him. For the having the idea of anything in our\nmind no more proves the existence of that thing, than the picture of\na man evidences his being in the world, or the visions of a dream\nmake thereby a true history. It is therefore the actual receiving of\nideas from without, that gives us notice of the existence of other\nthings, and makes us know that something doth exist at that time\nwithout us, which causes that idea in us, though perhaps we neither\nknow nor consider how it does it; for it takes not from the certainty\nof our senses and the ideas we receive by them, that we know not the\nmanner wherein they are produced; _e.g._ whilst I write this, I have,\nby the paper affecting my eyes, that idea produced in my mind, which,\nwhatever object causes, I call _white_; by which I know that the\nquality or accident (_i.e._ whose appearance before my eyes always\ncauses that idea) doth really exist, and hath a being without me. And\nof this the greatest assurance. I can possibly have, and to which\nmy faculties can attain, is the testimony of my eyes, which are the\nproper and sole judges of this thing, whose testimony I have reason\nto rely on, as so certain, that I can no more doubt whilst I write\nthis, that I see white and black, and that something really exists\nthat causes that sensation in me, than that I write and move my\nhand.\u2019 (Book IV. chap. xi. secs. 1, 2.)\n\nReasons why its testimony must be trusted.\n\n58. Reasons are afterwards given for the assurance that the\n\u2018perceptions\u2019 in question are produced in us by \u2018exterior causes\naffecting our senses.\u2019 The first _(a)_ is, that \u2018those that want\nthe organs of any sense never can have the ideas belonging to that\nsense produced in their mind.\u2019 The next _(b)_, that whereas \u2018if\nI turn my eyes at noon toward the sun, I cannot avoid the ideas\nwhich the light or the sun then produces in me;\u2019 on the other hand,\n\u2018when my eyes are shut or windows fast, as I can at pleasure recall\nto my mind the ideas of light or the sun, which former sensations\nhad lodged in my memory, so I can at pleasure lay them by.\u2019 Again\n_(c)_, \u2018many of those ideas are produced in us with pain which\nafterwards we remember without the least offence. Thus the pain of\nheat or cold, when the idea of it is revived in our minds, gives\nus no disturbance; which, when felt, was very troublesome, and is\nagain, when actually repeated; which is occasioned by the disorder\nthe external object causes in our body, when applied to it.\u2019 Finally\n(d), \u2018our senses in many cases bear witness to the truth of each\nother\u2019s report, concerning the existence of sensible things without\nus. He that sees a fire may, if he doubt whether it be anything\nmore than a bare fancy, feel it too.\u2019 Then comes the conclusion,\ndangerously qualified: \u2018When our senses do actually convey into our\nunderstandings any idea, we cannot but be satisfied that there doth\nsomething at that time really exist without us, which doth affect\nour senses, and by them give notice of itself to our apprehensive\nfaculties, and actually produce that idea which we then perceive;\nand we cannot so far distrust their testimony as to doubt that such\ncollections of simple ideas, as we have observed by our senses to\nbe united together, actually exist together. But this knowledge\nextends as far as the present testimony of our senses, employed about\nparticular objects, that do then affect them, and no farther. For\nif I saw such a collection of simple ideas as is wont to be called\nman, existing together one minute since, and am now alone; I cannot\nbe certain that the same man exists now, since there is no necessary\nconnexion of his existence a minute since with his existence now. By\na thousand ways he may cease to be, since I had the testimony of my\nsenses for his existence.\u2019 (Book IV. chap. xi. sec. 9.)\n\nHow does this account fit Locke\u2019s definition of knowledge?\n\n59. Upon the \u2018knowledge of the existence of things,\u2019 thus\nestablished, it has to be remarked in the first place that, after\nall, according to Locke\u2019s explicit statement, it is not properly\nknowledge. It is \u2018an assurance that deserves the name of knowledge\u2019\n(Book IV. chap. ii. sec. 14, and xi. sec. 3), yet being neither\nitself an intuition of agreement between ideas, nor resoluble into a\nseries of such intuitions, the definition of knowledge excludes it.\nOnly if existence were itself an \u2018idea,\u2019 would the consciousness of\nthe agreement of the idea with it be a case of knowledge; but to make\nexistence an idea is to make the whole question about the agreement\nof ideas, as such, with existence, as such, unmeaning. To seek escape\nfrom this dilemma by calling the consciousness of the agreement\nin question an \u2018assurance\u2019 instead of knowledge is a mere verbal\nsubterfuge. There can be no assurance of agreement between an idea\nand that which is no object of consciousness at all. If, however,\nexistence is an object of consciousness, it can, according to Locke,\nbe nothing but an idea, and the question as to the _assurance_\nof agreement is no less unmeaning than the question as to the\n_knowledge_ of it. The raising of the question in fact, as Locke puts\nit, implies the impossibility of answering it. It cannot be raised\nwith any significance, unless existence is external to and other than\nan idea. It cannot be answered unless existence is, or is given in,\nan object of consciousness, _i.e._ an idea.\n\nLocke\u2019s account of the testimony of sense renders his question as to\nits veracity superfluous.\n\n60. As usual, Locke disguises this difficulty from himself, because\nin answering the question he alters it. The question, _as he asks\nit_, is whether, given the idea, we can have posterior assurance of\nsomething else corresponding to it. The question, _as he answers\nit_, is whether the idea includes the consciousness of a real thing\nas a constituent; and the answer consists in the simple assertion,\nvariously repeated, that it does. It is clear, however, that this\nanswer to the latter question does not answer, but renders unmeaning,\nthe question as it is originally asked. If, according to Locke\u2019s own\nshowing, there is nowhere for anything to be found by us but in our\n\u2018ideas\u2019 or our consciousness--if the _thing_ is given in and with the\nidea, so that the idea is merely the thing _ex parte nostr\u00e2_--then to\nask if the idea agrees with the thing is as futile as to ask whether\nhearing agrees with sound, or the voice with the words it utters.\nThat the thing is so given is implied throughout Locke\u2019s statement\nof the \u2018assurance we have of the existence of material beings,\u2019 as\nwell as of the confirmations of this assurance. If the \u2018idea which\nI call white\u2019 means the knowledge that \u2018the property or accident\n(_i.e._ whose appearance before my eyes always causes that idea)\ndoth really exist and hath a being without me,\u2019 then consciousness\nof existence--outward, permanent, substantive, and causative\nexistence--is involved in the idea, and no ulterior question of\nagreement between idea and existence can properly arise. But unless\nthe simple idea is so interpreted, the senses have no testimony to\ngive. If it is so interpreted, no extraneous \u2018reason to rely upon\nthe testimony\u2019 can be discovered, for such reason can only be a\nrepetition of the testimony itself.\n\nConfirmations of the testimony turn upon the distinction between\n\u2018impression\u2019 and \u2018idea\u2019.\n\n61. This becomes clearer upon a view of the confirmations of the\ntestimony, as Locke gives them. They all, we may remark by the way,\npresuppose a distinction between the simple idea as originally\nrepresented and the same as recalled or revived. This distinction,\nfixed by the verbal one between \u2018impression\u2019 and \u2018idea,\u2019 we shall\nfind constantly maintained and all-important in Hume\u2019s system; but in\nLocke, though upon it (as we shall see) rests his distinction between\nreal and nominal essence and his confinement of general knowledge\nto the latter, it seems only to turn up as an afterthought. In the\naccount of the reality and adequacy of ideas it does not appear at\nall. There the distinction is merely between the simple idea, as\nsuch, and the complex, as such, without any further discrimination of\nthe simple idea as originally produced from the same as recalled. So,\ntoo, in the opening account of the reception of simple ideas (Book\nII. chap. xii. sec. 1), \u2018Perception,\u2019 \u2018Retention,\u2019 and \u2018Discerning\u2019\nare all reckoned together as alike forms of the _passivity_ of the\nmind, in contrast with its activity in combination and abstraction,\nthough retention and discerning have been previously described\nin terms which imply activity. In the \u2018confirmations\u2019 before us,\nhowever, the distinction between the originally produced and the\nrevived is essential.\n\nThey depend on language which presupposes the ascription of sensation\nto an outward cause.\n\n62. The first turns upon the impossibility of producing an idea _de\nnovo_ without the action of sensitive organs; the two next upon\nthe difference between the idea as produced through these organs\nand the like idea as revived at the will of the individual. It is\nhence inferred that the idea as originally produced is the work of\na thing, which must exist _in rerum natur\u00e2_, and by way of a fourth\n\u2018confirmation\u2019 the man who doubts this in the case of one sensation\nis invited to try it in another. If, on seeing a fire, he thinks it\n\u2018bare fancy,\u2019 _i.e._ doubts whether his idea is caused by a thing,\nlet him put his hand into it. This last \u2018confirmation\u2019 need not be\nfurther noticed here, since the operation of a producing thing is\nas certain or as doubtful for one sensation as for another. [1] Two\ncertainties are not more sure than one, nor can two doubts make a\ncertainty. The other \u2018confirmations\u2019 alike lie in the words \u2018product\u2019\nand \u2018organ.\u2019 A man has a certain \u2018idea:\u2019 afterwards he has another\nlike it, but differing in liveliness and in the accompanying pleasure\nor pain. If he already has, or if the ideas severally bring with\nthem, the idea of a producing outward thing to which parts of his\nbody are organs, on the one hand, and of a self \u2018having power\u2019 on the\nother, then the liveliness, and the accompanying pleasure or pain,\nmay become indications of the action of the thing, as their absence\nmay be so of the action of the man\u2019s self; but not otherwise. Locke\nthroughout, in speaking of the simple ideas as produced or recalled,\nimplies that they carry with them the consciousness of a cause,\neither an outward thing or the self, and only by so doing can he find\nin them the needful \u2018confirmations\u2019 of the \u2018testimony of the senses.\u2019\nThis testimony is confirmed just because it distinguishes of itself\nbetween the work of \u2018nature,\u2019 which is real, and the work of the\nman, which is a fiction. In other words, the confirmation is nothing\nelse than the testimony itself--a testimony which, as we have seen,\nsince it supposes consciousness, as such, to be consciousness _of a\nthing_, eliminates by anticipation the question as to the agreement\nof consciousness with things, as with the extraneous.\n\n[1] To feel the object, in the sense of touching it, had a special\nsignificance for Locke, since touch with him was the primary\n\u2018revelation\u2019 of body, as the solid. More will be said of this when we\ncome to consider his doctrine of \u2018real essence,\u2019 as constituted by\nprimary qualities of body. See below, paragraph 101.\n\nThis ascription means the clothing of sensation with invented\nrelations.\n\n63. The distinction between the real and the fantastic, according\nto the passages under consideration, thus depends upon that between\nthe work of nature and the work of man. It is the confusion between\nthe two works that renders the fantastic possible, while it is the\nconsciousness of the distinction that sets us upon correcting it.\nWhere all is the work of man and professes to be no more, as in the\ncase of \u2018mixed modes,\u2019 there is no room for the fantastic (Book II.\nchap. xxx. sec. 4, and Book IV. chap. iv. sec. 7); and where there\nis ever so much of the fantastic, it would not be so for us, unless\nwe were conscious of a \u2018work of nature,\u2019 to which to oppose it. But\non looking a little closer we find that to be conscious of an idea\nas the work of nature, in opposition to the work of man, is to be\nconscious of it under relations which, according to Locke, are the\ninventions of man. It is nothing else than to be conscious of it as\nthe result of \u2018something having power to produce it\u2019 (Book II. chap.\nxxxi. sec. 2), _i.e._ of a substance, to which it is related as a\nquality. \u2018Nature\u2019 is just the \u2018something we know not what,\u2019 which\nis substance according to the \u2018_abstract_ idea\u2019 thereof. Producing\nideas, it exercises powers, as it essentially belongs to substance\nto do, according to our _complex_ idea of it. (Book II. chap, xxiii.\nsecs. 9, 10.) But substance, according to Locke, whether as abstract\nor complex idea, is the \u2018workmanship of the mind,\u2019 and power, as a\nrelation (Book II. chap. xxi. sec. 3, and chap. xxv. sec. 8), \u2018is\nnot contained in the real existence of things.\u2019 Again, the idea of\nsubstance, as a source of power, is the same as the idea of cause.\n\u2018Whatever is considered by us to operate to the producing any\nparticular simple idea, which did not before exist, hath thereby in\nour minds the relation of a cause.\u2019 (Book II. chap. xxvi. sec. 1.)\nBut the idea of cause is not one \u2018that the mind has of things as they\nare in themselves,\u2019 but one that it gets by its own act in \u2018bringing\nthings to, and setting them by, one another.\u2019 (Book II. chap. xxv.\nsec. 1.) Thus it is with the very ideas, which are the workmanship\nof man, that the simple idea has to be clothed upon, in order to\n\u2018testify\u2019 to its being real, _i.e._ (in Locke\u2019s sense) not the work\nof man.\n\nWhat is meant by restricting the testimony of sense to _present_\nexistence?\n\n64. Thus invested, the simple idea has clearly lost its simplicity.\nIt is not the momentary, isolated consciousness, but the\nrepresentation of a thing determined by relations to other things\nin an order of nature, and causing an infinite series of resembling\nsensations to which a common name is applied. Thus in all the\ninstances of sensuous testimony mentioned in the chapter before\nus, it is not really a simple sensation that is spoken of, but a\nsensation referred to a thing--not a mere smell, or taste, or sight,\nor feeling, but the smell of a rose, the taste of a pine-apple, the\nsight of the sun, the feeling of fire. (Book IV. chap. xi. secs.\n4-7.) Immediately afterwards, however, reverting or attempting to\nrevert to his strict doctrine of the mere individuality of the\nsimple idea, he says that the testimony of the senses is a \u2018present\ntestimony employed about particular objects, that do then affect\nthem,\u2019 and that sensitive knowledge extends no farther than such\ntestimony. This statement, taken by itself, is ambiguous. Does it\nmean that sensation testifies to the momentary presence to the\nindividual of a continuous existence, or is the existence itself as\nmomentary as its presence to sense? The instance that follows does\nnot remove the doubt. \u2018If I saw such a collection of simple ideas\nas is wont to be called _man_, existing together one minute since,\nand am now alone; I cannot be certain that the same man exists now,\nsince there is no necessary connection of his existence a minute\nsince with his existence now.\u2019 (Book IV. chap. xi. sec. 9.) At first\nsight, these words might seem to decide that the existence is merely\ncoincident with the presence of the sensation--a decision fatal to\nthe distinction between the real and fantastic, since, if the thing\nis only present with the sensation, there can be no combination\nof qualities in reality other than the momentary coincidence of\nsensations in us. Memory or imagination, indeed, might recall these\nin a different order from that in which they originally occurred;\nbut, if this original order had no being after the occurrence, there\ncould be no ground for contrasting it with the order of reproduction\nas the real with the merely apparent.\n\nSuch restriction, if maintained, would render the testimony unmeaning.\n\n65. In the very sentence, however, where Locke restricts the\ntestimony of sensation to existence present along with it, he uses\nlanguage inconsistent with this restriction. The particular existence\nwhich he instances as \u2018testified to\u2019 is that of \u2018such a collection\nof simple ideas as is wont to be called man.\u2019 But these ideas can\nonly be present in succession. (See Book II. chap. vii. sec. 9, and\nchap. xiv. sec. 3.) Even the surface of the man\u2019s body can only be\ntaken in by successive acts of vision; and, more obviously, the\nstates of consciousness in which his qualities of motion and action\nare presented occupy separate times. If then sensation only testifies\nto an existence present along with it, how can it testify to the\nco-existence (say) of an erect attitude, of which I have a present\nsight, with the risibility which I saw a minute ago? How can the\n\u2018collection of ideas wont to be called man,\u2019 _as co-existing_, be\nformed at all? and, if it cannot, how can the present existence of\nan object so-called be testified to by sense any more than the past?\nThe same doctrine, which is fatal to the supposition of \u2018a necessary\nconnexion between the man\u2019s existence a minute since and his\nexistence now,\u2019 is in fact fatal to the supposition of his existence\nas a complex of qualities at all. It does not merely mean that, for\nanything we know, the man may have died. Of course he may, and yet\nthere may be continuity of existence according to natural laws,\nthough not one for which we have the testimony of present sense,\nbetween the living body and the dead. What Locke had in his mind\nwas the notion that, as existence is testified to only by present\nsensation, and each sensation is merely individual and momentary,\nthere could be no testimony to the continued existence of anything.\nHe could not, however, do such violence to the actual fabric of\nknowledge as would have been implied in the logical development of\nthis doctrine, and thus he allowed himself to speak of sense as\ntestifying to the co-existence of sensible qualities in a thing,\nthough the individual sensation could only testify to the presence of\none at a time, and could never testify to their _nexus_ in a common\ncause at all. This testimony to co-existence in a present thing once\nadmitted, he naturally allowed himself in the further assumption\nthat the testimony, on its recurrence, is a testimony to the same\nco-existence and the same thing. The existence of the same man (he\nevidently supposes), to which sensation testified an hour ago, may be\ntestified to by a like sensation now. This means that resemblance of\nsensation becomes identity of a thing--that like sensations occurring\nat different times are interpreted as representing the same thing,\nwhich continuously exists, though not testified to by sense, between\nthe times.\n\nBut it is not maintained: the testimony is to operation of permanent\nidentical things.\n\n66. In short, as we have seen the simple idea of sensation emerge\nfrom Locke\u2019s inquiry as to the beginning of knowledge transformed\ninto the judgment, \u2018I have an idea different from other ideas\nwhich I did not make for myself,\u2019 so now from the inquiry as to\nthe correspondence between knowledge and reality it emerges as the\nconsciousness of a thing now acting upon me, which has continued\nto exist since it acted on me before, and in which, as in a common\ncause, have existed together powers to affect me which have never\naffected me together. If in the one form the operation of thought in\nsense, the \u2018creation of the understanding\u2019 within the simple idea,\nis only latent or potential, in the other it is actual and explicit.\nThe relations of substance and quality, of cause and effect, and of\nidentity--all \u2018inventions of the mind\u2019--are necessarily involved in\nthe immediate, spontaneous testimony of passive sense.\n\nLocke\u2019s treatment of relations of cause and identity.\n\n67. It will be noticed that it is upon the first of these, the\nrelation of substance and quality, that our examination of Locke\u2019s\nEssay has so far chiefly gathered. In this it follows the course\ntaken by Locke himself. Of the idea of substance, _eo nomine_, he\ntreats at large: of cause and identity (apart from the special\nquestion of personal identity) he says little. So, too, the \u2018report\nof the senses\u2019 is commonly exhibited as announcing the sensible\nqualities of a thing rather than the agency of a cause or continuity\nof existence. The difference, of course, is mainly verbal. Sensible\nqualities being, as Locke constantly insists, nothing but \u2018powers\nto operate on our senses\u2019 directly or indirectly, the substance\nor thing, as the source of these, takes the character of a cause.\nAgain, as the sensible quality is supposed to be one and the same in\nmanifold separate cases of being felt, it has identity in contrast\nwith the variety of these cases, even as the thing has, on its part,\nin contrast with the variety of its qualities. Something, however,\nremains to be said of Locke\u2019s treatment of the ideas of cause and\nidentity in the short passages where he treats of them expressly.\nHere, too, we shall find the same contrast between the given and the\ninvented, tacitly contradicted by an account of the given in terms of\nthe invented.\n\nThat from which he derives idea of cause pre-supposes it.\n\n68. The relation of cause and effect, according to Locke\u2019s general\nstatement as to relation, must be something \u2018not contained in the\nreal existence of things, but extraneous and superinduced.\u2019 (Book II.\nchap. xxv. sec. 8.) It is a \u2018complex idea,\u2019 not belonging to things\nas they are in themselves, which the mind makes by its own act.\n(Book II. chap xii. secs. 1, 7, and chap. xxv. sec. 1.) Its origin,\nhowever, is thus described:--\u2018In the notice that our senses take\nof the constant vicissitude of things, we cannot but observe that\nseveral particular, both qualities and substances, begin to exist;\nand that they receive this their existence from the due application\nand operation of some other being. From this observation we get\nour ideas of cause and effect. That which produces any simple or\ncomplex idea we denote by the general name cause; and that which is\nproduced, effect. Thus, finding that in that substance which we call\nwax, fluidity, which is a simple idea that was not in it before,\nis constantly produced by the application of a certain degree of\nheat, we call the simple idea of heat, in relation to fluidity in\nwax, the cause of it, and fluidity the effect. So, also, finding\nthat the substance, wood, which is a certain collection of simple\nideas so-called, by the application of fire is turned into another\nsubstance called ashes, _i.e._ another complex idea, consisting of a\ncollection of simple ideas, quite different from that complex idea\nwhich we call wood; we consider fire, in relation to ashes, as cause,\nand the ashes as effect.\u2019 Here we find that the \u2018given,\u2019 upon which\nthe relation of cause and effect is \u2018superinduced\u2019 or from which the\n\u2018idea of it is got\u2019 (to give Locke the benefit of both expressions),\nprofessedly, according to the first sentence of the passage quoted,\ninvolves the complex or derived idea of substance. The sentence,\nindeed, is a remarkable instance of the double refraction which\narises from redundant phraseology. Our senses are supposed to \u2018take\nnotice of a constant vicissitude of things,\u2019 or substances. Thereupon\nwe observe, what is necessarily implied in this vicissitude, a\nbeginning of existence in substances or their qualities, \u2018received\nfrom the due application or operation of some other being.\u2019 Thereupon\nwe infer, what is simply another name for existence thus given\nand received, a relation of cause and effect. Thus not only does\nthe _datum_ of the process of \u2018invention\u2019 in question, _i.e._ the\nobservation of change in a thing, involve a _derived_ idea, but a\nderived idea which presupposes just this process of invention.\n\nRationale of this \u2018petitio principii\u2019.\n\n69. Here again it is necessary to guard against the notion that\nLocke\u2019s obvious _petitio principii_ might be avoided by a better\nstatement without essential change in his doctrine of ideas. It is\ntrue that \u2018a notice of the vicissitude of things\u2019 includes that\n\u2018invention of the understanding\u2019 which it is supposed to suggest,\nbut state the primary knowledge otherwise--reduce the vicissitude of\nthings, as it ought to be reduced, in order to make Locke consistent,\nto the mere multiplicity of sensations--and the appearance of\nsuggestion ceases. Change or \u2018vicissitude\u2019 is quite other than mere\ndiversity. It is diversity relative to something which maintains an\nidentity. This identity, which ulterior analysis may find in a \u2018law\nof nature,\u2019 Locke found in \u2018things\u2019 or \u2018substances.\u2019 By the same\nunconscious subreption, by which with him a sensible thing takes\nthe place of sensation, \u2018vicissitude of things\u2019 takes the place of\nmultiplicity of sensations, carrying with it the observation that\nthe changed state of the thing is due to something else. The mere\nmultiplicity of sensations could convey no such \u2018observation,\u2019 any\nmore than the sight of counters in a row would convey the notion\nthat one \u2018received its existence\u2019 from the other. Only so far as the\nmanifold appearances are referred, as its vicissitudes, to something\nwhich remains one, does any need of accounting for their diverse\nexistence, or in consequence any observation of its derivation \u2018from\nsome other being,\u2019 arise. Locke, it is true, after stating that it\nis upon a notice of the vicissitude of things that the observation\nin question rests, goes on to speak as if an _origination_ of\nsubstances, which is just the opposite of their vicissitude, might be\nobserved; and the second instance of production which he gives--that\nof ashes upon the burning of wood--seems intended for an instance\nof the production of a substance, as distinct from the production\nof a quality. He is here, however, as he often does, using the term\n\u2018substance\u2019 loosely, for \u2018a certain collection of simple ideas,\u2019\nwithout reference to the \u2018substratum wherein they do subsist,\u2019 which\nhe would have admitted to be ultimately the same for the wood and\nfor the ashes. The conception, indeed, of such a substratum, whether\nvaguely as \u2018nature,\u2019 or more precisely as a \u2018real constitution of\ninsensible parts\u2019 (Book III. chap. iii. secs. 18, &c.), governed all\nhis speculation, and rendered to him what he here calls _substance_\nvirtually a _mode_, and its production properly a \u2018vicissitude.\u2019\n\nRelation of cause has to be put into sensitive experience in order to\nbe got from it.\n\n70. We thus find that it is only so far as simple ideas are referred\nto things--only so far as each in turn, to use Locke\u2019s instance,\nis regarded as an appearance \u2018in a substance which was not in it\nbefore\u2019--that our sensitive experience, the supposed _datum_ of\nknowledge, is an experience of the vicissitudes of things; and again,\nthat only as an experience of such vicissitude does it furnish the\n\u2018observation from which we get our ideas of cause and effect.\u2019 But\nthe reference of a sensation to a sensible thing means its reference\nto a cause. In other words, the invented relation of cause and effect\nmust be found in the primary experience in order that it may be got\nfrom it. [1]\n\n[1] Locke\u2019s contradiction of himself in regard to this relation might\nbe exhibited in a still more striking light by putting side by side\nwith his account of it his account of the idea of power. The two are\nprecisely similar, the idea of power being represented as got by a\nnotice of the alteration of simple ideas in things without (Book II.\nchap. xxi. sec. 1), just as the idea of cause and effect is. Power,\ntoo, he expressly says, is a relation. Yet, although the idea of it,\nboth as derived and as of a relation, ought to be complex, he reckons\nit a simple and original one, and by using it interchangeably with\n\u2018sensible quality\u2019 makes it a primary _datum_ of sense.\n\nOrigin of the idea of identity according to Locke.\n\n71. The same holds of that other \u2018product of the mind,\u2019 the\nrelation of identity. This \u2018idea\u2019 according to Locke, is formed\nwhen, \u2018considering anything as existing at any determined time and\nplace, we compare it with itself existing at another time.\u2019 \u2018In this\nconsists identity,\u2019 he adds, \u2018when the ideas it is attributed to,\nvary not at all from what they were that moment wherein we consider\ntheir former existence, and to which we compare the present; for we\nnever finding nor conceiving it possible that two things of the same\nkind should exist in the same place at the same time, we rightly\nconclude that whatever exists anywhere, at any time, excludes all of\nthe same kind, and is there itself alone. When, therefore, we demand\nwhether anything be the same or no? it refers always to something\nthat existed such a time in such a place, which it was certain at\nthat instant was the same with itself, and no other; from whence it\nfollows that one thing cannot have two beginnings of existence, nor\ntwo things one beginning; it being impossible for two things of the\nsame kind to be or exist in the same instant in the very same place,\nor one and the same thing in different places. That, therefore, that\nhad one beginning, is the same thing; and that which had a different\nbeginning in time and place from that is not the same, but diverse.\u2019\nHe goes on to inquire about the _principium individuationis_, which\nhe decides is \u2018existence itself, which determines a being of any sort\nto a particular time and place, incommunicable to two beings of the\nsame kind ... for being at that instant what it is and nothing else,\nit is the same, and so must continue as long as its existence is\ncontinued; for so long it will be the same, and no other.\u2019 (Book II.\nchap, xxvii. secs. 1-3).\n\nRelation of identity not to be distinguished from idea of it.\n\n72. It is essential to bear in mind with regard to identity, as\nwith regard to cause and effect, that no distinction according to\nLocke can legitimately be made between the relation and the idea\nof the relation. As to substance, it is true, he was driven in his\ncontroversy with Stillingfleet to distinguish between \u2018the being\nand the idea thereof,\u2019 but in dealing with relation he does not\nattempt any such violence to his proper system. Between the \u2018idea\u2019 as\nsuch and \u2018being\u2019 as such, his \u2018new way of ideas,\u2019 as Stillingfleet\nplaintively called it, left no fair room for distinction. In\nthis indeed lay its permanent value for speculative thought. The\ndistinction by which alone it could consistently seek to replace the\nold one, so as to meet the exigencies of language and knowledge,\nwas that between simple ideas, as given and necessarily real, and\nthe reproductions or combinations in which the mind may alter them.\nBut since every relation implies a putting together of ideas, and\nis thus always, as Locke avows, a complex idea or the work of the\nmind, a distinction between its being and the idea thereof, in that\nsense of the distinction in which alone it can ever be consistently\nadmitted by Locke, was clearly inadmissible. Thus in the passages\nbefore us the relation of identity is not explicitly treated as an\noriginal \u2018being\u2019 or \u2018existence,\u2019 It is an idea formed by the mind\nupon a certain \u2018consideration of things\u2019 being or existent. But on\nlooking closely at Locke\u2019s account, we find that it is only so far as\nit already belongs to, nay constitutes, the things, that it is formed\nupon consideration of them.\n\nThis \u2018invented\u2019 relation forms the \u2018very being of things\u2019.\n\n73. When it is said that the idea of identity, or of any other\nrelation, is formed upon consideration of things as existing in a\ncertain way, this is naturally understood to mean--indeed, otherwise\nit is unmeaning--that the things are first _known_ as existing, and\nthat afterwards the idea of the relation in question is formed. But\naccording to Locke, as we have seen, [1] the first and simplest act\nof knowledge possible is the perception of identity between ideas.\nEither then the \u2018things,\u2019 upon consideration of which the idea of\nidentity is formed, are not known at all, or the knowledge of them\ninvolves the very idea afterwards formed on consideration of them.\nLocke, having at whatever cost of self-contradiction to make his\ntheory fit the exigencies of language, virtually adopts the latter\nalternative, though with an ambiguity of expression which makes a\ndefinite meaning difficult to elicit. We have, however, the positive\nstatement to begin with, that the comparison in which the relation\noriginates, is of a thing with itself as existing at another time.\nAgain, the \u2018ideas\u2019 (used interchangeably with \u2018things\u2019), to which\nidentity is attributed, \u2018vary not at all from what they were at\nthat moment wherein we consider their former existence.\u2019 It is here\nclearly implied that \u2018things\u2019 or \u2018ideas\u2019 _exist_, _i.e._ are given\nto us in the spontaneous consciousness which we do not make, as each\none and the same throughout a multiplicity of times. This, again,\nmeans that the relation of identity or sameness, _i.e._ unity of\nthing under multiplicity of appearance, belongs to or consists in\nthe \u2018very being\u2019 of those given objects of consciousness, which\nare in Locke\u2019s sense the real, and upon which according to him all\nrelation is superinduced by an after-act of thought. So long as each\nsuch object \u2018continues to exist,\u2019 so long its \u2018sameness with itself\nmust continue,\u2019 and this sameness is the complex idea, the relation,\nof identity. Just as before, following Locke\u2019s lead, we found the\nsimple idea, as the element of knowledge, become complex--a perceived\nidentity of ideas; so now mere existence, the \u2018very being of things\u2019\n(which with Locke is only another name for the simple idea), resolves\nitself into a relation, which it requires \u2018consideration by the mind\u2019\nto constitute.\n\n[1] See above, paragraph 25.\n\nLocke fails to distinguish between identity and mere unity.\n\n74. The process of self-contradiction, by which a \u2018creation of the\nmind\u2019 finds its way into the real or given, must also appear in a\ncontradictory conception of the real itself. Kept pure of all that\nLocke reckons intellectual fiction, it can be nothing but a simple\nchaos of individual units: only by the superinduction of relation\ncan there be sameness, or continuity of existence, in the minutest\nof these for successive moments. Locke presents it arbitrarily under\nthe conception of mere individuality or of continuity, according\nas its distinction from the work of the mind, or its intelligible\ncontent, happens to be before him. A like see-saw in his account of\nthe individuality and generality of ideas has already been noticed.\n[1] In his discussion of identity the contradiction is partly\ndisguised by a confusion between mere unity on the one hand, and\nsameness or unity in difference, on the other. Thus, after starting\nwith an account of identity as belonging to ideas which are the same\n_at different times_, he goes on to speak of a thing as the same\nwith itself, _at a single instant_. So, too, by the _principium\nindividuationis_, he understands \u2018existence itself, which determines\na being of any sort to a particular time and place.\u2019 As it is clear\nfrom the context that by the _principium individuationis_ he meant\nthe source of identity or sameness, it will follow that by \u2018sameness\u2019\nhe understood singleness of a thing in a single time and place.\nWhence then the plurality, without which \u2018sameness\u2019 is unmeaning? In\nfact, Locke, having excluded it in his definition, covertly brings\nit back again in his instance, which is that of \u2018an atom, _i.e._\na continued body under one immutable superficies, existing in a\ndetermined time and place.\u2019 This, \u2018considered in any instant of its\nexistence, is in that instant the same with itself.\u2019 But it is so\nbecause--and, if we suppose the consideration of plurality of _times_\nexcluded, only because--it is a \u2018_continued_\u2019 body, which implies,\nthough its place be determined, that it exists _in a plurality of\nparts of space_. Either this plurality, or that of instants of its\nexistence, must be recognised in contrast with the unity of body,\nif this unity is to become \u2018sameness with itself.\u2019 In adding that\nnot only at the supposed instant is the atom the same, but \u2018so must\ncontinue as long as its existence continues,\u2019 Locke shows that he\nreally thought of the identical body under a plurality of times _ex\nparte post_, if not _ex parte ante_.\n\n[1] See above, paragraphs 43, and the following.\n\nFeelings are the real, and do not admit of identity. How then can\nidentity be real?\n\n75. But how is this continuity, or sameness of existence in plurality\nof times or spaces, compatible with the constitution of \u2018real\nexistence\u2019 by mere _individua_? The difficulty is the same, according\nto Locke\u2019s premisses, whether the simple ideas by themselves are\ntaken for the real _individua_, or whether each is taken to represent\na single separate thing. In his chapter on identity he expressly\nsays that \u2018things whose existence is in succession\u2019 do not admit\nof identity. Such, he adds, are motion and thought; \u2018because, each\nperishing the moment it begins, they cannot exist in different times\nor in different places as permanent beings can at different times\nexist in distant places.\u2019 (Book I. chap, xxvii. sec. 2.) What he\nhere calls \u2018thought\u2019 clearly includes the passive consciousness in\nwhich alone, according to his strict doctrine, reality is given. So\nelsewhere (Book II. chap. vii. sec. 9), in accounting for the \u2018simple\nidea of succession,\u2019 he says generally that \u2018if we look immediately\ninto ourselves we shall find our ideas always, whilst we have any\nthought, passing in train, one going and another coming, without\nintermission.\u2019 [1] No statement of the \u2018perpetual flux\u2019 of ideas,\nas each having a separate beginning and end, and ending in the very\nmoment when it begins, can be stronger than the above. If \u2018ideas\u2019 of\nany sort, according to this account of them, are to constitute real\nexistence, no sameness can be found in reality. It must indeed be a\nrelation \u2018invented by the mind.\u2019\n\n[1] It is true that in this place Locke distinguishes between the\n\u2018suggestion by our senses\u2019 of the idea of succession, and that which\npasses in our \u2018minds,\u2019 by which it is \u2018more constantly offered us.\u2019\nBut since, according to him, the idea of sensation must be \u2018produced\nin the mind\u2019 if there is to be any either sensation or idea at all\n(Book II. chap, ix. secs. 3 and 4), the distinction between the\n\u2018suggestion by our senses\u2019 and what passes in our minds\u2019 cannot be\nmaintained.\n\nYet it is from reality that the idea of it is derived.\n\n76. This, it may be said, is just the conclusion that was wanted in\norder to make Locke\u2019s doctrine of the particular relation of identity\ncorrespond with his general doctrine of the fictitiousness of\nrelations. To complete the consistency, however, his whole account of\nthe origin of the relation (or of the idea in which it consists) must\nbe changed, since it supposes it to be derived from an observation\nof things or existence, which again is to suppose sameness to be\nin the things or to be real. This change made, philosophy would\nhave to start anew with the problem of accounting for the origin of\nthe fictitious idea. It would have to explain how it comes to pass\nthat the mind, if its function consists solely in reproducing and\ncombining given ideas, or again in \u2018abstracting\u2019 combined ideas from\neach other, should be able to invent a relation which is neither a\ngiven idea, nor a reproduction, combination, or abstract residuum\nof given ideas. This is the great problem which we shall find Hume\nattempting. Locke really never saw its necessity, because the\ndominion of language--a dominion which, as he did not recognise it,\nhe had no need to account for--always, in spite of his assertion that\nsimple ideas are the sole _data_ of consciousness, held him to the\nbelief in another _datum_ of which ideas are the appearances, viz., a\nthing having identity, because the same with itself in the manifold\ntimes of its appearance. This _datum_, under various guises, but\nin each demonstrably, according to Locke\u2019s showing, a \u2018creation of\nthought,\u2019 has met us in all the modes of his theory, as the condition\nof knowledge. As the \u2018abstract idea\u2019 of substance it renders\n\u2018perishing\u2019 ideas into qualities by which objects may be discerned.\n(Book II. chap. xi. sec. 1.) As the relative idea of cause, it makes\nthem \u2018affections\u2019 to be accounted for. As the fiction of a universal,\nit is the condition of their mutual qualification as constituents of\na whole. Finally, as the \u2018superinduced\u2019 relation of sameness, the\ndirect negative of the perpetual beginning and ending of \u2018ideas,\u2019 it\nconstitutes the \u2018very being of things.\u2019\n\nTransition to Locke\u2019s doctrine of essence.\n\n77. \u2018The very being of things,\u2019 let it be noticed, according to\nwhat Locke reckoned their \u2018real,\u2019 as distinct from their \u2018nominal,\u2019\nessence. The consideration of this distinction has been hitherto\npostponed; but the discussion of the relation of identity, as\nsubsisting between the parts of a \u2018continued body,\u2019 brings us upon\nthe doctrine of matter and its \u2018primary qualities,\u2019 which cannot be\nproperly treated except in connection with the other doctrine (which\nLocke unhappily kept apart) of the two sorts of \u2018essence.\u2019 So far,\nit will be remembered, the \u2018facts\u2019 or _given_ ideas, which we have\nfound him unawares converting into theories or \u2018invented\u2019 ideas, have\nbeen those of the \u2018secondary qualities of body.\u2019 [1] It is these\nwhich are united into things or substances, having been already\n\u2018found in them:\u2019 it is from these that we \u2018infer\u2019 the relation of\ncause and effect, because as \u2018vicissitudes of things\u2019 or \u2018affections\nof sense\u2019 they presuppose it: it is these again which, as \u2018received\nfrom without,\u2019 testify the present existence of something, because\nin being so received they are already interpreted as \u2018appearances\nof something.\u2019 That the \u2018thing,\u2019 by reference to which these ideas\nare judged to be \u2018real,\u2019 \u2018adequate,\u2019 and \u2018true\u2019--or, in other words,\nbecome elements of a knowledge--is yet itself according to Locke\u2019s\ndoctrine of substance and relation a \u2018fiction of thought,\u2019 has been\nsufficiently shown. That it is so no less according to his doctrine\nof essence will also appear. The question will then be, whether\nby the same showing the ideas of body, of the self, and of God,\ncan be other than fictions, and the way will be cleared for Hume\u2019s\nphilosophic adventure of accounting for them as such.\n\n[1] See above, paragraph 20.\n\nThis repeats the inconsistency found in his doctrine of substance.\n\n78. In Locke\u2019s doctrine of \u2018ideas of substances,\u2019 the \u2018thing\u2019\nappeared in two inconsistent positions: on the one hand, as that in\nwhich they \u2018are found;\u2019 on the other, as that which results from\ntheir concretion, or which, such concretion having been made, we\naccustom ourselves to suppose as its basis. This inconsistency,\nlatent to Locke himself in the theory of substance, comes to the\nsurface in the theory of essence, where it is (as he thought)\novercome, but in truth only made more definite, by a distinction of\nterms.\n\nPlan to be followed.\n\n79. This latter theory has so far become part and parcel of the\n\u2018common sense\u2019 of educated men, that it might seem scarcely to need\nrestatement. It is generally regarded as completing the work, which\nBacon had begun, of transferring philosophy from the scholastic\nbondage of words to the fruitful discipline of facts. In the\nprocess of transmission and popular adaptation, however, its true\nsignificance has been lost sight of, and it has been forgotten that\nto its original exponent implicitly--explicitly to his more logical\ndisciple--though it did indeed distinguish effectively between things\nand the meaning of words, it was the analysis of the latter only,\nand not the understanding of things, that it left as the possible\nfunction of knowledge. It will be well, then, in what follows, first\nbriefly to restate the theory in its general form; then to show\nhow it conflicts with the actual knowledge which mankind supposes\nitself to have attained; and finally to exhibit at once the necessity\nof this conflict as a result of Locke\u2019s governing ideas, and the\nambiguities by which he disguised it from himself.\n\nWhat Locke understood by essence.\n\n80. The essence of a thing with Locke, in the only sense in which\nwe can know or intelligibly speak of it, is the meaning of its\nname. This, again, is an \u2018abstract or general idea,\u2019 which means\nthat it is an idea \u2018separated from the circumstances of time and\nplace, and any other ideas that may determine it to this or that\nparticular existence. By this way of abstraction it is made capable\nof representing more individuals than one; each of which, having\nin it a conformity to that abstract idea, is (as we call it) of\nthat sort.\u2019 (Book III. chap. iii. sec. 6.) That which is given in\nimmediate experience, as he proceeds to explain, is this or that\n\u2018particular existence,\u2019 Peter or James, Mary or Jane, such particular\nexistence being already a complex idea. [1] That it should be so is\nindeed in direct contradiction to his doctrine of the primariness of\nthe simple idea, but is necessary to his doctrine of abstraction.\nSome part of the complex idea (it is supposed)--less or more--we\nproceed to leave out. The minimum of subtraction would seem to\nbe that of the \u2018circumstances of time and place,\u2019 in which the\nparticular existence is given. This is the \u2018separation of ideas,\u2019\nfirst made, and alone suffices to constitute an \u2018abstract idea,\u2019 even\nthough, as is the case with the idea of the sun, there is only one\n\u2018particular substance\u2019 to agree with it. (Book III. chap. vi. sec.\n1.) In proportion as the particular substances compared are more\nvarious, the subtraction of ideas is larger, but, be it less or more,\nthe remainder is the abstract idea, to which a name--_e.g._ man--is\nannexed, and to which as a \u2018species\u2019 or \u2018standard\u2019 other particular\nexistences, on being \u2018found to agree with it,\u2019 may be referred, so\nas to be called by the same name. These ideas then, \u2018tied together\nby a name,\u2019 form the essence of each particular existence, to which\nthe same name is applied (Book III. chap. iii. secs. 12 and the\nfollowing.) Such essence, however, according to Locke, is \u2018nominal,\u2019\nnot \u2018real.\u2019 It is a complex--fuller or emptier--of ideas in us,\nwhich, though it is a \u2018uniting medium between a general name and\nparticular beings,\u2019 [2] in no way represents the qualities of the\nlatter. These, consisting in an \u2018internal constitution of insensible\nparts,\u2019 form the \u2018real essence\u2019 of the particular beings; an essence,\nhowever, of which we can know nothing. (Book III. chap. vi. sec. 21,\nand ix. sec. 12.)\n\n[1] Book III. chap, iii, sec. 7, at the end.\n\n[2] Book III chap. iii. sec. 13.\n\nOnly to nominal essences that general propositions relate, _i.e._\nonly to abstract ideas having no real existence.\n\n81. It is the formation of \u2018nominal essences\u2019 that renders general\npropositions possible. \u2018General certainty,\u2019 says Locke, \u2018is never to\nbe found but in our ideas. Whenever we go to seek it elsewhere in\nexperiment or observation without us, our knowledge goes not beyond\nparticulars. It is the contemplation of our own abstract ideas, that\nalone is able to afford us general knowledge.\u2019 (Book IV. chap. vi.\nsec. 16.) \u2018General knowledge,\u2019 he says again, \u2018lies only in our own\nthoughts.\u2019 [1] This use of \u2018our ideas\u2019 and \u2018our own thoughts\u2019 as\nequivalent phrases, each antithetical to \u2018real existence,\u2019 tells\nthe old tale of a deviation from \u2018the new way of ideas\u2019 into easier\npaths. According to this new way in its strictness, as we have\nsufficiently seen, there is nowhere for anything to be found but \u2018in\nour ideas.\u2019 It therefore in no way distinguishes general knowledge\nor certainty that it cannot be found elsewhere. Locke, however,\nhaving allowed himself in the supposition that simple ideas report a\nreal existence, other than themselves, but to which they are related\nas ectype to archetype, tacitly proceeds to convert them into real\nexistences, to which ideas in general, as mere thoughts of our own,\nmay be opposed. Along with this conversion, there supervenes upon\nthe original distinction between simple and complex ideas, which\nalone does duty in the Second Book of the Essay, another distinction,\nessential to Locke\u2019s doctrine of the \u2018reality\u2019 of knowledge--that\nbetween the idea, whether simple or complex, as originally given in\nsensation, and the same as retained or reproduced in the mind. It is\nonly in the former form that the idea, however simple, reports, and\nthus (with Locke) itself is, a real existence. Such real existence\nis a \u2018particular\u2019 existence, and our knowledge of it a \u2018particular\u2019\nknowledge. In other words, according to the only consistent doctrine\nthat we have been able to elicit from Locke, [2] \u2018it is a knowledge\nwhich consists in a consciousness, upon occasion of a present\nsensation--say, a sensation of redness--that some object is present\nhere and now causing the sensation; an object which, accordingly,\nmust be \u2018particular\u2019 or transitory as the sensation. The \u2018here and\nnow,\u2019 as in such a case they constitute the particularity of the\nobject of consciousness, so also render it a real existence. Separate\nthese (\u2018the circumstances of time and place\u2019 [3]) from it, and it\nat once loses its real existence and becomes an \u2018abstract idea,\u2019\none of \u2018our own thoughts,\u2019 of which as \u2018in the mind\u2019 agreement or\ndisagreement with some other abstract idea can be asserted in a\ngeneral proposition; _e.g._ \u2018red is not blue.\u2019 (Book IV. chap. vii.\nsec. 4.) [4]\n\n[1] Book IV. chap. vi. sec. 13, cf. Book IV. chap. iii. sec. 31.\n\n[2] See above, paragraph 56.\n\n[3] Book III. chap. iii. sec. 6.\n\n[4] In case there should be any doubt as to Locke\u2019s meaning in this\npassage, it may be well to compare Book IV. chap. ix. sec. 1. There\nhe distinctly opposes the consideration of ideas in the understanding\nto the knowledge of real existence. Here (Book IV. chap. vii. sec.\n4) he distinctly speaks of the proposition \u2018red is not blue\u2019 as\nexpressing a consideration of ideas in the understanding. It follows\nthat it is not a proposition as to real existence.\n\nAn abstract idea may be a simple one.\n\n82. It is between simple ideas, it will be noticed, that a relation\nis here asserted, and in this respect the proposition differs from\nsuch an one as may be formed when simple ideas have been compounded\ninto the nominal essence of a thing, and in which some one of these\nmay be asserted of the thing, being already included within the\nmeaning of its name; _e.g._ \u2018a rose has leaves.\u2019 But as expressing a\nrelation between ideas \u2018abstract\u2019 or \u2018in the mind,\u2019 in distinction\nfrom present sensations received from without, the two sorts of\nproposition, according to the doctrine of Locke\u2019s Fourth Book, stand\non the same footing.\u2019 [1] It is a nominal essence with which both\nalike are concerned, and on this depends the general certainty or\nself-evidence, by which they are distinguished from \u2018experiment\nor observation without us.\u2019 These can never \u2018reach with certainty\nfarther than the bare instance\u2019 (Book IV. chap. vi. sec. 7): _i.e._,\nthough the only channels by which we can reach real existence, they\ncan never tell more than the presence of this or that sensation as\ncaused by an unknown thing without, or the present disagreement of\nsuch present sensations with each other. As to the recurrence of such\nsensations, or any permanently real relation between them, they can\ntell us nothing. Nothing as to their recurrence, because, though in\neach case they show the presence of something causing the sensations,\nthey show nothing of the real essence upon which their recurrence\ndepends. [2] Nothing as to any permanently real relation between\nthem, because, although the disagreement between ideas of blue and\nred, and the agreement between one idea of red and another, _as in\nthe mind_, is self-evident, yet as thus in the mind they are not\n\u2018actual sensations\u2019 at all (Book IV. chap. xi. sec. 6), nor do they\nconvey that \u2018sensitive knowledge of particular existence,\u2019 which is\nthe only possible knowledge of it. (Book IV. chap. iii. sec. 21.)\nAs actual sensations and indices of reality, they do indeed differ\nin this or that \u2018bare instance,\u2019 but can convey no certainty that\nthe real thing or \u2018parcel of matter\u2019 (Book III. chap. iii. sec. 18),\nwhich now causes the sensation of (and thus _is_) red, may not at\nanother time cause the sensation of (and thus _be_) blue.\u2019 [3]\n\n[1] Already in Book II. (chap. xxxi. sec. 12), the simple idea, as\nabstract, is spoken of as a nominal essence.\n\n[2] Cf. Book IV. chap. vi. sec. 5. \u2018If we could certainly know (which\nis impossible) where a real essence, which we know not, is--_e.g._ in\nwhat parcels of matter the real essence of gold is; yet could we not\nbe sure, that this or that quality could with truth be affirmed of\ngold; since it is impossible for us to know that this or that quality\nor idea has a necessary connexion with a real essence, of which we\nhave no idea at all.\u2019\n\nSeveral passages, of course, can be adduced from Locke which are\ninconsistent with the statement in the text: _e.g._ Book IV. chap.\niv. sec. 12. \u2018To make knowledge real concerning substances, the ideas\nmust be taken from the real existence of things. Whatever simple\nideas have been found to coexist in any substance, these we may\nwith confidence join together again, and so make abstract ideas of\nsubstances. For whatever have once had an union in nature, may be\nunited again.\u2019 In all such passages, however, as will appear below,\nthe strict opposition between the real and the mental is lost sight\nof, the \u2018nature\u2019 or \u2018substance,\u2019 in which ideas \u2018have a union,\u2019 or\nare \u2018found to coexist,\u2019 being a system of relations which, according\nto Locke, it requires a mind to constitute, and thus itself a\n\u2018nominal essence.\u2019\n\n[3] Cf. Book IV. chap. iii. sec. 29; Book IV. chap. vi. sec. 14; Book\nIV. chap. xi. sec. 11.\n\nHow then is science of nature possible?\n\n83. We thus come upon the crucial antithesis between relations of\nideas and matters of fact, with the exclusion of general certainty as\nto the latter, which was to prove such a potent weapon of scepticism\nin the hands of Hume. Of its incompatibility with recognized science\nwe can have no stronger sign than the fact that, after more than a\ncentury has elapsed since Locke\u2019s premisses were pushed to their\nlegitimate conclusion, the received system of logic among us is\none which, while professing to accept Locke\u2019s doctrine of essence,\nand with it the antithesis in question, throughout assumes the\npossibility of general propositions as to matters of fact, and seeks\nin their methodical discovery and proof that science of nature which\nLocke already \u2018suspected\u2019 to be impossible. (Book IV. chap. xii. sec.\n10.)\n\nNo \u2018uniformities of phenomena\u2019 can be known.\n\n84. That, so far as any inference from past to future uniformities\nis necessary to the science of nature, his doctrine does more than\njustify such \u2018suspicion,\u2019 is plain enough. Does it, however, leave\nroom for so much as a knowledge of past uniformities of fact, in\nwhich the natural philosopher, accepting the doctrine, might probably\nseek refuge? At first sight, it might seem to do so. \u2018As, when our\nsenses are actually employed about any object, we do know that it\ndoes exist; so by our memory we may be assured that heretofore things\nthat affected our senses have existed--and thus we have knowledge\nof the past existence of several things, whereof our senses having\ninformed us, our memories still retain the ideas.\u2019 (Book IV. chap.\nxi. sec. 11.) Let us see, however, how this knowledge is restricted.\n\u2018Seeing water at this instant, it is an unquestionable truth to me\nthat water doth exist; and remembering that I saw it yesterday, it\nwill also be always true, and as long as my memory retains it, always\nan undoubted proposition to me, that water did exist the 18th of\nJuly, 1688; as it will also be equally true that a certain number of\nvery fine colours did exist, which at the same time I saw on a bubble\nof that water; but being now quite out of sight both of the water and\nbubbles too, it is no more certainly known to me that the water doth\nnow exist, than that the bubbles and colours therein do so; it being\nno more necessary that water should exist to-day because it existed\nyesterday, than that the colours or bubbles exist to-day because they\nexisted yesterday.\u2019 (_Ibid_.)\n\nLocke not aware of the full effect of his own doctrine ...\n\n85. The result is that though I may enumerate a multitude of past\nmatters-of-fact about water, I cannot gather them up in any general\nstatement about it as a real existence. So soon as I do so, I pass\nfrom water as a real existence to its \u2018nominal essence,\u2019 _i.e._,\nto the ideas retained in my mind and put together in a fictitious\nsubstance, to which I have annexed the name \u2018water.\u2019 If we proceed to\napply this doctrine to the supposed past matters-of-fact themselves,\nwe shall find these too attenuating themselves to nonentity.\nSubtract in every case from the \u2018particular existence\u2019 of which we\nhave \u2018sensitive knowledge\u2019 the qualification by ideas which, as\nretained in the mind, do not testify to a present real existence,\nand what remains? There is a certainty, according to Locke (Book IV.\nchap. xi. sec. 11), not, indeed, that water exists to-day because\nit existed yesterday--this is only \u2018probable\u2019--but that it has,\nas a past matter-of-fact, at this time and that \u2018continued long\nin existence,\u2019 because this has been \u2018observed;\u2019 which must mean\n(Book IV. chap. ii. secs. 1, 5, and 9), because there has been a\ncontinued \u2018actual sensation\u2019 of it. \u2018Water,\u2019 however, is a complex\nidea of a substance, and of the elements of this complex idea\nthose only which at any moment are given in \u2018actual sensation\u2019 may\nbe accounted to \u2018really exist.\u2019 First, then, must disappear from\nreality the \u2018something,\u2019 that unknown substratum of ideas, of which\nthe idea is emphatically \u2018abstract.\u2019 This gone, we naturally fall\nback upon a fact of co-existence between ideas, as being a reality,\nthough the \u2018thing\u2019 be a fiction. But if this co-existence is to be\nreal or to represent a reality, the ideas between which it obtains\nmust be \u2018actual sensations.\u2019 These, whatever they may be, are at\nleast opposed by Locke to ideas retained in the mind, which only\nform a nominal essence. But it is the association of such nominal\nessence, in the supposed observation of water, with the actual\nsensation that alone gives the latter a meaning. Set this aside as\nunreal, and the reality, which the sensation reveals, is at any\nrate one of which nothing can be said. It cannot be a relation\nbetween sensations, for such relation implies a consideration of\nthem by the mind, whereby, according to Locke, they must cease to\nbe \u2018real existences.\u2019 (Book II. chap. xxv. sec. 1.) It cannot even\nbe a single sensation _as continuously observed_, for every present\nmoment of such observation has at the next become a past, and thus\nthe sensation observed in it has lost its \u2018actuality,\u2019 and cannot,\n_as a \u2018real existence_,\u2019 qualify the sensation observed in the next.\nRestrict the \u2018real existence,\u2019 in short, as Locke does, to an \u2018actual\npresent sensation,\u2019 which can only be defined by opposition to an\nidea retained in the mind, and at every instant of its existence it\nhas passed into the mind and thus ceased really to exist. Reality is\nin perpetual process of disappearing into the unreality of thought.\nNo point can be fixed either in the flux of time or in the imaginary\nprocess from \u2018without\u2019 to \u2018within\u2019 the mind, on the one side of which\ncan be placed \u2018real existence,\u2019 on the other the \u2018mere idea.\u2019 It is\nonly because Locke unawares defines to himself the \u2018actual sensation\u2019\nas representative of a real essence, of which, however, according\nto him, as itself unknown, the presence is merely inferred from the\nsensation, that the \u2018actual sensation\u2019 itself is saved from the limbo\nof nominal essence, to which ideas, as abstract or in the mind, are\nconsigned. Only, again, so far as it is thus illogically saved, are\nwe entitled to that distinction between \u2018facts\u2019 and \u2018things of the\nmind,\u2019 which Locke once for all fixed for English philosophy.\n\n... which is to make the real an abstract residuum of consciousness.\n\n86. By this time we are familiar with the difficulties which this\nantithesis has in store for a philosophy which yet admits that it\nis only in the mind or in relation to consciousness--in one word,\nas \u2018ideas\u2019--that facts are to be found at all, while by the \u2018mind\u2019\nit understands an abstract generalization from the many minds which\nseverally are born and grow, sleep and wake, with each of us. The\nantithesis itself, like every other form in which the impulse after\ntrue knowledge finds expression, implies a distinction between\nthe seeming and the real; or between that which exists for the\nconsciousness of the individual and that which really exists. But\noutside itself consciousness cannot get. It is there that the real\nmust, at any rate, manifest itself, if it is to be found at all. Yet\nthe original antithesis between the mind and its unknown opposite\nstill prevails, and in consequence that alone which, though indeed\nin the mind, is yet given to it by no act of its own, is held to\nrepresent the real. This is the notion which dominates Locke. He\nstrips from the formed content of consciousness all that the mind\nseems to have done for itself, and the abstract residuum, that of\nwhich the individual cannot help being conscious at each moment of\nhis existence, is or \u2018reports\u2019 the real, in opposition to the mind\u2019s\ncreation. This is Feeling; or more strictly--since it exists, and\nwhatever does so must exist as one in a number (Book II. chap. vii.\nsec. 7)--it is the multitude of single feelings, \u2018each perishing the\nmoment it begins\u2019 (Book II. chap, xxvii. sec. 2), from which all the\ndefiniteness that comes of composition and relation must be supposed\nabsent. Thus, in trying to get at what shall be the mere fact in\ndetachment from mental accretions, Locke comes to what is still\nconsciousness, but the merely indefinite in consciousness. He seeks\nthe real and finds the void. Of the real as outside consciousness\nnothing can be said; and of that again within consciousness, which is\nsupposed to represent it, nothing can be said.\n\nGround of distinction between actual sensation and ideas in the mind\nis itself a thing of the mind.\n\n87. We have already seen how Locke, in his doctrine of secondary\nqualities of substances, practically gets over this difficulty; how\nhe first projects out of the simple ideas, under relations which it\nrequires a mind to constitute, a cognisable system of things, and\nthen gives content and definiteness to the simple ideas in us by\ntreating them as manifestations of this system of things. In the\ndoctrine of propositions, the proper correlative to the reduction\nof the real to the present simple idea, as that of which we cannot\nget rid, would be the reduction of the \u2018real proposition\u2019 to the\nmere \u2018it is now felt.\u2019 If the matter-of-fact is to be that in\nconsciousness which is independent of the \u2018work of the mind\u2019 in\ncomparing and compounding, this is the only possible expression\nfor it. It states the only possible \u2018real essence,\u2019 which yet is\nan essence of nothing, for any reference of it to a thing, if the\nthing is outside consciousness, is an impossibility; and if it is\nwithin consciousness, implies an \u2018invention of the mind\u2019 both in\nthe creation of a thing, \u2018always the same with itself,\u2019 out of\nperishing feelings, and in the reference of the feelings to such a\nthing. Thus carried out, the antithesis between \u2018fact\u2019 and \u2018creation\nof the mind\u2019 becomes self-destructive, for, one feeling being as\nreal as another, it leaves no room for that distinction between the\nreal and fantastic, to the uncritical sense of which it owes its\nbirth. To avoid this fusion of dream-land and the waking world,\nLocke avails himself of the distinction between the idea (_i.e._\nfeeling) as in the mind, which is not convertible with reality,\nand the idea as somewhere else, no one can say where--\u2018the actual\nsensation\u2019--which is so convertible. The distinction, however, must\neither consist in degrees of liveliness, in which case there must be\na corresponding infinity of degrees of reality or unreality, or else\nmust presuppose a real existence from which the feeling, if \u2018actual\nsensation,\u2019 _is_--if merely \u2018in the mind\u2019 _is not_--derived. Such a\nreal existence either is an object of consciousness, or is not. If it\nis not, no distinction between one kind of feeling and another can\nfor consciousness be derived from it. If it is, then, granted the\ndistinction between given feelings and creations of the mind, it must\nfall to the latter, and a \u2018thing of the mind\u2019 turns out to be the\nground upon which \u2018fact\u2019 is opposed to \u2018things of the mind.\u2019\n\nTwo meanings of real essence.\n\n88. It remains to exhibit briefly the disguises under which these\ninherent difficulties of his theory of essence appear in Locke.\nThroughout, instead of treating \u2018essence\u2019 altogether as a fiction\nof the mind--as it must be if feelings in simplicity and singleness\nare alone the real--he treats indeed as a merely \u2018nominal essence\u2019\nevery possible combination of ideas of which we can speak, but still\nsupposes another essence which is \u2018real.\u2019 But a real essence of\nwhat? Clearly, according to his statements, of the same \u2018thing\u2019 of\nwhich the combination of ideas in the mind is the nominal essence.\nIndeed, there is no meaning in the antithesis unless the \u2018something,\u2019\nof which the latter essence is so nominally, is that of which it is\nnot so really. So says Locke, \u2018the nominal essence of gold is that\ncomplex idea the word gold stands for; let it be, for instance, a\nbody yellow, of a certain weight, malleable, fusible, and fixed. But\nthe real essence is the constitution of the insensible parts _of\nthat body_, on which those qualities and all the other properties of\ngold depend.\u2019 (Book III. chap. vi. sec. 2.) Here the notion clearly\nis that of one and the same thing, of which we can only say that it\nis a \u2018body,\u2019 a certain complex of ideas--yellowness, fusibility,\n&c.--is the nominal, a certain constitution of insensible parts the\nreal, essence. It is on the real essence, moreover, that the ideas\nwhich constitute the nominal depend. Yet while they are known, the\nreal essence (as appears from the context) is wholly unknown. In this\ncase, it would seem, the cause is not known from its effects.\n\nAccording to one, it is a collection of ideas as qualities of a thing:\n\n89. There are lurking here two opposite views of the relation between\nthe nominal essence and the real thing. According to one view, which\nprevails in the later chapters of the Second Book and in certain\npassages of the third, the relation between them is that with which\nwe have already become familiar in the doctrine of substance--that,\nnamely, between ideas as in us and the same as in the thing. (Book\nII, chap. xxiii. secs. 9 and 10.) No distinction is made between the\n\u2018idea in the mind\u2019 and the \u2018actual sensation.\u2019 The ideas in the mind\nare also in the thing, and thus are called its qualities, though for\nthe most part they are so only secondarily, _i.e._ as effects of\nother qualities, which, as copied directly in our ideas, are called\nprimary, and relatively to these effects are called powers. These\npowers have yet innumerable effects to produce in us which they have\nnot yet produced. (Book II. chap. xxxi. sec. 10.) Those which have\nbeen so far produced, being gathered up in a complex idea to which\na name is annexed, form the \u2018nominal essence\u2019 of the thing. Some of\nthem are of primary qualities, more are of secondary. The originals\nof the former, the powers to produce the latter, together with powers\nto produce an indefinite multitude more, will constitute the \u2018real\nessence,\u2019 which is thus \u2018a standard made by nature,\u2019 to which the\nnominal essence is opposed merely as the inadequate to the adequate.\nThe ideas, that is to say, which are indicated by the name of a\nthing, have been really \u2018found in it\u2019 or \u2018produced by it,\u2019 but are\nonly a part of those that remain to be found in it or produced by it.\nIt is in this sense that Locke opposes the adequacy between nominal\nand real essence in the case of mixed modes to their perpetual\ninadequacy in the case of ideas of substances. The combination in\nthe one case is artificially made, in the other is found and being\nperpetually enlarged. This he illustrates by imagining the processes\nwhich led Adam severally to the idea of the mixed mode \u2018jealousy\u2019 and\nthat of the substance \u2018gold.\u2019 In the former process Adam \u2018put ideas\ntogether only by his own imagination, not taken from the existence\nof anything ... the standard there was of his own making.\u2019 In the\nlatter, \u2018he has a standard made by nature; and therefore being to\nrepresent that to himself by the idea he has of it, even when it is\nabsent, he puts no simple idea into his complex one, but what he has\nthe perception of from the thing itself. He takes care that his idea\nbe conformable to this archetype.\u2019 (Book III. chap. vi. secs. 46,\n47.) \u2018It is plain,\u2019 however, \u2018that the idea made after this fashion\nby this archetype will be always inadequate.\u2019\n\n... about real essence in this sense there may be general knowledge.\n\n90. The nominal essence of a thing, then, according to this view,\nbeing no other than the \u2018complex idea of a substance,\u2019 is a\ncopy of reality, just as the simple idea is. It is a picture or\nrepresentation in the mind of a thing that does exist by ideas of\nthose qualities that are discoverable in it.\u2019 (Book II. chap. xxxi.\nsecs. 6, 8.) It only differs from the simple idea (which is itself,\nas abstract, a nominal essence) [1] in respect of reality, because\nthe latter is a copy or effect produced singly and involuntarily,\nwhereas we may put ideas together, as if in a thing, which have\nnever been so presented together, and, on the other hand, never can\nput together all that exist together. (Book II. chap. xxx. sec. 5,\nand xxxi. 10.) So far as Locke maintains this view, the difficulty\nabout general propositions concerning real existence need not arise.\nA statement which affirmed of gold one of the qualities included\nin the complex idea of that substance, would not express merely an\nanalysis of an idea in the mind, but would represent a relation\nof qualities in the existing thing from which the idea \u2018has been\ntaken.\u2019 These qualities, as in the thing, doubtless would not be,\nas in us, feelings (or, as Locke should rather have said in more\nrecent phraseology, possibilities of feeling), but powers to produce\nfeeling, nor could any relation between these, as in the thing,\nbe affirmed but such as had produced its copy or effect in actual\nexperience. No coexistence of qualities could be truly affirmed,\nwhich had not been found; but, once found--being a coexistence of\nqualities and not simply a momentary coincidence of feelings--it\ncould be affirmed as permanent in a general proposition. That a\nrelation can be stated universally between ideas collected in\nthe mind, no one denies, and if such collection \u2018is taken from a\ncombination of simple ideas _existing together constantly in things_\u2019\n(Book II. chap, xxxii. sec. 18), the statement will hold equally of\nsuch existence. Thus Locke contrasts mixed modes, which, for the\nmost part, \u2018being actions which perish in the birth, are not capable\nof a lasting duration,\u2019 with \u2018substances, which are the actors; and\nwherein the simple ideas that make up the complex ideas designed by\nthe name have a lasting union.\u2019 (Book III. chap. vi. sec. 42.)\n\n[1] Book II. chap. xxxi. sec. 12.\n\nBut such real essence a creature of thought.\n\n91. In such a doctrine Locke, starting whence he did, could not\nremain at rest. We need not here repeat what has been said of it\nabove in the consideration of his doctrine of substance. Taken\nstrictly, it implies that \u2018real existence\u2019 consists in a permanent\nrelation of ideas, said to be of secondary qualities, to each other\nin dependence on other ideas, said to be of primary qualities. In\nother words, in order to constitute reality, it takes ideas out of\nthat particularity in time and place, which is yet pronounced the\ncondition of reality, to give them an \u2018abstract generality\u2019 which is\nfictitious, and then treats them as constituents of a system of which\nthe \u2018invented\u2019 relations of cause and effect and of identity are\nthe framework. In short, it brings reality wholly within the region\nof thought, distinguishing it from the system of complex ideas or\nnominal essences which constitute our knowledge, not as the unknown\nopposite of all possible thought, but only as the complete from the\nincomplete. To one who logically carried out this view, the ground\nof distinction between fact and fancy would have to be found in the\nrelation between thought as \u2018objective,\u2019 or in the world, and thought\nas so far communicated to us. Here, however, it could scarcely be\nfound by Locke, with whom \u2018thought\u2019 meant simply a faculty of the\n\u2018thinking thing,\u2019 called a \u2018soul,\u2019 which might ride in a coach with\nhim from Oxford to London. (Book II. chap, xxiii. sec. 20.) Was the\ndistinction then to disappear altogether?\n\nHence another view of real essence as unknown qualities of unknown\nbody.\n\n92. It is saved, though at the cost of abandoning the \u2018new way of\nideas,\u2019 as it had been followed in the Second Book, by the transfer\nof real existence from the thing in which ideas are found, and whose\nqualities the complex of ideas in us, though inadequate, represents,\nto something called \u2018body,\u2019 necessarily unknown, because no ideas\nin us are in any way representative of it. To such an unknown body\nunknown qualities are supposed to belong under the designation \u2018real\nessence.\u2019 The subject of the nominal essence, just because its\nqualities, being matter of knowledge, are ideas in our minds, is a\nwholly different and a fictitious thing.\n\nHow Locke mixes up these two meanings in ambiguity about body.\n\n93. This change of ground is of course not recognized by Locke\nhimself. It is the perpetual crossing of the inconsistent doctrines\nthat renders his \u2018immortal Third Book\u2019 a web of contradictions.\nAs was said above, he constantly speaks as if the subject of the\nreal essence were the same with that of the nominal, and never\nexplicitly allows it to be different. The equivocation under which\nthe difference is disguised lies in the use of the term \u2018body.\u2019\nA \u2018particular body\u2019 is the subject both of the nominal and real\nessence \u2018gold\u2019 But \u2018body,\u2019 as that in which \u2018ideas are found,\u2019 and\nin which they permanently coexist according to a natural law, is one\nthing; \u2018body,\u2019 as the abstraction of the unknown, is quite another.\nIt is body in the former sense that is the real thing when nominal\nessence (the complex of ideas in us) is treated as representative,\nthough inadequately so, of the real thing; it is body in the latter\nsense that is the real thing when this is treated as wholly outside\npossible consciousness, and its essence as wholly unrepresented by\npossible ideas. By a jumble of the two meanings Locke obtains an\namphibious entity which is at once independent of relation to ideas,\nas is body in the latter sense, and a source of ideas representative\nof it, as is body in the former sense--which thus carries with it\nthat opposition to the mental which is supposed necessary to the\nreal, while yet it seems to manifest itself in ideas. Meanwhile a\nthird conception of the real keeps thrusting itself upon the other\ntwo--the view, namely, that body in both senses is a fiction of\nthought, and that the mere present feeling is alone the real.\n\nBody as \u2018parcel of matter\u2019 without essence.\n\n94. Where Locke is insisting on the opposition between the real\nessence and any essence that can be known, the former is generally\nascribed either to a \u2018particular being\u2019 or to a \u2018parcel of matter.\u2019\nThe passage which brings the opposition into the strongest relief is\nperhaps the following:--\u2018I would ask any one, what is sufficient to\nmake an essential difference in nature between any two particular\nbeings, without any regard had to some abstract idea, which is looked\nupon as the essence and standard of a species? All such patterns\nand standards being quite laid aside, particular beings, considered\nbarely in themselves, will be found to have all their qualities\nequally essential; and everything, in each individual, will be\nessential to it, or, which is more, nothing at all. For though it\nmay be reasonable to ask whether obeying the magnet be essential\nto iron; yet I think it is very improper and insignificant to ask\nwhether it be essential to the particular parcel of matter I cut my\npen with, without considering it under the name _iron_, or as being\nof a certain species.\u2019 (Book III. chap. vi. sec. 5.) [1] Here, it\nwill be seen, the exclusion of the abstract idea from reality carries\nwith it the exclusion of that \u2018standard made by nature,\u2019 which\naccording to the passages already quoted, is the \u2018thing itself from\nwhich the abstract idea is taken, and from which, if correctly taken,\nit derives reality. This exclusion, again, means nothing else than\nthe disappearance from \u2018nature\u2019 (which with Locke is interchangeable\nwith \u2018reality\u2019) of all essential difference. There remain, however,\nas the \u2018real,\u2019 \u2018particular beings,\u2019 or \u2018individuals,\u2019 or \u2018parcels of\nmatter.\u2019 In each of these, \u2018considered barely in itself, everything\nwill be essential to it, or, which is more, nothing at all.\u2019\n\n[1] To the same purpose is a passage in Book III. chap. x. sec. 19,\ntowards the end.\n\nIn this sense body is the mere individuum.\n\n95. We have already seen, [1] that if by a \u2018particular being\u2019 is\nmeant the mere _individuum_, as it would be upon abstraction of all\nrelations which according to Locke are fictitious, and constitute\na community or generality, it certainly can have no essential\nqualities, since it has no qualities at all. It is a something which\nequals nothing. The notion of this bare _individuum_ being the\nreal is the \u2018protoplasm\u2019 of Locke\u2019s philosophy to which, though he\nnever quite recognized it himself, after the removal of a certain\nnumber of accretions we may always penetrate. It is so because his\nunacknowledged method of finding the real consisted in abstracting\nfrom the formed content of consciousness till he came to that which\ncould not be got rid of. This is the momentarily present relation\nof subject and object, which, considered on the side of the object,\ngives the mere atom, and on the side of the subject, the mere \u2018it is\nfelt.\u2019 Even in this ultimate abstraction the \u2018fiction of thought\u2019\nstill survives, for the atom is determined to its mere individuality\nby relation to other individuals, and the feeling is determined to\nthe present moment or \u2018the now\u2019 by relation to other \u2018nows.\u2019\n\n[1] See above, paragraph 45.\n\nBody as qualified by circumstances of time and place.\n\n96. To this ultimate abstraction, however, Locke, though constantly\non the road to it, never quite penetrates. He is farthest from\nit--indeed, as far from it as possible--where he is most acceptable\nto common sense, as in his ordinary doctrine of abstraction, where\nthe real, from which the process of abstraction is supposed to begin,\nis already the individual in the fullness of its qualities, James\nand John, this man or this gold. He is nearest to it when the only\nqualification of the \u2018particular being,\u2019 which has to be removed by\nthought in order to its losing its reality and becoming an abstract\nidea, is supposed to consist in \u2018circumstances of time and place.\u2019\n\nSuch body Locke held to be subject of \u2018primary qualities\u2019: but are\nthese compatible with particularity in time?\n\n97. It is of these circumstances, as the constituents of the real,\nthat he is thinking in the passage last quoted. As qualified by\n\u2018circumstances of place\u2019 the real is a parcel of matter, and under\nthis designation Locke thought of it as a subject of \u2018primary\nqualities of body.\u2019 [1] These, indeed, as he enumerates them, may be\nshown to imply relations going far beyond that of simple distinctness\nbetween atoms, and thus to involve much more of the creative action\nof thought; but we need be the less concerned for this usurpation on\nthe part of the particular being, since that which he illegitimately\nconveys to it as derived from \u2018circumstances of place,\u2019 he virtually\ntakes away from it again by limitation in time. The \u2018particular\nbeing\u2019 has indeed on the one hand a real essence, consisting of\ncertain primary qualities, but on the other it has no continued\nidentity. It is only real as present to feeling at this or that\ntime. The particular being of one moment is not the particular being\nof the next. Thus the primary qualities which are a real essence,\n_i.e._ an essence of a particular being, at one moment, are not its\nreal essence at the next, because, while they as represented in the\nmind remain the same, the \u2018it,\u2019 the particular being is different.\nAn _immutable_ essence for that very reason cannot be real. The\nimmutability can only lie in a relation between a certain abstract\n(_i.e._ unreal) idea and a certain sound. (Book III. chap. iii. sec.\n19.) \u2018The real constitution of things,\u2019 on the other hand, \u2018begin\nand perish with them. All things that exist are liable to change.\u2019\n(_Ibid_.) Locke, it is true (as is implied in the term _change_ [2])\nnever quite drops the notion of there being a real identity in some\nunknown background, but this makes no difference in the bearing of\nhis doctrine upon the possibility of \u2018real\u2019 knowledge. It only means\nthat for an indefinite particularity of \u2018beings\u2019 there is substituted\none \u2018being\u2019 under an indefinite peculiarity of forms. Though the\nreality of the thing _in itself_ be immutable, yet its reality _for\nus_ is in perpetual flux. \u2018In itself\u2019 it is a substance without\nan essence, a \u2018something we know not what\u2019 without any ideas to\n\u2018support;\u2019 a \u2018parcel of matter,\u2019 indeed, but one in which no quality\nis really essential, because its real essence, consisting in its\nmomentary presentation to sense, changes with the moments. [3]\n\n[1] According to Locke\u2019s ordinary usage of the terms, no distinction\nappears between \u2018matter\u2019 and \u2018body.\u2019 In Book III. chap. X. sec. 15,\nhowever, he distinguishes matter from body as the less determinate\nconception from the more. The one implies solidity merely, the\nother extension and figure also, so that we may talk of the \u2018matter\nof bodies,\u2019 but not of the \u2018body of matters.\u2019 But since solidity,\naccording to Locke\u2019s definition, involves the other \u2018primary\nqualities,\u2019 this distinction does not avail him much.\n\n[2] See above, paragraph 69.\n\n[3] Cf. Book III. chap. vi. sec. 4: \u2018Take but away the abstract\nideas by which we sort individuals and rank them under common names,\nand then the thought of anything essential to any of them instantly\nvanishes,\u2019 &c.\n\nHow Locke avoids this question.\n\n98. We have previously noticed [1] Locke\u2019s pregnant remark, that\n\u2018things whose existence is in succession\u2019 do not admit of identity.\n(Book II. chap, xxvii. sec. 2.) So far, then, as the \u2018real,\u2019 in\ndistinction from the \u2018abstract,\u2019 is constituted by particularity in\ntime, or has its existence in succession, it excludes the relation\nof identity. \u2018It perishes in every moment that it begins.\u2019 Had Locke\nbeen master of this notion, instead of being irregularly mastered\nby it, he might have anticipated all that Hume had to say. As it\nis, even in passages such as those to which reference has just been\nmade, where he follows its lead the farthest, he is still pulled up\nby inconsistent conceptions with which common sense, acting through\ncommon language, restrains the most adventurous philosophy. Thus,\neven from his illustration of the liability of all existence to\nchange--\u2018that which was grass to-day is to-morrow the flesh of a\nsheep, and within a few days after will become part of a man\u2019 [2]--we\nfind that, just as he does not pursue the individualization of the\nreal in space so far but that it still remains \u2018a constitution\nof parts,\u2019 so he does not pursue it in time so far but that a\ncoexistence of real elements over a certain duration is possible. To\na more thorough analysis, indeed, there is no alternative between\nfinding reality in relations of thought, which, because relations of\nthought, are not in time and therefore are immutable, and submitting\nit to such subdivision of time as excludes all real coexistence\nbecause what is real, as present, at one moment is unreal, as past,\nat the next. This alternative could not present itself in its\nclearness to Locke, because, according to his method of interrogating\nconsciousness, he inevitably found in its supposed beginning, which\nhe identified with the real, those products of thought which he\nopposed to the real, and thus read into the simple feeling of the\nmoment that which, if it were the simple feeling of the moment, it\ncould not contain. Thus throughout the Second Book of the Essay the\nsimple idea is supposed to represent either as copy or as effect a\npermanent reality, whether body or mind: and in the later books, even\nwhere the _representation_ of such reality in knowledge comes in\nquestion, its existence as constituted by \u2018primary qualities of body\u2019\nis throughout assumed, though general propositions with regard to it\nare declared impossible. It is a feeling referred to body, or, in the\nlanguage of subsequent psychology, a feeling of the _outward_ sense,\n[3] that Locke means by an \u2018actual present sensation,\u2019 and it is\nproperly in virtue of this reference that such sensation is supposed\nto be, or to report, the real.\n\n[1] See above, paragraph 75.\n\n[2] Book III. chap. iii. sec. 10.\n\n[3] For the germs of the distinction between outer and inner sense,\nsee Locke\u2019s Essay, Book II. chap. i. sec. 14: \u2018This source of ideas\n(the perception of the operations of the mind) every man has wholly\nin himself; and though it be not sense, as having nothing to do with\nexternal objects, yet it is very like it, and might properly enough\nbe called internal sense.\u2019 For the notion of outer sense cf. Book\nII. chap. ix. sec. 6, where he is distinguishing the ideas of hunger\nand warmth, which he supposes children to receive in the womb from\nthe \u2018innate principles which some contend for.\u2019 \u2018These (the ideas\nof hunger and warmth) being the effects of sensation, are only from\nsome affections of the body which happen to them there, and so depend\non something exterior to the mind, not otherwise differing in their\nmanner of production from other ideas derived from sense, but only in\nthe precedency of time.\u2019\n\nBody and its qualities supposed to be outside consciousness.\n\n99. According to the doctrine of primary qualities, as originally\nstated, the antithesis lies between body as it is in itself and\nbody as it is for us, not between body as it is for us in \u2018actual\nsensation,\u2019 and body as it is for us according to \u2018ideas in the\nmind.\u2019 The primary qualities \u2018are in bodies whether we perceive them\nor no.\u2019 (Book II. chap. viii. sec. 23.) As he puts it elsewhere (Book\nII. chap. xxxi. sec. 2), it is just because \u2018solidity and extension\nand the termination of it, figure, with motion and rest, whereof we\nhave the ideas, would be really in the world as they are whether\nthere were any sensible being to perceive them or no,\u2019 that they are\nto be looked on as the _real_ modifications of matter. A change in\nthem, unlike one in the secondary qualities, or such as is relative\nto sense, is a _real_ alteration _in body_. \u2018Pound an almond, and the\nclear white colour will be altered into a dirty one, and the sweet\ntaste into an oily one. What alteration can the beating of the pestle\nmake in any body, but an alteration of the texture of it?\u2019 (Book II.\nchap. viii. sec. 20.) It is implied then in the notion of the real as\nbody that it should be outside consciousness. It is that which seems\nto remain when everything belonging to consciousness has been thought\naway. Yet it is brought within consciousness again by the supposition\nthat it has qualities which copy themselves in our ideas and are \u2018the\nexciting causes of all our various sensations from bodies.\u2019 (Book\nII. chap. xxxi. sec. 3.) Again, however, the antithesis between the\nreal and consciousness prevails, and the qualities of matter or body\nhaving been brought within the latter, are opposed to a \u2018substance\nof body\u2019--otherwise spoken of as \u2018the nature, cause, or manner of\nproducing the ideas of primary qualities\u2019--which remains outside it,\nunknown and unknowable. (Book II. chap, xxiii. sec. 30, &c.)\n\nHow can primary qualities be outside consciousness, and yet knowable?\n\n100. The doctrine of primary qualities was naturally the one upon\nwhich the criticism of Berkeley and Hume first fastened, as the\nmost obvious aberration from the \u2018new way of ideas.\u2019 That the very\nnotion of the senses as \u2018reporting\u2019 anything, under secondary no\nless than under primary qualities, implies the presence of \u2018fictions\nof thought\u2019 in the primitive consciousness, may become clear upon\nanalysis; but it lies on the surface and is avowed by Locke himself\n(Book II. chap. viii. secs. 2, 7), that the conception of primary\nqualities is only possible upon distinction being made between ideas\nas in our minds, and the \u2018nature of things existing without us,\u2019\nwhich cannot be given in the simple feeling itself. This admitted,\nthe distinction might either be traced to the presence within\nintelligent consciousness of another factor than simple ideas, or\nbe accounted for as a gradual \u2018invention of the mind.\u2019 In neither\nway, however, could Locke regard it and yet retain his distinction\nbetween fact and fancy, as resting upon that between the nature of\nthings and the mind of man. The way of escape lay in a figure of\nspeech, the figure of the wax or the mirror. \u2018The ideas of primary\nqualities are resemblances of them.\u2019 (Book II. chap, viii. sec. 15.)\nThese qualities then may be treated, according to occasion, either\nas primitive data of consciousness, or as the essence of that which\nis the unknown opposite of consciousness--in the latter way when the\nantithesis between nature and mind is in view, in the former when\nnature has yet to be represented as knowable.\n\nLocke answers that they copy themselves in ideas--Berkeley\u2019s\nrejoinder. Locke gets out of the difficulty by his doctrine of\nsolidity.\n\n101. How, asked Berkeley, can an idea be like anything that is not\nan idea? Put the question in its proper strength--How can an idea be\nlike that of which the sole and simple determination is just that it\nis not an idea (and such with Locke is body \u2018in itself\u2019 or as the\nreal)--and it is clearly unanswerable. The process by which Locke\nwas prevented from putting it to himself is not difficult to trace.\n\u2018Body\u2019 and \u2018the solid\u2019 are with him virtually convertible terms.\nEach indifferently holds the place of the substance, of which the\nprimary qualities are so many determinations. [1] It is true that\nwhere solidity has to be defined, it is defined as an attribute of\nbody, but conversely body itself is treated as a \u2018texture of solid\nparts,\u2019 _i.e._ as a mode of the solid. Body, in short, so soon as\nthought of, resolves itself into a relation of bodies, and the\nsolid into a relation of solids, but Locke, by a shuffle of the two\nterms--representing body as a relation between solids and the solid\nas a relation between bodies--gains the appearance of explaining\neach in turn by relation to a simpler idea. Body, as the unknown,\nis revealed to us by the idea of solidity, which sense conveys to\nus; while solidity is explained by reference to the idea of body.\nThe idea of solidity, we are told, is a simple idea which comes\ninto the mind solely by the sense of touch. (Book II. chap. iii.\nsec. 1.) But no sooner has he thus identified it with an immediate\nfeeling than, in disregard of his own doctrine, that \u2018an idea which\nhas no composition\u2019 is undefinable (see Book III. chap. iv. sec.\n7.), he converts it into a theory of the cause of that feeling. \u2018It\narises from the resistance which we find in body to the entrance of\nany other body into the place it possesses till it has left it;\u2019\nand he at once proceeds to treat it as the consciousness of such\nresistance. \u2018Whether we move or rest, in what posture soever we are,\nwe always feel something under us that supports us, and hinders our\nfarther sinking downwards: and the bodies which we daily handle make\nus perceive that whilst they remain between them, they do by an\ninsurmountable force hinder the approach of the parts of our hands\nthat press them. That which then hinders the approach of two bodies,\nwhen they are moving one towards another, I call solidity.\u2019 [2]\n\n[1] See Book II. chap. viii. sec. 23: The primary \u2018qualities that\nare in bodies, are the bulk, figure, number, situation, and motion\nor rest, _of their solid parts_.\u2019 Cf. Book II. chap. xiii. sec. 11:\n\u2018Solidity is so inseparable an idea from body, that upon that depends\nits filling of space, its contact, impulse, and communication of\nmotion upon impulse.\u2019\n\n[2] Book II. chap. iv. sec 7.\n\nIn which he equivocates between body as unknown opposite of mind and\nbody as a \u2018nominal essence\u2019.\n\n102. Now \u2018body\u2019 in this theory is by no means outside consciousness.\nIt is emphatically \u2018in the mind,\u2019 a \u2018nominal essence,\u2019 determined\nby the relation which the theory assigns to it, and which, like\nevery relation according to Locke, is a \u2018thing of the mind.\u2019 This\nrelation is that of outwardness to other bodies, and among these to\nthe sensitive body through which we receive \u2018ideas of sensation\u2019--a\nbody which, on its side, as determined by the relation, has its\nessence from the mind. It is, then, not as the unknown opposite\nof the mind, but as determined by an intelligible relation which\nthe mind constitutes, and of which the members are each \u2018nominal\nessences,\u2019 that body is outward to the sensitive subject. But to\nLocke, substituting for body as a nominal essence body as the unknown\nthing in itself, and identifying the sensitive subject with the\nmind, outwardness in the above sense--an outwardness constituted by\nthe mind--becomes outwardness to the mind of an unknown opposite of\nthe mind. Solidity, then, and the properties which its definition\ninvolves (and it involves all the \u2018primary qualities\u2019), become\nsomething wholly alien to the mind, which \u2018would exist without any\nsensible being to perceive them.\u2019 As such, they do duty as a real\nessence, when the opposition of this to everything in the mind has\nto be asserted. Yet must they be in some sort ideas, for of these\nalone (as Locke fully admits) can we think and speak; and if ideas,\nin the mind. How is this contradiction to be overcome? By the notion\nthat though not in or of the mind, they yet copy themselves upon it\nin virtue of an impulse in body, correlative to that resistance of\nwhich touch conveys the idea. (Book II. chap. viii. sec. 11). [1]\nThis explanation, however, is derived from the equivocation between\nthe two meanings of mind and body respectively. The problem to be\nexplained is the relation between the mind and that which is only\nqualified as the negation of mind; and the explanation is found in\na relation, only existing for the mind, between a sensitive and a\nnon-sensitive body.\n\n[1] Cf. also the passage from Book II. chap. xiii. sec. 11, quoted\nabove, paragraph 101, note [1].\n\nRationale of these contradictions.\n\n103. The case then stands as follows. All that Locke says of body\nas the real thing-in-itself, and of its qualities as the essence of\nsuch thing, comes according to his own showing of an action of the\nmind which he reckons the source of fictions. \u2018Body in itself\u2019 is a\nsubstratum of ideas which the mind \u2018accustoms itself to suppose.\u2019\nIt perpetually recedes, as what was at first a substance becomes in\nturn a complex of qualities for which a more remote substratum has\nto be supposed--a \u2018substance of body,\u2019 a productive cause of matter.\nBut the substance, however remote, is determined by the qualities to\nwhich it is correlative, as the cause by its effects; and every one\nof these--whether the most primary, solidity, or those which \u2018the\nmind finds inseparable from every particle of matter,\u2019 _i.e._ from\nthe \u2018solid parts of a body,\u2019 [1]--as defined by Locke, is a relation\nsuch as the mind, \u2018bringing one thing to and setting it by another\u2019\n(Book II. chap. xxv. sec. 1), can alone constitute. To Locke,\nhowever, overcome by the necessity of intelligence, as gradually\ndeveloping itself in each of us, to regard the intelligible world as\nthere before it is known, the real must be something which would be\nwhat it is if thought were not. Strictly taken, this must mean that\nit is that of which nothing can be said, and some expression must be\nfound by means of which it may do double duty as at once apart from\nconsciousness and in it. This is done by converting \u2018the primary\nqualities of body, though obviously complex ideas of relation, into\nsimple feelings of touch,\u2019 [2] and supposing the subject of this\nsensation to be related to its object as wax to the seal. If we\nsuppose this relation, again, which is really within the mind and\nconstituted by it, to be one between the mind itself, as passive,\nand the real, we obtain a \u2018real\u2019 which exists apart from the mind,\nyet copies itself upon it. The mind, then, so far as it takes such a\ncopy, becomes an \u2018outer sense,\u2019 as to which it may be conveniently\nforgotten that it is a mode of mind at all. Thus every modification\nof it, as an \u2018actual present sensation,\u2019 comes to be opposed to every\nidea of memory or imagination, as that which is not of the mind to\nthat which is; though there is no assignable difference between one\nand the other, except an indefinite one in degree of vivacity, that\nis not derived from the action of the mind in referring the one to an\nobject, constituted by itself, to which it does not refer the other.\n\n[1] Cf Book II. chap. viii. sec. 9. The primary qualities of body are\n\u2018such as sense constantly finds in every particle of matter, which\nhas bulk enough to be perceived, and the mind finds inseparable from\nevery particle of matter, though less than to make itself singly be\nperceived by our senses.\u2019\n\n[2] I write advisedly \u2018touch\u2019 only, not \u2018sight and touch,\u2019 because,\nthough Locke (Book II. chap, v.) speaks of the ideas of extension,\nfigure, motion, and rest of bodies, as received both by sight and\ntouch, these are all involved in the previous definition of solidity,\nof which the idea is ascribed to touch only.\n\nWhat knowledge can feeling, even as referred to a \u2018solid\u2019 body,\nconvey?\n\n104. Let us now consider whether by this reference to body, feeling\nbecomes any the more a source of general knowledge concerning\nmatters of fact. As we have seen, if we identify the real with\nfeeling simply, its distinction from \u2018bare vision\u2019 disappears. This\ndifficulty it is sought to overcome by distinguishing feeling as\nmerely in the mind from actually present sensation. But on reflection\nwe find that sensation after all is feeling, and that one feeling\nis as much present as another, though present only to become at\nthe next moment past, and thus, if it is the presence that is the\ncondition of reality, unreal. The distinction then must lie in the\n_actuality_ of the sensation. But does not this actuality mean simply\nderivation from the real, _i.e._ derivation from the idea which has\nto be derived from it? If, in the spirit of Locke, we answer, \u2018No,\nit means that the feeling belongs to the outer sense\u2019; the rejoinder\nwill be that this means either that it is a feeling of touch--and\nwhat should give the feeling of touch this singular privilege over\nother feelings of not being in the mind while they are in it?--or\nthat it is a feeling referred to body, which still implies the\npresupposition of the real, only under the special relations of\nresistance and impulse. The latter alternative is the one which\nLocke virtually adopts, and in adopting it he makes the actuality,\nby which sensation is distinguished from \u2018feelings in the mind,\u2019\nitself a creation of the mind. But though it is by an intellectual\ninterpretation of the feeling of touch, not by the feeling itself,\nthat there is given that idea of body, by reference to which actual\nsensation is distinguished from the mere idea, still with Locke the\nfeeling of touch is necessary to the interpretation. Thus, supposing\nhis notion to be carried out consistently, the actual present\nsensation, as reporting the real, must either be a feeling of touch,\nor, if of another sort, _e.g._, sight or hearing, must be referable\nto an object of touch. In other words, the real will exist for us\nso long only as it is touched, and ideas in us will constitute a\nreal essence so long only as they may be referred to an object now\ntouched. Let the object cease to be touched, and the ideas become\na nominal essence in the mind, the knowledge which they constitute\nceases to be real, and the proposition which expresses it ceases to\nconcern matter of fact. Truth as to matters of fact or bodies, then,\nmust be confined to singular propositions such as \u2018this is touched\nnow,\u2019 \u2018that was touched then;\u2019 \u2018what is touched now is bitter,\u2019 \u2018what\nwas then touched was red.\u2019 [1]\n\n[1] Thus the conviction that an object seen is not \u2018bare fancy,\u2019\nwhich is gained by \u2018putting the hand to it\u2019 (Book IV. chap. xi. sec.\nl7), as it conveys the idea of solidity, is properly, according to\nLocke\u2019s doctrine, not one among other \u2018confirmations of the testimony\nof the senses,\u2019 but the source of all such testimony, as a testimony\nto the real, _i.e._ to body. See above, paragraph 62.\n\nOnly the knowledge that something is, not what it is.\n\n105. All that is gained, then, by the conversion of the feeling of\ntouch, pure and simple, into the idea of a body touched, is the\nsupposition that _there is_ a real existence which does not come and\ngo with the sensations. As to _what_ this existence is, as to its\nreal essence, we can have no knowledge but such as is given in a\npresent sensation. [1] Any essence of it, otherwise known, could only\nbe a nominal essence, a relation of ideas in our minds: it would lack\nthe condition in virtue of which alone a datum of consciousness can\nclaim to be representative of reality, that of being an impression\nmade by a body now operating upon us. (Book III. chap. v. sec. 2, and\nBook IV. chap. xi. sec. 1.) The memory of such impression, however\nfaithful, will still only report a _past_ reality. It will itself\nbe merely \u2018an idea in the mind.\u2019 Neither it nor its relation to any\npresent sensation result from the immediate impact of body, and in\nconsequence neither \u2018really exists.\u2019 All that can be known, then, of\nthe real, in other words, the whole real essence of body, as it is\nfor us, reduces itself to that which can at any moment be \u2018revealed\u2019\nin a single sensation apart from all relation to past sensations; and\nthis, as we have seen, is nothing at all.\n\n[1] Cf. Book III. chap. vi. sec. 6: \u2018As to the real essences of\nsubstances, we only suppose their being, without precisely knowing\nwhat they are.\u2019 The appearance of the qualification \u2018precisely,\u2019 as\nwe shall see below, marks an oscillation from the view, according\nto which \u2018real essence\u2019 is the negation of the knowable to the view\naccording to which our knowledge of it is merely inadequate.\n\nHow it is that the real essence of things, according to Locke,\nperishes with them, yet is immutable.\n\n106. Thus that reduction of reality to that of which nothing can be\nsaid, which follows from its identification with particularity in\ntime, follows equally from its identification with the resistance\nof body, or (which comes to the same) from the notion of an \u2018outer\nsense\u2019 being its organ; since it is only that which _now_ resists,\nnot a general possibility of resistance nor a relation between the\nresistances of different times, that can be regarded as outside the\nmind. In Locke\u2019s language, it is only a particular parcel of matter\nthat can be so regarded. Of such a parcel, as he rightly says, it is\nabsurd to ask what is its essence, for it can have none at all. (See\nabove, paragraph 94.) As real, it has no quality save that of being\na body or of being now touched--a quality, which as all things real\nhave it and have none other, cannot be a _differentia_ of it. When\nwe consider that this quality may be regarded equally as immutable\nand as changing from moment to moment, we shall see the ground of\nLocke\u2019s contradiction of himself in speaking of the real thing\nsometimes as indestructible, sometimes as in continual dissolution.\n\u2018The real constitutions of things begin and perish with them.\u2019 (Book\nIII. chap. iii. sec. 19.) That is, the thing at one moment makes an\nimpact on the sensitive tablet--in the fact that it does so lie at\nonce its existence and its essence--but the next moment the impact\nis over, and with it thing and essence, _as real_, have disappeared.\nAnother impact, and thus another thing, has taken its place. But of\nthis the real essence is just the same as that of the previous thing,\nnamely, that it may be touched, or is solid, or a body, or a parcel\nof matter; nor can this essence be really lost, since than it there\nis no other reality, all difference of essence, as Locke expressly\nsays, [1] being constituted by abstract ideas and the work of the\nmind. It follows that _real_ change is impossible. A parcel of matter\nat one time is a parcel of matter at all times. Thus we have only to\nforget that the relation of continuity between the parcels, not being\nan idea caused by impact, should properly fall to the unreal--though\nonly on the same principle as should that of distinctness between the\ntimes--and we find the real in a continuity of matter, unchangeable\nbecause it has no qualities to change. It may seem strange that when\nthis notion of the formless continuity of the real being gets the\nbetter of Locke, a man should be the real being which he takes as his\ninstance. \u2018Nothing I have is essential to me. An accident or disease\nmay very much alter my colour or shape; a fever or fall may take away\nmy reason or memory, or both; and an apoplexy leave neither sense nor\nunderstanding, no, nor life.\u2019 (Book III. chap. vi. sec. 4.) But as\nthe sequel shows, the man or the \u2018I\u2019 is here considered simply as \u2018a\nparticular corporeal being,\u2019 _i.e._ as the \u2018parcel of matter\u2019 which\nalone (according to the doctrine of reality now in view) can be the\nreal in man, and upon which all qualities are \u2018superinductions of the\nmind.\u2019 [2]\n\n[1] Book III. chap. vi. sec. 4: \u2018Take but away the abstract ideas by\nwhich we sort individuals, and then the thought of anything essential\nto any of them instantly vanishes.\u2019\n\n[2] See a few lines below the passage quoted: \u2018So that if it be\nasked, whether it be essential to me, or any other particular\ncorporeal being, to have reason? I say, no; no more than it is\nessential to this white thing I write on to have words in it.\u2019\n\nOnly about qualities of matter, as distinct from matter itself, that\nLocke feels any difficulty.\n\n107. We may now discern the precise point where the qualm as to\nclothing reality with such superinductions commonly returns upon\nLocke. The conversion of feeling into body felt and of the particular\ntime of the feeling into an individuality of the body, and, further,\nthe fusion of the individual bodies, manifold as the times of\nsensation, into one continued body, he passes without scruple. So\nlong as these are all the traces of mental fiction which \u2018matter,\u2019\nor \u2018body,\u2019 or \u2018nature\u2019 bears upon it, he regards it undoubtingly as\nthe pure \u2018privation\u2019 of whatever belongs to the mind. But so soon\nas cognisable qualities, forming an essence, come to be ascribed to\nbody, the reflection arises that these qualities are on our side\nideas, and that so far as they are permanent or continuous they are\nnot ideas of the sort which can alone represent body as the \u2018real\u2019\nopposite of mind; they are not the result of momentary impact; they\nare not \u2018actually present sensations.\u2019 Suppose them, however, to have\nno permanence--suppose their reality to be confined to the fleeting\n\u2018now\u2019--and they are no qualities, no essence, at all. There is then\nfor us no _real_ essence of body or nature; what we call so is a\ncreation of the mind.\n\nThese, as knowable, must be our ideas, and therefore not a \u2018real\nessence\u2019.\n\n108. This implies the degradation of the \u2018primary qualities of\nbody\u2019 from the position which they hold in the Second Book of the\nEssay, as the real, _par excellence_, to that of a nominal essence.\nIn the Second Book, just as the complex of ideas, received and to\nbe received from a substance, is taken for the real thing without\ndisturbance from the antithesis between reality and \u2018ideas in the\nmind,\u2019 so the primary qualities of body are taken not only as real,\nbut as the sources of all other reality. Body, the real thing,\ncopying itself upon the mind in an idea of sensation (that of\nsolidity), carries with it from reality into the mind those qualities\nwhich \u2018the mind finds inseparable from it,\u2019 with all their modes. \u2018A\npiece of manna of a sensible bulk is able to produce in us the idea\nof a round or square figure, and, by being removed from one place\nto another, the idea of motion. This idea of motion represents it,\nas it really is in the manna, moving; a circle or square are the\nsame, whether in idea or existence, in the mind or in the manna;\nand this both motion and figure are really in the manna, whether\nwe take notice of them or no.\u2019 (Book II. chap. viii. sec. 18.) To\nthe unsophisticated man, taking for granted that the \u2018sensible\nbulk\u2019 of the manna is a \u2018real essence,\u2019 this statement will raise\nno difficulties. But when he has learnt from Locke himself that the\n\u2018sensible bulk,\u2019 so far as we can think and speak of it, must consist\nin the ideas which it is said to produce, the question as to the real\nexistence of these must arise. It turns out that they \u2018really exist,\u2019\nso far as they represent the impact of a body copying itself in\nactually present sensation, and that from their reality, accordingly,\nmust be excluded all qualities that accrue to the present sensation\nfrom its relation to the past. Can the \u2018primary qualities\u2019 escape\nthis exclusion?\n\nAre the \u2018primary qualities\u2019 then, a \u2018nominal essence\u2019?\n\n109. To obtain a direct and compendious answer to this question\nfrom Locke\u2019s own mouth is not easy, owing to the want of adjustment\nbetween the several passages where he treats of the primary\nqualities. They are originally enumerated as the \u2018bulk, figure,\nnumber, situation, and motion or rest of the solid parts of bodies\u2019\n(Book II. chap. viii. sec. 23), and, as we have seen, are treated\nas all involved in that idea of solidity which is given in the\nsensation of touch. We have no further account of them till we come\nto the chapters on \u2018simple modes of space and duration\u2019 (Book II.\nchaps. xiii. &c.), which are introduced by the remark, that in the\nprevious part of the book simple ideas have been treated \u2018rather\nin the way that they come into the mind than as distinguished from\nothers more compounded.\u2019 As the simple idea, according to Locke,\nis that which comes first into the mind, the two ways of treatment\nought to coincide; but there follows an explanation of the simple\nmodes in question, of which to a critical reader the plain result is\nthat the idea of body, which, according to the imaginary theory of\n\u2018the way that it came into the mind\u2019 is simple and equivalent to the\nsensation of touch, turns out to be a complex of relations of which\nthe simplest is called space.\n\nAccording to Locke\u2019s account they are relations, and thus inventions\nof the mind.\n\n110. To know what space itself is, \u2018we are sent to our senses\u2019 of\nsight and touch. It is \u2018as needless to go to prove that men perceive\nby their sight a distance between bodies of different colours,\nor between the parts of the same body, as that they see colours\nthemselves; nor is it less obvious that they can do so in the dark by\nfeeling and touch.\u2019 (Book II. chap. xiii. sec. 2.) Space being thus\nexplained by reference to distance, and distance _between bodies_,\nit might be supposed that distance and body were simpler ideas. In\nthe next paragraph, however, distance is itself explained to be a\nmode of space. It is \u2018space considered barely in length between any\ntwo beings,\u2019 and is distinguished _(a)_ from \u2018capacity\u2019 or \u2018space\nconsidered in length, breadth, and thickness;\u2019 _(b)_ from \u2018figure,\nwhich is nothing but the relation which the parts of the termination\nof extension, or circumscribed space, have among themselves;\u2019 _(c)_\nfrom \u2018place, which is the relation of distance between anything and\nany two or more points which are considered as keeping the same\ndistance one with another, and so as at rest.\u2019 It is then shown at\nlarge (Book II. chap. xiii. sec. 11), as against the Cartesians,\nthat extension, which is \u2018space in whatsoever manner considered,\u2019 is\na \u2018distinct idea from body.\u2019 The ground of the distinction plainly\nlies in the greater complexity of the idea of body. Throughout the\ndefinition just given \u2018space\u2019 is presupposed as the simpler idea of\nwhich capacity, figure, and place are severally modifications; and\nthese again, as \u2018primary qualities,\u2019 though with a slight difference\nof designation, [1] are not only all declared inseparable from body,\nbut are involved in it under a further modification as \u2018_qualities\nof its solid parts_\u2019 _i.e._, of parts so related to each other\nthat each will change its place sooner than admit another into it.\n(Book II. chap. iv. sec. 2, and chap. viii. sec. 23.) Yet, though\nbody is thus a complex of relations--all, according to Locke\u2019s\ndoctrine of relation, inventions of the mind--and though it must\nbe proportionately remote from the simple idea which \u2018comes first\ninto the mind,\u2019 yet, on the other hand, it is in body, as an object\npreviously given, that these relations are said to be found, and\nfound by the senses. (Book II. chap. xiii. secs. 2, 27.) [2]\n\n[1] In the enumeration of primary qualities, \u2018capacity\u2019 is\nrepresented by \u2018bulk,\u2019 \u2018place\u2019 by \u2018situation.\u2019\n\n[2] In the second of the passages referred to, it will be seen that\n\u2018matter\u2019 is used interchangeably with \u2018body.\u2019\n\nBody is the complex in which they are found. Do we derive the idea of\nbody from primary qualities, or the primary qualities from idea of\nbody?\n\n111. It will readily be seen that \u2018body\u2019 here is a mode of the\nidea of substance, and, like it, [1] appears in two inconsistent\npositions as at once the beginning and the end of the process of\nknowledge--as on the one hand that in which ideas are found and from\nwhich they are abstracted, and on the other hand that which results\nfrom their complication. As the attempt either to treat particular\nqualities as given and substance as an abstraction gradually made,\nor conversely to treat the \u2018thing\u2019 as given, and relations as\ngradually superinduced, necessarily fails for the simple reason\nthat substance and relations each presuppose the other, so body\npresupposes the primary qualities as so many relations which form\nits essence or make it what it is, while these again presuppose body\nas the matter which they determine, It is because Locke substitutes\nfor this intellectual order of mutual presupposition a succession of\nsensations in time, that he finds himself in the confusion we have\nnoticed--now giving the priority to sensations in which the idea\nof body is supposed to be conveyed, and from it deriving the ideas\nof the primary qualities, now giving it to these ideas themselves,\nand deriving the idea of body from their complication. This is just\nsuch a contradiction as it would be to put to-day before yesterday.\n_We_ may escape it by the consideration that in the case before us\nit is not a succession of sensations in time that we have to do with\nat all; that \u2018the real\u2019 is an intellectual order, or mind, in which\nevery element, being correlative to every other, at once presupposes\nand is presupposed by every other; but that this order communicates\nitself to us piecemeal, in a process of which the first condition on\nour part is the conception that there _is_ an order, or something\nrelated to something else; and that thus the conception of qualified\nsubstance, which in its definite articulation is the end of all our\nknowledge, is yet in another form, that may be called indifferently\neither abstract or confused, [2] its beginning. This way of escape,\nhowever, was not open to Locke, because with him it was the condition\nof reality in the idea of the body and its qualities that they\nshould be \u2018actually present sensations.\u2019 The priority then of body\nto the relations of extension, distance, &c., as of that in which\nthese relations are found, must, if body and extension are to be\nmore than nominal essences, be a priority of sensations in time.\nBut, on the other hand, the priority of the idea of space to the\nideas of its several modes, and of these again to the idea of body,\nas of the simpler to the more complex, must no less than the other,\nif the ideas in question are to be real, be one in time. Locke\u2019s\ncontradiction, then, is that of supposing that of two sensations each\nis actually present, of two impacts on the sensitive tablet each is\nactually made, before the other.\n\n[1] See above, paragraph 39.\n\n[2] \u2018Indifferently either abstract or confused,\u2019 because of the\nconception that is most confused the least can be said; and it is\nthus most abstract.\n\nMathematical ideas, though ideas of \u2018primary qualities of body,\u2019 have\n\u2018barely an ideal existence\u2019.\n\n112. From such a contradiction, even though he was not distinctly\naware of it, he could not but seek a way of escape, From his point\nof view two ways might at first sight seem to be open--the priority\nin sensitive experience, and with it reality, might be assigned\nexclusively either to the idea of body or to that of space. To\nwhichever of the two it is assigned, the other must become a nominal\nessence. If it is the idea of body that is conveyed to the mind\ndirectly from without through sensation, then it must be by a process\nin the mind that the spatial relations are abstracted from it; and\nconversely, if it is the latter that are given in sensation, it must\nbe by a mental operation of compounding that the idea of body is\nobtained from them. Now, according to Locke\u2019s fundamental notion,\nthat the reality of an idea depends upon its being in consciousness\na copy _through impact_ of that which is not in consciousness, any\nattempt to retain it in the idea of space while sacrificing it in\nthat of body would be obviously self-destructive. Nor, however we\nmight re-write his account of the relations of space as \u2018found in\nbodies,\u2019 could we avoid speaking of them as relations of some sort;\nand if relations, then derived from the \u2018mind\u2019s carrying its view\nfrom one thing to another,\u2019 and not \u2018actually present sensations.\u2019\nWe shall not, then, be surprised to find Locke tending to the other\nalternative, and gradually forgetting his assertion that \u2018a circle\nor a square are the same whether in idea or in existence,\u2019 and his\nelaborate maintenance of the \u2018real existence\u2019 of a vacuum, _i.e._,\nextension without body. (Book II. chap. xiii. secs. 21 and the\nfollowing, and xvii. 4.) In the Fourth Book it is body alone that has\nreal existence, an existence revealed by actually present sensation,\nwhile all mathematical ideas, the ideas of the circle and the square,\nhave \u2018barely an ideal existence\u2019 (Book IV. chap. iv. sec. 6); and\nthis means nothing else than the reduction of the primary qualities\nof body to a nominal essence. Our ideas of them are general (Book IV.\nchap. iii. sec. 24), or merely in the mind. \u2018There is no individual\nparcel of matter, to which any of these qualities are so annexed as\nto be essential to it or inseparable from it.\u2019 (Book III. chap. vi.\nsec. 6.) How should there be, when the \u2018individual parcel\u2019 means that\nwhich copies itself by impact in the present sensation, while the\nqualities in question are relations which cannot be so copied? Yet,\nexcept as attaching to such a parcel, they have no \u2018real existence;\u2019\nand, conversely, the \u2018body,\u2019 from which they _are_ inseparable, not\nbeing an individual parcel of matter in the above sense, must itself\nbe unreal and belong merely to the mind. The \u2018body\u2019 which is real\nhas for us no qualities, and that reference to it of the \u2018actually\npresent sensation\u2019 by which such sensation is distinguished from\nother feeling, is a reference to something of which nothing can be\nsaid. It is a reference which cannot be stated in any proposition\n_really_ true; and the difference which it constitutes between \u2018bare\nvision\u2019 and the feeling to which reality corresponds, must be either\nitself unreal or unintelligible.\n\nSummary view of Locke\u2019s difficulties in regard to the real.\n\n113. We have now pursued the antithesis between reality and the work\nof the mind along all the lines which Locke indicates, and find\nthat it everywhere eludes us. The distinction, which only appeared\nincidentally in the doctrine of substance, between \u2018the being and\nthe idea thereof--between substance as \u2018found\u2019 and substance as\nthat which \u2018we accustom ourselves to suppose\u2019--becomes definite and\nexplicit as that between real and nominal essence, but it does so\nonly that the essence, which is merely real, may disappear. Whether\nwe suppose it the quality of a mere sensation, as such, or of mere\nbody, as such, we find that we are unawares defining it by relations\nwhich are themselves the work of the mind, and that after abstraction\nof these nothing remains to give the antithesis to the work of the\nmind any meaning. Meanwhile the attitude of thought, when it has\ncleared the antithesis of disguise, but has not yet found that each\nof the opposites derives itself from thought as much as the other,\nis so awkward and painful that an instinctive reluctance to make\nthe clearance is not to be wondered at. Over against the world of\nknowledge, which is the work of the mind, stands a real world of\nwhich we can say nothing but that it is there, that it makes us aware\nof its presence in every sensation, while our interpretation of what\nit is, the system of relations which we read into it, is our own\ninvention. The interpretation is not even to be called a shadow, for\na shadow, however dim, still reflects the reality; it is an arbitrary\nfiction, and a fiction of which the possibility is as unaccountable\nas the inducement to make it. It is commonly presented as consisting\nin abstraction from the concrete. But the concrete, just so far as\nconcrete, _i.e._, a complex world of relations, cannot be the real\nif the separation of the real from the work of the mind is to be\nmaintained. It must itself be the work of the compounding mind, which\nmust be supposed again in \u2018abstraction\u2019 to decompose what it has\npreviously compounded. Now, it is of the essence of the doctrine in\nquestion that it denies all power of origination to the mind except\nin the way of compounding and abstracting given impressions. Its\nsupposition is, that whatever precedes the work of composition and\nabstraction must be real [1] because the mind passively receives it:\na supposition which, if the mind could originate, would not hold.\nHow, then, does it come to pass that a \u2018nominal essence,\u2019 consisting\nof definite qualities, is constructed by a mind, which originates\nnothing, out of a \u2018real\u2019 matter, which, apart from such construction,\nhas no qualities at all? And why, granted the construction, should\nthe mind in \u2018abstraction\u2019 go through the Penelopean exercise of\nperpetually unweaving the web which it has just woven?\n\n[1] \u2018Simple ideas, since the mind can by no means make them to\nitself, must necessarily be the product of things operating on the\nmind.\u2019 (Book IV. chap. v. sec. 4.)\n\nWhy they do not trouble him more.\n\n114. It is Hume\u2019s more logical version of Locke\u2019s doctrine that\nfirst forces these questions to the front. In Locke himself they\nare kept back by inconsistencies, which we have already dwelt upon.\nFor the real, absolutely void of intelligible qualities, because\nthese are relative to the mind, he is perpetually substituting a\nreal constituted by such qualities, only with a complexity which we\ncannot exhaust. By so doing, though at the cost of sacrificing the\nopposition between the real and the mental, he avoids the necessity\nof admitting that the system of the sciences is a mere language,\nwell-or ill-constructed, but unaccountably and without reference\nto things. Finally, he so far forgets the opposition altogether as\nto find the reality of \u2018moral and mathematical\u2019 knowledge in their\n\u2018bare ideality\u2019 itself. (Book IV. chap. iv. sec. 6, &c.) Thus with\nhim the divorce between knowledge and reality is never complete,\nand sometimes they appear in perfect fusion. A consideration of his\ndoctrine of propositions will show finally how the case between them\nstands, as he left it.\n\nThey re-appear in his doctrine of propositions.\n\n115. In the Fourth Book of the Essay the same ground has to be\nthrice traversed under the several titles of \u2018knowledge,\u2019 \u2018truth,\u2019\nand \u2018propositions.\u2019 Knowledge being the perception of agreement or\ndisagreement between ideas, the proposition is the putting together\nor separation of words, as the signs of ideas, in affirmative or\nnegative sentences (Book IV. chap. v. sec. 5), and truth--the\nexpression of certainty [1]--consists in the correspondence between\nthe conjunction or separation of the signs and the agreement or\ndisagreement of the ideas. (Book IV. chap. v. sec. 2.) Thus, the\nquestion between the real and the mental affects all these. Does this\nor that perception of agreement between ideas represent an agreement\nin real existence? Is its certainty a real certainty? Does such\nor such a proposition, being a correct expression of an agreement\nbetween ideas, also through this express an agreement between things?\nIs its truth real, or merely verbal?\n\n[1] All knowledge is certain according to Locke (Cf. IV. chap. vi.\nsec. 13, \u2018certainty is requisite to knowledge\u2019), though the knowledge\nmust be expressed before the term \u2018certainty\u2019 is naturally applied to\nit. (Book IV. chap. vi. sec. 3.) \u2018Certainty of knowledge\u2019 is thus a\npleonastic phrase, which only seems not to be so because we conceive\nknowledge to have a relation to things which Locke\u2019s definition\ndenies it, and by \u2018certainty,\u2019 in distinction from this, understand\nits relation to the subject.\n\n\u2018Certainty of truth\u2019 is, in like manner, a pleonastic phrase, there\nbeing no difference between the definition of it (Book IV. chap. vi.\nsec. 3) and that of \u2018truth\u2019 simply, given in Book IV. chap. V. sec. 2.\n\nThe knowledge expressed by a proposition, though certain, may not be\nreal ...\n\n116. To answer these questions, according to Locke, we must consider\nwhether the knowledge, or the proposition which expresses it,\nconcerns substances, _i.e._, \u2018the co-existence of ideas in nature,\u2019\non the one hand; or, on the other, either the properties of a\nmathematical figure or \u2018moral ideas.\u2019 If it is of the latter sort,\nthe agreement of the ideas in the mind is itself their agreement\nin reality, since the ideas themselves are archetypes. (Book IV.\nchap. iv. secs. 6, 7.) It is only when the ideas are ectypes, as is\nthe case when the proposition concerns substances, that the doubt\narises whether the agreement between them represents an agreement\nin reality. The distinction made here virtually corresponds to that\nwhich appears in the chapters on the reality and adequacy of ideas in\nthe Second Book, and again in those on \u2018names\u2019 in the Third. There\nthe \u2018complex ideas of modes and relation\u2019 are pronounced necessarily\nreal adequate and true, because, \u2018being themselves archetypes, they\ncannot differ from their archetypes.\u2019 (Book II. chap. XXX. sec.\n4.) [1] With them are contrasted simple ideas and complex ideas of\nsubstances, which are alike ectypes, but with this difference from\neach other, that the simple ideas cannot but be faithful copies\nof their archetypes, while the ideas of substances cannot but be\notherwise. (Book II. chap. xxxi. secs. 2, 11, &c.) Thus, \u2018the names\nof simple ideas and substances, with the abstract ideas in the mind\nwhich they immediately signify, intimate also some real existence,\nfrom which was derived their original pattern. But the names of\nmixed modes terminate in the idea that is in the mind.\u2019 (Book III.\nchap. iv. sec. 2.) \u2018The names of simple ideas and modes,\u2019 it is\nadded, \u2018signify always the real as well as nominal essence of their\nspecies\u2019--a statement which, if it is to express Locke\u2019s doctrine\nstrictly, must be confined to names of simple ideas, while in respect\nof modes it should run, that \u2018the nominal essence which the names of\nthese signify is itself the real.\u2019\n\n[1] cf. Book II. chap. xxxi. sec. 3, and xxxii. sec. 17.\n\n... when the knowledge concerns substances. In this case general\ntruth must be merely verbal. Mathematical truths, since they concern\nnot substances, may be both general and real.\n\n117. But though the distinction between different kinds of knowledge\nin regard to reality cannot but rest on the same principle as that\ndrawn between different kinds of ideas in the same regard, it is\nto be noticed that in the doctrine of the Fourth Book \u2018knowledge\nconcerning substances,\u2019 in contrast with that in which \u2018our thoughts\nterminate in the abstract ideas,\u2019 has by itself to cover the ground\nwhich, in the Second and Third Book, simple ideas and complex\nideas of substances cover together. This is to be explained by the\nobservation, already set forth at large, [1] that the simple idea\nhas in Locke\u2019s Fourth Book become explicitly what in the previous\nbooks it was implicitly, not a feeling proper, but the conscious\nreference of a feeling to a thing or substance. Only because it is\nthus converted, as we have seen, can it constitute the beginning\nof a knowledge which is not a simple idea but a conscious relation\nbetween ideas, or have (what yet it must have if it can be expressed\nin a proposition) that capacity of being true or false, which implies\n\u2018the reference by the mind of an idea to something extraneous to it.\u2019\n(Book II. chap, xxxii. sec. 4.) Thus, what is said of the \u2018simple\nidea\u2019 in the Second and Third Books, is in the Fourth transferred to\none form of knowledge concerning substances, to that, namely, which\nconsists in \u2018particular experiment and observation,\u2019 and is expressed\nin singular propositions, such as \u2018this is yellow,\u2019 \u2018this gold is\nnow solved in aqua regia.\u2019 Such knowledge cannot but be real, the\nproposition which expresses it cannot but have _real_ certainty,\nbecause it is the effect of a \u2018body actually operating upon us\u2019\n(Book IV. chap. xi. sec. 1), just as the simple idea is an ectype\ndirectly made by an archetype. It is otherwise with complex ideas of\nsubstances and with general knowledge or propositions about them.\nA group of ideas, each of which, when first produced by a \u2018body,\u2019\nhas been real, when retained in the mind as representing the body,\nbecomes unreal. The complex idea of gold is only a nominal essence\nor the signification of a name; the qualities which compose it are\nmerely ideas in the mind, and that general truth which consists in a\ncorrect statement of the relation between one of them and another or\nthe whole--_e.g._, \u2018gold is soluble in aqua regia\u2019--holds merely for\nthe mind; [2] but it is not therefore to be classed with those other\nmental truths, which constitute mathematical and moral knowledge,\nand which, just because \u2018merely ideal,\u2019 are therefore real. Its\nmerely mental character renders it in Locke\u2019s language a \u2018trifling\nproposition,\u2019 but does not therefore save it from being _really_\nuntrue. It is a \u2018trifling proposition,\u2019 for, unless solubility in\naqua regia is included in the complex idea which the sound \u2018gold\u2019\nstands for, the proposition which asserts it of gold is not certain,\nnot a truth at all. If it is so included, then the proposition is\nbut \u2018playing with sounds.\u2019 It may serve to remind an opponent of\na definition which he has made but is forgetting, but \u2018carries no\nknowledge with it but of the signification of a word, however certain\nit be.\u2019 (Book IV. chap. viii. secs. 5 & 9.) Yet there is a real gold,\noutside the mind, of which the complex idea of gold in the mind must\nneeds try to be a copy, though the conditions of real existence are\nsuch that no \u2018complex idea in the mind\u2019 can possibly be a copy of\nit. Thus the verbal truth, which general propositions concerning\nsubstances express, is under a perpetual doom of being really untrue.\nThe exemption of mathematical and moral knowledge from this doom\nremains an unexplained mercy. Because merely mental, such knowledge\nis real--there being no reality for it to _mis_represent--and yet\nnot trifling. The proposition that \u2018the external angle of all\ntriangles is bigger than either of the opposite internal angles,\u2019\nhas that general certainty which is never to be found but in our\nideas, yet \u2018conveys instructive real knowledge,\u2019 the predicate\nbeing \u2018a necessary consequence of precise complex idea\u2019 which forms\nthe subject, yet \u2018not contained in it.\u2019 (Book IV. chap. viii. sec.\n8.) [3] The same might be said apparently, according to Locke\u2019s\njudgment (though he is not so explicit about this), of a proposition\nin morals, such as \u2018God is to be feared and obeyed by man.\u2019 (Book\nIV. chap. xi. sec. 13.) [4] But how are such propositions, at once\nabstract and real, general and instructive, to be accounted for?\nThere is no \u2018workmanship of the mind\u2019 recognised by Locke but that\nwhich consists in compounding and abstracting (_i.e._, separating)\nideas of which \u2018it cannot originate one.\u2019 The \u2018abstract ideas\u2019\nof mathematics, the \u2018mixed modes\u2019 of morals, just as much as the\nideas of substances, must be derived by such mental artifice from\na material given in simple feeling, and \u2018real\u2019 because so given.\nYet, while this derivation renders ideas of substances unreal in\ncontrast with their real \u2018originals,\u2019 and general propositions about\nthem \u2018trifling,\u2019 because, while \u2018intimating an existence,\u2019 they\ntell nothing about it, on the other hand it actually constitutes\nthe reality of moral and mathematical ideas. Their relation to an\noriginal disappears; they are themselves archetypes, from which the\nmind, by its own act, can elicit other ideas not already involved\nin the meaning of their names. But this can only mean that the mind\nhas some other function than that of uniting what it has \u2018found\u2019 in\nseparation, and separating again what it has thus united--that it can\nitself originate.\n\n[1] See above, paragraph 25.\n\n[2] Book IV. chap. xi. sec. 13, xii. 9, &c.\n\n[3] Just as according to Kant such a proposition expresses a judgment\n\u2018synthetical,\u2019 yet \u2018\u00e1-priori.\u2019\n\n[4] Cf. Book IV. chap. iii. sec. 18, and Book III. chap, xi. sec. 16.\n\nSignificance of this doctrine.\n\n118. A genius of such native force as Locke\u2019s could not be applied\nto philosophy without determining the lines of future speculation,\neven though to itself they remained obscure. He stumbles upon truths\nwhen he is not looking for them, and the inconsistencies or accidents\nof his system are its most valuable part. Thus, in a certain sense,\nhe may claim the authorship at once of the popular empiricism of\nthe modern world, and of its refutation. He fixed the prime article\nof its creed, that thought has nothing to do with the constitution\nof facts, but only with the representation of them by signs and the\nrehearsal to itself of what its signs have signified--in brief, that\nits function is merely the analytical judgment; yet his admissions\nabout mathematical knowledge rendered inevitable the Kantian\nquestion, \u2018How are synthetic judgments \u00e1-priori possible?\u2019--which was\nto lead to the recognition of thought as constituting the objective\nworld, and thus to get rid of the antithesis between thought and\nreality. In his separation of the datum of experience from the work\nof thought he was merely following the Syllogistic Logic, which\nreally assigns no work to the thought, whose office it professes to\nmagnify, but the analysis of given ideas. Taking the work as that\nLogic conceived it (and as it must be conceived if the separation is\nto be maintained) he showed--conclusively as against Scholasticism--\nthe \u2018trifling\u2019 character of the necessary and universal truths with\nwhich it dealt. Experience, the manifestation of the real, regarded\nas a series of events which to us are sensations, can only yield\npropositions singular as the events, and having a truth like them\ncontingent. By consequence, necessity and universality of connection\ncan only be found in what the mind does for itself, without reference\nto reality, when it analyses the complex idea which it retains as\nthe memorandum of its past single experiences; _i.e._, in a relation\nbetween ideas or propositions of which one explicitly includes the\nother. Upon this relation syllogistic reasoning rests, and, except\nso far as it may be of use for convicting an opponent (or oneself)\nof inconsistency, it has nothing to say against such nominalism\nas the above. Hence, with those followers of Locke who have been\nmost faithful to their master, it has remained the standing rule\nto make the generality of a truth consist in its being analytical\nof the meaning of a name, and its necessity in its being included\nin one previously conceded. Yet if such were the true account of\nthe generality and necessity of mathematical propositions, their\ntruth according to Locke\u2019s explicit statement would be \u2018verbal and\ntrifling,\u2019 not, as it is, \u2018real and instructive.\u2019\n\nFatal to the notion that mathematical truths, though general, are got\nfrom experience:\n\n119. The point of this, the most obvious, contradiction inherent\nin Locke\u2019s empiricism, is more or less striking according to the\nfidelity with which the notion of matter-of-fact, or of the reality\nthat is not of the mind, proper to that system, is adhered to.\nWhen the popular Logic derived from Locke has so far forgotten\nthe pit whence it was digged as to hold that propositions of a\ncertainty at once real and general can be derived from experience,\nand to speak without question of \u2018general matters-of-fact\u2019 in a\nsense which to Locke almost, to Hume altogether, would have been\na contradiction in terms, it naturally finds no disturbance in\nregarding mathematical certainty as different not in kind, but only\nin degree, from that of any other \u2018generalisation from experience.\u2019\nNot aware that the distinction of mathematical from empirical\ngenerality is the condition upon which, according to Locke, the\nformer escapes condemnation as \u2018trifling,\u2019 it does not see any need\nfor distinguishing the sources from which the two are derived, and\nhence goes on asserting against imaginary or insignificant opponents\nthat mathematical truth is derived from \u2018experience;\u2019 which, if\n\u2018experience\u2019 be so changed from what Locke understood by it as to\nyield general propositions concerning matters-of-fact of other than\nanalytical purport, no one need care to deny. That it can yield such\npropositions is, doubtless, the supposition of the physical sciences;\nnor, we must repeat, is it the _correctness_ of this supposition\nthat is in question, but the validity, upon its admission, of that\nantithesis between experience and the work of thought, which is the\n\u2018be-all and end-all\u2019 of the popular Logic.\n\n... and to received views of natural science: but Locke not so clear\nabout this.\n\n120. Locke, as we have seen, after all the encroachments made\nunawares by thought within the limits of that experience which he\nopposes to it--or, to put it conversely, after all that he allows\n\u2018nature\u2019 to take without acknowledgment from \u2018mind\u2019--is still so far\nfaithful to the opposition as to \u2018suspect a science of nature to be\nimpossible.\u2019 This suspicion, which is but a hesitating expression\nof the doctrine that general propositions concerning substances are\nmerely verbal, is the exact counterpart of the doctrine pronounced\nwithout hesitation that mathematical truths, being at once real and\ngeneral, do not concern nature at all. Real knowledge concerning\nnature being given by single impressions of bodies at single times\noperating upon us, and by consequence being expressible only in\nsingular propositions, any reality which general propositions state\nmust belong merely to the mind, and a mind which can originate a\nreality other than nature\u2019s cannot be a passive receptacle of natural\nimpressions. Locke admits the real generality of mathematical truths,\nbut does not face its consequences. Hume, seeing the difficulty, will\nnot admit the real generality. The modern Logic, founded on Locke,\nbelieving in the possibility of propositions at once real and general\nconcerning nature. does not see the difficulty at all. It reckons\nmathematical to be the same in kind with natural knowledge, each\nalike being real notwithstanding its generality; not aware that by\nso doing, instead of getting rid, as it fancies, of the originative\nfunction of thought in respect of mathematical knowledge, it only\nnecessitates the supposition of its being originative in respect of\nthe knowledge of nature as well.\n\nAmbiguity as to real essence causes like ambiguity as to science of\nnature. Particular experiment cannot afford general knowledge.\n\n121. It may find some excuse for itself in the hesitation with\nwhich Locke pronounces the impossibility of real generality in\nthe knowledge of nature--an hesitation which necessarily results\nfrom the ambiguities, already noticed, in his doctrine of real and\nnominal essence. So far as the opposition between the nominal and\nreal essences of substances is maintained in its absoluteness, as\nthat between every possible collection of ideas on the one side, and\nsomething wholly apart from thought on the other, this impossibility\nfollows of necessity. But so far as the notion is admitted of\nthe nominal essence being in some way, however inadequately,\nrepresentative of the real, there is an opening, however indefinite,\nfor general propositions concerning the latter. On the one hand we\nhave the express statement that \u2018universal propositions, of whose\ntruth and falsehood we can have certain knowledge, concern not\nexistence\u2019 (Book IV. chap. ix. sec. 1). They are founded only on the\n\u2018relations and habitudes of abstract ideas\u2019 (Book IV. chap xii. sec.\n7); and since it is the proper operation of the mind in abstraction\nto consider an idea under no other existence but what it has in the\nunderstanding, they represent no knowledge of _real_ existence at all\n(Book IV. chap. ix. sec. 1). Here Locke is consistently following\nhis doctrine that the \u2018particularity in time,\u2019 of which abstraction\nis made when we consider ideas as in the understanding, is what\nspecially distinguishes the real; which thus can only be represented\nby \u2018actually present sensation.\u2019 It properly results from this\ndoctrine that the proposition representing particular experiment and\nobservation is only true of real existence so long as the sensation,\nin which the experiment consists, continues present. Not only is the\npossibility excluded of such experiment yielding a certainty which\nshall be general as well as real, but the particular proposition\nitself can only be _really_ true so far as the qualities, whose\nco-existence it asserts, are present sensations. The former of these\nlimitations to real truth we find Locke generally recognising, and\nconsequently suspecting a science of nature to be impossible; but\nthe latter, which would be fatal to the supposition of there being a\nreal nature at all, even when he carries furthest the reduction of\nreality to present feeling, he virtually ignores. On the other hand,\nthere keeps appearing the notion that, inasmuch as the combination\nof ideas which make up the nominal essence of a substance is taken\nfrom a combination in nature or reality, whenever the connexion\nbetween any of these is necessary, it warrants a proposition\n_universally_ true in virtue of the necessary connexion between the\nideas, and _really_ true in virtue of the ideas being taken from\nreality. According to this notion, though \u2018the certainty of universal\npropositions concerning substances is very narrow and scanty,\u2019 it is\nyet possible (Book IV. chap. vi. sec. 13). It is not recognised as\ninvolving that contradiction which it must involve if the antithesis\nbetween reality and ideas in the mind is absolutely adhered to. Nay,\ninasmuch as certain ideas of primary qualities, _e.g._ those of\nsolidity and of the receiving or communicating motion upon impulse,\nare necessarily connected, it is supposed actually to exist (Book\nIV. chap iii. sec. 14). It is only because, as a matter of fact, our\nknowledge of the relation between secondary qualities and primary is\nso limited that it cannot be carried further. That they are related\nas effects and causes, it would seem, we know; and that the \u2018causes\nwork steadily, and effects constantly flow from them,\u2019 we know also;\nbut \u2018their connexions and dependencies are not discoverable in our\nideas\u2019 (Book IV. chap. iii. sec. 29). That, if discoverable in our\nideas, just because there discovered, the connexion would not be a\nreal co-existence, Locke never expressly says. He does not so clearly\narticulate the antithesis between relations of ideas and matters\nof fact. If he had done so, he must also have excluded from real\nexistence those abstract ideas of body which constitute the scanty\nknowledge of it that according to him we do possess (Book IV. chap.\niii. sec. 24). He is more disposed to sigh for discoveries that would\nmake physics capable of the same general certainty as mathematics,\nthan to purge the former of those mathematical propositions--really\ntrue only because having no reference to reality--which to him formed\nthe only scientific element in them.\n\nWhat knowledge it can afford, according to Locke.\n\n122. The ambiguity of his position will become clearer if we resort\nto his favourite \u2018instances in gold.\u2019 The proposition, \u2018all gold is\nsoluble in aqua regia,\u2019 is certainly true, if such solubility is\nincluded in the complex idea which the word \u2018gold\u2019 stands for, and\nif such inclusion is all that the proposition purports to state. It\nis equally certain and equally trifling with the proposition, \u2018a\ncentaur is four-footed.\u2019 But, in fact, as a proposition concerning\nsubstance, it purports to state more than this, viz. that a \u2018body\nwhose complex idea is made up of yellow, very weighty, ductile,\nfusible, and fixed,\u2019 is always soluble in aqua regia. In other words,\nit states the invariable co-existence in a body of the complex idea,\n\u2018solubility in aqua regia,\u2019 with the group of ideas indicated by\n\u2018gold.\u2019 Thus understood--as instructive or synthetical--it has not\nthe certainty which would belong to it if it were \u2018trifling,\u2019 or\nanalytical, \u2018since we can never, from the consideration of the ideas\nthemselves, with certainty affirm\u2019 their co-existence (Book IV. chap.\nvi. sec 9). If we see the solution actually going on, or can recall\nthe sight of it by memory, we can affirm its co-existence with the\nideas in question in that \u2018bare instance;\u2019 and thus, on the principle\nthat \u2018whatever ideas have once been united in nature may be so united\nagain\u2019 (Book IV. chap. iv. sec. 12), infer a capacity of co-existence\nbetween the ideas, but that is all. \u2018Constant observation may assist\nour judgments in guessing\u2019 an invariable actual co-existence (Book\nIV. chap. viii. sec. 9); but beyond guessing we cannot get. If our\ninstructive proposition concerning co-existence is to be general\nit must remain problematical. It is otherwise with mathematical\npropositions. \u2018If the three angles of a triangle were once equal to\ntwo right angles, it is certain that they always will be so;\u2019 but\nonly because such a proposition concerns merely \u2018the habitudes and\nrelations of ideas.\u2019 \u2018If the perception that the same ideas will\neternally have the same habitudes and relations be not a sufficient\nground of knowledge, there could be no knowledge of general\npropositions in mathematics; for no mathematical demonstration\ncould be other than particular: and when a man had demonstrated any\nproposition concerning one triangle and circle, his knowledge would\nnot reach beyond that particular diagram\u2019 (Book IV. chap. i. sec. 9).\n\nNot the knowledge which is now supposed to be got by induction. Yet\nmore than Locke was entitled to suppose it could give.\n\n123. To a reader, fresh from our popular treatises on Logic, such\nlanguage would probably at first present no difficulty. He would\nmerely lament that Locke, as a successor of Bacon, was not better\nacquainted with the \u2018Inductive methods,\u2019 and thus did not understand\nhow an observation of co-existence in the bare instance, if the\ninstance be of the right sort, may warrant a universal affirmation.\nOr he may take the other side, and regard Locke\u2019s restriction upon\ngeneral certainty as conveying, not any doubt as to the validity\nof the inference from an observed case to all cases where the\nconditions are ascertainably the same, but a true sense of the\ndifficulty of ascertaining in any other case that the conditions\nare the same. On looking closer, however, he will see that, so far\nfrom Locke\u2019s doctrine legitimately allowing of such an adaptation\nto the exigencies of science, it is inconsistent with itself in\nadmitting the reality of most of the conditions in the case supposed\nto be observed, and thus in allowing the real truth even of the\nsingular proposition. This purports to state, according to Locke\u2019s\nterminology, that certain \u2018ideas\u2019 do now or did once co-exist in a\nbody. But the ideas, thus stated to co-exist, according to Locke\u2019s\ndoctrine that real existence is only testified to by actual present\nsensation, differ from each other as that which _really_ exists\nfrom that which does not. In the particular experiment of gold\nbeing solved in aqua regia, from the complex idea of solubility an\nindefinite deduction would have to be made for qualification by ideas\nretained in the understanding before we could reach the present\nsensation; and not only so, but the group of ideas indicated by\n\u2018gold,\u2019 to whose co-existence with solubility the experiment is said\nto testify, as Locke himself says, form merely a nominal essence,\nwhile the body to which we ascribe this essence is something which we\n\u2018accustom ourselves to suppose,\u2019 not any \u2018parcel of matter\u2019 having\na real existence in nature. [1] In asserting the co-existence of\nthe ideas forming such a nominal essence with the actual sensation\nsupposed to be given in the experiment, we change the meaning of\n\u2018existence,\u2019 between the beginning and end of the assertion, from\nthat according to which all ideas exist to that according to which\nexistence has no \u2018connexion with any other of our ideas but those of\nourselves and God,\u2019 but is testified to by present sensation. [2]\nThis paralogism escapes Locke just as his equivocal use of the term\n\u2018idea\u2019 escapes him. The distinction, fixed in Hume\u2019s terminology as\nthat between impression and idea, forces itself upon him, as we have\nseen, in the Fourth book of the Essay, where the whole doctrine of\nreal existence turns upon it, but alongside of it survives the notion\nthat ideas, though \u2018in the mind\u2019 and forming a nominal essence, are\nyet, if rightly taken from things, ectypes of reality. Thus he does\nnot see that the co-existence of ideas, to which the particular\nexperiment, as he describes it, testifies, is nothing else than the\nco-existence of an event with a conception--of that which is in a\nparticular time, and (according to him) only for that reason real,\nwith that which is not in time at all but is an unreal abstraction\nof the mind\u2019s making. [3] The reality given in the actual sensation\ncannot, as a matter of fact, be discovered to have a necessary\nconnexion with the ideas that form the nominal essence, and therefore\ncannot be asserted universally to co-exist with them; but with better\nfaculties, he thinks, the discovery might be made (Book IV. chap.\niii. sec. 16). It does not to him imply such a contradiction as it\nmust have done if he had steadily kept in view his doctrine that of\nparticular (_i.e._ real) existence our \u2018knowledge\u2019 is not properly\nknowledge at all, but simply sensation--such a contradiction as was\nto Hume involved in the notion of deducing a matter of fact.\n\n[1] See above, paragraphs 35, 94, &c.\n\n[2] See above, paragraph 30 and the following.\n\n[3] See above, paragraphs 45, 80, 85, 97.\n\nWith Locke mathematical truths, though ideal, true also of nature.\n\n124. It results that those followers of Locke, who hold the\ndistinction between propositions of mathematical certainty and those\nconcerning real existence to be one rather of degree than of kind,\nthough they have the express words of their master against them, can\nfind much in his way of thinking on their side. This, however, does\nnot mean that he in any case drops the antithesis between matters of\nfact and relations of ideas in favour of matters of fact, so as to\nadmit that mathematical propositions concern matters of fact, but\nthat he sometimes drops it in favour of relations of ideas, so as\nto represent real existence as consisting in such relations. If the\nmatter of fact, or real existence, is to be found only in the event\nconstituted or reported by present feeling, such a relation of ideas,\nby no manner of means reducible to an event, as the mathematical\nproposition states, can have no sort of connection with it. But if\nreal existence is such that the relations of ideas, called primary\nqualities of matter, constitute it, and the qualities included in our\nnominal essences are its copies or effects, then, as on the one side\nour complex ideas of substances only fail of reality through want of\nfulness, or through mistakes in the process by which they are \u2018taken\nfrom things,\u2019 so, on the other side, the mental truth of mathematical\npropositions need only fail to be real because the ideas, whose\nrelations they state, are considered in abstraction from conditions\nwhich qualify them in real existence. \u2018If it is true of the idea of\na triangle that its three angles equal two right ones, it is true\nalso of a triangle, wherever it really exists\u2019 (Book IV. chap. iv.\nsec. 6). There is, then, no incompatibility between the idea and\nreal existence. Mathematical ideas might fairly be reckoned, like\nthose of substances, to be taken from real existence; but though,\nlike these, inadequate to its complexity, to be saved from the\nnecessary infirmities which attach to ideas of substances because not\nconsidered as so taken, but merely as in the mind. There is language\nabout mathematics in Locke that may be interpreted in this direction,\nthough his most explicit statements are on the other side. It is not\nour business to adjust them, but merely to point out the opposite\ntendencies between which a clear-sighted operator on the material\ngiven by Locke would find that he had to choose.\n\nTwo lines of thought in Locke, between which a follower would have to\nchoose.\n\n125. On the one hand there is the identification of real existence\nwith the momentary sensible event. This view, of which the proper\nresult is the exclusion of predication concerning real existence\naltogether, appears in Locke\u2019s restriction of such predication to the\nsingular proposition, and in his converse assertion that propositions\nof mathematical certainty \u2018concern not existence\u2019 (Book IV. chap.\niv. sec. 8). The embarrassment resulting from such a doctrine is\nthat it leads round to the admission of the originativeness of\nthought and of the reality of its originations, with the denial of\nwhich it starts. [1] It leads Locke himself along a track, which his\nlater followers scarcely seem to have noticed, when he treats the\n\u2018never enough to be admired discoveries of Mr. Newton\u2019 as having to\ndo merely with the relations of ideas in distinction from things,\nand looks for a true extension of knowledge--neither in syllogism\nwhich can yield no instructive, nor in experiment which can yield no\ngeneral, certainty--but only in a further process of \u2018singling out\nand laying in order intermediate ideas,\u2019 which are \u2018real as well as\nnominal essences of their species,\u2019 because they have no reference\nto archetypes elsewhere than in the mind (Book IV. chap. vii. sec.\n11, and Book IV. chap. xii. sec. 7). On the other hand there is the\nnotion that ideas, without distinction between \u2018actual sensation\u2019\nand \u2018idea in the mind,\u2019 are taken from permanent things, and are\nreal if correctly so taken. From this it results that propositions,\nuniversally true as representing a necessary relation between ideas\nof primary qualities, are true also of real existence; and that an\nextension of such real certainty through the discovery of a necessary\nconnexion between ideas of primary and those of secondary qualities,\nthough scarcely to be hoped for, has no inherent impossibility. It\nis this notion, again, that unwittingly gives even that limited\nsignificance to the particular experiment which Locke assigns to it,\nas indicating a co-existence between ideas present as sensations\nand those which can only be regarded as in the mind. Nor is it the\nintrinsic import so much as the expression of this notion that is\naltered when Locke substitutes an order of nature for substance\nas that in which the ideas co-exist. In his Fourth Book he so far\ndeparts from the doctrine implied in his chapters on the reality and\nadequacy of ideas and on the names of substances, as to treat the\nnotion of several single subjects in which ideas co-exist (which he\nstill holds to be the proper notion of substances), as a fiction\nof thought. There are no such single subjects. What we deem so are\nreally \u2018retainers to other parts of nature.\u2019 \u2018Their observable\nqualities, actions, and powers are owing to something without them;\nand there is not so complete and perfect a part that we know of\nnature, which does not owe the being it has, and the excellencies\nof it, to its neighbours\u2019 (Book IV. chap. vi. sec. 11). As thus\nconceived of, the \u2018objective order\u2019 which our experience represents\nis doubtless other than that collection of fixed separate \u2018things,\u2019\nimplied in the language about substances which Locke found in vogue,\nbut it remains an objective order still--an order of \u2018qualities,\nactions, and powers\u2019 which no multitude of sensible events could\nconstitute, but apart from which no sensible event could have such\nsignificance as to render even a singular proposition of real truth\npossible.\n\n[1] See above, paragraph 117, sub. fin.\n\nTransition to doctrine of God and the soul.\n\n126. It remains to inquire how, with Locke, the ideas of self and\nGod escape subjection to those solvents of reality which, with\nmore or less of consistency and consciousness, he applied to the\nconceptions on which the science of nature rests. Such an enquiry\nforms the natural transition to the next stage in the history of his\nphilosophy. It was Berkeley\u2019s practical interest in these ideas that\nheld him back from a development of his master\u2019s principles, in which\nhe would have anticipated Hume, and finally brought him to attach\nthat other meaning to the \u2018new way of ideas\u2019 faintly adumbrated\nin the later sections of his \u2018Siris,\u2019 which gives to Reason the\nfunctions that Locke had assigned to Sense.\n\nThinking substance--source of the same ideas as outer substance.\n\n127. The dominant notion of the self in Locke is that of the inward\nsubstance, or \u2018substratum of ideas,\u2019 co-ordinate with the outward,\n\u2018wherein they do subsist, and from which they do result.\u2019 \u2018Sensation\nconvinces that there are solid extended substances, and reflection\nthat there are thinking ones\u2019 (Book II. chap, xxiii. sec. 29). We\nhave already seen how, without disturbance from his doctrine of the\nfictitiousness of universals, he treats the simple idea as carrying\nwith it the distinction of outward and inward, or relations severally\nto a \u2018thing\u2019 and to a \u2018mind.\u2019 It reports itself ambiguously as a\nquality of each of these separate substances. It is now, or was to\nbegin with, the result of an outward thing \u2018actually operating upon\nus;\u2019 for \u2018of simple ideas the mind cannot make one to itself:\u2019 on the\nother hand, it is a \u2018perception,\u2019 and perception is an \u2018operation\nof the mind.\u2019 In other words it is at once a modification of the\nmind by something of which it is consciously not conscious, and a\nmodification of the mind by itself--the two sources of one and the\nsame modification being each determined only as the contradictory\nof the other. Thus, when we come to probe the familiar metaphors\nunder which Locke describes Reflection, as a \u2018fountain of ideas\u2019\nother than sensation, we find that the confusions which we have\nalready explored in dealing with the ideas of sensation recur under\nadded circumstances of embarrassment. Not only does the simple\nidea of reflection, like that of sensation, turn out to be already\ncomplicated in its simplicity with the superinduced ideas of cause\nand relation, but the causal substance in question turns out to\nbe one which, from being actually nothing, becomes something by\nacting upon itself; while all the time the result of this action is\nindistinguishable from that ascribed to the opposite, the external,\ncause.\n\nOf which substance is perception the effect?\n\n128. To a reader to whom Locke\u2019s language has always seemed to be--as\nindeed it is--simply that of common sense and life, in writing the\nabove we shall seem to be creating a difficulty where none is to be\nfound. Let us turn, then, to one of the less prolix passages, in\nwhich the distinction between the two sources of ideas is expressed:\n\u2018External objects furnish the mind with the ideas of sensible\nqualities, which are all those different perceptions they produce in\nus; and the mind furnishes the understanding with ideas of its own\noperations\u2019 (Book II. chap. i. sec. 5). We have seen already that\nwith Locke perception and idea are equivalent terms. It only needs\nfurther to be pointed out that no distinction can be maintained\nbetween his usage of \u2018mind\u2019 and of \u2018understanding,\u2019 [1] and that the\nsimple ideas of the mind\u2019s own operations are those of perception\nand power, which must be given in and with every idea of a sensible\nquality.\u2019 [2] Avoiding synonyms, then, and recalling the results of\nour examination of the terms involved in the first clause of the\npassage before us, we may re-write the whole thus: \u2018Creations of the\nmind, which yet are external to it, produce in it those perceptions\nof their qualities which they do produce; and the mind produces in\nitself the perception of these, its own, perceptions.\u2019\n\n[1] As becomes apparent on examination of such passages, as Book II.\nchap. i. sec. 1, sub. fin.; and Book II. chap. i. sec. 23.\n\n[2] See above, paragraphs 11, 12, 16.\n\nThat which is the source of substantiation cannot be itself a\nsubstance.\n\n129. This attempt to present Locke\u2019s doctrine of the relation between\nthe mind and the world, as it would be without phraseological\ndisguises, must not be ascribed to any polemical interest in making\na great writer seem to talk nonsense, The greatest writer must fall\ninto confusions when he brings under the conceptions of cause and\nsubstance the self-conscious thought which is their source; and\nnothing else than this is involved in Locke\u2019s avowed enterprise\nof knowing that which renders knowledge possible as he might know\nany other object. The enterprise naturally falls into two parts,\ncorresponding to that distinction of subject and object which\nself-consciousness involves. Hitherto we have been dealing with\nit on the objective side--with the attempt to know knowledge as a\nresult of experience received through the senses--and have found the\nsupposed source of thought already charged with its creations; with\nthe relations of inner and outer, of substance and attribute, of\ncause and effect, of appearance and reality. The supposed \u2018outward\u2019\nturns out to have its outwardness constituted by thought, and thus\nto be inward. The \u2018outer sense\u2019 is only an outer sense at all so\nfar as feelings, by themselves neither outward nor inward, are by\nthe mind referred to a thing or cause which \u2018the mind supposes;\u2019\nand only thus have its reports a prerogative of reality over the\n\u2018fantasies,\u2019 supposed merely of the mind. Meanwhile, unable to\nignore the subjective side of self-consciousness, Locke has to put\nan inward experience as a separate, but co-ordinate, source of\nknowledge alongside of the outer. But this inward experience, simply\nas a succession of feelings, does not differ from the outer: it only\nso differs as referred to that very \u2018thinking thing,\u2019 called the\nmind, which by its supposition of causal substance has converted\nfeeling into an experience of an outer thing. \u2018Mind\u2019 thus, by\nthe relations which it \u2018invents,\u2019 constitutes both the inner and\nouter, and yet is treated as itself the inner \u2018substratum which it\naccustoms itself to suppose.\u2019 It thus becomes the creature of its\nown suppositions. Nor is this all. This, indeed, is no more than the\nfate which it must suffer at the hands of every philosopher who,\nin Kantian language, brings the source of the Categories under the\nCategories. But with Locke the constitution of the outer world by\nmental supposition, however uniformly implied, is always ignored; and\nthus mind, as the inward substance, is not only the creature of its\nown suppositions, but stands over against a real existence, of which\nthe reality is held to consist just in its being the opposite of all\nsuch suppositions: while, after all, the effect of these mutually\nexclusive causes is one and the same experience, one and the same\nsystem of sequent and co-existent ideas.\n\nTo get rid of the inner source of ideas in favour of the outer would\nbe false to Locke.\n\n130. Is it then a case of _joint_-effect? Do the outer and inner\nsubstances combine, like mechanical forces, to produce the psychical\nresult? Against such a supposition a follower of Locke would find\nnot only the language of his master, with whom perception appears\n_indifferently_ as the result of the outer or inner cause, but\nthe inherent impossibility of analysing the effect into separate\nelements. The \u2018Law of Parsimony,\u2019 then, will dictate to him that\none or other of the causes must be dispensed with; nor, so long as\nhe takes Locke\u2019s identification of the outward with the real for\ngranted, will he have much doubt as to which of the two must go.\nTo get rid of the causality of mind, however, though it might not\nbe untrue to the tendency of Locke, would be to lose sight of his\nessential merit as a formulator of what everyone thinks, which is\nthat, at whatever cost of confusion or contradiction, he at least\nformulates it fully. In him the \u2018Dialectic,\u2019 which popular belief\nimplicitly involves, goes on under our eyes. If the primacy of\nself-conscious thought is never recognized, if it remains the victim\nof its own misunderstood creations, there is at least no attempt\nto disguise the unrest which attaches to it in this self-imposed\nsubjection.\n\nThe mind, which Locke opposes to matter, perpetually shifting.\n\n131. We have already noticed how the inner \u2018tablet,\u2019 on which the\nouter thing is supposed to act, is with Locke perpetually receding.\n[1] It is first the brain, to which the \u2018motion of the outward parts\u2019\nmust be continued in order to constitute sensation (Book II. chap.\nix. sec. 3). Then perception is distinguished from sensation, and\nthe brain itself, as the subject of sensation, becomes the outward\nin contrast with the understanding as the subject of perception. [2]\nThen perception, from being simply a reception, is converted into an\n\u2018operation,\u2019 and thus into an efficient of ideas. The \u2018understanding\u2019\nitself, as perceptive, is now the outward which makes on the \u2018mind,\u2019\nas the inner \u2018tablet,\u2019 that impression of its own operation in\nperception which is called an idea of reflection. [3] Nor does\nthe regressive process--the process of finding a mind within the\nmind--stop here, though the distinction of inner and outer is not any\nfurther so explicitly employed in it. From mind, as receptive of, and\noperative about, ideas, _i.e._ consciousness, is distinguished mind\nas the \u2018substance within us\u2019 of which consciousness is an \u2018operation\u2019\nthat it sometimes exercises, sometimes (_e.g._ when it sleeps) does\nnot (Book II. chap. i. secs. 10-12); and from this thinking substance\nagain is distinguished the man who \u2018finds it in himself and carries\nit about with him in a coach or on horseback (Book II. chap, xxiii.\nsec. 20)--the person, \u2018consisting of soul and body,\u2019 who is prone to\nsleep and in sound sleep is unconscious, but whose personal identity\nstrangely consists in sameness of consciousness, sameness of an\noccasional operation of part of himself. [3]\n\n[1] See above, paragraph 14.\n\n[2] Book II., chap. i. sec. 23. \u2018Sensation is such an impression\nmade in some part of the body, as produces some perception in the\nunderstanding.\u2019\n\n[3] Locke speaks indifferently of the mind impressing the\nunderstanding, and of the understanding impressing the mind, with\nideas of reflection, but as he specially defines \u2018understanding\u2019 as\nthe \u2018perceptive power\u2019 (Book II. chap. xxi. sec. 25.), I have written\nas above.\n\n[4] Cf. II. chap. i. secs. 11 and 14, with II. chap, xxvii. sec. 9.\nIt is difficult to see what ingenuity could reconcile the doctrine\nstated in Book II. chap. xxvii. sec. 9, that personal identity is\nidentity of consciousness, with the doctrine implied in Book II.\nchap. i. sec. 11, that the waking Socrates is the same person with\nSocrates asleep, _i.e._ (according to Locke) not conscious at all.\n\nTwo ways out of such difficulties. \u2018Matter\u2019 and \u2018mind\u2019 have the same\nsource in self-consciousness.\n\n132. In the history of subsequent philosophy two typical methods\nhave appeared of dealing with this chaos of antinomies. One, which\nwe shall have to treat at large in writing of Hume, affects to\ndispose of both the outward and the inward synthesis--both of\nthe unity of feelings in a subject matter and of their unity in\na subject mind--as \u2018fictions of thought.\u2019 This method at once\nsuggests the vital question whether a mind which thus invents has\nbeen effectively suppressed--whether, indeed, the theory can be so\nmuch as stated without a covert assumption of that which it claims\nto have destroyed. The other method, of which Kant is the parent,\ndoes not attempt to efface the apparent contradictions which beset\nthe \u2018relation between mind and matter;\u2019 but regarding them as in\na certain sense inevitable, traces them to their source in the\napplication to the thinking Ego itself of conceptions, which it\ndoes indeed constitute in virtue of its presence to phenomena given\nunder conditions of time, but under which for that very reason it\ncannot itself be known. It is in virtue of the presence of the\nself-conscious unit to the manifold of feeling, according to this\ndoctrine, that the latter becomes an order of definite things, each\nexternal to the other; and it is only by a false inclusion within\nthis order of that which constitutes it that the Ego itself becomes\na \u2018thinking thing\u2019 with other things outside it. The result of such\ninclusion is that the real world, which it in the proper sense makes,\nbecomes a reality external to it, yet apart from which it would\nnot be actually anything. Thus with Locke, though the mind has a\npotential existence of its own, it is experience of \u2018things without\nit\u2019 that \u2018furnishes\u2019 it or makes it what it actually is. But the\nrelation of such outer things to the mind cannot be spoken of without\ncontradiction. If supposed outward as bodies, they have to be brought\nwithin consciousness as objects of sensation; if supposed outward\nas sensation, they have to be brought within consciousness--to find\na home in the understanding--as ideas of sensation. Meanwhile the\nconsideration returns that after all the \u2018thinking thing\u2019 contributes\nsomething to that which it thinks about; and, this once admitted, it\nis as impossible to limit its work on one side as that of the outer\nthing on the other. Each usurps the place of its opposite. Thus with\nLocke the understanding produces effects on itself, but the product\nis one and the same \u2018perception\u2019 otherwise treated as an effect of\nthe outer world. One and the same self-consciousness, in short, [1]\ninvolving the correlation of subject and object, becomes the result\nof two separate \u2018things,\u2019 each exclusive of the other, into which the\nopposite poles of this relation have been converted--the extended\nthing or \u2018body\u2019 on the one side, and the thinking thing or \u2018mind\u2019 on\nthe other.\n\n[1] For the equivalence of perception with self-consciousness in\nLocke, see above, paragraph 24, et infra.\n\nDifficulties in the way of ascribing reality to substance as matter,\nre-appear in regard to substance as mind.\n\n133. To each of these supposed \u2018things\u2019 thought transfers its own\nunity and self-containedness, and thereupon finds itself in new\ndifficulties. These, so far as they concern the outward thing, have\nalready been sufficiently noticed. We have seen how the single\nself-contained thing on the one hand attenuates itself to the bare\natom, presented in a moment of time, which in its exclusiveness\nis actually nothing: [1] how, on the other, it spreads itself, as\neverything which for one moment we regard as independent turns out\nin the next to be a \u2018retainer\u2019 to something else, into a series that\ncannot be summed. [2] A like consequence follows when the individual\nman, conceiving of the thought, which is not mine but me, and which\nis no less the world without which I am not I, as a thinking thing\nwithin him, limited by the limitations of his animal nature, seeks\nin this thinking thing, exclusive of other things, that unity and\nself-containedness, which only belong to the universal \u2018I.\u2019 He\nfinds that he \u2018thinks not always;\u2019 that during a fourth part of his\ntime he neither thinks nor perceives at all; and that even in his\nwaking hours his consciousness consists of a succession of separate\nfeelings, whose recurrence he cannot command. [3] Thought being\nthus broken and dependent, substantiality is not to be found in it.\nIt is next sought in the \u2018thing\u2019 of which thought is an occasional\noperation--a thing of which it may readily be admitted that its\nnature cannot be known (Book II. chap, xxiii. sec. 29, etc.), since\nit has no nature, being merely that which remains of the thinking\nthing upon abstraction of its sole determination. It is in principle\nnothing else than the supposed basis of sensible qualities remaining\nafter these have been abstracted--the \u2018parcel of matter\u2019 which has\nno essence--with which accordingly Locke sometimes himself tends\nto identify it. [4] But meanwhile, behind this unknown substance,\nwhether of spirit or of body, the self-consciousness, which has been\ntreated as its occasional unessential operation, re-asserts itself as\nthe self which claims both body and spirit, the immaterial no less\nthan the material substance, as its own, and throughout whatever\ndiversity in these maintains its own identity.\n\n[1] See above, paragraph 94 and the following.\n\n[2] See above, paragraph 125.\n\n[3] Locke, Essay ii. chap. i. sec. 10, etc.\n\n[4] See above, paragraph 106, near the end.\n\nWe think not always, yet thought constitutes the self.\n\n134. Just, then, as Locke\u2019s conception of outward reality grows\nunder his hands into a conception of nature as a system of relations\nwhich breaks through the limitations of reality as constituted\nby mere _individua_, so it is with the self, as he conceived it.\nIt is not a simple idea. It is not one of the train that is for\never passing, \u2018one going and another coming,\u2019 for it looks on this\nsuccession as that which it experiences, being itself the same\nthroughout the successive differences (Book II. chap. vii. sec. 9,\nand chap. xxvii. sec. 9). As little can it be adjusted to any of\nthe conditions of real \u2018things,\u2019 thinking or unthinking, which he\nordinarily recognises. It has no \u2018particularity in space and time.\u2019\nThat which is past in \u2018reality\u2019 is to it present. It is \u2018in its\nnature indifferent to any parcel of matter.\u2019 It is the same with\nitself yesterday and to-day, here and there. That \u2018with which its\nconsciousness can join itself is one self with it,\u2019 and it can so\njoin itself with substances apart in space and remote in time (Book\nII. chap, xxvii. secs. 9, 13, 14, 17). For speaking of it as eternal,\nindeed, we could find no warrant in Locke. He does not so clearly\ndistinguish it from the \u2018thinking thing\u2019 supposed to be within each\nman, that has \u2018had its determinate time and place of beginning to\nexist, relation to which determines its identity so long as it\nexists\u2019 (Book II. chap. xxvii. sec. 2). Hence he supposed an actual\nlimit to the past which it could make present--a limit seemingly\nfixed for each man at the farthest by the date of his birth--though\nhe talks vaguely of the possibility of its range being extended (Book\nII. chap. xxvii. sec. 16). In the discussion of personal identity,\nhowever, the distinction gradually forces itself upon him, and he at\nlast expressly says (sec. 16), that if the same Socrates, sleeping\nand waking, do not partake of the same consciousness (as according to\nBook II. chap. i. sec. 11 he certainly does not), \u2018Socrates sleeping\nand waking is not the same person;\u2019 whereas the \u2018thinking thing\u2019--the\nsubstance of which consciousness is a power sometimes exercised,\nsometimes not--is the same in the sleeping as in the waking Socrates.\nThis is a pregnant admission, but it brings nothing to the birth in\nLocke himself. The inference which it suggests to his reader, that\na self which does not slumber or sleep is not one which is born or\ndies, does not seem to have occurred to him. Taking for his method\nthe imaginary process of \u2018looking into his own breast,\u2019 instead\nof the analysis of knowledge and morality, he could not find the\neternal self which knowledge and morality pre-suppose, but only the\ncontradiction of a person whose consciousness is not the same for two\nmoments together, and often ceases altogether, but who yet, in virtue\nof an identity of this very consciousness, is the same in childhood\nand in old age.\n\nLocke neither disguises these contradictions, nor attempts to\novercome them.\n\n135. Here as elsewhere we have to be thankful that the contradiction\nhad not been brought home so strongly to Locke as to make him\nseek the suppression of either of its alternatives. He was aware\nneither of the burden which his philosophy tended to put upon the\nself which \u2018can consider itself as itself in different times and\nplaces\u2019--the burden of replacing the stable world, when \u2018the new way\nof ideas\u2019 should have resolved the outward thing into a succession of\nfeelings--nor of the hopelessness of such a burden being borne by a\n\u2018perishing\u2019 consciousness, \u2018of which no two parts exist together, but\nfollow each other in succession.\u2019 [1] When he \u2018looked into himself,\u2019\nhe found consciousness to consist in the succession of ideas,\n\u2018one coming and another going:\u2019 he also found that \u2018consciousness\nalone makes what we call self,\u2019 and that he was the same self at\nany different points in the succession. He noted the two \u2018facts of\nconsciousness\u2019 at different stages of his enquiry, and was apparently\nnot struck by their contradiction. He could describe them both, and\nwhatever he could describe seemed to him to be explained. Hence they\ndid not suggest to him any question either as to the nature of the\nobserved object or as to the possibility of observing it, such as\nmight have diverted philosophy from the method of self-observation.\nHe left them side by side, and, far from disguising either, put\nalongside of them another fact--the presence among the perpetually\nperishing ideas of that of a consciousness identical with itself, not\nmerely in different times and places, but in all times and places.\nSuch an idea, under the designation of an eternal wise Being, he was\n\u2018sure he had\u2019 (Book II. chap. xvii. sec. 14).\n\n[1] Cf. Book II. chap. xiv. sec. 32--\u2018by observing what passes in our\nminds, how our ideas there in train constantly some vanish and others\nbegin to appear, we come by the idea of succession; and by observing\na distance in the parts of this succession, we get the idea of\nduration\u2019--with chap. xv. sec. 12. \u2018Duration is the idea we have of\nperishing distance, of which no two parts exist together, but follow\neach other in succession.\u2019\n\nIs the idea of God possible to a consciousness given in time?\n\n136. The remark will at once occur that the question concerning the\nrelation between our consciousness, as in succession, and the idea\nof God, is essentially different from that concerning the relation\nbetween this consciousness and the self identical throughout it,\ninasmuch as the relation in the one case is between a fact and an\nidea, in the other between conflicting facts. The identity of the\nself, which Locke asserts, is one of \u2018real being,\u2019 and this is\nfound to lie in consciousness, in apparent conflict with the fact\nthat consciousness is a succession, of which \u2018no two parts exist\ntogether.\u2019 There is no such conflict, it will be said, between the\n_idea_ of a conscious being, who is the same yesterday, to-day, and\nfor ever--the correspondence to which of any reality is a farther\nquestion--and the _fact_ of our consciousness being in succession.\nAllowing for the moment the validity of this distinction, we will\nconsider first the difficulties that attach to Locke\u2019s account of the\nidea of God, as an idea.\n\nLocke\u2019s account of this idea.\n\n137. This idea, with him, is a \u2018complex idea of substance.\u2019 It is\nthe idea each man has of the \u2018thinking thing within him, enlarged\nto infinity.\u2019 It is beset then in the first place with all the\ndifficulties which we have found to belong to his doctrine of\nsubstance generally and of the thinking substance in particular.\n[1] These need not be recalled in detail. When God is the thinking\nsubstance they become more obvious. It is the antithesis to \u2018material\nsubstance,\u2019 as the source of ideas of sensation, that alone with\nLocke gives a meaning to \u2018thinking substance,\u2019 as the source of ideas\nof reflection: and if, as we have seen, the antithesis is untenable\nwhen it is merely the source of human ideas that is in question,\nmuch more must it be so in regard to God, to whom any opposition\nof material substance must be a limitation of his perfect nature.\nOf the generic element in the above definition, then, no more need\nhere be said. It is the qualification of \u2018enlargement to infinity,\u2019\nby which the idea of man as a thinking substance is represented as\nbecoming the idea of God, that is the special difficulty now before\nus. Of this Locke writes as follows:--\u2018The complex idea we have\nof God is made up of the simple ones we receive from reflection.\nIf I find that I know some few things, and some of them, or all\nperhaps, imperfectly, I can frame an idea of knowing twice as many:\nwhich I can double again as often as I can add to number, and thus\nenlarge my ideas of knowledge by extending its comprehension to all\nthings existing or possible. The same I can do of knowing them more\nperfectly, _i.e._ all their qualities, powers, causes, consequences,\nand relations; and thus frame the idea of infinite or boundless\nknowledge. The same also may be done of power till we come to that\nwe call infinite; and also of the duration of existence without\nbeginning or end; and so frame the idea of an eternal being. ...\nAll which is done by enlarging the simple ideas we have taken from\nthe operation of our own minds by reflection, or by our senses from\nexterior things, to that vastness to which infinity can extend them.\nFor it is infinity which joined to our ideas of existence, power,\nknowledge, &c., makes that complex idea whereby we represent to\nourselves the supreme being\u2019 (Book II. chap. xxiii. sec. 33--35).\nWhat is meant by this \u2018joining of infinity\u2019 to our ideas?\n\n[1] See above, paragraph 35 and the following, and 127 and the\nfollowing.\n\n\u2018Infinity,\u2019 according to Locke\u2019s account of it, only applicable to\nGod, if God has parts.\n\n138. \u2018Finite and infinite,\u2019 says Locke, \u2018are looked upon by the mind\nas the modes of quantity, and are to be attributed primarily only\nto those things that have parts and are capable of increase by the\naddition of any the least part\u2019 (Book II. chap. xvii. sec. 1). Such\nare \u2018duration and expansion.\u2019 The applicability then of the term\n\u2018infinite\u2019 in its proper sense to God implies that he has expansion\nor duration; and it is characteristic of Locke that though he was\nclear about the divisibility of expansion and duration, as the above\npassage shows, he has no scruple about speaking of them as attributes\nof God, of whom as being \u2018in his own essence simple and uncompounded\u2019\nhe would never have spoken as \u2018having parts.\u2019 \u2018Duration is the idea\nwe have of perishing distance, of which no parts exist together\nbut follow each other in succession; as expansion is the idea of\nlasting distance, all whose parts exist together.\u2019 Yet of duration\nand expansion, thus defined, he says that \u2018in their full extent\u2019\n(_i.e._ as severally \u2018eternity and immensity\u2019) \u2018they belong only to\nthe Deity\u2019 (Book II. chap. xv. secs. 8 and 12). \u2018A full extent\u2019 of\nthem, however, is in the nature of the case impossible. With a last\nmoment duration would cease to be duration; without another space\nbeyond it space would not be space. Locke is quite aware of this.\nWhen his conception of infinity is not embarrassed by reference to\nGod, it is simply that of unlimited \u2018addibility\u2019--a juxtaposition\nof space to space, a succession of time upon time, to which we can\nsuppose no limit so long as we consider space and time \u2018as having\nparts, and thus capable of increase by the addition of parts,\u2019 and\nwhich therefore excludes the very possibility of a totality or\n\u2018full extent\u2019 (Book II. chap. xvi. sec. 8, and xvii. sec. 13). The\nquestion, then, whether infinity of expansion and duration in this,\nits only proper, sense can be predicated of the perfect God, has\nonly to be asked in order to be answered in the negative. Nor do we\nmend the matter if, instead of ascribing such infinity to God, we\nsubstitute another phrase of Locke\u2019s, and say that He \u2018fills eternity\nand immensity\u2019 (Book II. chap. xv. sec. 8). Put for eternity and\nimmensity their proper equivalents according to Locke, viz. unlimited\n\u2018addibility\u2019 of times and spaces, and the essential unmeaningness of\nthe phrase becomes apparent.\n\nCan it be applied to him \u2018figuratively\u2019 in virtue of the indefinite\nnumber of His acts?\n\n139. In regard to any other attributes of God than those of his\nduration and expansion, [1] Locke admits that the term \u2018infinite\u2019 is\napplied \u2018figuratively\u2019 (Book II. chap. xvii. sec. 1). \u2018When we call\nthem (_e.g._ His power, wisdom, and goodness) infinite, we have no\nother idea of this infinity but what carries with it some reflection\non, or intimation of, that number or extent of the acts or objects of\nGod\u2019s wisdom, &c., which can never be supposed so great or so many\nwhich these attributes will not always surmount, let us multiply them\nin our thoughts as far as we can with all the infinity of endless\nnumber.\u2019 What determination, then, according to this passage, of our\nconception of God\u2019s goodness is represented by calling it infinite?\nSimply its relation to a number of acts and objects of which the sum\ncan always be increased, and which, just for that reason, cannot\nrepresent the perfect God. Is it then, it may be asked, of mere\nperversity that when thinking of God under attributes that are not\nquantitative, and therefore do not carry with them the necessity of\nincompleteness, we yet go out of our way by this epithet \u2018infinite\u2019\nto subject them to the conditions of quantity and its \u2018progressus ad\ninfinitum?\u2019\n\n[1] In the passages referred to, Locke speaks of \u2018duration and\n_ubiquity_.\u2019 The proper counterpart, however, of \u2018duration\u2019 according\nto him is \u2018expansion\u2019--this being to space what duration is to\ntime. Under the embarrassment, however, which necessarily attends\nthe ascription of expansion to God, he tacitly substitutes for it\n\u2018ubiquity,\u2019 a term which does not match \u2018duration,\u2019 and can only mean\npresence throughout the _whole_ of expansion, presence throughout the\nwhole of that which does not admit of a whole.\n\nAn act, finite in its nature, remains so, however often repeated.\n\n140. Retaining Locke\u2019s point of view, our answer of course must\nbe that our ideas of the Divine attributes, being primarily our\nown ideas of reflection, are either ideas of the single successive\nacts that constitute our inward experience or formed from these\nby abstraction and combination. In parts our experience is given,\nin parts only can we recall it. Our complex or abstract ideas are\nsymbols which only take a meaning so far as we resolve them into\nthe detached impressions which in the sum they represent, or recall\nthe objects, each with its own before and after, from which they\nwere originally taken. So it is with the ideas of wisdom, power, and\ngoodness, which from ourselves we transfer to God. They represent an\nexperience given in succession and piece-meal--a numerable series of\nacts and events, which like every other number is already infinite\nin the only sense of the word of which Locke can give a clear\naccount, as susceptible of indefinite repetition (Book II. chap.\nvi. sec. 8.) When we \u2018join infinity\u2019 to these ideas, then, unless\nsome other meaning is given to infinity, we merely state explicitly\nwhat was originally predicable of the experience they embody. Nor\nwill it avail us much to shift the meaning of infinite, as Locke\ndoes when he applies it to the divine attributes, from that of\nindefinite \u2018addibility\u2019 to that of exceeding any sum which indefinite\nmultiplication can yield us. Let us suppose an act of consciousness,\nfrom which we have taken an abstract idea of an attribute--say of\nwisdom--to be a million times repeated; our idea of the attribute\nwill not vary with the repetition. Nor if, having supposed a limit\nto the repetition, we then suppose the act indefinitely repeated\nbeyond this limit and accordingly speak of the attribute as infinite,\nwill our idea of the attribute vary at all from what it was to\nbegin with. Its content will be the same. There will be nothing\nto be said of it which could not have been said of the experience\nfrom which it was originally abstracted, and of which the essential\ncharacteristic--that it is one of a series of events of which no two\ncan be present together--is incompatible with divine perfection.\n\nGod only infinite in a sense in which time is _not_ infinite, and\nwhich Locke could not recognize ...\n\n141. It appears then that it is the subjection of our experience\nto the form of time which unfits the ideas derived from it for\nany combination into an idea of God; nor by being \u2018joined with\nan infinity,\u2019 which itself merely means the absence of limit to\nsuccession in time, is their unfitness in any way modified. On the\ncontrary, by such conjunction from being latent it becomes patent.\nIn one important passage Locke becomes so far aware of this that,\nthough continuing to ascribe infinite duration to God, he does it\nunder qualifications inconsistent with the very notion of duration.\n\u2018Though we cannot conceive any duration without succession, nor put\nit together in our thoughts that any being does now exist to-morrow\nor possess at once more than the present moment of duration; yet\nwe can conceive the eternal duration of the Almighty far different\nfrom that of man, or any other finite being: because man comprehends\nnot in his knowledge or power all past and future things ... what\nis once past he can never recall, and what is yet to come he cannot\nmake present. ... God\u2019s infinite duration being accompanied with\ninfinite knowledge and power, he sees all things past and to come\u2019\n(Book II. chap. xv. sec 12). It is clear that in this passage\n\u2018infinite\u2019 changes its meaning; that it is used in one sense--the\nproper sense according to Locke--when applied to duration, and in\nsome wholly different sense, not a figurative one derived from the\nformer, when applied to knowledge and power; and that the infinite\nduration of God, as \u2018accompanied by infinite power and knowledge,\u2019 is\nno longer in any intelligible sense duration at all. It is no longer\n\u2018the idea we have of perishing distance,\u2019 derived from our fleeting\nconsciousness in which \u2018what is once past can never be recalled,\u2019 but\nthe attribute of a consciousness of which, if it is to be described\nin terms of time at all, in virtue of its \u2018seeing all things past\nand to come\u2019 at once, it can only be said that it \u2018does now exist\nto-morrow.\u2019 If it be asked, What meaning can we have in speaking\nof such a consciousness? into what simple ideas can it be resolved\nwhen all our ideas are determined by a before and after?--the answer\nmust be, Just as much or as little meaning as we have when, in like\ncontradiction to the successive presentation of ideas, we speak\nof a self, constituted by consciousness, as identical with itself\nthroughout the years of our life.\n\n... the same sense in which the self is infinite.\n\n142. A more positive answer it is not our present business to give.\nOur concern is to show that \u2018eternity and immensity,\u2019 according to\nany meaning that Locke recognises, or that the observation of our\nideas could justify, do not express any conception that can carry\nus beyond the perpetual incompleteness of our experience; but that\nin his doctrine of personal identity he does admit a conception\nwhich no observation of our ideas of reflection--since these are\nin succession and could not be observed if they were not--can\naccount for; and that it is just this conception, the conception of\na constant presence of consciousness to itself incompatible with\nconditions of space and time, that can alone give such meaning to\n\u2018eternal and infinite\u2019 as can render them significant epithets of\nGod. Such a conception (we say it with respect) Locke admits when\nit is wanted without knowing it. It must indeed always underlie\nthe idea of God, however alien to it may be attempted adaptations\nof the other \u2018infinite\u2019--the _progressus ad indefinitum_ in space\nand time--by which, as with Locke, the idea is explained. But it is\none for which the psychological method of observing what happens in\noneself cannot account, and which therefore this method, just so far\nas it is thoroughly carried out, must tend to discard. That which\nhappens, whether we reckon it an inward or an outward, a physical or\na psychical event--and nothing but an event can, properly speaking,\nbe observed--is as such in time. But the presence of consciousness to\nitself, though, as the true \u2018punctum stans,\u2019 [1] it is the condition\nof the observation of events in time, is not such an event itself.\nIn the ordinary and proper sense of \u2018fact,\u2019 it is not a fact at all,\nnor yet a possible abstraction from facts. To the method, then, which\ndeals with phrases about the mind by ascertaining the observable\n\u2018mental phenomena\u2019 which they represent, it must remain a mere\nphrase, to be explained as the offspring of other phrases whose real\nimport has been misunderstood. It can only recover a significance\nwhen this method, as with Hume, has done its worst, and is found to\nleave the possibility of knowledge, without such \u2018punctum stans,\u2019\nstill unaccounted for.\n\n[1] Locke, Essay II. chap. xvii. sec. 16.\n\nHow do I know my own real existence?--Locke\u2019s answer.\n\n143. We have finally to notice the way in which Locke maintains\nour knowledge of the \u2018real existence\u2019 of thinking substance, both\nas that which \u2018we call our mind,\u2019 and as God. Of the former first.\n\u2018Experience convinces us that we have an intuitive knowledge of our\nown existence.... If I know I feel pain, it is evident I have as\ncertain perception of my own existence as of the pain I feel. If I\nknow I doubt, I have as certain perception of the existence of the\nthing doubting as of that thought which I call doubt\u2019 (Book IV. chap.\nix. sec. 3). Upon this the remark must occur that the existence of a\npainful feeling is one thing; the existence of a permanent subject,\nremaining the same with itself, when the feeling is over, and\nthrough the succession of other feelings, quite another. The latter\nis what is meant by my own existence, of which undoubtedly there\nis a \u2018certain perception,\u2019 if the feeling of pain has become the\n\u2018knowledge that I feel pain,\u2019 and if by the \u2018I\u2019 is understood such\na permanent subject. That the feeling, as \u2018simple idea,\u2019 is taken\nto begin with by Locke for the knowledge that I feel something, we\nhave sufficiently seen. [1] Just as, in virtue of this conversion,\nit gives us \u2018assurance\u2019 of the real existence of the outer thing or\nmaterial substance on the one side, so of the thinking substance\non the other. It carries with it the certainty at once that I have\na feeling, and that something makes me feel. But whereas, after\nthe conversion of feeling into a felt thing has been throughout\nassumed--as indeed otherwise feeling could not be spoken of--a\nfurther question is raised, which causes much embarrassment, as to\nthe real existence of such thing; on the contrary, the reference of\nthe feeling to the _thinking_ thing is taken as carrying with it the\nreal existence of such thing. The question whether it really exists\nor no is only once raised, and then summarily settled by the sentence\nwe have quoted, while the reality whether of existence or of essence\non the part of the outward thing, as we have found to our cost, is\nthe main burden of the Third and Fourth Books.\n\n[1] See above, paragraphs 26 and following, and 59 and following.\n\nIt cannot be known consistently with Locke\u2019s doctrine of real\nexistence.\n\n144. In principle, indeed, the answer to both questions, as given\nby Locke, is the same: for the reasons which he alleges for being\nassured of the \u2018existence of a thing without us corresponding to\nthe idea of sensation\u2019 reduce themselves, as we have seen, to the\nreiteration of that reference of the idea to a thing, which according\nto him is originally involved in it, and which is but the correlative\nof its reference to a subject. This, however, is what he was not\nhimself aware of. To him the outer and the inner substance were\nseparate and independent things, for each of which the question of\nreal existence had to be separately settled. To us, according to the\nview already indicated, it is the presence of self-consciousness, or\nthought as an object-to-itself, to feeling that converts it into a\nrelation between feeling thing and felt thing, between \u2018cogitative\nand incogitative substance.\u2019 The source of substantiation upon\neach side being the same, the question as to the real existence of\neither substance must be the same, and equally so the answer to it.\nIt is an answer that must be preceded by a counter question.--Does\nreal existence mean existence independent of thought? To suppose\nsuch existence is to suppose an impossibility--one which is not the\nless so though the existence be supposed material, if \u2018material\u2019\nmeans in \u2018space\u2019 and space itself is a relation constituted by the\nmind, \u2018bringing things to and setting them by one another.\u2019 Yet is\nthe supposition itself but a mode of the logical substantiation\nwe have explained, followed by an imaginary abstraction of the\nwork of the mind from this, its own creation. Does real existence\nmean a possible feeling? If so, it is as clear that what converts\nfeeling into a relation between felt thing and feeling subject\ncannot in this sense be real, as it is that without such conversion\nno distinction between real and fantastic would be possible. Does\nit, finally, mean individuality, in such a sense that unless I can\nsay this or that is substance, thinking or material, substance does\nnot really exist? If it does, the answer is that substance, being\nconstituted by a relation by which self-conscious thought is for ever\ndetermining feelings, and which every predication represents, cannot\nbe identified with any \u2018this or that,\u2019 though without it there could\nbe no \u2018this or that\u2019 at all.\n\nBut he ignores this in treating of the self.\n\n145. We have already found that Locke accepts each of the above as\ndeterminations of real existence, and that, though in spite of them\nhe labours to maintain the real existence of outward things, he is\nso far faithful to them as to declare real essence unknowable. In\nanswering the question as to \u2018his own existence\u2019 he wholly ignores\nthem. He does not ask how the real existence of the thinking Ego\nsorts with his ordinary doctrine that the real is what would be in\nthe world whether there were a mind or no; or its real identity,\npresent throughout the particulars of experience, with his ordinary\ndoctrine of the fictitiousness of \u2018generals.\u2019 A real existence of the\nmind, however, founded on the logical necessity of substantiation,\nrests on a shifting basis, so long as by the mind is understood a\nthinking thing, different in each man, to which his inner experience\nis referred as accidents to a substance. The same law of thought\nwhich compels such reference requires that the thinking thing in\nits turn, as that which is born grows and dies, be referred as an\naccident to some ulterior substance. \u2018A fever or fall may take away\nmy reason or memory, or both; and an apoplexy leave neither sense\nnor understanding, no, nor life.\u2019 [1] Just as each outer thing turns\nout to be a \u2018retainer to something else,\u2019 so is it with the inner\nthing. Such a dependent being cannot be an ultimate substance; nor\ncan any natural agents to which we may trace its dependence really\nbe so either. The logical necessity of further substantiation would\naffect them equally, appearing in the supposition of an unknown\nsomething beyond, which makes them what they are. It is under such\nlogical necessity that Locke, in regard to all the substances which\nhe commonly speaks of as ultimate--God, spirit, body--from time\nto time gives warning of something still ulterior and unknowable,\nwhether under the designation of substance or real essence (Book II.\nchap. xxiii. secs. 30 and 36). If, then, it will be said, substance\nis but the constantly-shifting result of a necessity of thought--so\nshifting that there is nothing of which we can finally say, \u2018This\nis substance, not accident\u2019--there can be no evidence of the \u2018real\nexistence\u2019 of a permanent Ego in the necessary substantiation therein\nof my inner experience.\n\n[1] Locke, Book III. chap. vi. sec. 4.\n\nSense in which the self is truly real.\n\n146. The first result of such a consideration in a reader of Locke\nwill naturally be an attempt to treat the inner synthesis as a\nfiction of thought or figure of speech, and to confine real existence\nto single feelings in the moments of their occurrence. This, it\nwill seem, is to be faithful to Locke\u2019s own clearer mind, as it\nfrequently emerges from the still-returning cloud of scholasticism.\nThe final result will rather be the discovery that the single feeling\nis nothing real, but that the synthesis of appearances, which alone\nfor us constitutes reality, is never final or complete: that thus\nabsolute reality, like ultimate substance, is never to be found by\nus--in a thinking as little as in a material thing--belonging as it\ndoes only to that divine self-consciousness, of which the presence\nin us is the source and bond of the ever-growing synthesis called\nknowledge, but which, because it is the source of that synthesis and\nnot one of its partial results, is neither real nor knowable in the\nsame sense as is any other object. It is this presence which alone\ngives meaning to \u2018proofs of the being of God;\u2019 to Locke\u2019s among\nthe rest. For it is in a sense true, as he held, that \u2018my own real\nexistence\u2019 is evidence of the existence of God, since the self, in\nthe only sense in which it is absolutely real or an ultimate subject,\nis already God. [1]\n\n[1] See below, paragraph 152.\n\nLocke\u2019s proof of the real existence of God. There must have been\nsomething from eternity to cause what now is.\n\n147. Our knowledge of God\u2019s existence, according to him, is\n\u2018demonstrative,\u2019 based on the \u2018intuitive\u2019 knowledge of our own.\nStrictly taken, according to his definitions, this must mean that the\nagreement of the idea of God with existence is perceived mediately\nthrough the agreement of the idea of self with existence, which is\nperceived immediately; that thus the idea of God and the idea of\nself \u2018agree\u2019. [1] We need not, however, further dwell either on\nthe contradiction implied in the knowledge of real existence, if\nknowledge is a perception of agreement between ideas and if real\nexistence is the antithesis of ideas; or on the embarrassments which\nfollow when a definition of reasoning, only really applicable to the\ncomparison of quantities, is extended to other regions of knowledge.\nLocke virtually ignores his definitions in the passage before us. \u2018If\nwe know there is some real being\u2019 (as we do know in the knowledge\nof our own existence) \u2018and that non-entity cannot produce any real\nbeing, it is an evident demonstration that from eternity there has\nbeen something; since what was not from eternity had a beginning,\nand what had a beginning must be produced by something else\u2019 (Book\nIV. chap. x. sec. 3). Next as to the qualities of this something\nelse. \u2018What had its being and beginning from another must also have\nall that which is in, and belongs to, its being from another too\u2019\n(Ibid, sec. 4.). From this is deduced the supreme power and perfect\nknowledge of the eternal being upon the principle that whatever is in\nthe effect must also be in the cause--a principle, however, which has\nto be subjected to awkward limitations in order that, while proving\nenough, it may not prove too much, it might seem that, according to\nit, since the real being, from which as effect the eternal being as\ncause is demonstrated, is \u2018both material and cogitative\u2019 or \u2018made up\nof body and spirit,\u2019 matter as well as thought must belong to the\neternal being too. That thought must belong to him, Locke is quite\nclear. It is as impossible, he holds, that thought should be derived\nfrom matter, or from matter and motion together, as that something\nshould be derived from nothing. \u2018If we will suppose nothing first\nor eternal, matter can never begin to be: if we suppose bare matter\nwithout motion eternal, motion can never begin to be: if we suppose\nonly matter and motion first or eternal, thought can never begin\nto be\u2019 (Book IV. chap. x. sec. 10). The objection which is sure to\noccur, that it must be equally impossible for matter to be derived\nfrom thought, he can scarcely be said to face. He takes refuge in the\nsupreme power of the eternal being, as that which is able to create\nmatter out of nothing. He does not anticipate the rejoinder to which\nhe thus lays himself open, that this power in the eternal being to\nproduce one effect not homogeneous with itself, viz. matter, may\nextend to another effect, viz. thought, and that thus the argument\nfrom thought in the effect to thought in the cause becomes invalid,\nand nothing but blind power, we know not what, remains as the\nattribute of the eternal being. Nor does he remember, when he meets\nthe objection drawn from the inconceivability of matter being made\nout of nothing by saying that what is inconceivable is not therefore\nimpossible (_ibid_. sec. 19), that it is simply the inconceivability\nof a sequence of something upon nothing that has given him his\n\u2018evident demonstration\u2019 of an eternal being.\n\n[1] See above, paragraphs 25 and 24.\n\nHow \u2018eternity\u2019 must be understood if this argument is to be valid:\n\n148. The value of the first step in Locke\u2019s argument--the inference,\nnamely, from there being something now to there having been something\nfrom eternity--must be differently estimated according to the meaning\nattached to \u2018something\u2019 and \u2018from eternity.\u2019 If the existence of\nsomething means the occurrence of an event, of this undoubtedly\nit can always be said that it follows another event, nor to this\nsequence can any limit be supposed, for a first event would not be\nan event at all. It would be a contingency contingent upon nothing.\nThus understood, the argument from a something now to a something\nfrom eternity is merely a statement of the infinity of time according\nto that notion of infinity, as a \u2018progressus ad indefinitum,\u2019 which\nwe have already seen to be Locke\u2019s. [1] It is the exact reverse of\nan argument to a creation or a first cause. If we try to change\nits character by a supplementary consideration that infinity in\nthe series of events is inconceivable, the rejoinder will be that\na first event is not for that reason any less of a contradiction,\nand that the infinity which Locke speaks of only professes to be a\nnegative idea, representing the impossibility of conceiving a first\nevent (Book II. chap. xvii. sec. 13, &c.). In truth, however, when\nLocke speaks of \u2018something from eternity\u2019 he does not mean--what\nwould clearly be no God at all--a series of events to which, because\n_of events_, and therefore in time, no limit can be supposed; but a\nbeing which is neither event nor series of events, to which there is\nno before or after. The inference to such a being is not of a kind\nwith the transition from one event to another habitually associated\nwith it; and if this be the true account of reasoning from effect to\ncause, no such reasoning can yield the result which Locke requires.\nAs we have seen, however, this is not his account of it, [2] however\nlegitimately it may follow from his general doctrine.\n\n[1] See above, paragraph 138.\n\n[2] See above, paragraph 68.\n\n... and how \u2018cause\u2019.\n\n149. The inference of cause with him is the inference from a change\nto something having power to produce it. [1] The value of this\ndefinition lies not in the notion of efficient power, but in that\nof an order of nature, which it involves. If instead of \u2018something\nhaving power to produce it\u2019 we read \u2018something that accounts for the\nchange,\u2019 it expresses the inference on which all science rests, but\nwhich is as far as possible from being merely a transition from one\nevent to another that usually precedes it. An event, interpreted as\na change of something that remains constant, is no longer a mere\nevent. It is no longer merely in time, a present which next moment\nbecomes a past. It takes its character from relation to the thing or\nsystem of things of which it is an altered appearance, but which in\nitself is always the same. Only in virtue of such a relation does it\nrequire to be accounted for, to be referred to a \u2018cause\u2019 which is in\ntruth the conception that holds together or reconciles the endless\nflux of events with eternal unity. The cause of a \u2018phenomenon,\u2019 even\naccording to the authoritative exponent of the Logic which believes\nitself to follow Hume, is the \u2018sum total of its conditions.\u2019 In its\nfulness, that is, it is simply that system of things, conceived\nexplicitly, of which there must already have been an implicit\nconception in order that the event might be regarded as a change\nand thus start the search for a cause. An event in time, apart from\nreference to something not in time, could suggest no enquiry into\nthe sum of its conditions. Upon occurrence of a certain feeling\nthere might indeed be spontaneous recollection of a feeling usually\nprecedent, spontaneous expectation of another usually sequent. But\nsuch association of feelings can never explain that conception of\ncause in virtue of which, when accounting for a phenomenon, we set\naside the event which in our actual experience has usually preceded\nit, for one which we only find to precede it in the single case of\na crucial experiment. That we do so shows that it is not because of\nantecedence in time, however apparently uniform, that an educated man\nreckons a certain event to be the cause of another, but that, because\nof its sole sufficiency under the sum of known conditions to account\nfor the given event, he decides it to be its uniform antecedent,\nhowever much ordinary appearances may tell to the contrary. Thus,\nthough he may still strangely define cause as a uniformly antecedent\nevent (in spite of its being a definition that would prevent him\nfrom speaking of gravity as the cause of the fall of a stone), it\nis clear that by such event he means one determined by a complex of\nconditions in an unchanging universe. These conditions, again, he may\nspeak of as contingencies, _i.e._ as events contingent upon other\nevents in endless series, but he must add \u2018contingent in accordance\nwith the uniformity of nature\u2019--in other words, he must determine the\ncontingencies by relation to what is not contingent; he must suppose\nnature unchanging, though our experience of it through sensation\nbe a \u2018progressus ad indefinitum\u2019--if he is to allow a possibility\nof knowledge at all. In short, if events were merely events,\nfeelings that happen to me now and next moment are over, no \u2018law\nof causation\u2019 and therefore no knowledge would be possible. If the\nknowledge founded on this law actually exists, then the \u2018argumentum\na contingenti\u00e2 mundi\u2019 rightly understood--the \u2018inference\u2019 from\nnature to a being neither in time nor contingent but self-dependent\nand eternal, that constant reality of which events are the changing\nappearances--is valid because the conception of nature, of a\nworld to be known, already implies such a being. To the rejoinder\nthat implication in the conception of nature does not prove real\nexistence, the answer must be the question. What meaning has real\nexistence, the antithesis of illusion, except such as is equivalent\nto this conception?\n\n[1] cf. Book II. chap. xxvi. sec. 1, and chap. xxi. sec. 1.\n\nThe world which is to prove an eternal God must be itself eternal.\n\n150. The value, then, of Locke\u2019s demonstration of the existence of\nGod, as an argument from there being something now to an eternal\nbeing from which the real existence that we know \u2018has all which is in\nand belongs to it,\u2019 depends on our converting it into the \u2018argumentum\na contingenti\u00e2 mundi,\u2019 stated as above. In other words, it depends on\nour interpreting it in a manner which may be warranted by his rough\naccount of causation, and by one of the incompatible views of the\nreal that we have found in him, [1] but which is inconsistent with\nhis opposition of reality to the work of the mind, and his reduction\nof it to \u2018particular existence,\u2019 as well as with his ordinary view\nthat \u2018infinite\u2019 and \u2018eternal\u2019 can represent only a \u2018progressus ad\nindefinitum.\u2019 If by \u2018real existence corresponding to an idea\u2019 is\nmeant its presentation in a particular \u2018here and now,\u2019 an attempt\nto find a real existence of God can bring us to nothing but such a\ncontradiction in terms as a first event. To prove it from the real\nexistence of the self is to prove one impossibility from another.\nIf, on the other hand, real existence implies the determination\nof our ideas by an order of nature--if it means ideas \u2018in ordine\nad universum\u2019 (to use a Baconian phrase), in distinction from \u2018in\nordine ad nos\u2019--then the argument from a present to an eternal real\nexistence is valid, but simply in the sense that the present is\nalready real, and \u2018has all that is in and belongs to it,\u2019 only in\nvirtue of the relation to the eternal.\n\n[1] See above, paragraphs 49 and 91.\n\nBut will the God, whose existence is so proven, be a thinking being?\n\n151. This, it may be said, is to vindicate Locke\u2019s \u2018proof\u2019 only\nby making it Pantheistic. It gives us an eternity of nature, but\nnot God. Our present concern, however, is not with the distinction\nbetween Pantheism and true Theism, but with the exposition of Locke\u2019s\ndoctrine according to the only development by which it can be made\nto show the real existence of an eternal being at all. It is only\nby making the most of certain Cartesian elements that appear in his\ndoctrine, irreconcileable with its general purport, that we can find\nfair room in it for such a being, even as the system of nature.\nAny attempt to exhibit (in Hegelian phrase) \u2018Spirit as the truth\nof nature,\u2019 would be to go wholly beyond our record; yet without\nthis the \u2018ens realissimum\u2019 cannot be the God whose existence Locke\nbelieves himself to prove--a _thinking_ being from whom matter and\nmotion are derived, but in whom they are not. It is true that,\naccording to the context, it is the real existence of the self from\nwhich that of the eternal being is proved. This is because, in the\nFourth Book, where the \u2018proof\u2019 occurs, following the new train of\nenquiry started by the definition of knowledge, Locke has for the\ntime left in abeyance his fundamental doctrine that all simple ideas\nare types of reality, and is writing as if \u2018my own real existence\u2019\nwere the only one known with intuitive certainty. This, however,\nmakes no essential difference in the effect of his argument. The\ngiven existence, from which the divine is proved, is treated\nexpressly as _both_ \u2018material and cogitative:\u2019 nor, since according\nto Locke the world is both and man is both, and even the \u2018thinking\nthing\u2019 takes its content from impressions made by matter, could it be\notherwise. To have taken thought by itself as the basis of the proof\nwould have been to leave the other part of the world, as he conceived\nit, to be referred to another God. The difficulty then arises, either\nthat there is no inference possible from the nature of the effect\nto the nature of the eternal being, its cause; in which case no\nattribute whatever can be asserted of the latter: or that to it too,\nlike the effect, matter as well as thought must belong.\n\nYes, according to the true notion of the relation between thought and\nmatter.\n\n152. As we have seen, neither of these alternative views is really\nmet by Locke. To the former we may reply that the relation between\ntwo events, of which neither has anything in common with the other,\nbut which we improperly speak of as effect and cause (_e.g._ death\nand a sunstroke), has no likeness to that which we have explained\nbetween the world in its contingency and the world as an eternal\nsystem--a relation according to which the cause is the effect in\nunity. Whatever is part of the reality of the world must belong, it\nwould seem, to the \u2018ens realissimum,\u2019 its cause. We are thus thrown\nback on the other horn of the dilemma. Is not matter part of the\nreality of the world? This is a question to which the method of\nobserving the individual consciousness can give none but a delusive\nanswer. A true answer cannot be given till for this method has been\nsubstituted the enquiry, How knowledge is possible, and it has been\nfound that it is only possible as the progressive actualisation in\nus of a self-consciousness in itself complete, and which in its\ncompleteness includes the world as its object. From the point of view\nthus attained the question as to matter will be, How is it related to\nthis self-consciousness?--a question to which the answer must vary\naccording to what is understood by \u2018matter.\u2019 If it means the abstract\nopposite of thought--that which is supposed void of all determination\nthat comes of thinking--we must pronounce it simply a delusion, the\ncreation of self-consciousness in one stage of its communication to\nus. If it means the world as in space and time, this we may allow to\nbe real enough as a stage in the process by which self-consciousness\nconstitutes reality. Thus understood, we may speak of it roughly as\npart of the \u2018ens realissimum\u2019 which the complete self-consciousness,\nor God, includes as its object, without any limitation of the divine\nperfectness. The limitation only seems to arise so far as we, being\nourselves (as our knowledge and morality testify), though formally\nself-conscious, yet parts of this partial world, interpret it amiss\nand ascribe to it a reality, in abstraction from the self-conscious\nsubject, which it only derives from relation to it. Thus while on\nthe one hand it is the presence in us of God, as the self-conscious\nsource of reality, that at once gives us the idea of God and of an\neternal self, and renders superfluous the further question as to\ntheir real existence; on the other hand it is because, for all this\npresence, we are but emerging from nature, of which as animals we are\nparts, that to us there must seem an incompatibility of existence\nbetween God and matter, between the self and the flux of events which\nmakes our life. This necessary illusion is our bondage, but when the\nsource of illusion is known, the bondage is already being broken.\n\nLocke\u2019s antinomies--Hume takes one side of them as true.\n\n153. We have now sufficiently explored the system which it was Hume\u2019s\nmission to try to make consistent with itself. We have found that it\nis governed throughout by the antithesis between what is given to\nconsciousness--that in regard to which the mind is passive--as the\nsupposed real on the one side, and what is \u2018invented,\u2019 \u2018created,\u2019\n\u2018superinduced\u2019 by the mind on the other: while yet this \u2018real\u2019 in all\nits forms, as described by Locke, has turned out to be constituted by\nsuch ideas as, according to him, are not given but invented. Stripped\nof these superinductions, nothing has been found to remain of it\nbut that of which nothing can be said--a chaos of unrelated, and\ntherefore unmeaning, _individua_. Turning to the theory of the mind\nitself, the source of the superinduction, we have found this to be a\nreduplication of the prolonged inconsistency which forms the theory\nof the \u2018real.\u2019 It impresses itself with that which, according to the\nother theory, is the impress of matter, and it really exists as that\nwhich it itself invents. The value of Hume\u2019s philosophy lies in its\nbeing an attempt to carry out the antithesis more rigorously--to\nclear the real, whether under the designation of mind or of its\nobject, of all that could not be reckoned as given in feelings which\noccur to us \u2018whether we will or no.\u2019 The consequence is a splendid\nfailure, a failure which it might have been hoped would have been\ntaken as a sufficient proof that a theory, which starts from that\nantithesis, cannot even be stated without implicitly contradicting\nitself.\n\nHume\u2019s scepticism fatal to his own premises. This derived from\nBerkeley.\n\n154. Such a doctrine--a doctrine founded on the testimony of\nthe senses, which ends by showing that the senses testify to\nnothing--cannot be criticised step by step according to the order in\nwhich its author puts it, for its characteristic is that, in order\nto state itself, it has to take for granted popular notions which it\nafterwards shows to be unmeaning. Its power over ordinary thinkers\nlies just in this, that it arrives at its destructive result by\nmeans of propositions which every one believes, but to the validity\nof which its result is really fatal. An account of our primitive\nconsciousness, which derives its plausibility from availing itself of\nthe conceptions of cause and substance, is the basis of the argument\nwhich reduces these conceptions to words misunderstood. It cannot,\ntherefore, be treated by itself, as it stands in the first part of\nthe Treatise on the Understanding, but must be taken in connection\nwith Part IV., especially with the section on \u2018Scepticism with regard\nto the Senses;\u2019 not upon the plan of discrediting a principle by\nreference to the \u2018dangerous\u2019 nature of its consequences, but because\nthe final doctrine brings out the inconsistencies lurking in that\nassumed to begin with. On this side of his scepticism Hume mainly\nfollowed the orthodox Berkeley, of whose criticism of Locke, made\nwith a very different purpose, some account must first be given.\nThe connection between the two authors is instructive in many ways;\nnot least as showing that when the most pious theological purpose\nexpresses itself in a doctrine resting on an inadequate philosophical\nprinciple, it is the principle and not the purpose that will regulate\nthe permanent effect of the doctrine.\n\nBerkeley\u2019s religious interest in making Locke consistent.\n\n155. Berkeley\u2019s treatises, we must remember, though professedly\nphilosophical, really form a theological polemic. He wrote as the\nchampion of orthodox Christianity against \u2018mathematical atheism,\u2019\nand, like others of his order, content with the demolition of the\nrival stronghold, did not stay to enquire whether his own untempered\nmortar could really hold together the fabric of knowledge and\nrational religion which he sought to maintain. He found practical\nungodliness and immorality excusing themselves by a theory of\n\u2018materialism\u2019--a theory which made the whole conscious experience of\nman dependent upon \u2018unperceiving matter.\u2019 This, whatever it might\nbe, was not an object which man could love or reverence, or to\nwhich he could think of himself as accountable. Berkeley, full of\ndevout zeal for God and man, and not without a tincture of clerical\nparty-spirit (as appears in his heat against Shaftesbury, whom he\nought to have regarded as a philosophical yoke-fellow), felt that\nit must be got rid of. He saw, or thought he saw, that the \u2018new\nway of ideas\u2019 had only to be made consistent with itself, and the\noppressive shadow must vanish. Ideas, according to that new way (or,\nto speak less ambiguously, feelings) make up our experience, and\nthey are not matter. Let us get rid, then, of the self-contradictory\nassumption that they are either copies of matter--copies of that,\nof which it is the sole and simple differentia that it is not an\nidea, or its effects--effects of that which can only be described\nas the unknown opposite of the only efficient power with which we\nare acquainted--and what becomes of the philosopher\u2019s blind and\ndead substitute for the living and knowing God? It was one thing,\nhowever, to show the contradictions involved in Locke\u2019s doctrine of\nmatter, another effectively to replace it. To the latter end Berkeley\ncannot be said to have made any permanent contribution. That explicit\nreduction of ideas to feelings \u2018particular in time,\u2019 which was his\ngreat weapon of destruction, was incompatible with his doing so.\nHe adds nothing to the philosophy, which he makes consistent with\nitself, while by making it consistent he empties it of three parts of\nits suggestiveness. His doctrine, in short, is merely Locke purged,\nand Locke purged is no Locke.\n\nWhat is meant by relation of mind and matter?\n\n156. The question which he mainly dealt with may be stated in general\nterms as that of the relation between the mind and the external\nworld. Under this general statement, however, are covered several\ndistinct questions, the confusion between which has been a great\nsnare for philosophers--questions as to the relations _(a)_ between\na sensitive and non-sensitive body, _(b)_ between thought and its\nobject, _(c)_ between thought and something only qualified as the\nnegation of thought. The last question, it will be observed, is what\nthe second becomes upon a certain notion being formed of what the\nobject of thought must be. Upon this notion being discarded a further\nquestion _(d)_, also covered by the above general statement, must\nstill remain as to the relation between thought, as in each man, and\nthe world which he does not make, but which, in some sort, makes\nhim what he is. In what follows, these questions, for the sake of\nbrevity, will be referred to symbolically.\n\nConfusions involved in Locke\u2019s materialism.\n\n157. Locke\u2019s doctrine of matter, as we have seen, involves a\nconfusion between _(a)_ and _(b)_. The feeling of touch in virtue of\nan intellectual interpretation--_intellectual_ because implying the\naction of the mind as (according to Locke) the source of ideas of\nrelation--becomes the idea of solidity, _i.e._ the idea of a relation\nbetween bodies in the way of impulse and resistance. But the function\nof the intellect in constituting the relation is ignored. Under cover\nof the ambiguous \u2018idea,\u2019 which stands alike for a nervous irritation\nand the intellectual interpretation thereof, the feeling of touch\nand conception of solidity are treated as one and the same. Thus the\ntrue _conceived_ outwardness of body to body--an outwardness which\nthought, as the source of relations, can alone constitute--becomes\nfirst an imaginary _felt_ outwardness of body to the organs of\ntouch, and then, by a further fallacy--these organs being confused\nwith the mind--an outwardness of body to mind, which we need only\nkick a stone to be sure of. Meanwhile the consideration of question\n_(d)_ necessitates the belief that the real world does not come and\ngo with each man\u2019s fleeting consciousness, and no distinction being\nrecognised between consciousness as fleeting and consciousness as\npermanent, or between feeling and thought, the real world comes to\nbe regarded as the absolute opposite of thought and its work. This\nopposition combines with the supposed externality of body to mind to\ngive the notion that body is the real. The qualities which \u2018the mind\nfinds inseparable from body\u2019 thus become qualities which would exist\nall the same \u2018whether there were a perceiving mind or no,\u2019 and are\nprimarily real; while such as consist in our feelings, though real\nin so far as, \u2018not being of our own making, they imply the action\nof things without us,\u2019 are yet only secondarily so because this\naction is relative to something which is not body. Then, finally, by\na renewed confusion of the relation between thought and its object\nwith that between body and body, qualities, which are credited with\na primary reality as independent of and antithetical to the mind,\nare brought within it again as ideas. They are supposed to copy\nthemselves upon it by impact and impression; and that not in touch\nmerely, but (visual feelings being interpreted by help of the same\nconception) in sight also.\n\nTwo ways of dealing with it. Berkeley chooses the most obvious.\n\n158. Such \u2018materialism\u2019 invites two different methods of attack.\nOn the one hand its recognised principle, that all intellectual\n\u2018superinduction\u2019 upon simple feeling is a departure from the real,\nmay be insisted on, and it may be shown that it is only by such\nsuperinduction that simple feeling becomes a feeling of body. Matter,\nthen, with all its qualities, is a fiction except so far as these\ncan be reduced to simple feelings. Such in substance was Berkeley\u2019s\nshort method with the materialists. In his early life it seemed\nto him sufficient for the purposes of orthodox \u2018spiritualism,\u2019\nbecause, having posed the materialist, he took the moral and\nspiritual attributes of God as \u2018revealed,\u2019 without enquiring into the\npossibility of such revelation to a merely sensitive consciousness.\nAs he advanced, other questions, fatal to the constructive value of\nhis original method, began to force themselves upon him. Granting\nthat intellectual superinduction = fiction, how is the fiction\npossible to a mind which cannot originate? Exclude from reality all\nthat such fiction constitutes, and what remains to be real? These\nquestions, however, though their effect on his mind appears in the\nlater sections of his \u2018Siris,\u2019 he never systematically pursued. He\nthus missed the true method of attack on materialism--the only one\nthat does not build again that which it destroys--the method which\nallows that matter is real but only so in virtue of that intellectual\nsuperinduction upon feeling without which there could be for us no\nreality at all: that thus it is indeed opposed to thought, but only\nby a position which is thought\u2019s own act. For the development of\nsuch views Berkeley had not patience in his youth nor leisure in his\nmiddle life. Whatever he may have suggested, all that he logically\nachieved was an exposure of the equivocation between feeling and felt\nbody; and of this the next result, as appears in Hume, was a doctrine\nwhich indeed delivers mind from dependence on matter, but only by\nreducing it in effect to a succession of feelings which cannot know\nthemselves.\n\nHis account of the relation between visible and tangible extension.\nWe do not see bodies without the mind ...\n\n159. It was upon the extension of the metaphor of impression to\nsight as well as touch, and the consequent notion that body, with\nits inseparable qualities, revealed itself through both senses,\nthat Berkeley first fastened. Is it evident, as Locke supposed it\nto be, that men \u2018perceive by their sight\u2019 not colours merely, but\n\u2018a distance between bodies of different colours and between parts\nof the same body\u2019; [1] in other words, situation and magnitude? To\nshow that they do not is the purpose of Berkeley\u2019s \u2018Essay towards a\nnew Theory of Vision.\u2019 He starts from two principles which he takes\nas recognised: one, that the \u2018proper and immediate object of sight\nis colour\u2019; the other, that distance from the eye, or distance in\nthe line of vision, is not immediately seen. If, then, situation and\nmagnitude are \u2018properly and immediately\u2019 seen, they must be qualities\nof colour. Now in one sense, according to Berkeley, they are so: in\nother words, there is such a thing as _visible_ extension. We see\nlights and colours in \u2018sundry situations\u2019 as well as \u2018in degrees\nof faintness and clearness, confusion and distinctness.\u2019 (_Theory\nof Vision_, sec. 77.) We also see objects as made up of certain\n\u2018quantities of coloured points,\u2019 _i.e._ as having visible magnitude.\n(Ibid. sec. 54.) But situation and magnitude _as visible_ are not\nexternal, not \u2018qualities of body,\u2019 nor do they represent by any\n_necessary_ connection the situation and magnitude that are truly\nqualities of body, the mind, \u2018without the mind and at a distance.\u2019\nThese are tangible. Distance in all its forms--as distance from the\neye; as distance between parts of the same body, or magnitude; and\nas distance of body from body, or situation--is tangible. What a man\nmeans when he says that \u2018he sees this or that thing at a distance\u2019 is\nthat \u2018what he sees suggests to his understanding that after having\npassed a certain distance, to be measured by the motion of his body\nwhich is perceivable by touch, he shall come to perceive such and\nsuch tangible ideas which have been usually connected with such and\nsuch visible ideas\u2019 (Ibid. sec. 45). On the same principle we are\nsaid to see the magnitude and situation of bodies. Owing to long\nexperience of the connection of these tangible ideas with visible\nones, the magnitude of the latter and their degrees of faintness and\nclearness, of confusion and distinctness, enable us to form a \u2018sudden\nand true\u2019 estimate of the magnitude of the former (_i.e._ of bodies);\neven as visible situation enables us to form a like estimate of the\n\u2018situation of things outward and tangible\u2019 (Ibid. secs. 56 and 99).\nThe connection, however, between the two sets of ideas, Berkeley\ninsists, is habitual only, not necessary. As Hume afterwards said of\nthe relation of cause and effect, it is not constituted by the nature\nof the ideas related. [2] The visible ideas, that as a matter of fact\n\u2018suggest to us the various magnitudes of external objects before we\ntouch them, might have suggested no such thing.\u2019 That would really\nhave been the case had our eyes been so framed as that the _maximum\nvisibile_ should be less than the _minimum tangibile_; and, as a\nmatter of constant experience, the greater visible extension suggests\nsometimes a greater, sometimes a less, tangible extension according\nto the degree of its strength or faintness, \u2018being in its own nature\nequally fitted to bring into our minds the idea of small or great\nor no size at all, just as the words of a language are in their own\nnature indifferent to signify this or that thing, or nothing at all.\u2019\n(Ibid. secs. 62-64.)\n\n[1] Locke, Essay Book II. chap. xiii. sec. 2.\n\n[2] See below, paragraph 283\n\n... nor yet feel them. The \u2018esse\u2019 of body is the \u2018percipi\u2019.\n\n160. So far, then, the conclusion merely is that body as external,\nand space as a relation between bodies or parts of a body, are not\nboth seen and felt, but felt only; in other words, that it is only\nthrough the organs of touch that we receive, strictly speaking,\nimpressions from without. This is all that the Essay on Vision goes\nto show; but according to the \u2018Principles of Human Knowledge\u2019 this\nconclusion was merely provisional. The object of touch does not,\nany more than the object of sight, \u2018exist without the mind,\u2019 nor is\nit \u2018the image of an external thing.\u2019 \u2018In strict truth the ideas of\nsight, when by them we apprehend distance and things placed at a\ndistance, do not suggest or mark out to us things actually existing\nat a distance, but only admonish us what ideas of touch will be\nimprinted in our minds at such and such distances of time, and in\nconsequence of such and such actions\u2019 (\u2018Principles of H. K.\u2019 sec.\n44). Whether, then, we speak of visible or tangible objects, the\nobject _is_ the idea, its \u2018esse is the percipi.\u2019 Body is not a thing\nseparate from the idea of touch, yet revealed by it; so far as it\nexists at all, it must either be that idea or be a succession of\nideas of which that idea is suggestive. It follows that the notion of\nthe real which identifies it with matter, as something external to\nand independent of consciousness, and which derives the reality of\nideas from their relation to body as thus outward, must disappear.\nMust not, then, the distinction between the real and fantastic,\nbetween dreams and facts, disappear with it? What meaning is there in\nasking whether any given idea is real or not, unless a reference is\nimplied to something other than the idea itself?\n\n[There are no paragraphs 161-169 in any edition or reprint. Tr]\n\nWhat then becomes of distinction between reality and fancy?\n\n170. Berkeley\u2019s theory, no less than Locke\u2019s, requires such\nreference. He insists, as much as Locke does, on the difference\nbetween ideas of imagination which do, and those of sense which\ndo not, depend on our own will. \u2018It is no more than willing, and\nstraightway this or that idea arises in my fancy; and by the same\npower it is obliterated and makes way for another.\u2019 But \u2018when in\nbroad daylight I open my eyes, it is not in my power to choose\nwhether I shall see or no, or to determine what particular objects\nshall present themselves to my view.\u2019 Moreover \u2018the ideas of sense\nare more strong, lively and distinct than those of the imagination;\nthey have likewise a steadiness, order, and coherence, and are not\nexcited at random as those which are the effects of human wills often\nare, but in a regular train and series\u2019 (Ibid. secs. 28-30). These\ncharacteristics of ideas of sense, however, do not with Berkeley, any\nmore than with Locke, properly speaking, _constitute_ their reality.\nThis lies in their relation to something else, of which these\ncharacteristics are the tests. The difference between the two writers\nlies in their several views as to what this \u2018something else\u2019 is. With\nLocke it was body or matter, as proximately, though in subordination\nto the Divine Will, the \u2018imprinter\u2019 of those most lively ideas\nwhich we cannot make for ourselves. His followers insisted on the\nproximate, while they ignored the ultimate, reference. Hence, as\nBerkeley conceived, their Atheism, which he could cut from under\ntheir feet by the simple plan of eliminating the proximate reference\naltogether, and thus showing that God, not matter, is the immediate\n\u2018imprinter\u2019 of ideas on the senses and the suggester of such ideas of\nimagination as the ideas of sense, in virtue of habitual association,\nconstantly introduce (Ibid. sec. 33).\n\nThe real = ideas that God causes.\n\n171. To eliminate the reference to matter might seem to be more\neasy than to substitute for it a reference to God. If the object\nof the idea is only the idea itself, does not all determination by\nrelation logically disappear from the idea, except (perhaps) such\nas consists in the fact of its sequence or antecedence to other\nideas? This issue was afterwards to be tried by Hume--with what\nconsequences to science and religion we shall see. Berkeley avoids\nit by insisting that the \u2018percipi,\u2019 to which \u2018esse\u2019 is equivalent,\nimplies reference to a mind. At first sight this reference, as common\nto all ideas alike, would not seem to avail much as a basis either\nfor a distinction between the real and fantastic or for any Theism\nexcept such as would \u2018entitle God to all our fancies.\u2019 If it is to\nserve Berkeley\u2019s purpose, we must suppose the idea to carry with it\nnot merely a relation to mind but a relation to it as its effect, and\nthe conscious subject to carry with him such a distinction between\nhis own mind and God\u2019s as leads him to refer his ideas to God\u2019s mind\nas their cause when they are lively, distinct and coherent, but when\nthey are otherwise, to his own. And this, in substance, is Berkeley\u2019s\nsupposition. To show the efficient power of mind he appeals to our\nconsciousness of ability to produce at will ideas of imagination; to\nshow that there is a divine mind, distinct from our own, he appeals\nto our consciousness of inability to produce ideas of sense.\n\nIs it then a succession of feelings?\n\n172. Even those least disposed to \u2018vanquish Berkeley with a grin\u2019\nhave found his doctrine of the real, which is also his doctrine of\nGod, \u2018unsatisfactory.\u2019 By the real world they are accustomed to\nunderstand something which--at least in respect of its \u2018elements\u2019\nor \u2018conditions\u2019 or \u2018laws\u2019--permanently is; though the combinations\nof the elements, the events which flow from the conditions, the\nmanifestations of the laws, may never be at one time what they\nwill be at the next. But according to the Berkeleian doctrine the\npermanent seems to disappear: the \u2018is\u2019 gives place to a \u2018has been\u2019\nand \u2018will be.\u2019 If I say (\u03b4\u03b5\u03b9\u03ba\u03c4\u03b9\u03ba\u1ff6\u03c2) [1] \u2018there is a body,\u2019 I must\nmean according to it that a feeling has just occurred to me, which\nhas been so constantly followed by certain other feelings that it\nsuggests a lively expectation of these. The suggestive feeling\nalone _is_, and it is ceasing to be. If this is the true account of\npropositions suggested by everyone\u2019s constantly-recurrent experience,\nwhat are we to make of scientific truths, _e.g._ \u2018a body will change\nits place sooner than let another enter it,\u2019 \u2018planets move in\nellipses,\u2019 \u2018the square on the hypotheneuse is equal to the squares on\nthe sides.\u2019 In these cases, too, does the present reality lie merely\nin a feeling experienced by this or that scientific man, and to him\nsuggestive of other feelings? Does the proposition that \u2018planets move\nin ellipses\u2019 mean that to some watcher of the skies, who understands\nKepler\u2019s laws, a certain perception of \u2018visible extension\u2019 (_i.e._\nof colour or light and shade) not only suggests, as to others, a\nparticular expectation of other feelings, which expectation is called\na planet, but a further expectation, not shared by the multitude, of\nfeelings suggesting successive situations of the visible extension,\nwhich further expectation is called elliptical motion? Such an\nexplanation of general propositions would be a form of the doctrine\nconveniently named after Protagoras--\u2018\u1f00\u03bb\u03b7\u03b8\u1f72\u03c2 \u1f43 \u1f11\u03ba\u03ac\u03c3\u03c4\u1ff3 \u1f11\u03ba\u03ac\u03c3\u03c4\u03bf\u03c4\u03b5 \u03b4\u03bf\u03ba\u03b5\u1fd6\u2019\n[2]--a doctrine which the vindicators of Berkeley are careful to\ntell us we must not confound with his. The question, however, is not\nwhether Berkeley himself admits the doctrine, but whether or no it is\nthe logical consequence of the method which he uses for the overthrow\nof materialists and \u2018mathematical Atheists\u2019?\n\n[1] [Greek \u03b4\u03b5\u03b9\u03ba\u03c4\u03b9\u03ba\u1ff6\u03c2 (deiktikos) = \u201caffirmatively\u201d or \u201ccapable of\nbeing proven\u201d _i.e._ not merely hypothetically. Tr.]\n\n[2] [Greek \u1f00\u03bb\u03b7\u03b8\u1f72\u03c2 \u1f43 \u1f11\u03ba\u03ac\u03c3\u03c4\u1ff3 \u1f11\u03ba\u03ac\u03c3\u03c4\u03bf\u03c4\u03b5 \u03b4\u03bf\u03ba\u03b5\u1fd6 (alethes ho hekasto\nhekastote dokei) = the truth for each man is as it appears to him.\nTr.]\n\nBerkeley goes wrong from confusion between thought and feeling. For\nLocke\u2019s \u2018idea of a thing\u2019 he substitutes \u2018idea\u2019 simply.\n\n173. His purpose was the maintenance of Theism, and a true instinct\ntold him that pure Theism, as distinct from nature-worship and\ndaemonism, has no philosophical foundation, unless it can be shown\nthat there is nothing real apart from thought. But in the hurry\nof theological advocacy, and under the influence of a misleading\nterminology, he failed to distinguish this true proposition--there\nis nothing real apart from thought--from this false one, its virtual\ncontradictory--there is nothing other than feeling. The confusion\nwas covered, if not caused, by the ambiguity, often noticed, in\nthe use of the term \u2018idea.\u2019 This to Berkeley\u2019s generation stood\nalike for feeling proper, which to the subject that merely feels\nis neither outer nor inner, because not referring itself to either\nmind or thing, and for conception, or an object thought of under\nrelations. According to Locke, pain, colour, solidity, are all ideas\nequally with each other and equally with the _idea of_ pain, _idea\nof_ colour, _idea of_ solidity. If all alike, however, were feelings\nproper, there would be no world either to exist or be spoken of.\nLocke virtually saves it by two suppositions, each incompatible with\nthe equivalence of idea to feeling, and implying the conversion of\nit into conception as above defined. One is that there are abstract\nideas; the other that there are primary qualities of which ideas\nare copies, but which do not come and go with our feelings. The\nlatter supposition gives a world that \u2018really exists,\u2019 the former\na world that may be known and spoken of; but neither can maintain\nitself without a theory of conception which is not forthcoming in\nLocke himself. We need not traverse again the contradictions which\naccording to his statement they involve--contradictions which, under\nwhatever disguise, must attach to every philosophy that admits a\nreality either in things as apart from thought or in thought as apart\nfrom things, and only disappear when the thing as thought of, and\nthrough thought individualised by the relations which constitute\nits community with the universe, is recognised as alone the real.\nMisled by the phrase \u2018idea of a thing,\u2019 we fancy that idea and thing\nhave each a separate reality of their own, and then puzzle ourselves\nwith questions as to how the idea can represent the thing--how the\nideas of primary qualities can be copies of them, and how, if the\nreal thing of experience be merely individual, a general idea can\nbe abstracted from it. These questions Berkeley asked and found\nunanswerable. There were then two ways of dealing with them before\nhim. One was to supersede them by a truer view of thought and its\nobject, as together in essential correlation constituting the real;\nbut this way he did not take. The other was to avoid them by merging\nboth thing and idea in the indifference of simple feeling. For a\nmerely sentient being, it is true--for one who did not think upon\nhis feelings--the oppositions of inner and outer, of subjective\nand objective, of fantastic and real, would not exist; but neither\nwould knowledge or a world to be known. That such oppositions,\nmisunderstood, may be a heavy burden on the human spirit, the\nexperience of current controversy and its spiritual effects might\nalone suffice to convince us; but the philosophical deliverance can\nonly lie in the recognition of thought as their author, not in the\nattempt to obliterate them by the reduction of thought and its world\nto feeling--an attempt which contradicts itself, since it virtually\nadmits their existence while it renders them unaccountable.\n\nWhich, if idea = feeling, does away with space and body.\n\n174. That Berkeley\u2019s was such an attempt, looking merely to his\ntreatment of primary qualities and abstract ideas, we certainly could\nnot doubt: though, since language does not allow of its consistent\nstatement, and Berkeley was quite ready to turn the exigencies of\nlanguage to account, passages logically incompatible with it may\neasily be found in him. The hasty reader, when he is told that body\nor distance are suggested by feelings of sight and touch rather than\nimmediately seen, accepts the doctrine without scruple, because he\nsupposes that which is suggested to be a present reality, though\nnot at present felt. But if not at present felt it is not according\nto Berkeley an idea, therefore \u2018without the mind,\u2019 therefore an\nimpossibility. [1] That which is suggested, then, must itself be\na feeling which consists in the expectation of other feelings.\nDistance, and body, _as suggested_, can be no more than such an\nexpectation; and as _actually existing_, no more than the actual\nsuccession of the expected feelings--a succession of which, as of\nevery succession, \u2018no two parts exist together.\u2019 [2] There is no\ntime, then, at which it can be said that distance and body exist.\n\n[1] Reference is here merely made to the doctrine by which Berkeley\ndisposes of \u2018matter,\u2019 the consideration of its reconcilability with\nhis doctrine of \u2018spirits\u2019 and \u2018relations\u2019 as objects of knowledge\nbeing postponed.\n\n[2] Locke, Book II. chap. xv. sec. 1.\n\nHe does not even retain them as \u2018abstract ideas\u2019.\n\n175. This, it may seem, however inconsistent with the doctrine of\nprimary qualities, is little more than the result which Locke himself\ncomes to in his Fourth Book; since, if \u2018actual present succession\u2019\nforms our only knowledge of real existence, there could be no time\nat which distance and body might be _known_ as really existing. But\nLocke, as we have seen, is able to save mathematical, though not\nphysical, knowledge from the consequences of this admission by his\ndoctrine of abstract ideas--\u2018ideas removed in our thoughts from\nparticular existence\u2019--whose agreement or disagreement is stated\nin propositions which \u2018concern not existence,\u2019 and for that reason\nmay be general without becoming either uncertain or uninstructive.\nThis doctrine Berkeley expressly rejects on the ground that he\ncould not perceive separately that which could not exist separately\n(\u2018Principles of Human Knowledge,\u2019 Introduction, sec. 10); a ground\nwhich to the ordinary reader seems satisfactory because he has no\ndoubt, and Berkeley\u2019s instances do not suggest a doubt, as to the\npresent existence of \u2018individual objects\u2019--this man, this horse,\nthis body. But with Berkeley to exist means to be felt (\u2018Principles\nof Human Knowledge,\u2019 sec. 3), and the feelings, which I name a body,\nbeing successive, its existence must be in succession likewise. The\nlimitation, then, of possibility of \u2018conception\u2019 by possibility of\nexistence, means that \u2018conception,\u2019 too, is reduced to a succession\nof feelings.\n\nOn the same principle all permanent relations should disappear.\n\n176. Berkeley, then, as a consequence of the methods by which he\ndisposes at once of the \u2018real existence\u2019 and \u2018abstract idea of\nmatter,\u2019 has to meet the following questions:--How are either reality\nor knowledge possible without permanent relations? and, How can\nfeelings, of which one is over before the next begins, constitute\nor represent a world of permanent relations? The difficulty becomes\nmore obvious, though not more serious, when the relations in question\nare not merely themselves permanent, as are those between natural\nphenomena, but are relations between permanent parts like those of\nspace. It is for this reason that its doctrine of geometry is the\nmost easily assailable point of the \u2018sensational\u2019 philosophy. Locke\ndistinguishes the ideas of space and of duration as got, the one\nfrom the permanent parts of space, the other \u2018from the fleeting and\nperpetually perishing parts of succession.\u2019 [1] He afterwards prefers\nto oppose the term \u2018expansion\u2019 to \u2018duration,\u2019 as bringing out more\nclearly than \u2018space\u2019 the opposition of relation between permanent\nfacts to that between \u2018fleeting successive facts which never exist\ntogether.\u2019 How, then, can a consciousness, consisting simply of\n\u2018fleeting successive facts,\u2019 either be or represent that of which the\ndifferentia is that its facts are permanent and co-exist?\n\n[1] Book II. chap. xiv. sec. 1.\n\nBy making colour = relations of coloured points, Berkeley represents\nrelation as seen.\n\n177. This crucial question in regard to extension does not seem even\nto have suggested itself to Berkeley. The reason why is not far to\nseek. Professor Fraser, in his valuable edition, represents him as\nmeaning by visible extension \u2018coloured experience in sense,\u2019 and by\ntangible extension \u2018resistent experience in sense.\u2019 [1] No fault can\nbe found with this interpretation, but the essential question, which\nBerkeley does not fairly meet, is whether the experience in each\ncase is complete in a single feeling or consists in a succession of\nfeelings. If in a single feeling, it clearly is not extension, as a\nrelation between parts, at all; if in a succession of feelings, it\nis only extension because a synthetic principle, which is not itself\none of the feelings, but equally present to them all, transforms\nthem into permanent parts of which each qualifies the other by\noutwardness to it. Berkeley does not see the necessity of such a\nprinciple, because he allows himself to suppose extension--at any\nrate visible extension--to be constituted by a single feeling.\nHaving first pronounced that the proper object of sight is colour,\nhe quietly substitutes for this _situations_ of colour, degrees of\nstrength and faintness in colour, and quantities of coloured points,\nas if these, interchangeably with mere colour, were properly objects\nof sight and perceived in single acts of vision. Now if by object of\nsight were meant something other than the sensation itself--something\nwhich to a thinking being it suggests as its cause--there would be\nno harm in this language, but neither would there be any ground for\nsaying that the proper object of sight is colour, for distinguishing\nvisible from tangible extension, or for denying that the outwardness\nof body to body is seen. Such restrictions and distinctions have\nno meaning, unless by sight is meant the nervous irritation, the\naffection of the visual organ, as it is to a merely feeling subject;\nyet in the very passages where he makes them, by saying that we see\nsituations and degrees of colour, and quantities of coloured points,\nBerkeley converts sight into a judgment of extensive and intensive\nquantity. He thus fails to discern that the transition from colour\nto coloured extension cannot be made without on the one hand either\nthe presentation of successive pictures or (which comes to the\nsame) successive acts of attention to a single picture, and on the\nother hand a synthesis of the successive presentations as mutually\nqualified parts of a whole. In other words, he ignores the work of\nthought involved in the constitution alike of coloured and tangible\nextension, and in virtue of which alone either is extension at all.\n\n[1] See Fraser\u2019s Berkeley, \u2018Theory of Vision,\u2019 note 42. I may here\nsay that I have gone into less detail in my account of Berkeley\u2019s\nsystem than I should otherwise have thought necessary, because\nProfessor Fraser has supplied, in the way of explanation of it, all\nthat a student can require.\n\nStill he admits that space is constituted by a succession of feelings.\n\n178. But though he does not scruple to substitute for colour\nsituations and quantities of coloured points, these do not with him\nconstitute space, which he takes according to Locke\u2019s account of it\nto be \u2018distance between bodies or parts of the same body.\u2019 This,\naccording to his \u2018Theory of Vision,\u2019 is tangible extension, and\nthis again is alone the object of geometry. As in that treatise a\ndifference is still supposed between _tangible_ extension and the\nfeeling of touch, the question does not there necessarily arise\nwhether the tactual experience, that constitutes this extension, is\ncomplete in a single feeling or only in a succession of feelings;\nbut when in the subsequent treatise the difference is effaced, it is\ndecided by implication that the experience is successive: [1] and all\nreceived modifications of the theory, which assign to a locomotive or\nmuscular sense the office which Berkeley roughly assigned to touch,\nmake the same implication still more clearly. Now in the absence of\nany recognition of a synthetic principle, in relation to which the\nsuccessive experience becomes what it is not in itself, this means\nnothing else than that space is a succession of feelings, which again\nmeans that space is not space, not a qualification of bodies or parts\nof body by mutual externality, since to such qualification it is\nnecessary that bodies or their parts coexist. Thus, in his hurry to\nget rid of externality as independence of the mind, he has really got\nrid of it as a relation between bodies, and in so doing (however the\nresult may be disguised) has logically made a clean sweep of geometry\nand physics.\n\n[1] \u2018Principles of Human Knowledge,\u2019 sec. 44. It will be observed\nthat in that passage Berkeley uses the term \u2018distance\u2019 not \u2018space,\u2019\nand though with him the terms are strictly interchangeable, this\nmay have helped to disguise from him the full monstrosity of the\ndoctrine, \u2018space is a succession of feelings,\u2019 which, stated in that\nform, must surely have scandalised him.\n\nIf so, it is not space at all; but Berkeley thinks it is only not\n\u2018pure\u2019 space. _Space_ and _pure_ space stand or fall together.\n\n179. Of this result he himself shows no suspicion. He professes to\nbe able, without violence to his doctrine, to accept the sciences\nas they stand, except so far as they rest upon the needless and\nunmeaning assumptions (as he reckoned them) of _pure_ space and its\ninfinite divisibility. The truth seems to be that--at any rate in\nthe state of mind represented his earlier treatises--he was only\nable to work on the lines which Locke had laid. It did not occur\nto him to treat the primary qualities as relations constituted by\nthought, because Locke had not done so. Locke having treated them as\nexternal to the mind, Berkeley does so likewise, and for that reason\nfeels that they must be got rid of. The mode of riddance, again,\nwas virtually determined for him by Locke. Locke having admitted\nthat they copied themselves in feelings, the untenable element in\nthis supposition had only to be dropped and they became feelings\nsimply. It is thus only so far as space is supposed to exist after\na mode of which, according to Locke himself, sense could take no\ncopy--_i.e._ as exclusive not merely of all colour but of all body,\nand as infinitely divisible--that Berkeley becomes aware of its\nincompatibility with his doctrine. Pure space, or \u2018vacuum,\u2019 to him\nmeans space that can not be touched--a tangible extension that is not\ntangible--and is therefore a contradiction in terms. The notion that,\nthough not touched, it might be seen, he excludes, [1] apparently for\nthe same reason which prevents him from allowing _visible_ extension\nto be space at all; the reason, namely, that there is no \u2018outness\u2019\nor relation of externality between the parts of such extension.\nThe fact that there can be no such relation between the successive\nfeelings which alone, according to him, constitute \u2018tangible\nextension,\u2019 he did not see to be equally fatal to the latter being\nin any true sense space. In other words, he did not see that the\ntest of reduction to feeling, by which he disposed of the _vacuum_,\ndisposed of space altogether. If he had, he would have understood\nthat space and body were intelligible relations, which can be thought\nof apart from the feelings which through them become the world that\nwe know, since it is they that are the conditions of these feelings\nbecoming a knowledge, not the feelings that are the condition of the\nrelations being known. Whether they can be thought of apart from\neach other--whether the simple relation of externality between parts\nof a whole can be thought of without the parts being considered\nas solid--is of course a further question, and one which Berkeley\ncannot be said properly to discuss at all, since the abstraction of\nspace from body to him meant its abstraction from feelings of touch.\nThe answer to it ceases to be difficult as soon as the question is\nproperly stated.\n\n[1] \u2018Principles of Human Knowledge,\u2019 sec. 116.\n\nBerkeley disposes of space for fear of limiting God.\n\n180. As with vacuum, so with infinite divisibility. Once let it\nbe understood that extension is constituted by the relation of\nexternality between homogeneous parts, and it follows that there can\nbe no _least_ part of extension, none that does not itself consist\nof parts; in other words, that it is infinitely divisible: just as\nconversely it follows that there can be no _last_ part of it, not\nhaving another outside it; in other words, that (to use Locke\u2019s\nphrase) it is infinitely addible. Doubtless, as Berkeley held, there\nis a \u2018minimum visibile\u2019; but this means that there are conditions\nunder which any seen colour disappears, and disappearing, ceases to\nbe known under the relation of extension; but it is only through a\nconfusion of the relation with the colour that the disappearance of\nthe latter is thought to be a disappearance of so much extension. [1]\nIt was, in short, the same failure to recognise the true ideality of\nspace, as a relation constituted by thought, that on the one hand\nmade its \u2018purity\u2019 and infinity unmeaning to Berkeley, and on the\nother made him think that, if pure (_sc_. irreducible to feelings)\nand infinite, it must limit the Divine perfection, either as being\nitself God or as \u2018something beside God which is eternal, uncreated,\nand infinite\u2019 (\u2018Principles of Human Knowledge,\u2019 sec. 117). Fear of\nthis result set him upon that method of resolving space, and with it\nthe world of nature, into sequent feelings, which, if it had been\nreally susceptible of logical expression, would at best have given\nhim nothing but a \u03bc\u03ad\u03b3\u03b1 \u03b6\u1ff6\u03bf\u03bd [2] for God. If he had been in less of\na hurry with his philosophy, he might have found that the current\ntendency to \u2018bind God in nature or diffuse in space\u2019 required to be\nmet by a sounder than his boyish idealism--by an idealism which gives\nspace its due, but reflects that to make space God, or a limitation\non God, is to subject thought itself to the most superficial of the\nrelations by which it forms the world that it knows.\n\n[1] The same remark of course applies, _mutatis mutandis_, to the\n\u2018minimum tangibile.\u2019 See below, paragraphs 265 and 260.\n\n[2] [Greek \u03bc\u03ad\u03b3\u03b1 \u03b6\u1ff6\u03bf\u03bd (mega zoon) = great being. Tr.]\n\nHow he deals with possibility of general knowledge.\n\n181. So far we have only considered Berkeley\u2019s reduction of primary\nqualities, supposed to be sensible, to sensations as it affects the\nqualities themselves, rather than as it affects the possibility\nof universal judgments about them. If, indeed, as we have found,\nsuch reduction really amounts to the absolute obliteration of the\nqualities, no further question can remain as to the possibility of\ngeneral knowledge concerning them. As Berkeley, however, did not\nadmit the obliteration, the further question did remain for him:\nand the condition of his plausibly answering it was that he should\nrecognise in the \u2018idea,\u2019 as subject of predication, that intelligible\nqualification by relation which he did not recognise in it simply as\n\u2018idea,\u2019 and which essentially differences it from feeling proper. If\nany particular \u2018tangible extension,\u2019 _e.g._ a right-angled triangle,\nis only a feeling, or in Berkeley\u2019s own language, \u2018a fleeting\nperishable passion\u2019 [1] not existing at all, even as an \u2018abstract\nidea,\u2019 except when some one\u2019s tactual organs are being affected in\na certain way--what are we to make of such a general truth as that\nthe square on its base is always equal to the squares on its sides?\nOmitting all difficulties about the convertibility of a figure with\na feeling, we find two questions still remain--How such separation\ncan be made of the figure from the other conditions of the tactual\nexperience as that propositions should be possible which concern the\nfigure simply; and how a single case of tactual experience--that\nin which the mathematician finds a feeling called a right-angled\ntriangle followed by another which he calls equality between the\nsquares, &c.--leads in the absence of any \u2018necessary connexion\u2019\nto the expectation that the sequence will always be the same. [2]\nThe difficulty becomes the more striking when it is remembered\nthat though the geometrical proposition in question, according to\nBerkeley, concerns the tangible, the experience which suggests it is\nmerely visual.\n\n[1] \u2018Principles of Human Knowledge,\u2019 sec. 89.\n\n[2] See above, paragraph 122.\n\nHis theory of universals ...\n\n182. Berkeley\u2019s answer to these questions must be gathered from his\ntheory of general names. \u2018It is, I know,\u2019 he says, \u2018a point much\ninsisted on, that all knowledge and demonstration are about universal\nnotions, to which I fully agree: but then it does not appear to me\nthat those notions are formed by abstraction--_universality_, so\nfar as I can comprehend, not consisting in the absolute positive\nnature or conception of anything, but in the relation it bears to\nthe particulars signified or represented by it; by virtue whereof\nit is that things, names, or notions, being in their own nature\n_particular_, are rendered universal. Thus, when I demonstrate any\nproposition concerning triangles, it is to be supposed that I have in\nview the universal idea of a triangle; which is not to be understood\nas if I could frame an idea of a triangle which was neither\nequilateral nor scalene nor equicrural; but only that the particular\ntriangle I considered, whether of this or that sort it matters not,\ndoth equally stand for and represent all rectilinear triangles\nwhatsoever, and is in that sense universal.\u2019 Thus it is that \u2018a man\nmay consider a figure merely as triangular.\u2019 (\u2018Principles of Human\nKnowledge,\u2019 Introd. secs. 15 and 16.)\n\n... of value, as implying that universality of ideas lies in relation.\n\n183. In this passage appear the beginnings of a process of thought\nwhich, if it had been systematically pursued by Berkeley, might have\nbrought him to understand by the \u2018percipi,\u2019 to which he pronounced\n\u2018esse\u2019 equivalent, definitely the \u2018intelligi.\u2019 As it stands, the\nresult of the passage merely is that the triangle (for instance)\n\u2018in its own nature,\u2019 because \u2018particular,\u2019 is not a possible\nsubject of general predication or reasoning: that it is so only as\n\u2018considered\u2019 under a relation of resemblance to other triangles and\nby such consideration universalized. \u2018In its own nature,\u2019 or as a\n\u2018particular idea,\u2019 the triangle, we must suppose, is so much tangible\n(or visible, as symbolical of tangible) extension, and therefore\naccording to Berkeley a feeling. But a relation, as he virtually\nadmits, [1] is neither a feeling nor felt. The triangle, then, as\nconsidered under relation and thus a possible subject of general\npropositions, is quite other than the triangle in its own nature.\nThis, of course, is so far merely a virtual repetition of Locke\u2019s\nembarrassing doctrine that real things are not the things which we\nspeak of, and which are the subject of our sciences; but it is a\nrepetition with two fruitful differences--one, that the thing in\nits \u2018absolute positive nature\u2019 is more explicitly identified with\nfeeling; the other, that the process, by which the thing thought and\nspoken of is supposed to be derived from the real thing, is no longer\none of \u2018abstraction,\u2019 but consists in consideration of relation.\nIt is true that with Berkeley the mere feeling has a \u2018positive\nnature\u2019 apart from considered relations, [2] and that the considered\nrelation, by which the feeling is universalised, is only that of\nresemblance between properties supposed to exist independently of it.\nThe \u2018particular triangle,\u2019 reducible to feelings of touch, has its\ntriangularity (we must suppose) simply as a feeling. It is only the\nresemblance between the triangularity in this and other figures--not\nthe triangularity itself--that is a relation, and, as a relation, not\nfelt but considered; or in Berkeley\u2019s language, something of which we\nhave not properly an \u2018idea\u2019 but a \u2018notion.\u2019 [3]\n\n[1] See \u2018Principles of Human Knowledge,\u2019 sec. 89. (2nd edit.)\n\n[2] See below, paragraph 298.\n\n[3] \u2018Principles of Human Knowledge,\u2019 Ibid. This perhaps is the\nbest place for saying that it is not from any want of respect for\nDr. Stirling that I habitually use \u2018notion\u2019 in the loose popular\nway which he counts \u2018barbarous,\u2019 but because the barbarism is so\nprevalent that it seems best to submit to it, and to use \u2018conception\u2019\nas the equivalent of the German \u2018Begriff.\u2019\n\nBut he fancies that each idea has a positive nature apart from\nrelation.\n\n184. But though Berkeley only renders explicit the difficulties\nimplicit in Locke\u2019s doctrine of ideas, that is itself a great step\ntaken towards disposing of them. Once let the equivocation between\nsensible qualities and sensations be got rid of--once let it be\nadmitted that the triangle in its absolute nature, as opposed to\nthe triangle considered, is merely a feeling, and that relations\nare not feelings or felt--and the question must soon arise, What\nin the absence of all relation remains to be the absolute nature\nof the triangle? It is a question which ultimately admits of but\none answer. The triangularity of the given single figure must\nbe allowed to be just as much a relation as the resemblance,\nconsisting in triangularity, between it and other figures; and if a\nrelation, then not properly felt, but understood. The \u2018particular\u2019\ntriangle, if by that is meant the triangle as subject of a singular\nproposition, is no more \u2018particular in time,\u2019 no more constituted\nby the occurrence of a feeling, than is the triangle as subject of\na general proposition. It really exists as constituted by relation,\nand therefore only as \u2018considered\u2019 or understood. In its existence,\nas in the consideration of it, the relations indicated by the terms\n\u2018equilateral, equicrural and scalene,\u2019 presuppose the relation of\ntriangularity, not it them; and for that reason it can be considered\napart from them, though not they apart from it, without any breach\nbetween that which is considered and that which really exists. Thus,\ntoo, it becomes explicable that a single experiment should warrant\na universal affirmation; that the mathematician, having once found\nas the result of a certain comparison of magnitudes that the square\non the hypothenuse is equal to the square on the sides, without\nwaiting for repeated experience at once substitutes for the singular\nproposition, which states his discovery, a general one. If the\nsingular proposition stated a sensible event or the occurrence of a\nfeeling, such substitution would be inexplicable: for if that were\nthe true account of the singular proposition, a general one could but\nexpress such expectation of the recurrence of the event as repeated\nexperience of it can alone give. But a relation is not contingent\nwith the contingency of feeling. It is permanent with the permanence\nof the combining and comparing thought which alone constitutes it;\nand for that reason, whether it be recognised as the result of a\nmathematical construction or of a crucial experiment in physics, the\nproposition which states it must already be virtually universal.\n\nTraces of progress in his idealism.\n\n185. Of such a doctrine Berkeley is rather the unconscious forerunner\nthan the intelligent prophet. It is precisely upon the question\nwhether, or how far, he recognised the constitution of things by\nintelligible relations, that the interpretation of his early (which\nis his only developed) idealism rests. Is it such idealism as Hume\u2019s,\nor such idealism as that adumbrated in some passages of his own\n\u2018Siris\u2019? Is the idea, which is real, according to him a feeling or\na conception? Has it a nature of its own, consisting simply in its\nbeing felt, and which we afterwards for purposes of our own consider\nin various relations; or does the nature consist only in relations,\nwhich again imply the action of a mind that is eternal--present\nto that which is in succession, but not in succession itself? The\ntruth seems to be that this question in its full significance never\npresented itself to Berkeley, at least during the period represented\nby his philosophical treatises. His early idealism, as we learn\nfrom the commonplace-book brought to light by Professor Fraser, was\nmerely a cruder form of Hume\u2019s. By the time of the publication of\nthe \u2018Principles of Human Knowledge\u2019 he had learnt that, unless this\ndoctrine was to efface \u2018spirit\u2019 as well as \u2018matter,\u2019 he must modify\nit by the admission of a \u2018thing\u2019 that was not an \u2018idea,\u2019 and of which\nthe \u2018esse\u2019 was \u2018percipere\u2019 not \u2018percipi.\u2019 This admission carried with\nit the distinction between the object felt and the object known,\nbetween \u2018idea\u2019 and \u2018notion\u2019--a distinction which was more clearly\nmarked in the \u2018Dialogues.\u2019 Of \u2018spirit\u2019 we could have a \u2018notion,\u2019\nthough not an \u2018idea.\u2019 But it was only in the second edition of the\n\u2018Principles\u2019 that \u2018relation\u2019 was put along with \u2018spirit,\u2019 as that\nwhich could be known but which was no \u2018idea:\u2019 and then without any\nrecognition of the fact that the whole reduction of primary qualities\nto mere ideas was thereby invalidated. The objects, with which the\nmathematician deals, are throughout treated as in their own nature\n\u2018particular ideas,\u2019 into the constitution of which relation does not\nenter at all; in other words, as successive feelings.\n\nHis way of dealing with physical truths.\n\n186. If the truths of mathematics seemed to Berkeley explicable on\nthis supposition, those of the physical sciences were not likely to\nseem less so. As long as the relations with which these sciences deal\nare relations between \u2018sensible objects,\u2019 he does not notice that\nthey _are_ relations, and therefore not feelings or felt, at all.\nHe treats felt things as if the same as feelings, and ignores the\nrelations altogether. Thus a so-called \u2018sensible\u2019 motion causes him\nno difficulty. He would be content to say that it was a succession\nof ideas, not perceiving that motion implies a relation between\nspaces or moments as successively occupied by something that remains\none with itself--a relation which a mere sequence of feelings could\nneither constitute nor of itself suggest. It is only about a motion\nwhich does not profess to be \u2018seen,\u2019 such as the motion of the earth,\nthat any question is raised--a question easily disposed of by the\nconsideration that in a different position we should see it. \u2018The\nquestion whether the earth moves or not amounts in reality to no more\nthan this, to wit, whether we have reason to conclude from what hath\nbeen observed by astronomers, that if we were placed in such and such\ncircumstances, and such or such a position and distance both from the\nearth and sun, we should perceive the former to move among the choir\nof the planets, and appearing in all respects like one of them: and\nthis by the established rules of nature, which we have no reason to\nmistrust, is reasonably collected from the phenomena\u2019 (\u2018Principles of\nHuman Knowledge,\u2019 sec. 58). [1]\n\n[1] Cf. \u2018Dialogues,\u2019 page 147, in Prof. Fraser\u2019s edition.\n\nIf they imply permanent relations, his theory properly excludes them.\nHe supposes a divine decree that one feeling shall follow another.\n\n187. Now this passage clearly does not mean--as it ought to mean\nif the \u2018_esse_\u2019 of the motion were the \u2018_percipi_\u2019 by us--that the\nmotion of the earth would begin as soon as we were there to see it.\nIt means that it is now going on as an \u2018established law of nature,\u2019\nwhich may be \u2018collected from the phenomena.\u2019 In other words, it\nmeans that our successive feelings are so related to each other as\ndetermined by one present and permanent system, on which not they\nonly but all possible feelings depend, that by a certain set of\nthem we are led--not to expect a recurrence of them in like order\naccording to the laws of association, but, what is the exact reverse\nof this--to infer that certain other feelings, of which we have no\nexperience, would now occur to us if certain conditions of situation\non our part were fulfilled, because the \u2018ordo ad universum,\u2019 of\nwhich these feelings would be the \u2018ordo ad nos,\u2019 does now obtain.\nBut though Berkeley\u2019s words mean this for us, they did not mean it\nfor him. That such relation--merely intelligible, or according to\nhis phraseology not an idea or object of an idea at all, as he must\nhave admitted it to be--gives to our successive feelings the only\n\u2018nature\u2019 that they possess, he never recognised. By the relation of\nidea to idea, as he repeatedly tells us, he meant not a \u2018necessary\nconnexion,\u2019 _i.e._ not a relation without which, neither idea would\nbe what it is, but such _de facto_ sequence of one upon the other\nas renders the occurrence of one the unfailing but arbitrary sign\nthat the other is coming. It is thus according to him (and here Hume\nmerely followed suit) that feelings are symbolical--symbolical not\nof an order other than the feelings and which accounts for them, but\nsimply of feelings to follow. To Berkeley, indeed, unlike Hume, the\nsequence of feelings symbolical of each other is also symbolical of\nsomething farther, viz. the mind of God: but when we examine what\nthis \u2018mind\u2019 means, we find that it is not an intelligible order by\nwhich our feelings may be interpreted, or the spiritual subject of\nsuch an order, but simply the arbitrary will of a creator that this\nfeeling shall follow that.\n\nLocke had explained reality by relation of ideas to outward body.\nLiveliness in the idea evidence of this relation.\n\n188. Such a doctrine could not help being at once confused in its\naccount of reality, and insecure in its doctrine alike of the human\nspirit and of God. On the recognition of relations as constituting\nthe _nature_ of ideas rests the possibility of any tenable theory of\ntheir reality. An isolated idea could be neither real nor unreal.\nApart from a definite order of relation we may suppose (if we like)\nthat it would _be_, but it would certainly not be real; and as\nlittle could it be unreal, since unreality can only result from the\nconfusion in our consciousness of one order of relation with another.\nIt is diversity of relations that distinguishes, for instance, these\nletters as they now appear on paper from the same as I imagine them\nwith my eyes shut, giving each sort its own reality: just as upon\nconfusion with the other each alike becomes unreal. Thus, though\nwith Locke simple ideas are necessarily real, we soon find that even\naccording to him they are not truly so in their simplicity, but only\nas related to an external thing producing them. He is right enough,\nhowever inconsistent with himself, in making relation constitute\nreality; wrong in limiting this prerogative to the one relation of\nexternality. When he afterwards, in virtual contradiction to this\nlimitation, finds the reality of moral and mathematical ideas just\nin that sole relation to the mind, as its products, which he had\npreviously made the source of all unreality, he forces upon us the\nexplanation which he does not himself give, that unreality does not\nlie in either relation as opposed to the other, but in the confusion\nof any relation with another. It is for lack of this explanation\nthat Locke himself, as we have seen, finds in the liveliness\nand involuntariness of ideas the sole and sufficient tests (not\n_constituents_) of their reality; though they are obviously tests\nwhich put the dreams of a man in a fever upon the same footing with\nthe \u2018impressions\u2019 of a man awake, and would often prove that unreal\nafter dinner which had been proved real before. There is a well-known\nstory of a man who in a certain state of health commonly saw a\nparticular gory apparition, but who, knowing its origin, used to have\nhimself bled till it disappeared. The reality of the apparition lay,\nhe knew, in some relation between the circulation of his blood and\nhis organs of sight, in distinction from the reality existing in the\nnormal relations of his visual organs to the light: and in his idea,\naccordingly, there was nothing unreal, because he did not confuse the\none relation with the other. Locke\u2019s doctrine, however, would allow\nof no distinction between the apparition as it was for such a man\nand as it would be for one who interpreted it as an actual \u2018ghost.\u2019\nHowever interpreted, the liveliness and the involuntariness of the\nidea remain the same, as does its relation to an efficient cause. If\nin order to its reality the cause must be an \u2018outward body,\u2019 then it\nis no more real when rightly, than when wrongly, interpreted; while\non the ground of liveliness and involuntariness it is as real when\ntaken for a ghost as when referred to an excess of blood in the head.\n\nBerkeley retains this notion, only substituting \u2018God\u2019 for \u2018body\u2019.\n\n189. As has been pointed out above, it is in respect not of the\n\u2018ratio cognoscendi\u2019 but of the \u2018ratio essendi\u2019 that Berkeley\u2019s\ndoctrine of reality differs from Locke\u2019s. With him it is not as an\neffect of an outward body, but as an immediate effect of God, that\nan \u2018idea of sense\u2019 is real. Just as with Locke real ideas and matter\nserve each to explain the other, so with Berkeley do real ideas and\nGod. If he is asked, What is God? the answer is, He is the efficient\ncause of real ideas; if he is asked, What are real ideas? the answer\nis, Those which God produces, as opposed to those which we make\nfor ourselves. To the inevitable objection, that this is a logical\nsee-saw, no effective answer can be extracted from Berkeley but\nthis--that we have subjective tests of the reality of ideas apart\nfrom a knowledge of their cause. In his account of these Berkeley\nonly differs from Locke in adding to the qualifications of liveliness\nand involuntariness those of \u2018steadiness, order, and coherence\u2019 in\nthe ideas. This addition may mean either a great deal or very little.\nTo us it may mean that the distinction of real and unreal is one that\napplies not to feelings but to the conceived relations of feelings;\nnot to events as such, but to the intellectual interpretation of\nthem. The occurrence of a feeling taken by itself (it may be truly\nsaid) is neither coherent nor incoherent; nor can the sequence of\nfeelings one upon another with any significance be called coherence,\nsince in that case an incoherence would be as impossible as any\nfailure in the sequence. As little can we mean by such coherence an\nusual, by incoherence an unusual, sequence of feelings. If we did,\nevery sequence not before experienced--such, for instance, as is\nexhibited by a new scientific experiment--being unusual, would have\nto be pronounced incoherent, and therefore unreal. Coherence, in\nshort, we may conclude, is only predicable of a system of relations,\nnot felt but conceived; while incoherence arises from the attempt of\nan imperfect intelligence to think an object under relations which\ncannot ultimately be held together in thought. The qualification\nthen of \u2018ideas\u2019 as coherent has in truth no meaning unless \u2018idea\u2019 be\ntaken to mean not _feeling_ but _conception_: and thus understood,\nthe doctrine that coherent ideas _are_ (Berkeley happily excludes the\nnotion that they merely _represent_) the real, amounts to a clear\nidentification of the real with the world of conception.\n\nNot regarding the world as a system of intelligible relations, he\ncould not regard God as the subject of it.\n\n190. If such idealism were Berkeley\u2019s, his inference from the\n\u2018ideality\u2019 of the real to spirit and God would be more valid than\nit is. To have got rid of the notion that the world first exists\nand then is thought of--to have seen that it only really exists as\nthought of--is to have taken the first step in the only possible\n\u2018proof of the being of God,\u2019 as the self-conscious subject in\nrelation to which alone an intelligible world can exist, and the\npresence of which in us is the condition of our knowing it. [1] But\nthere is nothing to show that in adopting coherence as one test,\namong others, of the reality of ideas, he attached to it any of the\nsignificance exhibited above. He adopted it from ordinary language\nwithout considering how it affected his view of the world as a\nsuccession of feelings. That still remained to him a sufficient\naccount of the world, even when he treated it as affording intuitive\ncertainty of a soul \u2018naturally immortal,\u2019 and demonstrative certainty\nof God. He is not aware, while he takes his doctrine of such\ncertainty from Locke, that he has left out, and not replaced, the\nonly solid ground for it which Locke\u2019s system suggested.\n\n[1] See above, paragraphs 146 and 149-152.\n\nHis view of the soul as \u2018naturally immortal\u2019.\n\n191. The soul or self, as he describes it, does not differ from\nLocke\u2019s \u2018thinking substance,\u2019 except that, having got rid of\n\u2018extended matter\u2019 altogether, he cannot admit with Locke any\npossibility of the soul\u2019s being extended, and, having satisfied\nhimself that \u2018time was nothing abstracted from the succession\nof ideas in the mind,\u2019 [1] he was clear that \u2018the soul always\nthinks\u2019--since the time at which it did not think, being abstracted\nfrom a succession of ideas, would be no time at all. A soul which is\nnecessarily unextended and therefore \u2018indiscerptible,\u2019 and without\nwhich there would be no time, he reckons \u2018naturally immortal.\u2019\n\n[1] \u2018Principles of Human Knowledge,\u2019 sec. 98.\n\nEndless succession of feelings is not immortality in true sense.\nBerkeley\u2019s doctrine of matter fatal to a true spiritualism;\n\n192. Upon this the remark must occur that, if the fact of being\nunextended constituted immortality, all sounds and smells must be\nimmortal, and that the inseparability of time from the succession of\nfeelings may prove that succession endless, but proves no immortality\nof a soul unless there be one self-conscious subject of that\nsuccession, identical with itself throughout it. To the supposition\nof there being such a subject, which Berkeley virtually makes, his\nown mode of disposing of matter suggested ready objections. In Locke,\nas we have seen, the two opposite \u2018things,\u2019 thinking and material,\nalways appear in strict correlativity, each representing (though he\nwas not aware of this) the same logical necessity of substantiation.\n\u2018Sensation convinces us that there are solid extended substances,\nand reflection that there are thinking ones.\u2019 These are not two\nconvictions, however, but one conviction, representing one and the\nsame essential condition of knowledge. Such logical necessity indeed\nis misinterpreted when made a ground for believing the real existence\neither of a multitude of independent things, for everything is a\n\u2018retainer\u2019 to everything else; [1] or of a separation of the thinking\nfrom the material substance, since, according to Locke\u2019s own showing,\nthey at least everywhere overlap; [2] or of an absolutely last\nsubstance, which because last would be unknowable: but it is evidence\nof the action of a synthetic principle of self-consciousness without\nwhich all reference of feelings to mutually-qualified subjects\nand objects, and therefore all knowledge, would be impossible. It\nis idle, however, with Berkeley so to ignore the action of this\nprinciple on the one side as to pronounce the material world a mere\nsuccession of feelings, and so to take it for granted on the other\nas to assert that every feeling implies relation to a conscious\nsubstance. Upon such a method the latter assertion has nothing to\nrest on but an appeal to the individual\u2019s consciousness--an appeal\nwhich avails as much or as little for material as for thinking\nsubstance, and, in face of the apparent fact that with a knock on the\nhead the conscious independent substance may disappear altogether,\ncannot hold its own against the suggestion that the one substance no\nless than the other is reducible to a series of feelings, so closely\nand constantly sequent on each other as to seem to coalesce. We\ncannot substitute for this illusory appeal the valid method of an\nanalysis of knowledge, without finding that substantiation in matter\nis just as necessary to knowledge as substantiation in mind. If this\nmethod had been Berkeley\u2019s he would have found a better plan for\ndealing with the \u2018materialism\u2019 in vogue. Instead of trying to show\nthat material substance was a fiction, he would have shown that it\nwas really a basis of intelligible relations, and that thus all that\nwas fictitious about it was its supposed sensibility and consequent\nopposition to the work of thought. Then his doctrine of matter\nwould itself have established the necessity of spirit, not indeed\nas substance but as the source of all substantiation. As it was,\nmisunderstanding the true nature of the antithesis between matter and\nmind, in his zeal against matter he took away the ground from under\nthe spiritualism which he sought to maintain. He simply invited a\nsuccessor in speculation, of colder blood than himself, to try the\nsolution of spirit in the same crucible with matter.\n\n[1] Above, paragraph 125.\n\n[2] Above, paragraph 127.\n\n... as well as to a true Theism. His inference to God from necessity\nof a power to produce ideas;\n\n193. His doctrine of God is not only open to the same objection as\nhis doctrine of spiritual substance, but to others which arise from\nthe illogical restrictions that have to be put upon his notion of\nsuch substance, if it is to represent at once the God of received\ntheology and the God whose agency the Berkeleian system requires as\nthe basis of distinction between the real and unreal. Admitting the\nsupposition involved in his certainty of the \u2018natural immortality\u2019\nof the soul--the supposition that the succession of feelings which\nconstitutes the world, and which at no time was not, implies one\nfeeling substance--that substance we should naturally conclude was\nGod. Such a God, it is true (as has been already pointed out [1]),\nwould merely be the \u03bc\u03ad\u03b3\u03b1 \u03b6\u1ff6\u03bf\u03bd [2] of the crudest Pantheism, but it\nis the only God logically admissible--if any be admissible--in an\n\u2018ideal\u2019 system of which the text is not \u2018the world really exists\nonly as thought of,\u2019 but \u2018the world only exists as a succession of\nfeelings.\u2019 It was other than a _feeling_ substance, however, that\nBerkeley required not merely to satisfy his religious instincts,\nbut to take the place held by \u2018outward body\u2019 with Locke as the\nefficient of real ideas. The reference to this feeling substance, if\nnecessary for any idea, is necessary for all--for the \u2018fantastic\u2019\nas well as for those of sense--and can therefore afford no ground\nfor distinction between the real and unreal. Instead, however, of\nbeing thus led to a truer view of this distinction, as in truth a\ndistinction between the complete and incomplete conception of an\nintelligible world, he simply puts the feeling substance, when he\nregards it as God, under an arbitrary limitation, making it relative\nonly to those ideas of which with Locke \u2018matter\u2019 was the substance,\nas opposed to those which Locke had referred to the thinking thing.\nThe direct consequence of this limitation, indeed, might seem to\nbe merely to make God an animal of partial, instead of universal,\nsusceptibility; but this consequence Berkeley avoids by dropping\nthe ordinary notion of substance altogether, so as to represent\nthe ideas of sense not as subsisting in God but as effects of His\npower--as related to Him, in short, just as with Locke ideas of sense\nare related to the primary qualities of matter. \u2018There must be an\nactive power to produce our ideas, which is not to be found in ideas\nthemselves, for we are conscious that they are inert, nor in matter,\nsince that is but a name for a bundle of ideas; which must therefore\nbe in spirit, since of that we are conscious as active; yet not in\nthe spirit of which we are conscious, since then there would be no\ndifference between real and imaginary ideas; therefore in a Divine\nSpirit, to whom, however, may forthwith be ascribed the attributes of\nthe spirit of which we are conscious.\u2019 Such is the sum of Berkeley\u2019s\nnatural theology.\n\n[1] Above, paragraph 180.\n\n[2] [Greek \u03bc\u03ad\u03b3\u03b1 \u03b6\u1ff6\u03bf\u03bd (mega zoon) = great being. Tr.]\n\n... a necessity which Hume does not see. A different turn should have\nbeen given to his idealism, if it was to serve his purpose.\n\n194. From a follower of Hume it of course invites the reply that\nhe does not see the necessity of an active power at all, to which,\nsince, according to Berkeley\u2019s own showing, it is no possible \u2018idea\u2019\nor object of an idea, all his own polemic against the \u2018absolute\nidea\u2019 of matter is equally applicable; that the efficient power,\nof which we profess to be conscious in ourselves, is itself only a\nname for a particular feeling or impression which precedes certain\nother of our impressions; that, even if it were more than this, the\ntransition from the spiritual efficiency of which we are conscious\nto another, of which it is the special differentia that we are not\nconscious of it, would be quite illegitimate, and that thus in\nsaying that certain feelings are real because, being lively and\ninvoluntary, they must be the work of this unknown spirit, we in\neffect say nothing more than that they are real because lively and\ninvoluntary. Against a retort of this kind Berkeley\u2019s theistic armour\nis even less proof than Locke\u2019s. His \u2018proof of the being of God\u2019\nis in fact Locke\u2019s with the sole _nervus probandi_ left out. The\nvalue of Locke\u2019s proof, as an argument from their being something\nnow to their having been something from eternity, lay, we saw, in\nits convertibility into an argument from the world as a system of\nrelations to a present and eternal subject of those relations. For\nits being so convertible there was this to be said, that Locke, with\nwhatever inconsistency, at least recognised the constitution of\nreality by permanent relations, though he treated the mere relation\nof external efficiency--that in virtue of which we say of nature that\nit consists of bodies outward to and acting on each other--as if it\nalone constituted the reality of the world. Berkeley\u2019s reduction\nof the \u2018primary qualities of matter\u2019 to a succession of feelings\nlogically effaces this relation, and puts nothing intelligible,\nnothing but a name, in its place. The effacement of the distinction\nbetween the real and unreal, which would properly ensue, is only\nprevented by bringing back relation to something under the name of\nGod, either wholly unknown and indeterminate, or else, under a thin\ndisguise, determined by that very relation of external efficiency\nwhich, when ascribed to something only nominally different, had been\npronounced a gratuitous fiction. If Berkeley had dealt with the\nopposition of reality to thought by showing the primary qualities\nto be conceived relations, and the distinction between the real and\nunreal to be one between the fully and the defectively conceived, the\ncase would have been different. The real and God would alike have\nbeen logically saved. The peculiar embarrassment of Locke\u2019s doctrine\nwe have found to be that it involves the unreality of every object,\ninto the constitution of which there enters any idea of reflection,\nor any idea retained in the mind, as distinct from the present effect\nof a body acting upon us--_i.e._ of every object of which anything\ncan be said. With the definite substitution of full intelligibility\nof relations for present sensibility, as the true account of the\nreal, this embarrassment would have been got rid of. At the same\ntime there would have been implied an intelligent subject of these\nrelations; the ascription to whom, indeed, of moral attributes would\nhave remained a further problem, but who, far from being a \u2018Great\nUnknown,\u2019 would be at least determined by relation to that order of\nnature which is as necessary to Him as He to it. But in fact, as\nwe have seen, the notion of the reality of relations, not felt but\nunderstood, only appears in Berkeley\u2019s developed philosophy as an\nafter-thought, and the notion of an order of nature, other than our\nfeelings, which enables us to infer what feelings that have never\nbeen felt would be, is an unexplained intrusion in it. The same is\ntrue of the doctrine, which struggles to the surface in the Third\nDialogue, that the \u2018sensible world\u2019 is to God not felt at all,\nbut known; that to Him it is precisely not that which according\nto Berkeley\u2019s refutation of materialism it really is--a series or\ncollection of sensations. These \u2018after-thoughts,\u2019 when thoroughly\nthought out, imply a complete departure from Berkeley\u2019s original\ninterpretation of \u2018phenomena\u2019 as simple feelings; but with him,\nso far from being thought out, they merely suggested themselves\nincidentally as the conceptions of God and reality were found to\nrequire them. In other words, that interpretation of phenomena, which\nis necessary to any valid \u2018collection\u2019 from them of the existence\nof God, only appears in him as a consequence of that \u2018collection\u2019\nhaving been made. To pursue the original interpretation, so that all\nmight know what it left of reality, was the best way of deciding\nthe question of its compatibility with a rational belief in God--a\nquestion of too momentous an interest to be fairly considered in\nitself. Thus to pursue it was the mission of Hume.\n\nHume\u2019s mission. His account of impressions and ideas. Ideas are\nfainter impressions.\n\n195. Hume begins with an account of the \u2018perceptions of the human\nmind,\u2019 which corresponds to Locke\u2019s account of ideas with two main\nqualifications, both tending to complete that dependence of thought\non something other than itself which Locke had asserted, but not\nconsistently maintained. He distinguishes \u2018perceptions\u2019 (equivalent\nto Locke\u2019s ideas) into \u2018impressions\u2019 and \u2018ideas\u2019 accordingly as\nthey are originally produced in feeling or reproduced by memory and\nimagination, and he does not allow \u2018ideas of reflection\u2019 any place in\nthe _original_ \u2018furniture of the mind.\u2019 \u2018An impression first strikes\nupon the senses, and makes us perceive heat or cold, thirst or\nhunger, pleasure or pain, of some kind or other. Of this impression\nthere is a copy taken by the mind, which remains after the impression\nceases; and this we call an idea. This idea of pleasure or pain, when\nit returns upon the soul, produces the new impressions of desire and\naversion, hope and fear, which may properly be called impressions\nof reflection, because derived from it. These, again, are copied by\nthe memory and imagination, and become ideas; which, perhaps, in\ntheir turn give rise to other impressions and ideas; so that the\nimpressions are only antecedent to their correspondent ideas, but\nposterior to those of sensation and derived from them\u2019 (Part I. \u00a72).\nHe is at the same time careful to explain that the causes from which\nthe impressions of sensation arise are unknown (ibid.), and that by\nthe term \u2018impression\u2019 he is not to be \u2018understood to express the\nmanner in which our lively perceptions are produced in the soul,\nbut merely the perceptions themselves\u2019. [1] The distinction between\nimpression and idea he treats as equivalent to that between feeling\nand thinking, which, again, lies merely in the different degrees of\n\u2018force and liveliness\u2019 with which the perceptions, thus designated,\nseverally \u2018strike upon the mind.\u2019 [2] Thus the rule which he\nemphasises [3] \u2018that all our simple ideas in their first appearance\nare derived from simple impressions which are correspondent to them\nand which they exactly represent,\u2019 strictly taken, means no more\nthan that a feeling must be more lively before it becomes less so.\nAs the reproduced perception, or \u2018idea,\u2019 differs in this respect\nfrom the original one, so, according to the greater or less degree\nof secondary liveliness which it possesses, is it called \u2018idea of\nmemory,\u2019 or \u2018idea of imagination.\u2019 The only other distinction noticed\nis that, as might be expected, the comparative faintness of the\nideas of imagination is accompanied by a possibility of their being\nreproduced in a different order from that in which the corresponding\nideas were originally presented. Memory, on the contrary, \u2018is in a\nmanner tied down in this respect, without any power of variation\u2019;\n[4] which must be understood to mean that, when the ideas are faint\nenough to allow of variation in the order of reproduction, they are\nnot called \u2018ideas of memory.\u2019\n\n[1] p. 312, note [Introduction to the Treatise of Human Nature]\n\n[2] See pp. 327 and 375 [Book I, part I., sec. II. and part III. sec.\nII.]\n\n[3] p. 310 [Introduction to the Treatise of Human Nature]\n\n[4] p. 318 [Book I, part I., sec. III.]\n\n\u2018Ideas\u2019 that cannot be so represented must be explained as mere words.\n\n196. All, then, that Hume could find in his mind, when after Locke\u2019s\nexample he \u2018looked into it,\u2019 were, according to his own statement,\nfeelings with their copies, dividing themselves into two main\norders--those of sensation and those of reflection, of which the\nlatter, though results of the former, are not their copies. The\nquestion, then, that he had to deal with was, to what impressions\nhe could reduce those conceptions of relation--of cause and effect,\nsubstance and attribute, and identity--which all knowledge involves.\nFailing the impressions of sensation he must try those of reflection,\nand failing both he must pronounce such conceptions to be no \u2018ideas\u2019\nat all, but words misunderstood, and leave knowledge to take its\nchance. The vital nerve of his philosophy lies in his treatment\nof the \u2018association of ideas\u2019 as a sort of process of spontaneous\ngeneration, by which impressions of sensation issue in such\nimpressions of reflection, in the shape of habitual propensities,\u2019\n[1] as will account, not indeed for there being--since there really\nare not--but for there seeming to be, those formal conceptions which\nLocke, to the embarrassment of his philosophy, had treated as at once\nreal and creations of the mind.\n\n[1] Pp. 460 and 496 [Book I, part III., sec. XIV. and part IV., sec.\nII.]\n\nHume, taken strictly, leaves no distinction between impressions of\nreflection and of sensation.\n\n197. Such a method meets at the outset with the difficulty that\nthe impressions of sensation and those of reflection, if Locke\u2019s\ndetermination of the former by reference to an impressive matter\nis excluded, are each determined only by reference to the other.\nWhat is an impression of reflection? It is one that can only\ncome after an impression of sensation. What is an impression of\nsensation? It is one that comes before any impression of reflection.\nAn apparent determination, indeed, is gained by speaking of the\noriginal impressions as \u2018conveyed to us by our senses;\u2019 but this\nreally means determination by reference to the organs of our body\nas affected by outward bodies--in short, by a physical theory. But\nof the two essential terms of this theory, \u2018our own body,\u2019 and\n\u2018outward body,\u2019 neither, according to Hume, expresses anything\npresent to the original consciousness. \u2018Properly speaking, it is\nnot our body we perceive when we regard our limbs and members,\nbut certain impressions which enter by the senses.\u2019 Nor do any of\nour impressions \u2018inform us of distance and outness (so to speak)\nimmediately, and without a certain reasoning and experience\u2019. [1] In\nsuch admissions Hume is as much a Berkeleian as Berkeley himself, and\nthey effectually exclude any reference to body from those original\nimpressions, by reference to which all other modes of consciousness\nare to be explained.\n\n[1] p. 481 [Book I, part IV., sec. II.]\n\nLocke\u2019s theory of sensation disappears. Physiology won\u2019t answer the\nquestion that Locke asked.\n\n198. He thus logically cuts off his psychology from the support\nwhich, according to popular conceptions, its primary truths derive\nfrom physiology. We have already noticed how with Locke metaphysic\nbegs defence of physic; [1] how, having undertaken to answer by\nthe impossible method of self-observation the question as to what\nconsciousness is to itself at its beginning, he in fact tells us what\nit is to the natural philosopher, who accounts for the production of\nsensation by the impact of matter \u2018on the outward parts, continued\nto the brain.\u2019 To those, of course, who hold that the only possible\ntheory of knowledge and of the human spirit is physical, it must seem\nthat this was his greatest merit; that, an unmeaning question having\nbeen asked, it was the best thing to give an answer which indeed\nis no answer to the question, but has some elementary truth of its\nown. According to them, though he may have been wrong in supposing\nconsciousness to be to itself what the physiologist explains it to\nbe--since any supposition at all about it except as a phenomenon,\nto which certain other phenomena are invariably antecedent, is at\nbest superfluous--he was not wrong in taking the physiological\nexplanation to be the true and sufficient one. To such persons we can\nbut respectfully point out that they have not come in sight of the\nproblem which Locke and his followers, on however false a method,\nsought to solve; that, however certain may be the correlation between\nthe brain and thought, in the sense that the individual would be\nincapable of the processes of thought unless he had brain and nerves\nof a particular sort, yet it is equally certain that every theory of\nthe correlation must presuppose a knowledge of the processes, and\nleave that knowledge exactly where it was before; that thus their\nscience, valuable like every other science within its own department,\ntakes for granted just what metaphysic, as a theory of knowledge,\nseeks to explain. When the origin, for instance, of the conception\nof body or of that of an organic structure is in question, it is\nin the strictest sense preposterous to be told that body makes the\nconception of body, and that unless the brain were organic to thought\nI should not now be thinking. \u2018The brain is organic to thought;\u2019 here\nis a proposition involving conceptions within conceptions--a whole\nhierarchy of ideas. How am I enabled to re-think these in order, to\nmake my way from the simpler to the more complex, by any iteration\nor demonstration of the proposition, which no one disputes, or by\nthe most precise examination of the details of the organic structure\nitself?\n\n[1] See above, paragraph 17.\n\nThose who think it will don\u2019t understand the question.\n\n199. The quarrel of the physiologist with the metaphysician is, in\nfact, due to an _ignorantia elenchi_ on the part of the former, for\nwhich the behaviour of English \u2018metaphysicians,\u2019 in attempting to\nassimilate their own procedure to that of the natural philosophers,\nand thus to win the popular acceptance which these alone can fairly\nlook for, has afforded too much excuse. The question really at issue\nis not between two co-ordinate sciences, as if a theory of the human\nbody were claiming also to be a theory of the human soul, and the\ntheory of the soul were resisting the aggression. The question is,\nwhether the conceptions which all the departmental sciences alike\npresuppose shall have an account given of them or no. For dispensing\nwith such an account altogether (life being short) there is much\nto be said, if only men would or could dispense with it; but the\nphysiologist, when he claims that his science should supersede\nmetaphysic, is not dispensing with it, but rendering it in a\npreposterous way. He accounts for the formal conceptions in question,\nin other words for thought as it is common to all the sciences, as\nsequent upon the antecedent facts which his science ascertains--the\nfacts of the animal organisation. But these conceptions--the\nrelations of cause and effect, &c.--are necessary to constitute the\nfacts. They are not an _ex post facto_ interpretation of them, but an\ninterpretation without which there would be no ascertainable facts\nat all. To account for them, therefore, as the result of the facts\nis to proceed as a geologist would do, who should treat the present\nconformation of the earth as the result of a certain series of past\nevents, and yet, in describing these, should assume the present\nconformation as a determining element in each.\n\nHume\u2019s psychology will not answer it either.\n\n200. \u2018Empirical psychology,\u2019 however, claims to have a way of its\nown for explaining thought, distinct from that of the physiologist,\nbut yet founded on observation, though it is admitted that the\nobservation takes place under difficulties. Its method consists in\na history of consciousness, as a series of events or successive\nstates observed in the individual by himself. By tracing such a chain\nof _de facto_ sequence it undertakes to account for the elements\ncommon to all knowledge. Its first concern, then, must be, as we\nhave previously put it, to ascertain what consciousness is to itself\nat its beginning. No one with Berkeley before him, and accepting\nBerkeley\u2019s negative results, could answer this question in Locke\u2019s\nsimple way by making the primitive consciousness report itself as an\neffect of the operation of body. To do so is to transfer a later and\nhighly complex form of consciousness, whose growth has to be traced,\ninto the earlier and simple form from which the growth is supposed to\nbegin. This, upon the supposition that the process of consciousness\nby which conceptions are formed is a series of psychical events--a\nsupposition on which the whole method of empirical psychology\nrests--is in principle the same false procedure as that which we\nhave imagined in the case of a geologist above. But the question\nis whether, by any procedure not open to this condemnation, the\ntheory could seem to do what it professes to do--explain thought\nor \u2018cognition by means of conceptions\u2019 as something which happens\nin sequence upon previous psychical events. Does it not, however\nstated, carry with it an implication of the supposed later state in\nthe earlier, and is it not solely in virtue of this implication that\nit seems to be able to trace the genesis of the later? No one has\npursued it with stricter promises, or made a fairer show of being\nfaithful to them, than Hume. He will begin with simple feeling,\nas first experienced by the individual--unqualified by complex\nconceptions, physical or metaphysical, of matter or of mind--and\ntrace the process by which it generates the \u2018ideas of philosophical\nrelation.\u2019 If it can be shown, as we believe it can be, that, even\nwhen thus pursued, its semblance of success is due to the fact that,\nby interpreting the earliest consciousness in terms of the latest, it\nputs the latter in place of the former, some suspicion may perhaps\nbe created that a natural history of self-consciousness, and of the\nconceptions by which it makes the world its own, is impossible,\nsince such a history must be of events, and self-consciousness is\nnot reducible to a series of events; being already at its beginning\nformally, or potentially, or implicitly all that it becomes actually\nor explicitly in developed knowledge.\n\nIt only seems to do so by assuming the \u2018fiction\u2019 it has to account\nfor; by assuming that impression represents a real world.\n\n201. If Hume were consistent in allowing no other determination to\nthe impression than that of its having the maximum of vivacity, or\nto other modes of consciousness than the several degrees of their\nremoval from this maximum, he would certainly have avoided the\ndifficulties which attend Locke\u2019s use of the metaphor of impression,\nwhile at the same time he would have missed the convenience, involved\nin this use, of being able to represent the primitive consciousness\nas already a recognition of a thing impressing it, and thus an\n\u2018idea of a quality of body.\u2019 But at the outset he remarks that \u2018the\nexamination of our sensations\u2019 (_i.e._ our impressions of sensation)\n\u2018belongs more to anatomists and natural philosophers than to moral,\u2019\nand that for that reason he shall begin not with them but with\nideas. [1] Now this virtually means that he will begin, indeed,\nwith the feelings he finds in himself, but with these as determined\nby the notion that they are results of something else, of which\nthe nature is not for the present explained. Thus, while he does\nnot, like Locke, identify our earliest consciousness with a rough\nand ready physical theory of its cause, he gains the advantage of\nthis identification in the mind of his reader, who from sensation,\nthus apparently defined, transfers a definiteness to the ideas and\nsecondary impressions as derived from it, though in the sequel the\ntheory turns out, if possible at all, to be at best a remote result\nof custom and association. We shall see this more clearly if we look\nback to the general account of impressions and ideas quoted above.\n\u2018An impression first strikes upon the senses and makes us perceive\npleasure or pain, of which a copy is taken by the mind,\u2019 called\nan idea. Now if we set aside the notion of a body making impact\nupon a sensuous, and through it upon a mental, tablet, pleasure or\npain _is_ the impression, which, again, is as much or as little in\nthe mind as the idea. Thus the statement might be re-written as\nfollows:--\u2018Pleasure or pain makes the mind perceive pleasure or pain,\nof which a copy is taken by the mind.\u2019 This, of course, is nonsense;\nbut between this nonsense and the plausibility of the statement as it\nstands, the difference depends on the double distinction understood\nin the latter--the distinction _(a)_ between the producing cause of\nthe impression and the impression produced; and _(b)_ between the\nimpression as produced on the senses, and the idea as preserved by\nthe mind. This passage, as we shall see, is only a sample of many of\nthe same sort. Throughout, however explicitly Hume may give warning\nthat the difference between impression and idea is only one of\nliveliness, however little he may scruple in the sequel to reduce\nbody and mind alike to the succession of feelings, his system gains\nthe benefit of the contrary assumption which the uncritical reader\nis ready to make for him. As often as the question returns whether\na phrase, purporting to express an \u2018abstract conception,\u2019 expresses\nany actual idea or no, his test is, \u2018Point out the impression from\nwhich the idea, if there be any, is derived\u2019--a test which has\nclearly no significance if the impression is merely the idea itself\nat a livelier stage (for a person, claiming to have the idea,\nwould merely have to say that he had never known it more lively,\nand that, therefore, it was itself an impression, and the force of\nthe test would be gone), but which seems so satisfactory because\nthe impression is regarded as the direct effect of outward things,\nand thus as having a prerogative of reality over any perception to\nwhich the mind contributes anything of its own. By availing himself\nalternately of this popular conception of the impression of sensation\nand of his own account of it, he gains a double means of suppressing\nany claim of thought to originate. Every idea, by being supposed in a\nmore lively state, can be represented as derived from an impression,\nand thus (according to the popular notion) as an effect of something\nwhich, whatever it is, is not thought. If thereupon it is pointed out\nthat this outward something is a form of substance which, according\nto Hume\u2019s own showing, is a fiction of thought, there is an easy\nrefuge open in the reply that \u2018impression\u2019 is only meant to express\na lively feeling, not any dependence upon matter of which we know\nnothing.\n\n[1] p. 317 [Book I, part I., sec. III.]\n\nSo the \u2018Positivist\u2019 juggles with \u2018phenomena\u2019.\n\n202. Thus the way is prepared for the juggle which the modern popular\nlogic performs with the word \u2018phenomenon\u2019--a term which gains\nacceptance for the theory that turns upon it because it conveys the\nnotion of a relation between a real order and a perceiving mind,\nand thus gives to those who avail themselves of it the benefit\nof an implication of the \u2018noumena\u2019 which they affect to ignore.\nHume\u2019s inconsistency, however, stops far short of that of his later\ndisciples. For the purpose of detraction from the work of thought he\navailed himself, indeed, of that work as embodied in language, but\nonly so far as was necessary to his destructive purpose. He did not\nseriously affect to be reconstructing the fabric of knowledge on a\nbasis of fact. There occasionally appears in him, indeed, something\nof the charlatanry of common sense in passages, more worthy of\nBolingbroke than himself, where he writes as a champion of facts\nagainst metaphysical jargon. But when we get behind the mask of\nconcession to popular prejudice, partly ironical, partly due to his\nundoubted vanity, we find much more of the ancient sceptic than of\nthe \u2018positive philosopher.\u2019\n\nEssential difference, however, between Hume and the \u2018Positivist\u2019.\n\n203. The ancient sceptic (at least as represented by the ancient\nphilosophers), finding knowledge on the basis of distinction between\nthe real and apparent to be impossible, discarded the enterprise\nof arriving at general truth in opposition to what appears to\nthe individual at any particular instant, and satisfied himself\nwith noting such general tendencies of expectation and desire as\nwould guide men in the conduct of life and enable them to get\nwhat they wanted by contrivance and persuasion. [1] Such a state\nof mind excludes all motive to the \u2018interrogation of nature,\u2019\nfor it recognises no \u2018nature\u2019 but the present appearance to the\nindividual; and this does not admit of being interrogated. The\n\u2018positive philosopher\u2019 has nothing in common with it but the use,\nin a different sense, of the word \u2018apparent.\u2019 He plumes himself,\nindeed, on not going in quest of any \u2018thing-in-itself\u2019 other than\nwhat appears to the senses; but he distinguishes between a real\nand apparent in the order of appearance, and considers the real\norder of appearance, having a permanence and uniformity which\nbelong to no feeling as the individual feels it, to be the true\nobject of knowledge. No one is more severe upon \u2018propensities to\nbelieve,\u2019 however spontaneously suggested by the ordinary sequence\nof appearances, if they are found to conflict with the order of\nnature as ascertained by experimental interrogation; _i.e._ with a\nsequence observed (it may be) in but a single instance. Which of the\ntwo attitudes of thought is the more nearly Hume\u2019s, will come out as\nwe proceed. It was just with the distinction between the \u2018real and\nfantastic,\u2019 as Locke had left it, that he had to deal; and, as will\nappear, it is finally by a \u2018propensity to feign,\u2019 not by a uniform\norder of natural phenomena, that he replaces the real which Locke,\naccording to his first mind, had found in archetypal things and their\noperations on us.\n\n[1] Cf. Plato\u2019s \u2018Protagoras,\u2019 323, and \u2018Theaetetus,\u2019 167, with the\nconcluding paragraphs of the last part of the first book of Hume\u2019s\n\u2018Treatise on Human Nature.\u2019\n\nHe adopts Berkeley\u2019s doctrine of ideas, but without Berkeley\u2019s saving\nsuppositions,\n\n204. We have seen that Berkeley, having reduced \u2018simple ideas\u2019 to\ntheir simplicity by showing the illegitimacy of the assumption that\nthey report qualities of a matter which is itself a complex idea, is\nonly able to make his constructive theory march by the supposition of\nthe reality and knowability of \u2018spirit\u2019 and relations. \u2018Ideas\u2019 are\n\u2018fleeting, perishable passions;\u2019 but the relations between them are\nuniform, and in virtue of this uniformity the fleeting idea may be\ninterpreted as a symbol of a real order. But such relations, as real,\nimply the presence of the ideas to the constant mind of God, and, as\nknowable, their presence to a like mind in us. We have further seen\nhow little Berkeley, according to the method by which he disposed of\n\u2018abstract general ideas,\u2019 was entitled to such a supposition. Hume\nsets it aside; but the question is, whether without a supposition\nvirtually the same he can represent the association of ideas as doing\nthe work that he assigned to it.\n\n... in regard to \u2018spirit\u2019,\n\n205. His exclusion of Berkeley\u2019s supposition with regard to \u2018spirit\u2019\nis stated without disguise, though unfortunately not till towards\nthe end of the first book of the \u2018Treatise on Human Nature,\u2019 which\ncould not have run so smoothly if the statement had been made at\nthe beginning. It follows legitimately from the method, which he\ninherited, of \u2018looking into his mind to see how it wrought.\u2019 \u2018From\nwhat impression,\u2019 he asks, \u2018could the idea of self be derived? It\nmust be some one impression that gives rise to every real idea.\nBut self or person is not any impression, but that to which our\nseveral impressions and ideas are supposed to have a reference.\nIf any impression gives rise to the idea of self, that impression\nmust continue invariably the same through the whole course of our\nlives, since self is supposed to exist after that manner. But there\nis no impression constant and invariable. Pain and pleasure, grief\nand joy, passions and sensations succeed each other, and never\nall exist at the same time. It cannot, therefore, be from any of\nthese impressions, or from any other, that the idea of self is\nderived; and, consequently, there is no such idea.\u2019 Again: \u2018When I\nenter most intimately into what is called myself, I always stumble\non some particular perception of heat or cold, light or shade,\nlove or hatred, pain or pleasure. I never can catch myself at any\ntime without a perception, and never can observe anything but the\nperception. When my perceptions are removed for any time, as by\nsound sleep, so long am I insensible of myself, and may truly be\nsaid not to exist.\u2019 Thus \u2018men are nothing but a bundle or collection\nof different perceptions that succeed each other with inconceivable\nrapidity, and are in a perpetual flux or movement. Our eyes cannot\nturn in their sockets without varying our perceptions. Our thought\nis still more variable than our sight. ... nor is there any single\npower of the soul which remains unalterably the same perhaps for one\nmoment.... There is properly no simplicity in the mind at one time\nnor identity at different\u2019. [1]\n\n[1] pp. 533 and 534 [Book I., part IV., sec. VI.]\n\n... in regard to relations. His account of these.\n\n206. His position in regard to ideas of relation cannot be so\nsummarily exhibited. It is from its ambiguity, indeed, that his\nsystem derives at once its plausibility and its weakness. In the\nfirst place, it is necessary, according to him, to distinguish\nbetween \u2018natural\u2019 and \u2018philosophical relation.\u2019 The latter is one\nof which the idea is acquired by the comparison of objects, as\ndistinct from natural relation or \u2018the quality by which two ideas\nare connected together in the imagination, and the one naturally\u2019\n(_i.e._ according to the principle of association) \u2018introduces the\nother\u2019. [1] Of philosophical relation--or, according to another form\nof expression, of \u2018qualities by which the ideas of philosophical\nrelation are produced\u2019--seven kinds are enumerated; viz.\n\u2018resemblance, identity, relations of time and place, proportion in\nquantity and number, degrees in quality, contrariety, and causation\u2019.\n[2] Some of these do, some do not, _apparently_ correspond to the\nqualities by which the mind is _naturally_ \u2018conveyed from one idea\nto another;\u2019 or which, in other words, constitute the \u2018gentle force\u2019\nthat determines the order in which the imagination habitually puts\ntogether ideas. Freedom in the conjunction of ideas, indeed, is\nimplied in the term \u2018imagination,\u2019 which is only thus differenced\nfrom \u2018memory;\u2019 but, as a matter of fact, it commonly only connects\nideas which are related to each other in the way either of\nresemblance, or of contiguity in time and place, or of cause and\neffect. Other relations of the philosophical sort are the opposite\nof _natural_. Thus, \u2018distance will be allowed by philosophers to be\na true relation, because we acquire an idea of it by the comparing\nof objects; but in a common way we say, \u201cthat nothing can be more\ndistant than such or such things from each other; nothing can have\nless relation\u201d\u2019 (ibid.).\n\n[1] p. 322 [Book I, part I., sec. V.]\n\n[2] ibid., and p. 372 [Book I., part III., sec. I.]\n\nIt corresponds to Locke\u2019s account of the sorts of agreement between\nideas.\n\n207. Hume\u2019s classification of philosophical relations evidently\nserves the same purpose as Locke\u2019s, of the \u2018four sorts of agreement\nor disagreement between ideas,\u2019 in the perception of which\nknowledge consists; [1] but there are some important discrepancies.\nLocke\u2019s second sort, which he awkwardly describes as \u2018agreement or\ndisagreement in the way of relation,\u2019 may fairly be taken to cover\nthree of Hume\u2019s kinds; viz. relations of time and place, proportion\nin quantity or number, and degrees in any quality. About Locke\u2019s\nfirst sort, \u2018identity and diversity,\u2019 there is more difficulty. Under\n\u2018identity,\u2019 as was pointed out above, he includes the relations\nwhich Hume distinguishes as \u2018identity proper\u2019 and \u2018resemblance.\u2019\n\u2018Diversity\u2019 at first sight might seem to correspond to \u2018contrariety;\u2019\nbut the latter, according to Hume\u2019s usage, is much more restricted\nin meaning. Difference of number and difference of kind, which he\ndistinguishes as the opposites severally of identity and resemblance,\nthough they come under Locke\u2019s \u2018diversity,\u2019 are not by Hume\nconsidered relations at all, on the principle that \u2018no relation of\nany kind can subsist without some degree of resemblance.\u2019 They are\n\u2018rather a negation of relation than anything real and positive.\u2019\n\u2018Contrariety\u2019 he reckons only to obtain between ideas of existence\nand non-existence, \u2018which are plainly resembling as implying both of\nthem an idea of the object; though the latter excludes the object\nfrom all times and places in which it is supposed not to exist\u2019.\n[2] There remain \u2018cause and effect\u2019 in Hume\u2019s list; \u2018co-existence\u2019\nand \u2018real existence\u2019 in Locke\u2019s. \u2018Co-existence\u2019 is not expressly\nidentified by Locke with the relation of cause and effect, but it\nis with \u2018necessary connection.\u2019 It means specially, it will be\nremembered [3], the co-existence of ideas, not as constituents of a\n\u2018nominal essence,\u2019 but as qualities of real substances in nature;\nand our knowledge of this depends on our knowledge of necessary\nconnection between the qualities, either as one supposing the\nother (which is the form of necessary connection between primary\nqualities), or as one being the effect of the other (which is the\nform of necessary connection between the ideas of secondary qualities\nand the primary ones). Having no knowledge of necessary connection\nas in real substances, we have none of \u2018co-existence\u2019 in the above\nsense, but only of the present union of ideas in any particular\nexperiment. [4] The parallel between this doctrine of Locke\u2019s and\nHume\u2019s of cause and effect will appear as we proceed. To \u2018real\nexistence,\u2019 since the knowledge of it according to Locke\u2019s account is\nnot a perception of agreement between ideas at all, it is not strange\nthat nothing should correspond in Hume\u2019s list of relations.\n\n[1] See above, paragraph 25 and the passages from Locke there\nreferred to.\n\n[2] p. 323 [Book I, part I., sec. V.]\n\n[3] See above, paragraph 122.\n\n[4] Locke, Book IV. sec. iii. chap. xiv.; and above, paragraph 121\nand 122.\n\nCould Hume consistently admit idea of relation at all?\n\n208. It is his method of dealing with these ideas of philosophical\nrelation that is specially characteristic of Hume, Let us, then,\nconsider how the notion of relation altogether is affected by his\nreduction of the world of consciousness to impressions and ideas.\nWhat is an impression? To this, as we have seen, the only direct\nanswer given by him is that it is a feeling which must be more lively\nbefore it becomes less so. [1] For a further account of what is to\nbe understood by it we must look to the passages where the governing\nterms of \u2018school-metaphysics\u2019 are, one after the other, shown to be\nunmeaning, because not taken from impressions. Thus, when the idea\nof substance is to be reduced to an \u2018unintelligible chimaera,\u2019 it is\nasked whether it \u2018be derived from the impressions of sensation or\nreflection? If it be conveyed to us by our senses, I ask, which of\nthem, and after what manner? If it be perceived by the eyes, it must\nbe a colour; if by the ears, a sound; if by the palate, a taste; and\nso of the other senses. But I believe none will assert that substance\nis either a colour, or a sound, or a taste. The idea of substance\nmust therefore be derived from an impression of reflection, if it\nreally exist. But the impressions of reflection resolve themselves\ninto our passions and emotions\u2019. [2] From the polemic against\nabstract ideas we learn further that \u2018the appearance of an object to\nthe senses\u2019 is the same thing as an \u2018impression becoming present to\nthe mind\u2019. [3] That is to say, when we talk of an impression of an\nobject, it is not to be understood that the feeling is determined by\nreference to anything other than itself: it is itself the object.\nTo the same purpose, in the criticism of the notion of an external\nworld, we are told that \u2018the senses are incapable of giving rise\nto the notion of the continued existence of their objects, after\nthey no longer appear to the senses; for that is a contradiction\nin terms\u2019 (since the appearance _is_ the object); and that \u2018they\noffer not their impressions as the images of something distinct, or\nindependent, or external, because they convey to us nothing but a\nsingle perception, and never give us the least intimation of anything\nbeyond\u2019. [4] The distinction between impression of sensation and\nimpression of reflection, then, cannot, any more than that between\nimpression and idea, be regarded as either really or apparently a\ndistinction between outer and inner. \u2018All impressions are internal\nand perishing existences\u2019; [5] and, \u2018everything that enters the mind\nbeing in reality as the impression, \u2019tis impossible anything should\nto feeling appear different\u2019. [6]\n\n[1] See above, paragraphs 195 and 197.\n\n[2] p. 324 [Book I, part I., sec. VI.]\n\n[3] p. 327 [Book I, part I., sec. VII.]\n\n[4] p. 479 [Book I, part IV., sec. II.]\n\n[5] p. 483 [Book I, part IV., sec. II.]\n\n[6] p. 480 [Book I, part IV., sec. II.]\n\n209. This amounts to a full acceptance of Berkeley\u2019s doctrine\nof sense; and the question necessarily arises--such being the\nimpression, and all ideas being impressions grown weaker, can there\nbe an idea of relation at all? Is it not open to the same challenge\nwhich Hume offers to those who talk of an idea of substance or of\nspirit? \u2018It is from some one impression that every real idea is\nderived.\u2019 What, then, is the one impression from which the idea of\nrelation is derived? \u2018If it be perceived by the eyes, it must be a\ncolour; if by the ears, a sound; if by the palate, a taste; and so of\nthe other senses.\u2019 There remain \u2018our passions and emotions;\u2019 but what\npassion or emotion is a resemblance, or a proportion, or a relation\nof cause and effect?\n\nOnly in regard to identity and causation that he sees any difficulty.\nThese he treats as fictions resulting from \u2018natural relations\u2019 of\nideas: _i.e._ from resemblance and contiguity.\n\n210. Respect for Hume\u2019s thoroughness as a philosopher must be\nqualified by the observation that he does not attempt to meet this\ndifficulty in its generality, but only as it affects the relations\nof identity and causation. The truth seems to be that he wrote with\nBerkeley steadily before his mind; and it was Berkeley\u2019s treatment of\nthese two relations in particular as not sensible but intelligible,\nand his assertion of a philosophic Theism on the strength of their\nmere intelligibility, that determined Hume, since it would have been\nan anachronism any longer to treat them as sensible, to dispose of\nthem altogether. The condition of his doing so with success was\nthat, however unwarrantably, he should treat the other relations\nas sensible. The language, which seems to express ideas of the two\nquestionable relations, he has to account for as the result of\ncertain impressions of reflection, called \u2018propensities to feign,\u2019\nwhich in their turn have to be accounted for as resulting from the\n_natural_ relations of ideas according to the definition of these\nquoted above, [1] as \u2018the qualities by which one idea habitually\nintroduces another.\u2019 Among these, as we saw, he included not\nonly resemblance and contiguity in time or place, but \u2018cause and\neffect.\u2019 \u2018There is no relation,\u2019 he says, \u2018which produces a stronger\nconnection in the fancy than this.\u2019 But in this, as in much of the\nlanguage which gives the first two Parts their plausibility, he is\ntaking advantage of received notions on the part of the reader, which\nit is the work of the rest of the book to set aside. In any sense,\naccording to him, in which it differs from usual contiguity, the\nrelation of cause and effect is itself reducible to a \u2018propensity\nto feign\u2019 arising from the other natural relations; but when the\nreader is told of its producing \u2018a strong connection in the fancy,\u2019\nhe is not apt to think of it as itself nothing more than the product\nof such a connection. For the present, however, we have only to\npoint out that Hume, when he co-ordinates it with the other natural\nrelations, must be understood to do so provisionally. According to\nhim it is derived, while they are primary. Upon them, then, rested\nthe possibility of filling the gap between the occurrence of single\nimpressions, none \u2018determined by reference to anything other than\nitself,\u2019 and what we are pleased to call our knowledge, with its\nfictions of mind and thing, of real and apparent, of necessary as\ndistinct from usual connection.\n\n[1] See above, paragraph 206.\n\n211. We will begin with Resemblance. As to this, it will be said, it\nis an affectation of subtlety to question whether there can be an\nimpression of it or no. The difficulty only arises from our regarding\nthe perception of resemblance as different from, and subsequent to,\nthe resembling sensations; whereas, in fact, the occurrence of two\nimpressions of sense, such as (let us say) yellow and red, is itself\nthe impression of their likeness and unlikeness. Hume himself, it may\nbe further urged, at any rate in regard to resemblance, anticipates\nthis solution of an imaginary difficulty by his important division\nof philosophical relations into two classes [1]--\u2018such as depend\nentirely on the ideas which we compare together, and such as may be\nchanged without any change in the ideas\u2019--and by his inclusion of\nresemblance in the former class.\n\n[1] p. 372 [Book I, part III., sec. I.]\n\nIs resemblance then an impression?\n\n212. Now we gladly admit the mistake of supposing that sensations\nundetermined by relation first occur, and that afterwards we become\nconscious of their relation in the way of likeness or unlikeness.\nApart from such relation, it is true, the sensations would be\nnothing. But this admission involves an important qualification\nof the doctrine that impressions are single, and that the mind\n(according to Hume\u2019s awkward figure) is a \u2018bundle or collection of\nthese,\u2019 succeeding each other \u2018in a perpetual flux or movement.\u2019 It\nimplies that the single impression in its singleness is what it is\nthrough relation to another, which must therefore be present along\nwith it; and that thus, though they may occur in a perpetual flux\nof succession--every turn of the eyes in their sockets, as Hume\ntruly says, giving a new one--yet, just so far as they are qualified\nby likeness or unlikeness to each other, they must be taken out\nof that succession by something which is not itself in it, but is\nindivisibly present to every moment of it. This we may call soul, or\nmind, or what we will; but we must not identify it with the brain\n[1] either directly or by implication (as we do when we \u2018refer to\nthe anatomist\u2019 for an account of it), since by the brain is meant\nsomething material, _i.e._ divisible, which the unifying subject\nspoken of, as feeling no less than as thinking, cannot be. In short,\nany such modification of Hume\u2019s doctrine of the singleness and\nsuccessiveness of impressions as will entitle us to speak of their\ncarrying with them, though single and successive, the consciousness\nof their resemblance to each other, will also entitle us to speak of\ntheir carrying with them a reference to that which is not itself any\nsingle impression, but is permanent throughout the impressions; and\nthe whole ground of Hume\u2019s polemic against the idea of self or spirit\nis removed. [2]\n\n[1] It is, of course, quite a different thing to say that the brain\n(or, more properly, the whole body) is organic to it.\n\n[2] See above, paragraph 205.\n\nDistinction between resembling feelings and idea of resemblance.\n\n213. The above admission, however, does not dispose of the question\nabout ideas of resemblance. A feeling qualified by relation of\nresemblance to other feelings is a different thing from an idea\nof that relation--different with all the difference which Hume\nignores between feeling and thought, between consciousness and\nself-consciousness. The qualification of successive feelings by\nmutual relation implies, indeed, the presence to them of a subject\npermanent and immaterial (_i.e._ not in time or space); but it does\nnot imply that this subject presents them to itself as related\nobjects, permanent with its own permanence, which abide and may\nbe considered apart from \u2018the circumstances in time\u2019 of their\noccurrence. Yet such presentation is supposed by all language other\nthan interjectional. It is it alone which can give us names of\nthings, as distinct from noises prompted by the feelings as they\noccur. Of course it is open to any one to say that by an idea of\nresemblance he does not mean any thought involving the self-conscious\npresentation spoken of, but merely a feeling qualified by\nresemblance, and not at its liveliest stage. Thus Hume tells us that\nby \u2018idea\u2019 he merely means a feeling less lively than it has been, and\nthat by idea _of anything_ he implies no reference to anything other\nthan the idea, [1] but means just a related idea, _i.e._ a feeling\nqualified by \u2018natural relation\u2019 to other feelings. It is by this\nthoughtful abnegation of thought, as we shall find, that he arrives\nat his sceptical result. But language (for the reason mentioned)\nwould not allow him to be faithful to the abnegation. He could not\nmake such a profession without being false to it. This appears\nalready in his account of \u2018complex\u2019 and \u2018abstract\u2019 ideas.\n\n[1] See above, paragraph 208.\n\nSubstances = collections of ideas.\n\n214, His account of the idea of a substance [1] is simply Locke\u2019s, as\nLocke\u2019s would become upon elimination of the notion that there is a\nreal \u2018something\u2019 in which the collection of ideas subsist, and from\nwhich they result. It thus avoids all difficulties about the relation\nbetween nominal and real essence. Just as Locke says that in the case\nof a \u2018mixed mode\u2019 the nominal essence _is_ the real, so Hume would\nsay of a substance. The only difference is that while the collection\nof ideas, called a mixed mode, does not admit of addition without\na change of its name, that called a substance does. Upon discovery\nof the solubility of gold in aqua regia we add that idea to the\ncollection, to which the name \u2018gold\u2019 has previously been assigned,\nwithout disturbance in the use of the name, because the name already\ncovers not only the ideas of certain qualities, but also the idea of\na \u2018principle of union\u2019 between them, which will extend to any ideas\npresented along with them. As this principle of union, however, is\nnot itself any \u2018real essence,\u2019 but \u2018part of the complex idea,\u2019 the\nquestion, so troublesome to Locke, whether a proposition about gold\nasserts real co-existence or only the inclusion of an idea in a\nnominal essence, will be superfluous. How the \u2018principle of union\u2019 is\nto be explained, will appear below. [2]\n\n[1] p. 324 [Book I, part I., sec. VI.]\n\n[2] Paragraph 303, and the following.\n\nHow can ideas \u2018in flux\u2019 be collected?\n\n215. There are names, then, which represent \u2018collections of ideas.\u2019\nHow can we explain such collection if ideas are merely related\nfeelings grown\u2019 fainter? Do we, when we use one of these names\nsignificantly, recall, though in a fainter form, a series of feelings\nthat we have experienced in the process of collection? Does the\nchemist, when he says that gold is soluble in aqua regia, recall the\nvisual and tactual feeling which he experienced when he found it\nsoluble? If so, as that feeling took its character from relation to\na multitude of other \u2018complex ideas,\u2019 he must on the same principle\nrecall in endless series the sensible occurrences from which each\nconstituent of each constituent of these was derived; and a like\nprocess must be gone through when gold is pronounced ductile,\nmalleable, &c. But this would be, according to the figure which\nHume himself adopts, to recall a \u2018perpetual flux.\u2019 The very term\n\u2018collection of ideas,\u2019 indeed, if this be the meaning of ideas, is\nan absurdity, for how can a perpetual flux be collected? If we turn\nfor a solution of the difficulty to the chapter where Hume expressly\ndiscusses the significance of general names, we shall find that it\nis not the question we have here put, and which flows directly from\nhis account of ideas, that he is there treating, but an entirely\ndifferent one, and one that could not be raised till for related\nfeeling had been substituted the thought of an object under relations.\n\nAre there general ideas? Berkeley said, \u2018yes and no\u2019.\n\n216. The chapter mentioned concerns the question which arises out of\nLocke\u2019s pregnant statement that words and ideas are \u2018particular in\ntheir existence\u2019 even when \u2018general in their signification.\u2019 From\nthis statement we saw [1] that Berkeley derived his explanation of\nthe apparent generality of ideas--the explanation, namely, which\nreduces it to a relation, yet not such a one as would affect the\nnature of the idea itself, which is and remains \u2018particular,\u2019 but\na symbolical relation between it and other particular ideas for\nwhich it is taken to stand. An idea, however, that carries with it a\nconsciousness of symbolical relation to other ideas, cannot but be\nqualified by this relation. The generality must become part of its\n\u2018nature,\u2019 and, accordingly, the distinction between idea and thing\nbeing obliterated, of the nature of things. Thus Berkeley virtually\narrives at a result which renders unmeaning his preliminary exclusion\nof universality from \u2018the absolute, positive nature or conception of\nanything.\u2019 Hume seeks to avoid it by putting \u2018custom\u2019 in the place\nof the consciousness of symbolical relation. True to his vocation of\nexplaining away all functions of thought that will not sort with the\ntreatment of it as \u2018decaying sense,\u2019 he would resolve that idea of a\nrelation between certain ideas, in virtue of which one is taken to\nstand for the rest, into the _de facto_ sequence upon one of them of\nthe rest. Here, as everywhere else, he would make related feelings do\ninstead of relations of ideas; but whether the related feelings, as\nhe is obliged to describe them, do not already presuppose relations\nof ideas in distinction from feelings, remains to be seen.\n\n[1] Above, paragraphs 182 and 183.\n\nHume \u2018no\u2019 simply. How he accounts for the appearance of there being\nsuch.\n\n217. The question about \u2018generality of signification,\u2019 as he puts it,\ncomes to this. In every proposition, though its subject be a common\nnoun, we necessarily present to ourselves some one individual object\n\u2018with all its particular circumstances and proportions.\u2019 How then can\nthe proposition be general in denotation and connotation? How can it\nbe made with reference to a multitude of individual objects other\nthan that presented to the mind, and how can it concern only such of\nthe qualities of the latter as are common to the multitude? The first\npart of the question is answered as follows:-\u2018When we have found a\nresemblance among several objects that often occur to us, we apply\nthe same name to all of them ... whatever differences may appear\namong them. After we have acquired a custom of this kind, the hearing\nof that name revives the idea of one of these objects, and makes the\nimagination conceive it with all its particular circumstances and\nproportions. But as the same word is supposed to have been frequently\napplied to other individuals, that are different in many respects\nfrom that idea which is immediately present to the mind, the word\nnot being able to revive the idea of all these individuals, only\ntouches the soul and revives that custom which we have acquired by\nsurveying them. They are not really and in fact present to the mind,\nbut only in power. ... The word raises up an individual idea along\nwith a certain custom, and that custom produces any other individual\none for which we may have occasion. ... Thus, should we mention the\nword triangle and form the idea of a particular equilateral one to\ncorrespond to it, and should we afterwards assert _that the three\nangles of a triangle are equal to each other_, the other individuals\nof a scalenum and isosceles, which we overlooked at first,\nimmediately crowd in upon us and make us perceive the falsehood of\nthis proposition, though it be true with relation to that idea which\nwe had formed\u2019. [1]\n\n[1] p. 328 [Book I, part I., sec. VII.]\n\n218. Next, as to the question concerning connotation:--\u2018The mind\nwould never have dreamed of distinguishing a figure from the body\nfigured, as being in reality neither distinguishable nor different\nnor separable, did it not observe that even in this simplicity there\nmight be contained many different resemblances and relations. Thus,\nwhen a globe of white marble is presented, we receive only the\nimpression of a white colour disposed in a certain form, nor are\nwe able to distinguish and separate the colour from the form. But\nobserving afterwards a globe of black marble and a cube of white,\nand comparing them with our former object, we find two separate\nresemblances in what formerly seemed, and really is, perfectly\ninseparable. After a little more practice of this kind, we begin\nto distinguish the figure from the colour by a _distinction of\nreason_;--_i.e._ we consider the figure and colour together, since\nthey are, in effect, the same and indistinguishable; but still view\nthem in different aspects according to the resemblances of which they\nare susceptible. ... A person who desires us to consider the figure\nof a globe of white marble without thinking on its colour, desires an\nimpossibility; but his meaning is, that we should consider the colour\nand figure together, but still keep in our eye the resemblance to the\nglobe of black marble or that to any other globe whatever\u2019. [1]\n\n[1] p. 333 [Book I, part I., sec. VII.]\n\nHis account implies that \u2018ideas\u2019 are conceptions, not feelings.\n\n219. It is clear that the process described in these passages\nsupposes \u2018ab initio\u2019 the conversion of a feeling into a conception;\nin other words, the substitution of the definite individuality of\na thing, thought of under attributes, for the mere singleness in\ntime of a feeling that occurs after another and before a third. The\n\u2018finding of resemblances and differences among objects that often\noccur to us\u2019 implies that each object is distinguished as one and\nabiding from manifold occurrences, in the way of related feelings, in\nwhich it is presented to us, and that these accordingly are regarded\nas representing permanent relations or qualities of the object. Thus\nfrom being related feelings, whether more or less \u2018vivacious,\u2019 they\nhave become, in the proper sense, ideas of relation. The difficulty\nabout the use of general names, as Hume puts it, really arises just\nfrom the extent to which this process of determination by ideas of\nrelation, and with it the removal of the object of thought from\nsimple feeling, is supposed to have gone. It is because the idea is\nso complex in its individuality, and because this qualification is\nnot understood to be the work of thought, by comparison and contrast\naccumulating attributes on an object which it itself constitutes,\nbut is regarded as given ready-made in an impression (_i.e._ a\nfeeling), that the question arises whether a general proposition is\nreally possible or no. To all intents and purposes Hume decides that\nit is not. The mind is so tied down to the particular collection of\nqualities which is given to it or which it \u2018finds,\u2019 that it cannot\npresent one of them to itself without presenting all. Having never\nfound a triangle that is not equilateral or isosceles or scalene, we\ncannot imagine one, for ideas can only be copies of impressions, and\nthe imagination, though it has a certain freedom in combining what\nit finds, can invent nothing that it does not find. Thus the idea,\nrepresented by a general name and of which an assertion, general\nin form, is made, must always have a multitude of other qualities\nbesides those common to it with the other individuals to which the\nname is applicable. If any of these, however, were included in the\npredicate of the proposition, the sleeping custom, which determines\nthe mind to pass from the idea present to it to the others to which\nthe name has been applied, would be awakened, and it would be seen at\nonce that the predicate is not true of them. When I make a general\nstatement about \u2018the horse,\u2019 there must be present to my mind some\nparticular horse of my acquaintance, but if on the strength of this\nI asserted that \u2018the horse is a grey-haired animal,\u2019 the custom of\napplying the name without reference to colour would return upon me\nand correct me--as it would not if the predicate were \u2018four-footed.\u2019\n\nHe virtually yields the point in regard to the _predicate_ of\npropositions.\n\n220. It would seem then that the predicate may, though the subject\ncannot, represent either a single quality, or a set of qualities\nwhich falls far short even of those common to the class, much more of\nthose which characterise any individual. If I can think these apart,\nor have an idea of them, as the predicate of a proposition, why\nnot (it may be asked) as the subject? It may be said, indeed, with\ntruth, that it is a mistake to think of the subject as representing\none idea and the predicate another; that the proposition as a whole\nrepresents one idea, in the sense of a conception of relation between\nattributes, and that at bottom this account of it is consistent\nwith Locke\u2019s definition of knowledge as a perception of relation\nbetween \u2018ideas,\u2019 since with him \u2018ideas\u2019 and \u2018qualities\u2019 are used\ninterchangeably. [1] It is no less true, however, that the relation\nbetween attributes, which the proposition states, is a relation\nbetween them in an individual subject. It is the nature of the\nindividuality of this subject, then, that is really in question.\nMust it, as Hume supposed, be \u2018considered\u2019 under other qualities\nthan those to which the predicate relates? When the proposition\nonly concerns the relation between certain qualities of a spherical\nfigure, must the figure still be considered as of a certain colour\nand material?\n\n[1] See above, paragraph 17.\n\nAs to the subject, he equivocates between singleness of feeling and\nindividuality of conception.\n\n221. The possibility of such a question being raised implies that\nthe step has been already taken, which Hume ignored, from feeling\nto thought. His doctrine on the matter arises from that mental\nequivocation, of which the effects on Locke have been already\nnoticed, [1] between the mere singleness of a feeling in time and the\nindividuality of the object of thought as a complex of relations.\nIf the impression is the single feeling which disappears with a\nturn of the head, and the idea a weaker impression, every idea must\nindeed be in one sense \u2018individual,\u2019 but in a sense that renders\nall predication impossible because it empties the idea of all\ncontent. Really, according to Hume\u2019s doctrine of general names, it is\nindividual in a sense which is the most remote opposite of this, as a\nmultitude of \u2018different resemblances and relations\u2019 in \u2018simplicity.\u2019\nIt is just such an individual as Locke supposed to be found (so to\nspeak) ready-made in nature, and from which he supposed the mind\nsuccessively to abstract ideas less and less determinate. Such an\nobject Hume, coming after Berkeley, could not regard in Locke\u2019s\nfashion as a separate material existence outside consciousness. The\nidea with him is a \u2018copy\u2019 not of a thing but of an \u2018impression,\u2019\nbut to the impression he transfers all that individualization by\nqualities which Locke had ascribed to the substance found in nature;\nand from the impression again transfers it to the idea which \u2018is but\nthe weaker impression.\u2019 Thus the singleness in time of the impression\nbecomes the \u2018simplicity\u2019 of an object \u2018containing many different\nresemblances and relations,\u2019 and the individuality of the subject\nof a proposition, instead of being regarded in its true light as a\ntemporary isolation from other relations of those for the time under\nview--an individuality which is perpetually shifting its limits as\nthought proceeds--becomes an individuality fixed once for all by what\nis given in the impression. Because, as is supposed, I can only \u2018see\u2019\na globe as of a certain colour and material, I can only think of it\nas such. If the \u2018sight\u2019 of it had been rightly interpreted as itself\na complex work of thought, successively detaching felt things from\nthe \u2018flux\u2019 of feelings and determining these by relations similarly\ndetached, the difficulty of thinking certain of these--_e.g._ those\ndesignated as \u2018figure\u2019--apart from the rest would have disappeared.\nIt would have been seen that this was merely to separate in\nreflective analysis what had been gradually put together in the\nsuccessive synthesis of perception. But such an interpretation of the\nsupposed _datum_ of sense would have been to elevate thought from the\nposition which Hume assigned to it, as a \u2018decaying sense,\u2019 to that of\nbeing itself the organizer of the world which it knows. [2]\n\n[1] See above, paragraphs 47, 95, &c.\n\n[2] The phrase \u2018decaying sense\u2019 belongs to Hobbes, but its meaning is\nadopted by Hume.\n\nResult is a theory which admits predication, but only as singular.\n\n222. Here, then, as elsewhere, the embarrassment of Hume\u2019s doctrine\nis nothing which a better statement of it could avoid. Nay, so\ndexterous is his statement, that only upon a close scrutiny does\nthe embarrassment disclose itself. To be faithful at once to his\nreduction of the impression to simple feeling, and to his account of\nthe idea as a mere copy of the impression, was really impossible. If\nhe had kept his word in regard to the impression, he must have found\nthought filling the void left by the disappearance, under Berkeley\u2019s\ncriticism, of that outward system of things which Locke had commonly\ntaken for granted. He preferred fidelity to his account of the\nidea, and thus virtually restores the fiction which represents the\nreal world as consisting of so many, materially separate, bundles\nof qualities--a fiction which even Locke in his better moments\nwas beginning to outgrow--with only the difference that for the\nseparation of \u2018substances\u2019 in space he substitutes a separation of\n\u2018impressions\u2019 in time. That thought (the \u2018idea\u2019) can but faintly\ncopy feeling (the \u2018impression\u2019) he consistently maintains, but he\navails himself of the actual determination of feeling by reference\nto an object of thought--the determination expressed by such phrases\nas impression of a man, impression of a globe, &c.--to charge the\nfeeling with a content which it only derives from such determination,\nwhile yet he denies it. By this means predication can be accounted\nfor, as it could not be if our consciousness consisted of mere\nfeelings and their copies, but only in the form of the singular\nproposition; because the object of thought determined by relations,\nbeing identified with a single feeling, must be limited by the \u2018this\u2019\nor \u2018that\u2019 which expresses this singleness of feeling. It is really\n_this_ or _that_ globe, _this_ or _that_ man, that is the subject\nof the proposition, according to Hume, even when in form it is\ngeneral. It is true that the general name \u2018globe\u2019 or \u2018man\u2019 not merely\nrepresents a \u2018particular\u2019 globe or man, though that is all that is\npresented to the mind, but also \u2018raises up a custom which produces\nany other individual idea for which we may have occasion.\u2019 As this\ncustom, however, is neither itself an idea nor affects the singleness\nof the subject idea, it does not constitute any distinction between\nsingular and general propositions, but only between two sorts of the\nsingular proposition according as it does, or does not, suggest an\nindefinite series of other singular propositions, in which the same\nqualities are affirmed of different individual ideas to which the\nsubject-name has been applied.\n\nAll propositions restricted in same way as Locke\u2019s propositions about\nreal existence.\n\n223. A customary sequence, then, of individual ideas upon each other\nis the reality, which through the delusion of words (as we must\nsuppose) has given rise to the fiction of there being such a thing\nas general knowledge. We say \u2018fiction,\u2019 for with the possibility of\ngeneral propositions, as the Greek philosophers once for all pointed\nout, stands or falls the possibility of science. Locke was so far\naware of this that, upon the same principle which led him to deny the\npossibility of general propositions concerning real existence, he\n\u2018suspected\u2019 a science of nature to be impossible, and only found an\nexemption for moral and mathematical truth from this condemnation in\nits \u2018bare ideality.\u2019 Hume does away with the exemption. He applies\nto all propositions alike the same limitation which Locke applies to\nthose concerning real existence. With Locke there may very well be a\nproposition which to the mind, as well as in form, is general--one of\nwhich the subject is an \u2018abstract general idea\u2019--but such proposition\n\u2018concerns not existence.\u2019 As knowledge of real existence is limited\nto the \u2018actual present sensation,\u2019 so a proposition about such\nexistence is limited to what is given in such sensation. It is a\nreal truth that this piece of gold is now being dissolved in aqua\nregia, when the \u2018particular experiment\u2019 is going on under our eyes,\nbut the general proposition \u2018gold is soluble\u2019 is only an analysis of\na nominal essence. With Hume the distinction between propositions\nthat do, and those that do not, \u2018concern existence\u2019 disappears.\nEvery proposition is on the same footing in this respect, since it\nmust needs be a statement about an \u2018idea,\u2019 and every idea exists.\n\u2018Every object that is presented must necessarily be existent. ...\nWhatever we conceive, we conceive to be existent. Any idea we please\nto form is the idea of a being; and the idea of a being is any idea\nwe please to form\u2019. [1] But since, according to him, the idea cannot\nbe separated, as Locke supposed it could, from the conditions \u2018that\ndetermine it to this or that particular existence,\u2019 propositions\nof the sort which Locke understood by \u2018general propositions\nconcerning substances,\u2019 though if they were possible they would\n\u2018concern existence\u2019 as much as any, are simply impossible. Hume,\nin short, though he identifies the real and nominal essences which\nLocke had distinguished, yet limits the nominal essence by the same\n\u2018particularity in space and time\u2019 by which Locke had limited the real.\n\n[1] p. 370 [Book I, part II., sec. VI.]\n\nThe question, how the _singular_ proposition is possible, the vital\none.\n\n224. A great advance in simplification has been made when the false\nsort of \u2018conceptualism\u2019 has thus been got rid of--that conceptualism\nwhich opposes knowing and being under the notion that things,\nthough merely individual in reality, may be known as general. This\nriddance having been achieved, as it was by Hume, the import of the\nproposition becomes the central question of philosophy, the answer\nto which must determine our theory of real existence just as much\nas of the mind. The issue may be taken on the proposition in its\nsingular no less than in its general form. The weakness of Hume\u2019s\nopponents, indeed, has lain primarily in their allowing that his\ndoctrine would account for any significant predication whatever,\nas distinct from exclamations prompted by feelings as they occur.\nThis has been the inch, which once yielded, the full ell of his\nnominalism has been easily won; just as Locke\u2019s empiricism becomes\ninvincible as soon as it is admitted that qualified things are \u2018found\nin nature\u2019 without any constitutive action of the mind. As the only\neffective way of dealing with Locke is to ask,--After abstraction\nof all that he himself admitted to be the creation of thought, what\nremains to be merely found?--so Hume must be met _in limine_ by the\nquestion whether, apart from such ideas of relation as according to\nhis own showing are not simple impressions, so much as the singular\nproposition is possible. If not, then the singularity of such\nproposition does not consist in any singleness of presentation to\nsense; it is not the \u2018particularity in time\u2019 of a present feeling;\nand the exclusion of generality, whether in thoughts or in things,\nas following from the supposed necessity of such singleness or\nparticularity, is quite groundless.\n\nNot relations of resemblance only, but those of quantity also,\ntreated by Hume as feelings.\n\n225. Hitherto the idea of relation which we have had specially in\nview has been that of relation in the way of resemblance, and the\npropositions have been such as represent the most obvious \u2018facts of\nobservation\u2019--facts about this or that \u2018body,\u2019 man or horse or ball.\nWe have seen that these already suppose the thought of an object\nqualified, not transitory as are feelings, but one to which feelings\nare referred on their occurrence as resemblances or differences\nbetween it and other objects; but that by an equivocation, which\nunexamined phraseology covers, between the thought of such an object\nand feeling proper--as if because we talk of seeing a man, therefore\na man were a feeling of colour--Hume is able to represent them as\nmere data of sense, and thus to ignore the difference between related\nfeelings and ideas of relation. Thus the first step has been taken\ntowards transferring to the sensitive subject, as merely sensitive,\nthe power of thought and significant speech. The next is to transfer\nto it ideas of those other relations [1] which Hume classifies as\n\u2018relations of time and place, proportion in quantity or number,\ndegrees in any quality\u2019. [2] This done, it is sufficiently equipped\nfor achieving its deliverance from metaphysics. An animal, capable of\nexperiments concerning matter of fact, and of reasoning concerning\nquantity and number, would certainly have some excuse for throwing\ninto the fire all books which sought to make it ashamed of its\nanimality. [3]\n\n[1] The course which our examination of Hume should take was marked\nout, it will be remembered, by his enumeration of the \u2018_natural_\u2019\nrelations that regulate the association of ideas. It might seem\na departure from this course to proceed, as in the text, from\nthe relation of resemblance to \u2018relations of time and place,\nproportion in quantity or number, and degrees of any quality,\u2019\nsince these appear in Hume\u2019s enumeration, not of \u2018_natural_\u2019 but\nof \u2018_philosophical_\u2019 relations. Such departure, however, is the\nconsequence of Hume\u2019s own procedure. Whether he considered these\nrelations merely equivalent to the \u2018natural ones\u2019 of resemblance\nand contiguity, he does not expressly say; but his reduction of the\nprinciples of mathematics to data of sense implies that he did so.\nThe treatment of degrees in quality and proportions in quantity as\nsensible implies that the difference between resemblance and measured\nresemblance, between contiguity and measured contiguity, is ignored.\n\n[2] p. 368 [Book I, part II., sec. V.]\n\n[3] If we take in our hand any volume of divinity or\nschool-metaphysics, for instance, let us ask, _Does it contain\nany abstract reasoning for quantity or number?_ No. _Does it\ncontain any experimental reasoning concerning matter of fact and\nexistence?_ No. _Commit it then to the flames, for it can contain\nnothing but sophistry and illusion.\u2019_--\u2018Inquiry concerning the Human\nUnderstanding,\u2019 at the end.\n\nHe draws the line between certainty and probability at the same point\nas Locke; but is more definite as to probability,\n\n226. In thus leaving mathematics and a limited sort of experimental\nphysics (limited by the exclusion of all general inference from the\nexperiment) out of the reach of his scepticism, and in making them\nhis basis of attack upon what he conceived to be the more pretentious\nclaims of knowledge, Hume was again following the course marked out\nfor him by Locke. It will be remembered that Locke, even when his\n\u2018suspicion\u2019 of knowledge is at its strongest, still finds solid\nground _(a)_ in \u2018particular experiments\u2019 upon nature, expressed\nin singular propositions as opposed to assertions of universal or\nnecessary connexion, and _(b)_ in mathematical truths which are at\nonce general, certain, and instructive, because \u2018barely ideal.\u2019 All\nspeculative propositions that do not fall under one or other of these\nheads are either \u2018trifling\u2019 or merely \u2018probable.\u2019 Hume draws the line\nbetween certainty and probability at the same point, nor in regard\nto the ground of certainty as to \u2018matter of fact or existence\u2019 is\nthere any essential difference between him and his master. As this\nground is the \u2018actual present sensation\u2019 with the one, so it is the\n\u2018impression\u2019 with the other; and it is only when the proposition\nbecomes universal or asserts a necessary connection, that the\ncertainty, thus given, is by either supposed to fail. It is true that\nwith Locke this authority of the sensation is a derived authority,\ndepending on its reference to a \u2018body now operating upon us,\u2019 while\nwith Hume, so far as he is faithful to his profession of discarding\nsuch reference, it is original. But with each alike the fundamental\nnotion is that a feeling must be \u2018true _while it lasts_,\u2019 and that\nin regard to real existence or matter of fact no other truth can\nbe known but this. Neither perceives that a truth thus restricted\nis no truth at all--nothing that can be stated even in a singular\nproposition; that the \u2018particularity in time,\u2019 on which is supposed\nto depend the real certainty of the simple feeling, is just that\nwhich deprives it of significance [1]--because neither is really\nfaithful to the restriction. Each allows himself to substitute for\nthe momentary feeling an object qualified by relations, which are the\nexact opposite of momentary feelings. \u2018If I myself see a man walk\non the ice,\u2019 says Locke (IV, xv. 5), \u2018it is past probability, it is\nknowledge:\u2019 nor would Hume, though ready enough on occasion to point\nout that what is seen must be a colour, have any scruple in assuming\nthat such a complex judgment as the above so-called \u2018sight\u2019 has the\ncertainty of a simple impression. It is only in bringing to bear upon\nthe characteristic admission of Locke\u2019s Fourth Book, that no general\nknowledge of nature can be more than probable, a more definite notion\nof what probability is, and in exhibiting the latent inconsistency\nof this admission with Locke\u2019s own doctrine of ideas as effects of\na causative substance, that he modifies the theory of _physical_\ncertainty which he inherited. In their treatment of mathematical\ntruths on the other hand, of propositions involving relations of\ndistance, quantity and degree, a fundamental discrepancy appears\nbetween the two writers. The ground of certainty, which Hume admits\nin regard to propositions of this order, must be examined before we\ncan appreciate his theory of probability as it affects the relations\nof cause and substance.\n\n[1] See above, paragraphs 45 and 97.\n\n... and does not admit opposition of mathematical to physical\ncertainty--here following Berkeley.\n\n227. It has been shown [1] that Locke\u2019s opposition of mathematical to\nphysical certainty, with his ascription to the former of instructive\ngenerality on the ground of its bare ideality--the \u2018ideal\u2019 in this\nregard being opposed to what is found in sensation--strikes at the\nvery root of his system. It implies that thought can originate, and\nthat what it originates is in some sort real--nay, as being nothing\nelse than the \u2018primary qualities of matter,\u2019 is the source of all\nother reality. Here was an alien element which \u2018empiricism\u2019 could not\nassimilate without changing its character. Carrying such a conception\nalong with it, it was already charged with an influence which must\nultimately work its complete transmutation by compelling, not the\nadmission of an ideal world of guess and aspiration alongside of\nthe empirical, but the recognition of the empirical as itself ideal\nThe time for this transmutation, however, was not yet. Berkeley, in\nover-hasty zeal for God, had missed that only true way of finding God\nin the world which lies in the discovery that the world is Thought.\nHaving taken fright at the \u2018mathematical Atheism,\u2019 which seemed to\ngrow out of the current doctrines about primary qualities of matter,\ninstead of applying Locke\u2019s own admissions to show that these were\nintelligible and merely intelligible, he fancied that he had won the\nbattle for Theism by making out that they were merely feelings or\nsequences of feelings. From him Hume got the text for all he had to\nsay against the metaphysical mathematicians; but, for the reason that\nHume applied it with no theological interest, its true import becomes\nmore apparent with him than with Berkeley.\n\n[1] See above, paragraphs 117 and 125.\n\nHis criticisms of the doctrine of primary qualities.\n\n228. His account of mathematical truths, as contained in Part II.\nof the First Book of the \u2018Treatise on Human Nature,\u2019 cannot be\nfairly read except in connection with the chapters in Part IV.\non \u2018Scepticism with regard to the Senses,\u2019 and on \u2018the Modern\nPhilosophy.\u2019 The latter chapter is expressly a polemic against\nLocke\u2019s doctrine of primary qualities, and its drift is to reverse\nthe relations which Locke had asserted between them and sensations,\nmaking the primary qualities depend on sensations, instead of\nsensations on the primary qualities. In Locke himself we have\nfound that two inconsistent views on the subject perpetually cross\neach other. [1] According to one, momentary sensation is the sole\nconveyance to us of reality; according to the other, the real is\nconstituted by qualities of bodies which not only \u2018are in them\nwhether we perceive them or not,\u2019 but which only complex ideas\nof relation can represent. The unconscious device which covered\nthis inconsistency lay, we found, [2] in the conversion of the\nmere feeling of touch into the touch _of a body_, and thus into an\nexperience of solidity. By this conversion, since solidity according\nto Locke\u2019s account carries with it all the primary qualities,\nthese too become data of sensation, while yet, by the retention of\nthe opposition between them and ideas, the advantage is gained of\napparently avoiding that identification of what is real with simple\nfeeling, which science and common sense alike repel.\n\n[1] See above, paragraph 99 and following.\n\n[2] See above, paragraph 101.\n\nIt will not do to oppose bodies to our feeling when only feeling can\ngive idea of body.\n\n229. Hume makes a show of getting rid of this see-saw. Instead of\nassuming at once the reality of sensation on the strength of its\nrelation to the primary qualities and the reality of these on the\nstrength of their being given in tactual experience, he pronounces\nsensations alone the real, to which the primary qualities must be\nreduced, if they are not to disappear altogether. \u2018If colours,\nsounds, tastes, and smells be merely perceptions, nothing we\ncan conceive is possessed of a real, continued, and independent\nexistence\u2019. [1] That they are perceptions is of course undoubted.\nThe question is, whether there is a real something beside and\nbeyond them, contrast with which is implied in speaking of them\nas \u2018_merely_ perceptions.\u2019 The supposed qualities of such a real\nare \u2018motion, extension, and solidity\u2019. [2] To modes of these the\nother primary qualities enumerated by Locke are reducible; and of\nthese again motion and extension, according to Locke\u2019s account no\nless than Hume\u2019s own, presuppose solidity. What then do we assert\nof the real, in contrast with which we talk of perception, as\n_mere_ perception, when we say that it is solid? \u2018In order to form\nan idea of solidity we must conceive two bodies pressing on each\nother without any penetration. ... Now, what idea do we form of\nthese bodies? ... To say that we conceive them as solid is to run\non _ad infinitum_. To affirm that we paint them out to ourselves as\nextended, either resolves them all into a false idea or returns in a\ncircle; extension must necessarily be conceived either as coloured,\nwhich is a false idea, [3] or as solid, which brings us back to the\nfirst question.\u2019 Of solidity, then, the ultimate determination of\nthe supposed real, there is \u2018no idea to be formed\u2019 apart from those\nperceptions to which, as independent of our senses, it is opposed.\n\u2018After exclusion of colours, sounds, heat and cold from the rank of\nexternal existences, there remains nothing which can afford us a just\nand consistent idea of body.\u2019\n\n[1] p. 513 [Book I, part IV., sec. IV.]\n\n[2] Ibid.\n\n[3] \u2018A false idea,\u2019 that is, according to the doctrine that extension\nis a primary quality, while colour is only an idea of a secondary\nquality, not resembling the quality as it is in the thing.\n\nLocke\u2019s shuffle of \u2018body,\u2019 \u2018solidity,\u2019 and \u2018touch,\u2019 fairly exposed.\n\n230. Our examination of Locke has shown us how it is that his\ninterpretation of ideas by reference to body is fairly open to this\nattack. It is so because, in thus interpreting them, he did not know\nwhat he was really about. He thought he was explaining ideas of sense\naccording to the only method of explanation which he recognises--the\nmethod of resolving complex into simple ideas, and of \u2018sending a man\nto his senses\u2019 for a knowledge of the simple. In fact, however, when\nhe explained ideas of sense as derived from the qualities of body,\nhe was explaining simple ideas by reference to that which, according\nto his own showing, is a complex idea. To say that, as Locke\nunderstood the derivation in question, the primary qualities are\nan \u1f04\u03b9\u03c4\u03b9\u03bf\u03bd \u03b3\u03b5\u03bd\u03ad\u03c3\u03b5\u03c9\u03c2 to the ideas of secondary qualities, but not an\n\u1f04\u03b9\u03c4\u03b9\u03bf\u03bd \u03b3\u03bd\u03ce\u03c3\u03b5\u03c9\u03c2 [1]--that without our having ideas of them they cause\nthose ideas of sense from which afterwards our ideas of the primary\nqualities are formed--is to suppose an order of reality other than\nthe order of our sensitive experience, and thus to contradict Locke\u2019s\nfundamental doctrine that the genesis of ideas is to be found by\nobserving their succession in \u2018our own breasts.\u2019 It is not thus that\nLocke himself escapes the difficulty. As we have seen, he supposes\nour ideas of sense to be from the beginning ideas of the qualities of\nbodies, and virtually justifies the supposition by sending the reader\nto his sense of touch for that idea of solidity in which, as he\ndefines it, all the primary qualities are involved. That the sense in\nquestion does not really yield the idea is what Hume points out when\nhe says that, \u2018though bodies are felt by means of their solidity, yet\nthe feeling is quite a different thing from the solidity, nor have\nthey the least resemblance to each other.\u2019 In other words, having\ncome to suppose that there are solid bodies, we explain our feeling\nas due to their solidity; but we may not at once interpret feeling\nas the result of solidity, and treat solidity as itself a feeling.\nIt was by allowing himself so to treat it that Locke disguised\nfrom himself the objection to his interpretation of feeling. Hume\ntears off the disguise, and in effect gives him the choice of being\nconvicted either of reasoning in a circle or of explaining the simple\nidea by reference to the complex. The solidity, which is to explain\nfeeling, can itself only be explained by reference to body. If body\nis only a complex of ideas of sense, in referring tactual feeling to\nit we are explaining a simple idea by reference to a compound one.\nIf it is not, how is it to be defined except in the \u2018circular\u2019 way,\nwhich Locke in fact adopts when he makes body a \u2018texture of solid\nparts\u2019 and solidity a relation of bodies? [2]\n\n[1] [Greek \u1f04\u03b9\u03c4\u03b9\u03bf\u03bd \u03b3\u03b5\u03bd\u03ad\u03c3\u03b5\u03c9\u03c2 (aition geneseos) = cause of coming-to-be,\n\u1f04\u03b9\u03c4\u03b9\u03bf\u03bd \u03b3\u03bd\u03ce\u03c3\u03b5\u03c9\u03c2 (aition gnoseos) = cause of being known. Tr.]\n\n[2] See above, paragraph 101.\n\nTrue rationale of Locke\u2019s doctrine.\n\n231. This \u2018vicious circle\u2019 was nothing of which Locke need have been\nashamed, if only he had understood and avowed its necessity. Body\nis to solidity and to the primary qualities in general simply as a\nsubstance to the relations that determine it; and the \u2018circle\u2019 in\nquestion merely represents the logical impossibility of defining\na substance except by relations, and of defining these relations\nwithout presupposing a substance. It was only Locke\u2019s confusion\nof the order of logical correlation with the sequence of feelings\nin time, that laid him open to the charge of making body and the\nideas of primary qualities, and again the latter ideas and those\nof secondary qualities, at once precede and follow each other. To\navoid this confusion by recognising the logical order--the order of\nintellectual \u2018fictions\u2019--as that apart from which the sequence of\nfeelings would be no order of knowable reality at all, would be of\ncourse impossible for one who took Locke\u2019s antithesis of thought\nand fact for granted. The time for that was not yet. A way of\nescape had first to be sought in a more strict adherence to Locke\u2019s\nidentification of the sequence of feelings with the order of reality.\nHence Hume\u2019s attempt, reversing Locke\u2019s derivation of ideas of sense\nfrom primary qualities of body, to derive what with Locke had been\nprimary qualities, as compound impressions of sense, from simple\nimpressions and to reduce body itself to a name not for any \u2018just\nand consistent idea,\u2019 but for a \u2018propensity to feign,\u2019 the gradual\nproduct of custom and imagination. The question by which the value of\nsuch derivation and reduction is to be tried is our old one, whether\nit is not a tacit conversion of the supposed original impressions\ninto qualities of body that alone makes them seem to yield the result\nrequired of them. If the Fourth Book of the \u2018Treatise on Human\nNature,\u2019 with its elimination of the idea of body, had come before\nthe second, would not the plausibility of the account of mathematical\nideas contained in the latter have disappeared? And conversely, if\nthese ideas had been reduced to that which upon elimination of the\nidea of body they properly become, would not that \u2018propensity to\nfeign,\u2019 which is to take the place of the excluded idea, be itself\nunaccountable?\n\nWith Hume \u2018body\u2019 logically disappears. What then?\n\n232. \u2018After exclusion of colours, sounds, heat and cold, from the\nrank of external existences, there remains nothing which can afford\nus a just and consistent idea of body.\u2019 Now, no one can \u2018exclude\nthem from the rank of external existences\u2019 more decisively than\nHume. They are impressions, and \u2018all impressions are internal and\nperishing existences, and appear as such.\u2019 Nor does he shirk the\nconsequence, that we have no \u2018just and consistent idea of body.\u2019 It\nis true that we cannot avoid a \u2018belief in its existence\u2019--a belief\nwhich according to Hume consists in the supposition of \u2018a continued\nexistence of objects when they no longer appear to the senses, and of\ntheir existence as distinct from the mind and perceptions;\u2019 in other\nwords, as \u2018external to and independent of us.\u2019 This belief, however,\nas he shows, is not given by the senses. That we should feel the\nexistence of an object to be continued when we no longer feel it, is\na contradiction in terms; nor is it less so, that we should feel it\nto be distinct from the feeling. We cannot, then, have an impression\nof body; and, since we cannot have an idea which does not correspond\nto an impression or collection of impressions, it follows that we can\nhave no idea of it. How the \u2018belief in its existence\u2019 is accounted\nfor by Hume in the absence of any idea of it, is a question to be\nconsidered later. [1] Our present concern is to know whether the idea\nof extension can hold its ground when the idea of body is excluded.\n\n[1] See below, paragraph 303, and foll.\n\nCan space survive body? Hume derives idea of it from sight and\nfeeling. Significance with him of such derivation.\n\n233. \u2018The first notion of space and extension,\u2019 he says, \u2018is derived\nsolely from the senses of sight and feeling: nor is there anything\nbut what is coloured or tangible that has parts disposed after\nsuch a manner as to convey the idea.\u2019 Now, there may be a meaning\nof \u2018derivation,\u2019 according to which no one would care to dispute\nthe first clause of this sentence. Those who hold that _really_,\ni.e. _for a consciousness to which the distinction between real and\nunreal is possible_, there is no feeling except such as is determined\nby thought, are yet far from holding that the determination is\narbitrary; that any and every feeling is potentially any and every\nconception. Of the feelings to which the visual and tactual nerves\nare organic, as they would be for a merely feeling consciousness,\nnothing, they hold, can be said; in that sense they are an \u1f04\u03c0\u03b5\u03b9\u03c1\u03bf\u03bd;\n[1] but for the thinking consciousness, or (which is the same) as\nthey _really_ are, these feelings do, while those to which other\nnerves are organic do not, form the specific possibility of the\nconception of space. According to this meaning of the words, all\nmust admit that \u2018the first notion of space and extension is derived\nfrom the senses of sight and feeling;\u2019 though it does not follow\nthat a repeated or continued activity of either sense is necessary\nto the continued presence of the notion. With Hume, however, the\nderivation spoken of must mean that the notion of space is, to begin\nwith, simply a visual or tactual feeling, and that such it remains,\nthough with indefinite abatement and revival in the liveliness of the\nfeeling, according to the amount of which it is called \u2018impression\u2019\nor \u2018idea.\u2019 If we supposed him to mean, not that the notion of space\nwas either a visual or tactual feeling indifferently, but that it\nwas a compound result of both, [2] we should merely have to meet\na further difficulty as to the possibility of such composition of\nfeelings when their inward synthesis in a soul, and the outward in a\nbody, have been alike excluded. In the next clause of the sentence,\nhowever, we find that for visual and tactual feelings there are\nquietly substituted \u2018coloured and tangible objects, having parts so\ndisposed as to convey the idea of extension.\u2019 It is in the light of\nthis latter clause that the uncritical reader interprets the former.\nHe reads back the plausibility of the one into the other, and, having\ndone so, finds the whole plausible. Now this plausibility of the\nlatter clause arises from its implying a three-fold distinction--a\ndistinction of colour or tangibility on the one side from the\ndisposition of the parts on the other; a distinction of the colour,\ntangibility and disposition of parts alike from an object to which\nthey belong; and a distinction of this object from the idea that it\nconveys. In other words, it supposes a negative answer to the three\nfollowing questions:--Is the idea of extension the same as that of\ncolour or tangibility? Is it possible without reference to something\nother than a possible impression? Is the idea of extension itself\nextended? Yet to the two latter questions, according to Hume\u2019s\nexpress statements, the answer must be affirmative; nor can he avoid\nthe affirmative answer to the first, to which he would properly be\nbrought, except by equivocation.\n\n[1] [Greek \u1f04\u03c0\u03b5\u03b9\u03c1\u03bf\u03bd (apeiron) = unlimited, indefinite or infinite. Tr.]\n\n[2] It is not really in this sense that the impression of space\naccording to Hume is a \u2018compound\u2019 one, as will appear below.\n\nIt means, in effect, that colour and space are the same, and that\nfeeling may be extended.\n\n234. The _pi\u00e8ces justificatives_ for this assertion are not far to\nseek. Some of them have been adduced already. The idea of space, like\nevery other idea, must be a \u2018copy of an impression.\u2019 [1] To speak of\na feeling in its fainter stage as an \u2018image\u2019 of what it was in its\nlivelier stage may, indeed, seem a curious use of terms; but in this\nsense only, according to Hume\u2019s strict doctrine, can the idea of\nspace be spoken of as an \u2018image\u2019 of anything at all. The impression\nfrom which it is derived, _i.e._ the feeling at its liveliest,\ncannot properly be so spoken of, for \u2018no impression is presented\nby the senses as the image of anything distinct, or external, or\nindependent.\u2019 [2] If no impression is so presented, neither can\nany idea, which copies the impression, be so. It can involve no\nreference to anything which does not come and go with the impression.\nAccordingly no distinction is possible between space on the one hand,\nand either the impression or idea of it on the other. All impressions\nand ideas that can be said to be of extension must be themselves\nextended; and conversely, as Hume puts it, \u2018all the qualities of\nextension are qualities of a perception.\u2019 It should follow that space\nis either a colour or feeling of touch. In the terms which Hume\nhimself uses with reference to \u2018substance,\u2019 \u2018if it be perceived by\nthe eyes, it must be colour; if by the ears, a sound; and so on, of\nthe other senses.\u2019 As he expressly tells us that it is \u2018perceived by\nthe eyes,\u2019 the conclusion is inevitable.\n\n[1] P. 340 [Book I, part II., sec. III.]\n\n[2] P. 479 [Book I, part IV., sec. II.]\n\nThe parts of space are parts of a perception.\n\n235. Hume does not attempt to reject the conclusion directly. He\nhad too much eye to the appearance of consistency for that. But,\nin professing to admit it, he wholly alters its significance. The\npassage in question must be quoted at length. \u2018The table, which just\nnow appears to me, is only a perception, and all its qualities are\nqualities of a perception. Now, the most obvious of all its qualities\nis extension. The perception consists of parts. These parts are so\nsituated as to afford us the notion of distance and contiguity,\nof length, breadth, and thickness. The termination of these three\ndimensions is what we call figure. The figure is moveable, separable,\nand divisible. Mobility and separability are the distinguishing\nproperties of extended objects. And, to cut short all disputes, the\nvery idea of extension is copied from nothing but an impression,\nand consequently must perfectly agree to it. To say the idea of\nextension agrees to anything is to say it is extended.\u2019 Thus \u2018there\nare impressions and ideas that are really extended.\u2019 [1]\n\n[1] P. 523 [Book I, part IV., sec. V.]\n\nYet the parts of space are co-existent not successive.\n\n236. In order to a proper appreciation of this passage it is\nessential to bear in mind that Hume, so far as the usages of\nlanguage would allow him, ignores all such differences in modes of\nconsciousness as the Germans indicate by the distinction between\n\u2018Empfindung\u2019 and \u2018Vorstellung,\u2019 and by that between \u2018Anschauung\u2019\nand \u2018Begriff;\u2019 or, more properly, that he expressly merges them in\na mode of consciousness for which, according to the most consistent\naccount that can be gathered from him, the most natural term\nwould be \u2018feeling.\u2019 [1] It is true that Hume himself, admitting a\ndistinction in the degree of vivacity with which this consciousness\nis at different times presented, inclines to restrict the term\n\u2018feeling\u2019 to its more vivacious stage, and to use \u2018perception\u2019 as\nthe more general term, applicable whatever the degree of vivacity\nmay be. [2] We must not allow him, however, in using this term to\ngain the advantage of a meaning which popular theory does, but his\ndoes not, attach to it. \u2018Perception\u2019 with him covers \u2018idea\u2019 as well\nas \u2018impression;\u2019 but nothing can be said of idea that cannot be\nsaid of impression, save that it is less lively, nor of impression\nthat cannot be said of idea, save that it is more so. It is this\nexplicit reduction of all consciousness virtually, if not in name, to\nfeeling that brings to the surface the difficulties latent in Locke\u2019s\n\u2018idealism.\u2019 These we have already traced at large; but they may be\nsummed up in the question, How can feelings, as \u2018particular in time\u2019\nor (which is the same) in \u2018perpetual flux,\u2019 constitute or represent\na world of permanent relations? [3] The difficulty becomes more\nobvious, though not more real, when the relations in question are not\nmerely themselves permanent, like those between natural phenomena,\nbut are \u2018relations between permanent parts,\u2019 like those of space.\nIt is for this reason that its doctrine about geometry has always\nbeen found the most easily assailable point of the \u2018sensational\u2019\nphilosophy. Locke distinguishes the ideas of space and of duration\nas got, the one \u2018from the permanent parts of space,\u2019 the other \u2018from\nthe fleeting and perpetually perishing parts of succession.\u2019 [4] He\nafterwards prefers the term \u2018expansion\u2019 to space, as the opposite\nof duration, because it brings out more clearly the distinction\nof a relation between permanent parts from that between \u2018fleeting\nsuccessive parts which never exist together.\u2019 How, then, can a\nconsciousness consisting simply of \u2018fleeting successive parts\u2019 either\nbe or represent that of which the differentia is that its parts are\npermanent and co-exist?\n\n[1] As implying no distinction from, or reference to, a thing causing\nand a subject experiencing it. See above, paragraphs 195 and 208, and\nthe passages there referred to.\n\n[2] \u2018To hate, to love, to think, to feel, to see; all this is nothing\nbut to perceive.\u2019 p. 371 [Book I, part II., sec. IV.]. \u2018When I\nshut my eyes and _think_ of my chamber, the ideas I form are exact\nrepresentations of the impressions I _felt_.\u2019 p. 312 [Book I, part\nI., sec. I.].\n\n[3] See above, paragraphs 172 & 176.\n\n[4] Essay Book II. chap. xiv. sec. 1.\n\nHume cannot make space a \u2018perception\u2019 without being false to his own\naccount of perception;\n\n237. If this crux had been fairly faced by Hume, he must have seen\nthat the only way in which he could consistently deal with it was\nby radically altering, with whatever consequence to the sciences,\nLocke\u2019s account of space. As it was, he did not face it, but--whether\nintentionally or only in effect--disguised it by availing himself of\nthe received usages of language, which roughly represent a theory\nthe exact opposite of his own, to cover the incompatibility between\nthe established view of the nature of space, and his own reduction\nof it to feeling. A very little examination of the passage, quoted\nat large above, will show that while in it a profession is made\nof identifying extension and a certain sort of perception with\neach other, its effect is not really to reduce extension to such a\nperception as Hume elsewhere explains all perceptions to be, but\nto transfer the recognised properties of extension which with such\nreduction would disappear, to something which for the time he chooses\nto reckon a perception, but which he can only so reckon at the cost\nof contradicting his whole method of dealing with the ideas of God,\nthe soul, and the world. The passage, in fact, is merely one sample\nof the continued shuffle by which Hume on the one hand ascribes to\nfeeling that intelligible content which it only derives from relation\nto objects of thought, and on the other disposes of these objects\nbecause they are not feelings.\n\n... as appears if we put \u2018feeling\u2019 for \u2018perception\u2019 in the passages\nin question.\n\n238. \u2018The table, which just now appears to me, is only a perception,\nand all its qualities are qualities of a perception. Now, the most\nobvious of all its qualities is extension. The perception consists\nof parts. These parts are so situated as to afford us the notion\nof distance and contiguity, of length, breadth, and thickness,\u2019\n&c., &c. If, now, throughout this statement (as according to Hume\u2019s\ndoctrine we are entitled to do) we write _feeling_ for \u2018perception\u2019\nand \u2018notion,\u2019 it will appear that this table is a feeling, which\nhas another feeling, called extension, as one of its qualities; and\nthat this latter feeling consists of parts. These, in turn, must be\nthemselves feelings, since the parts of which a perception consists\nmust be themselves perceived, and, being perceived, must, according\nto Hume, be themselves perceptions which = feelings. These feelings,\nagain, afford us other feelings of certain relations--distance and\ncontiguity, &c.--feelings which, as Hume\u2019s doctrine allows of no\ndistinction between the feeling and that of which it is the feeling,\nmust be themselves relations. Thus it would seem that a feeling may\nhave another feeling as one of its qualities; that the feeling, which\nis thus a quality, has other feelings as its co-existent parts; and\nthat the feelings which are parts \u2018afford us\u2019 other feelings which\nare relations. Is that sense or nonsense?\n\nTo make sense of them, we must take perception to mean perceived\nthing,\n\n239. To this a follower of Hume, if he could be brought to admit the\nlegitimacy of depriving his master of the benefit of synonyms, might\nprobably reply, that the apparent nonsense only arises from our being\nunaccustomed to such use of the term \u2018feeling;\u2019 that the table is a\n\u2018bundle of feelings,\u2019 actual and possible, of which the actual one\nof sight suggests a lively expectation, easily confused with the\npresence, of the others belonging to the other senses; that any one\nof these may be considered a quality of the total impression formed\nby all; that the feeling thus considered, if it happens to be visual,\nmay not improperly be said to consist of other feelings, as a whole\nconsists of parts, since it is the result of impressions on different\nparts of the retina, and from a different point of view even itself\nto be the relation between the parts, just as naturally as a mutual\nfeeling of friendship may be said either to consist of the loves of\nthe two parties to the friendship, or to constitute the relation\nbetween them. Such language represents those modern adaptations of\nHume, which retain his identification of the real with the felt but\nignore his restrictions on the felt. Undoubtedly, if Hume allowed\nus to drop the distinction between feeling as it might be for a\nmerely feeling consciousness, and feeling as it is for a thinking\nconsciousness, the objection to his speaking of feeling in those\nterms, in which it must be spoken of if extension is to be a feeling,\nwould disappear; but so, likewise, would the objection to speaking of\nthought as constitutive of reality. To appreciate his view we must\ntake feeling not as we really know it--for we cannot know it except\nunder those conditions of self-consciousness, the logical categories,\nwhich in his attempt to get at feeling, pure and simple, Hume is\nconsistent enough to exclude--but as it becomes upon exclusion of\nall determination by objects which Hume reckons fictitious. What it\nwould thus become _positively_ we of course cannot say, for of the\nunknowable nothing can be said; but we can decide _negatively_ what\nit cannot be. Can that in any case be said of it, which must be said\nof it if a feeling may be extended, and if extension is a feeling?\nCan it be such a quality of an object, so consisting of parts, and\nsuch a relation, as we have found that Hume takes it to be in his\naccount of the perception of this table?\n\n... which it can only mean as the result of certain \u2018fictions\u2019.\n\n240. After having taken leave throughout the earlier part of the\n\u2018Treatise on Human Nature\u2019 to speak in the ordinary way of objects\nand their qualities--and otherwise of course he could not have\nspoken at all--in the fourth book he seems for the first time to\nbecome aware that his doctrine did not authorise such language.\nTo perceive qualities of an object is to be conscious of relation\nbetween a subject and object, of which neither perishes with the\nmoment of perception. Such consciousness is self-consciousness,\nand cannot be reduced to any natural observable event, since it is\nconsciousness of that of which we cannot say \u2018Lo, here,\u2019 or \u2018Lo,\nthere,\u2019 \u2018it is now but was not then,\u2019 or \u2018it was then but is not\nnow.\u2019 It is therefore something which the spirit of the Lockeian\nphilosophy cannot assimilate, and which Hume, as the most consistent\nexponent of that spirit, most consistently tried to get rid of.\nThe subject as self, the object as body, he professes to reduce to\nfigures of speech, to be accounted for as the result of certain\n\u2018propensities to feign:\u2019 nor will he allow that any impression or\nidea (and impressions and ideas with him, be it remembered, exhaust\nour consciousness) carries with it a reference to an object other\nthan itself, any more than do pleasure or pain to which \u2018in their\nnature\u2019 all perceptions correspond. [1] He cannot, indeed, avoid\nspeaking of the consciousness thus reduced to the level of simple\npain and pleasure, as being that which in fact it can only be when\ndetermined by relation to a self-conscious subject, _i.e._ as itself\nan object; but he is so far faithful in his attempt to avoid such\ndetermination, that he does not reckon the object more permanent\nthan the impression. It, too, is a \u2018perishing existence.\u2019 As the\nimpression disappears with a \u2018turn of the eye in its socket,\u2019 so does\nthe object, which really is the impression, and cannot appear other\nthan it is any more than a feeling can be felt to be what it is not.\n[2]\n\n[1] \u2018Every impression, external and internal, passions, affections,\nsensations, pains, and pleasures, are originally on the same footing;\nand, whatever other differences we may observe among them, appear,\nall of them, in their true colours, as impressions or perceptions.\u2019\np. 480. [Book I, part IV., sec. II.]\n\n\u2018All sensations are felt by the mind such as they really are; and,\nwhen we doubt whether they present themselves as distinct objects or\nas mere impressions, the difficulty is not concerning their nature,\nbut concerning their relations and situation.\u2019 p. 480. [ibid.]\n\n[2] See above, paragraph 208, with the passages there cited.\n\nIf felt thing is no more than feeling, how can it have qualities?\n\n241. Such being the only possible object, how can qualities of it\nbe perceived? We cannot here find refuge in any such propensity to\nfeign as that which, according to Hume, leads us to \u2018endow objects\nwith a continued existence, distinct from our perceptions.\u2019 If such\npropensities can give rise to impressions at all, it can only be to\nimpressions of reflection, and it cannot be in virtue of them that\nextension, an impression of sensation, is given as a quality of an\nobject. Now if there is any meaning in the phrase \u2018qualities of an\nobject,\u2019 it implies that the qualities co-exist with each other and\nthe object. Feelings, then, which are felt as qualities of another\nfeeling must co-exist with, _i.e._ (according to Hume) be felt at\nthe same time as, it and each other. Thus, if an impression of sight\nbe the supposed object, no feeling that occurs after this impression\nhas disappeared can be a quality of it. Accordingly, when Hume speaks\nof extension being seen as one of the qualities of this table, he\nis only entitled to mean that it is one among several feelings,\nexperienced at one and the same time, which together constitute\nthe table. Whatever is not so experienced, whether extension or\nanything else, can be no quality of that \u2018perception.\u2019 How much\nof the perception, then, will survive? Can any feelings, strictly\nspeaking, be cotemporaneous? Those received through different senses,\nas Hume is careful to show, may be; _e.g._ the smell, taste, and\ncolour of a fruit. [1] In regard to them, therefore, we may waive\nthe difficulty, How can feelings successive to each other be yet\nco-existent qualities? but only to find ourselves in another as to\nwhat the object may be of which the cotemporaneous feelings are\nqualities. It cannot, according to Hume, be other than one or all\nof the cotemporaneous feelings. Is, then, the taste of an apple a\nquality of its colour or of its smell, or of colour, smell, and taste\nput together? It will not help us to speak of the several feelings\nas qualities of the \u2018total impression;\u2019 for the \u2018total impression\u2019\neither merely means the several feelings put together, or else\ncovertly implies just that reference to an object other than these,\nwhich Hume expressly excludes.\n\n[1] \u2018The taste and smell of any fruit are inseparable from its other\nqualities of colour and tangibility, and ... \u2019tis certain they are\nalways co-existent. Nor are they only co-existent in general, but\nalso cotemporary in their appearance in the mind.\u2019 p. 521. (Contrast\np. 370, where existence and appearance are identified.) [Book I, part\nIV., sec. V. and part II, sec. IV.]\n\nThe thing will have ceased before the quality begins to be.\n\n242. In fact, however, when he speaks of the feeling, which is\ncalled extension, as a quality of the feeling, which is called\nsight, of the table, he has not even the excuse that he might have\nhad if the feelings in question, being of different senses, might be\ncotemporary. According to him they are feelings of the same sense.\nThe extension of the table he took to be a datum of sight just as\nproperly as its colour; yet he cannot call it the same as colour, but\nonly \u2018a quality of the coloured object.\u2019 As the \u2018coloured object,\u2019\nhowever, apart from \u2018propensities to feign,\u2019 can, according to him,\nbe no other than the feeling of colour, his doctrine can only mean\nthat, colour and extension being feelings of the same sense, the\nlatter is a quality of the former. Is this any more possible than\nthat red should be a quality of blue, or a sour taste of a bitter\none? Must not the two feelings be successive, however closely\nsuccessive, so that the one which is object will have disappeared\nbefore the other, which is to be its quality, will have occurred? [1]\n\n[1] It should be needless to point out that by taking extension to be\na quality of \u2018tangibility\u2019 or muscular effort we merely change the\ndifficulty. The question as to its relation to such feelings will be\nsimply a repetition of that, put in the text, as to its relation to\nthe feeling of colour.\n\nHume equivocates by putting \u2018coloured points\u2019 for colour.\n\n243. If we look to the detailed account which Hume gives of the\nrelation between extension and colour, we find that he avoids the\nappearance of making one feeling a quality of another, by in fact\nsubstituting for colour a superficies of coloured points, in which\nit is very easy to find extension as a quality because it already\nis extension as an object. To speak of extension, though a feeling,\nas made up of parts is just as legitimate or illegitimate as to\nspeak of the feeling of colour being made up of coloured points. The\nlegitimacy of this once admitted, there remains, indeed, a logical\nquestion as to how it is that a quality should be spoken of in terms\nthat seem proper to a substance--as is done when it is said to\nconsist of parts--and yet, again, should be pronounced a relation\nof these parts; but to one who professed to merge all logical\ndistinctions in the indifference of simple feeling, such a question\ncould have no recognised meaning. It is, then, upon the question\nwhether, according to Hume\u2019s doctrine of perception, the perception\nof an object made up of coloured points may be used interchangeably\nwith the perception of colour, that the consistency of his doctrine\nof extension must finally be tried.\n\n244. The detailed account is to the following effect:--\u2018Upon opening\nmy eyes and turning them to the surrounding objects, I perceive many\nvisible bodies; and upon shutting them again and considering the\ndistance betwixt these bodies, I acquire the idea of extension.\u2019\nFrom what impression, Hume proceeds to ask, is this idea derived?\n\u2018Internal impressions\u2019 being excluded, \u2018there remain nothing but the\nsenses which can convey to us this original impression.\u2019 ... \u2018The\ntable before me is alone sufficient by its view to give me the idea\nof extension. This idea, then, is borrowed from and represents some\nimpression which this moment appears to the senses. But my senses\nconvey to me only the impressions of coloured points, disposed in a\ncertain manner. ... We may conclude that the idea of extension is\nnothing but a copy of these coloured points and of the manner of\ntheir appearance.\u2019 [1]\n\n[1] Pp. 340 and 341. [Book I, part II., sec. III.]\n\nCan a \u2018disposition of coloured points\u2019 be an impression?\n\n245. If the first sentence of the above had been found by Hume in\nan author whom he was criticising, he would scarcely have been\nslow to pronounce it tautological. As it stands, it simply tells\nus that having seen things extended we consider their extension,\nand upon considering it acquire an idea of it. It is a fair sample\nenough of those \u2018natural histories\u2019 of the soul in vogue among us,\nwhich by the help of a varied nomenclature seem able to explain a\nsupposed later state of consciousness as the result of a supposed\nearlier one, because the terms in which the earlier is described\nin effect assume the later. It may be said, however, that it is\nonly by a misinterpretation of a carelessly written sentence that\nHume can be represented as deriving the idea of extension from the\nconsideration of distance; that, as the sequel shows, he regarded the\n\u2018consideration\u2019 and the \u2018idea\u2019 in question as equivalent, and derived\nfrom the same impression of sense. It is undoubtedly upon his account\nof this impression that his doctrine of extension depends. It is\ndescribed as \u2018an impression of coloured points disposed in a certain\nmanner.\u2019 To it the idea of extension is related simply as a copy;\nwhich, we have seen, properly means with Hume, as a feeling in a less\nlively stage is related to the same feeling in a more lively stage.\nIt is itself, we must note, the _impression_ of extension; and it is\nan impression of sense, about which, accordingly, no further question\ncan properly be raised. Hume, indeed, allows himself to speak as\nif it were included in a \u2018perception of visible bodies\u2019 other than\nitself; just as in the passage from the fourth book previously\nexamined, he speaks as if the perception, called extension, were a\nquality of some other perception. This we must regard as an exercise\nof the privilege which he claims of \u2018speaking with the vulgar while\nhe thought with the learned;\u2019 since, according to him, \u2018visible\nbody,\u2019 in any other sense than that of the impression of coloured\npoints, is properly a name for a \u2018propensity to feign\u2019 resulting from\na process posterior to all impressions of sense. The question remains\nwhether, in speaking of an impression as one of \u2018coloured points\ndisposed in a certain manner,\u2019 he is not introducing a \u2018fiction\nof thought\u2019 into the impression just as much as in calling it a\n\u2018perception of body.\u2019\n\nThe points must be themselves impressions, and therefore not\nco-existent.\n\n246. An impression, we know, can, according to Hume, never be _of_ an\nobject in the sense of involving a reference to anything other than\nitself. When one is said, then, to be _of_ coloured points, &c., this\ncan only mean that itself _is_, or consists of, such points. Thus the\nquestion we have to answer is only a more definite form of the one\npreviously put, Can a feeling consist of parts? In answering it we\nmust remember that the parts, here supposed to be coloured points,\nmust, according to Hume\u2019s doctrine, be themselves impressions or they\nare nothing. Consistently with this he speaks of extension as \u2018a\ncompound impression, consisting of parts or lesser impressions, that\nare indivisible to the eye or feeling, and may be called impressions\nof atoms or corpuscles, endowed with colour and solidity.\u2019 [1] Now,\nunless we suppose that a multitude of feelings of one and the same\nsense can be present together, these \u2018lesser impressions\u2019 must follow\neach other and precede the \u2018compound impression.\u2019 That is to say,\nnone of the parts of which extension consists will be in existence\nat the same time, and all will have ceased to exist before extension\nitself comes into being. Can we, then, adopt the alternative\nsupposition that a multitude of feelings of one and the same sense\ncan be present together? In answering this question according to\nHume\u2019s premisses we may not help ourselves by saying that in a case\nof vision there really are impressions on different parts of the\nretina. To say that it _really_ is so, is to say that it is so for\nthe _thinking_ consciousness--for a consciousness that distinguishes\nbetween what it feels and what it knows. To a man, as simply seeing\nand while he sees, his sight is not an impression on the retina at\nall, much less a combination of impressions on different parts of\nthe retina. It is so for him only as thinking on the organs of his\nsight; or, if we like, as \u2018seeing\u2019 them in another, but \u2018seeing\u2019 them\nin a way determined by sundry suppositions (bodies, rays, and the\nlike) which are not feelings, and therefore with Hume not possible\n\u2018perceptions,\u2019 at all. But it is the impression of sight, as it would\nbe for one simply seeing and while he sees, undetermined by reference\nto anything other than itself, whether subject or object--an\nimpression as it would be for a merely feeling consciousness or (in\nHume\u2019s language) \u2018on the same footing with pain and pleasure\u2019--that\nwe have to do with when, from Hume\u2019s point of view, we ask whether a\nmultitude of such impressions can be present at once, _i.e._ as one\nimpression.\n\n[1] P. 345 [Book I, part II., sec. IV.]\n\nA \u2018compound impression\u2019 excluded by Hume\u2019s doctrine of time.\n\n247. If this question had been brought home to Hume, he could\nscarcely have avoided the admission that to answer it affirmatively\ninvolved just as much of a contradiction as that which he recognises\nbetween the \u2018interrupted\u2019 and \u2018continuous\u2019 existence of objects;\n[1] and just as in the latter case he gets over the contradiction\nby taking the interrupted existence, because the datum of sense,\nto be the reality, and the continued existence to be a belief\nresulting from \u2018propensities to feign,\u2019 so in the case before us\nhe must have taken the multiplicity of successive impressions to\nbe the reality, and their co-existence as related parts to be a\nfigure of speech, which he must account for as best he could. As\nit is, he so plays fast and loose with the meaning of \u2018impression\u2019\nas to hide the contradiction which is involved in the notion of a\n\u2018compound impression\u2019 if impression is interpreted as feeling--the\ncontradiction, namely, that a single feeling should he felt to be\nmanifold--and in consequence loses the chance of being brought to\nthat truer interpretation of the compound impression, as the thought\nof an object under relations, which a more honest trial of its\nreduction to feeling might have shown to be necessary. To convict so\nskilful a writer of a contradiction in terms can never be an easy\ntask. He does not in so many words tell us that all impressions of\nsight must be successive, but he does tell us that \u2018the impressions\nof touch,\u2019 which, indifferently with those of sight, he holds to\nconstitute the compound impression of extension, \u2018change every moment\nupon us.\u2019 [2] And in the immediate sequel of the passage where he\nhas made out extension to be a compound of co-existent impressions,\nhe derives the idea of time \u2018from the succession of our perceptions\n_of every kind_, ideas as well as impressions, and impressions of\nreflection as well as of sensation.\u2019 The parts of time, he goes on to\nsay, cannot be co-existent; and, since \u2018time itself is nothing but\ndifferent ideas and impressions succeeding each other,\u2019 these parts,\nwe must conclude, are those \u2018perceptions of every kind\u2019 from which\nthe idea of time is derived. [3] It is only, in fact, by availing\nhimself of the distinction, which he yet expressly rejects, between\nthe impression and its object, that he disguises the contradiction in\nterms of first pronouncing certain impressions, as parts of space,\nco-existent, and then pronouncing all impressions, as parts of time,\nsuccessive. A statement that \u2018as from the coexistence of visual, and\nalso of tactual, perceptions we receive the idea of extension, so\nfrom the succession of perceptions of every kind we form the idea\nof time,\u2019 would arouse the suspicion of the most casual reader;\nwhile Hume\u2019s version of the same,--\u2018as \u2019tis from the disposition of\nvisible and tangible objects we receive the idea of space, so from\nthe succession of ideas and impressions we form the idea of time\u2019\n[4]--has the full ring of empirical plausibility.\n\n[1] P. 483 and following, and p. 486 [Book I, part IV., sec. II.]\n\n[2] P. 516 [Book I, part IV., sec. IV.].\n\n[3] Pp. 342, 343 [Book I, part II., sec. III.].\n\n[4] P. 342 [Book I, part II., sec. III.]\n\nThe fact that colours mix, not to the purpose.\n\n248. This plausibility depends chiefly on our reading into Hume\u2019s\ndoctrine a physical theory which, as implying a distinction between\nfeeling and its real but unfelt cause, is strictly incompatible with\nit. Is it not an undoubted fact, the reader asks, that two colours\nmay combine to produce a third different from both--that red and\nyellow, for instance, together produce orange? Is not this already an\ninstance of a compound impression? Why may not a like composition of\nunextended impressions of colour constitute an impression different\nfrom any one of the component impressions, viz. extended colour?\nA moment\u2019s consideration, however, will show that no one has a\nconscious sensation at once of red and yellow, and of orange as a\ncompound of the two. The elements which combine to produce the colour\ncalled orange are not--as they ought to be if it is to be a case of\ncompound impression in Hume\u2019s sense--feelings of the person who sees\nthe orange colour, but certain known causes of feeling, confused in\nlanguage with the feelings, which separately they might produce,\nbut which in fact they do not produce when they combine to give the\nsensation of orange; and to such causes of feeling, which are not\nthemselves feelings, Hume properly can have nothing to say.\n\nHow Hume avoids appearance of identifying space with colour, and\naccounts for the abstraction of space.\n\n249. So far we have been considering the composition of impressions\ngenerally, without special reference to extension. The contradiction\npointed out arises from the confusion between impressions as felt\nand impressions as thought of; colour, between feelings as they\nare in themselves, presented successively in time, and feelings as\ndetermined by relation to the thinking subject, which takes them out\nof the flux of time and converts them into members of a permanent\nwhole. It is in this form that the confusion is most apt to elude us.\nWhen the conceived object is one of which the qualities can really\nbe felt, _e.g._ colour, we readily forget that a felt quality is no\nlonger simply a feeling. But the case is different when the object\nis one, like extension, which forces on us the question whether its\nqualities can be felt, or presented in feeling, at all. A compound\nof impressions of colour, to adopt Hume\u2019s phraseology, even if such\ncomposition were possible, would still be nothing else than an\nimpression of colour. In more accurate language, the conception,\nwhich results from the action of thought upon feelings of colour,\ncan only be a conception of colour. Is extension, then, the same as\ncolour? To say that it was would imply that geometry was a science\nof colour; and Hume, though ready enough to outrage \u2018Metaphysics and\nSchool Divinity,\u2019 always stops reverently short of direct offence\nto the mathematical sciences. As has been said above, of the three\nmain questions about the idea of extension which his doctrine\nraises--Is it itself extended? Is it possible without reference to\nsomething other than a possible impression? Is it the same as the\nidea of colour or tangibility?--the last is the only one which he can\nscarcely even profess to answer in the affirmative. [1] Even when he\nhas gone so far as to speak of the parts of a perception, a sound\ninstinct compels him, instead of identifying the perception directly\nwith extension, to speak of it as \u2018affording through the situation\nof its parts the notion of\u2019 extension. [2] In like manner, when he\nhas asserted extension to be a compound of impressions, he avoids\nthe proper consequence of the assertion by speaking of the component\nimpressions as those, not of colour but, of coloured points, \u2018atoms\nor corpuscles endowed with colour and solidity;\u2019 and, again, does not\ncall extension the compound of these simply, but the compound of them\nas \u2018disposed in a certain manner.\u2019 When the idea which is a copy of\nthis impression has to be spoken of, the expression is varied again.\nIt is an \u2018idea of the coloured points _and of the manner of their\nappearance_,\u2019 or of their \u2018disposition.\u2019 The disposition of the parts\nhaving been thus virtually distinguished from their colour, it is\neasy to suppose that, finding a likeness in the disposition of points\nunder every unlikeness of their colour, \u2018we omit the peculiarities\nof colour, as far as possible, and found an abstract idea merely on\nthat disposition of points, or manner of appearance, in which they\nagree. Nay, even when the resemblance is carried beyond the objects\nof one sense, and the impressions of touch are found to be similar\nto those of sight in the disposition of their parts, this does not\nhinder the abstract idea from representing both on account of their\nresemblance\u2019. [3]\n\n[1] Above, paragraph 233. Though, as we shall see, he does so in one\npassage.\n\n[2] Above, paragraph 235.\n\n[3] P. 341 [Book I, part II., sec. III.]\n\nIn so doing, he implies that space is a relation, and a relation\nwhich is not a possible impression.\n\n250. If words have any meaning, the above must imply that the\ndisposition of points is at least a different idea from either colour\nor tangibility, however impossible it may be for us to experience\nit without one or other of the latter. Nor can we suppose that\nthis impression, other than colour, is one that first results from\nthe composition of colours, even if we admit that such composition\ncould yield a result different from colour. According to Hume,\nthe components of the compound impression are already impressions\nof coloured \u2018points, atoms, or corpuscles,\u2019 and such points imply\njust that limitation by mutual externality, which is already the\ndisposition in question. Is this \u2018disposition,\u2019 then, an impression\nof sensation? If so, \u2018through which of the senses is it received? If\nit be perceived by the eyes it must be a colour,\u2019 &c. &c.; [1] but\nfrom colour, the impression with which Hume would have identified it\nif he could, he yet finds himself obliged virtually to distinguish\nit. It is a relation, and not even one of those relations, such as\nresemblance, which in Hume\u2019s language, \u2018depending on the nature\nof the impressions related,\u2019 [2] may plausibly be reckoned to\nbe themselves impressions. The \u2018disposition\u2019 of parts and their\n\u2018situation\u2019 he uses interchangeably, and the situation of impressions\nhe expressly opposes to their \u2018nature\u2019 [3]--that nature in respect\nof which all impressions, call them what we like, are \u2018originally\non the same footing\u2019 with pain and pleasure. Consistently with this\nhe pronounces the \u2018external position\u2019 of objects--their position as\nbodies external to each other and to our body--to be no datum of\nsense, no impression or idea, at all. [4] Our belief in it has to\nbe accounted for as a complex result of \u2018propensities to feign.\u2019\nHow, then, can there be an impression of that which does not belong\nto the nature of any impression? What difference is there between\n\u2018bodies\u2019 and \u2018corpuscles endowed with colour and solidity,\u2019 that the\noutwardness of the latter to each other--also called their \u2018distance\u2019\nfrom each, other [5]--should be an impression, while it is admitted\nthat the same relation between \u2018bodies\u2019 cannot be so?\n\n[1] Above, paragraph 208.\n\n[2] P. 372 [Book I, part III., sec. I.], \u2018Philosophical relations\nmay be divided into two classes: into such as depend entirely on the\nideas which we compare together; and such as may be changed without\nany change in the ideas. ... The relations of contiguity and distance\nbetween two objects may be changed without any change in the objects\nthemselves or their ideas.\u2019\n\n[3] P. 480. [Book I, part IV., sec. II.] \u2018When we doubt whether\nsensations present themselves as distinct objects or as mere\nimpressions, the difficulty is not concerning their nature, but\nconcerning their relations and situation.\u2019\n\n[4] P. 481. [Book I, part IV., sec. II.] In there showing that the\nsenses alone cannot convince us of the external existence of body,\nhe remarks that \u2018sounds, tastes, and smells appear not to have any\nexistence in extension;\u2019 and (p. 483) [ibid] \u2018as far as the senses\nare judges, all perceptions are the same in the manner of their\nexistence.\u2019 Therefore perceptions of sight cannot have \u2018an existence\nin extension\u2019 any more than \u2018sounds, tastes, and smells;\u2019 and if so,\nhow can \u2018existence in extension\u2019 be a perception?\n\n[5] Above, paragraphs 235 and 244.\n\nNo logical alternative between identifying space with colour, and\nadmitting an idea not copied from an impression.\n\n251. To have plainly admitted that it was not an impression must\nhave compelled Hume either to discard the \u2018abstract idea\u2019 with which\ngeometry deals, or to admit the possibility of ideas other than\n\u2018fainter impressions.\u2019 It is a principle on which he insists with\nmuch emphasis and repetition, that whatever \u2018objects,\u2019 \u2018impressions,\u2019\nor \u2018ideas\u2019 are distinguishable are also separable. [1] Now if there\nis an abstract idea of extension, it can scarcely be other than\ndistinguishable, and consequently (according to Hume\u2019s account of\nthe relation of idea to impression) derived from a distinguishable\nand therefore separable impression. It would seem then that Hume\ncannot escape conviction of one of two inconsistencies; either that\nof supposing a separate impression of extension, which yet is not\nof the nature of any assignable sensation; or that of supposing an\nabstract idea of it in the absence of any such impression. We shall\nfind that he does not directly face either horn of the dilemma, but\nevades both of them. He admits that \u2018the ideas of space and time\nare no separate and distinct ideas, but merely those of the manner\nor order in which objects\u2019 (_sc._ impressions) \u2018exist\u2019. [2] In the\nFourth Book, where the equivalence of impression to feeling is more\nconsistently carried out, the fact that what is commonly reckoned an\nimpression is really a judgment about the \u2018manner of existence,\u2019 as\nopposed to the \u2018nature,\u2019 of impressions, is taken as sufficient proof\nthat it is no impression at all; and if not an impression, therefore\nnot an idea. [3] He thus involuntarily recognized the true difference\nbetween feeling and thought, between the mere occurrence of feelings\nand the presentation of that occurrence by the self-conscious subject\nto itself; and, if only he had known what he was about in the\nrecognition, might have anticipated Kant\u2019s distinction between the\nmatter and form of sensation. In the Second Book, however, he will\nneither say explicitly that space is an impression of colour or a\ncompound of colours--that would be to extinguish geometry; nor yet\nthat it is impression of sense separate from that of colour--that\nwould lay him open to the retort that he was virtually introducing\na sixth sense; nor on the other hand will he boldly avow of it, as\nhe afterwards does of body, that it is a fiction. He denies that it\nis a separate impression, so far as that is necessary for avoiding\nthe challenge to specify the sense through which it is received;\nhe distinguishes it from a mere impression of sight, when it is\nnecessary to avoid its simple identification with colour. By speaking\nof it as \u2018the manner in which objects exist\u2019--so long as he is not\nconfronted with the declarations of the Fourth Book or with the\nquestion how, the objects being impressions, their order of existence\ncan be at once that of succession in time and of co-existence in\nspace--he gains the credit for it of being a datum of sight, yet\nso far distinct from colour as to be a possible \u2018foundation for an\nabstract idea,\u2019 representative also of objects not coloured at all\nbut tangible. At the same time, if pressed with the question how it\ncould be an impression of sight and yet not interchangeable with\ncolour, he could put off the questioner by reminding him that he\nnever made it a \u2018separate or distinct impression, but one of the\nmanner in which objects exist.\u2019\n\n[1] Pp. 319, 326, 332, 335, 518. [Book I, part I., sec. IV and VII,\npart II, sec. I, and part IV., sec. V.]\n\n[2] P. 346. [Book I, part II., sec. IV.]\n\n[3] P. 480. [Book I, part IV., sec. II.]\n\nIn his account of the idea as _abstract_, Hume really introduces\ndistinction between feeling and conception;\n\n252. Disguise it as he might, however, the admission that there\nwas in some sense an abstract idea of space, which the existence\nof geometry required of him, really carried with it the admission\neither of a distinct impression of the same, or of some transmuting\nprocess by which the idea may become what the impression is not.\nHis way of evading this consequence has been already noticed in our\nexamination of his doctrine of \u2018abstract ideas\u2019 generally, though\nwithout special reference to extension. [1] It consists in asserting\nfigure and colour to be \u2018really,\u2019 or as an impression, \u2018the same and\nindistinguishable,\u2019 but different as \u2018relations and resemblances\u2019 of\nthe impression; in other words, different according to the \u2018light in\nwhich the impression is considered\u2019 or \u2018the aspect in which it is\nviewed.\u2019 Of these \u2018separate resemblances and relations,\u2019 however, are\nthere ideas or are there not? If there are not, they are according\nto Hume nothing of which we are conscious at all; if there are,\nthere must be distinguishable, and therefore separable, impressions\ncorresponding. To say then that figure and colour form one and the\nsame indistinguishable impression, and yet that they constitute\n\u2018different resemblances and relations,\u2019 without such explanation\nas Hume cannot consistently give, is in fact a contradiction in\nterms. The true explanation is that the \u2018impression\u2019 has a different\nmeaning, when figure and colour are said to be inseparable in the\nimpression, from that which it has when spoken of as a subject of\ndifferent resemblances and relations. In the former sense it is\nthe feeling pure and simple--_one_ as presented singly in time,\nafter another and before a third. In this sense it is doubtless\ninsusceptible of distinction into qualities of figure and colour,\nbecause (for reasons already stated) it can have no qualities at\nall. But the \u2018simplicity in which many different resemblances and\nrelations may be contained\u2019 is quite other than this singleness. It\nis the unity of an object thought of under manifold relations--a\nunity of which Hume, reducing all consciousness to \u2018impression\u2019 and\nimpression to feeling, has no consistent account to give. Failing\nsuch an account, the unity of the intelligible object, and the\nsingleness of the feeling in time, are simply confused with each\nother. It is only an object as thought of, not a feeling as felt,\nthat can properly be said to have qualities at all; while it is only\nbecause it is still regarded as a feeling that qualities of it, which\ncannot be referred to separate impressions, are pronounced the same\nand indistinguishable. If the idea of space is other than a feeling\ngrown fainter, the sole reason for regarding it as originally an\nimpression of colour disappears; if it _is_ such a feeling, it cannot\ncontain such \u2018different resemblances and relations\u2019 as render it\nrepresentative of objects not only coloured in every possible way,\nbut not coloured at all.\n\n[1] Above, paragraph 218.\n\n... yet avoids appearance of doing so, by treating \u2018consideration\u2019 of\nthe relations of a felt thing as if it were itself the feeling.\n\n253. It is thus by playing fast and loose with the difference between\nfeeling and conception that Hume is able, when the character of\nextension as an intelligible relation is urged, to reply that it is\nthe same with the feeling of colour; and on the other hand, when\nasked how there then can be an abstract idea of it, to reply that\nthis does not mean a separate idea, but coloured objects considered\nunder a certain relation, viz. under that which consists in the\ndisposition of their parts. The most effective way of meeting him on\nhis own ground is to ask him how it is, since \u2018consideration\u2019 can\nonly mean a succession of ideas, and ideas are fainter impressions,\nthat extension, being one and the same impression with colour, can\nby any \u2018consideration\u2019 become so different from it as to constitute\na resemblance to objects that are not coloured at all. The true\nexplanation, according to his own terminology, would be that the\nresemblance between the white globe and all other globes, being\na resemblance not of impressions but of such relations between\nimpressions as do not \u2018depend on the nature of the impressions\u2019\nrelated, is unaffected by the presence or absence of colour or any\nother sensation. Of such relations, however, there can properly,\nif ideas are fainter impressions, be no ideas at all. In regard\nto those of cause and identity Hume virtually admits this; but\nthe \u2018propensities to feign\u2019 by which in the case of these latter\nrelations he tries to account for the appearance of there being\nideas of them, cannot plausibly be applied to relations in space and\ntime, of which, as we shall see, ideas must be assumed in order to\naccount for the \u2018fictions\u2019 of body and necessary connexion. Since\nthen they cannot be derived from any separate impression without the\nintroduction in effect of a sixth sense, and since all constitutive\naction of thought as distinct from feeling is denied by Hume, the\nonly way to save appearances is to treat the order in which a\nmultitude of impressions present themselves as the same with each\nimpression, even though immediately afterwards it may have to be\nconfessed, that it is so independent of the nature of any or all of\nthe impressions as to be the foundation of an abstract idea, which is\nrepresentative of other impressions having nothing whatever in common\nwith them but the order of appearance. This once allowed--an abstract\nidea having been somehow arrived at which is not really the copy of\nany impression--it is easy to argue back from the abstract idea to an\nimpression, and because there is an idea of the composition of points\nto substitute a \u2018composition of coloured points\u2019 for colour as the\noriginal impression. From such impression, being already extension,\nthe idea of extension can undoubtedly be abstracted.\n\nSummary of contradictions in his account of extension.\n\n254. We now know what becomes of \u2018extended matter\u2019 when the doctrine,\nwhich has only to be stated to find acceptance, that we cannot \u2018look\nfor anything anywhere but in our ideas\u2019--in other words that for\nus there is no world but consciousness--is fairly carried out. Its\nposition must become more and more equivocal, as the assumption, that\nconsciousness reveals to us an alien matter, has in one after another\nof its details to be rejected, until a principle of synthesis within\nconsciousness is found to explain it. In default of this, the feeling\nconsciousness has to be made to take its place as best it may; which\nmeans that what is said of it as feeling has to be unsaid of it as\nextended, and _vice vers\u00e2_. As _feeling_, it carries no reference to\nanything other than itself, to an object of which it is a quality;\nas _extended_, it is a qualified object. As _extended_ again, its\nqualities are relations of coexistent parts; as _feeling_, it is an\nunlimited succession, and therefore, not being a possible whole, can\nhave no parts at all. Finally as _feeling_, it must in each moment of\nexistence either be \u2018on the same footing\u2019 with pain and pleasure or\nelse--a distinction between impressions of sensation and reflection\nbeing unwarrantably admitted--be a colour, a taste, a sound, a smell,\nor \u2018tangibility;\u2019 as _extended_, it is an \u2018order of appearance\u2019 or\n\u2018disposition of corpuscles,\u2019 which, being predicable indifferently\nat any rate of two of these sensations, can no more be the same with\neither than either can be the same with the other. It is not the\nfault of Hume but his merit that, in undertaking to maintain more\nstrictly than others the identification of extension with feeling,\nhe brought its impossibility more clearly into view. The pity is\nthat having carried his speculative enterprise so far before he was\nthirty, he allowed literary vanity to interfere with its consistent\npursuit, caring only to think out the philosophy which he inherited\nso far as it enabled him to pose with advantage against Mystics and\nDogmatists, but not to that further issue which is the entrance to\nthe philosophy of Kant.\n\nHe gives no account of quantity as such.\n\n255. As it was, he never came fairly to ask himself the fruitful\nquestion. How the sciences of quantity \u2018continuous and discreet,\u2019\nwhich undoubtedly do exist, are possible to a merely feeling\nconsciousness, because, while professedly reducing all consciousness\nto this form, he still allowed himself to interpret it in the terms\nof these sciences and, having done so, could easily account for their\napparent \u2018abstraction\u2019 from it. If colour is already for feeling a\nmagnitude, as is implied in calling it a \u2018composition of coloured\npoints,\u2019 the question, how a knowledge of magnitude is possible, is\nof course superfluous. It only remains to deal, as Hume professes to\ndo, with the apparent abstraction in mathematics of magnitude from\ncolour and the consequent suppositions of pure space and infinite\ndivisibility. Any ulterior problem he ignores. That magnitude is not\nany the more a feeling for being \u2018endowed with colour\u2019 he shows no\nsuspicion. He pursues his \u2018sensationalism\u2019 in short, in its bearing\non mathematics, just as far as Berkeley did and no further. The\nquestion at issue, as he conceived it, was not as to the possibility\nof magnitude altogether, but only as to the existence of a vacuum;\nnot as to the possibility of number altogether, but only as to the\ninfinity of its parts. Just as he takes magnitude for granted as\nfound in extension, and extension as equivalent to the feeling of\ncolour, so he takes number for granted, without indeed any explicit\naccount of the impression in which it is to be found, but apparently\nas found in time, which again is identified with the succession of\nimpressions. In the second part of the Treatise, though the idea of\nnumber is assumed and an account is given of it which is supposed\nto be fatal to the infinite divisibility of extension, we are told\nnothing of the impression or impressions from which it is derived.\nIn the Fourth Part, however, there is a passage in which a certain\nconsideration of time is spoken of as its source.\n\nHis account of the relation between Time and Number.\n\n256. In the latter passage, in order to account for the idea of\nidentity, he is supposing \u2018a single object placed before us and\nsurveyed for any time without our discovering in it any variation\nor interruption.\u2019 \u2018When we consider any two points of this time,\u2019\nhe proceeds, \u2018we may place them in different lights. We may either\nsurvey them at the very same instant; in which case they give us the\nidea of number, both by themselves and by the object, which must be\nmultiplied in order to be conceived at once, as existent in these\ntwo different points of time: or, on the other hand, we may trace\nthe succession of time by a like succession of ideas, and conceiving\nfirst one moment, along with the object then existent, imagine\nafterwards a change in the time without any variation or interruption\nin the object; in which case it gives us the idea of unity\u2019. [1]\n\n[1] P. 490. [Book I, part IV., sec. II.]\n\nWhat does it come to?\n\n257. A slight scrutiny of this passage will show that it is a\nprolonged tautology. The difference is merely verbal between the\nprocesses by which the ideas of number and unity are severally\nsupposed to be given, except that in the former process it is the\nmoment of surveying the times that is supposed to be one, while\nthe times themselves are many; in the latter it is the object that\nis supposed to be one, but the times many. According to the second\nversion of the former process--that according to which the different\ntimes surveyed together are said to give the idea of number \u2018by\ntheir object\u2019--even this difference disappears. The only remaining\ndistinction is that in the one case the object is supposed to be\ngiven as one, \u2018without interruption or variation,\u2019 but to become\nmultiple as conceived to exist in different moments; in the other the\nobjects are supposed to be given as manifold, being ideas presented\nin successive times, but to become one through the imaginary\nrestriction of the multiplicity to the times in distinction from the\nobject. Undoubtedly any one of these verbally distinct processes will\nyield indifferently the ideas of number and of unity, since these\nideas in strict correlativity are presupposed by each of them. \u2018Two\npoints of time surveyed at the same time\u2019 will give us the idea of\nnumber because, being a duality in unity, they are already a number.\nSo, too, and for the same reason, will the object, one in itself but\nmultiple as existent at different times. Nor does the idea given by\nimagining ideas, successively presented, to be \u2018one uninterrupted\nobject,\u2019 differ from the above more than many-in-one differs from\none-in-many. The real questions of course are, How two times can be\nsurveyed at one time; how a single object can be multiplied or become\nmany; how a succession of ideas can be imagined to be an unvaried and\nuninterrupted object. To these questions Hume has no answer to give.\nHis reduction of thought to feeling logically excluded an answer, and\nthe only alternative for him was to ignore or disguise them.\n\nUnites alone really exist: number a \u2018fictitious denomination\u2019. Yet\n\u2018unites\u2019 and \u2018number\u2019 are correlative; and the supposed fiction\nunaccountable.\n\n258. In the passage from part II. of the Treatise, already referred\nto, he distinctly tells us that the unity to which existence\nbelongs excludes multiplicity. \u2018Existence itself belongs to unity,\nand is never applicable to number but on account of the unites of\nwhich the number is composed. Twenty men may be said to exist, but\n\u2019tis only because one, two, three, four, &c., are existent. ... A\nunite, consisting of a number of fractions, is merely a fictitious\ndenomination, which the mind may apply to any quantity of objects\nit collects together; nor can such an unity any more exist alone\nthan number can, as being in reality a true number. But the unity\nwhich can exist alone, and whose existence is necessary to that of\nall number, is of another kind and must be perfectly indivisible\nand incapable of being resolved into any lesser unity\u2019. [1] What\nthen is the \u2018unity which can exist alone\u2019? The answer, according\nto Hume, must be that it is an impression separately felt and not\nresoluble into any other impressions. But then the question arises,\nhow a succession of such impressions can form a number or sum; and\nif they cannot, how the so-called real unity or separate impression\ncan in any sense be a unite, since a unite is only so as one of a\nsum. To put the question otherwise, Is it not the case that a unite\nhas no more meaning without number than number without unites, and\nthat every number is not only just such a \u2018fictitious denomination,\u2019\nas Hume pronounces a \u2018unite consisting of a number of fractions\u2019\nto be, but a fiction impossible for our consciousness according to\nHume\u2019s account of it? It will not do to say that such a question\ntouches only the fiction of \u2018abstract number,\u2019 but not the existence\nof numbered objects; that (to take Hume\u2019s instance) twenty men exist\nwith the existence of each individual man, each real unit, of the\nlot. It is precisely the numerability of objects--not indeed their\nexistence, if that only means their successive appearance, but their\nexistence _as a sum_--that is in question. If such numerability\nis possible for such a consciousness as Hume makes ours to be; in\nother words, if he can explain the fact that we count; \u2018abstract\nnumber\u2019 may no doubt be left to take care of itself. Is it then\npossible? \u2018Separate impressions\u2019 mean impressions felt at different\ntimes, which accordingly can no more co-exist than, to use Hume\u2019s\nexpression, \u2018the year 1737 can concur with the year 1738;\u2019 whereas\nthe constituents of a sum must, as such, co-exist. Thus when we are\ntold that \u2018twenty may be said to exist because one, two, three, &c.,\nare existent,\u2019 the alleged reason, understood as Hume was bound\nto understand it, is incompatible with the supposed consequence.\nThe existence of an object would, to him, mean no more than the\noccurrence of an impression; but that one impression should occur,\nand then another and then another, is the exact opposite of their\ncoexistence as a sum of impressions, and it is such co-existence that\nis implied when the impressions are counted and pronounced so many.\nThus when Hume tells us that a single object, by being \u2018multiplied\nin order to be conceived at once as existent in different points of\ntime,\u2019 gives us the idea of number, we are forced to ask him what\nprecisely it is which thus, being one, can become manifold. Is it a\n\u2018unite that can exist alone\u2019? That, having no parts, cannot become\nmanifold by resolution. \u2018But it may by repetition?\u2019 No, for it is\na separate impression, and the repetition of an impression cannot\nco-exist, so as to form one sum, with its former occurrence. \u2018But it\nmay be _thought of_ as doing so?\u2019 No, for that, according to Hume,\ncould only mean that feelings might concur in a fainter stage though\nthey could not in a livelier. Is the single object then a unite which\nalready consists of parts? But that is a \u2018fictitious denomination,\u2019\nand presupposes the very idea of number that has to be accounted for.\n\n[1] P. 338. [Book I, part II., sec. II.]\n\nIdea of time even more unaccountable on Hume\u2019s principles.\n\n259. The impossibility of getting number, as a many-in-one, out of\nthe succession of feelings, so long as the self is treated as only\nanother name for that succession, is less easy to disguise when the\nsupposed units are not merely given in succession, but are actually\nthe moments of the succession; in other words, when time is the\nmany-in-one to be accounted for. How can a multitude of feelings\nof which no two are present together, undetermined by relation to\nanything other than the feelings, be at the same time a consciousness\nof the relation between the moments in which the feelings are given,\nor of a sum which these moments form? How can there be a relation\nbetween \u2018objects\u2019 of which one has ceased before the other has\nbegun to exist? \u2018For the same reason,\u2019 says Hume, \u2018that the year\n1737 cannot concur with the present year 1738, every moment must be\ndistinct from, and posterior or antecedent to, another\u2019. [1] How\nthen can the present moment form one sum with all past moments,\nthe present year with all past years; the sum which we indicate\nby the number 1738? The answer of common sense of course will be\nthat, though the feeling of one moment is really past before that\nof another begins, yet thought retains the former, and combining\nit with the latter, gets the idea of time both as a relation and\nas a sum. Such an answer, however, implies that the retaining and\ncombining thought is other than the succession of the feelings, and\nwhile it takes this succession to be the reality, imports into it\nthat determination by the relations of past and present which it can\nonly derive from the retaining and combining thought opposed to it.\nIt is thus both inconsistent with Hume\u2019s doctrine, which allows no\nsuch distinction between thought, _i.e._ the succession of ideas,\nand the succession of impressions, and inconsistent with itself. Yet\nHume by disguising both inconsistencies contrives to avail himself\nof it. By tacitly assuming that a conception of \u2018the manner in which\nimpressions appear to the mind\u2019 is given in and with the occurrence\nof the impressions, he imports the consciousness of time, both as\nrelation and as numerable quantity, into the sequence of impressions.\nHe thus gains the advantage of being able to speak of this sequence\nindifferently under predicates which properly exclude each other.\nHe can make it now a consciousness in time, now a consciousness\nof itself as in time; now a series that cannot be summed, now a\nconception of the sum of the series. The sequence of feelings, then,\nhaving been so dealt with as to make it appear in effect that time\ncan be _felt_, that it should be _thought of_ can involve no further\ndifficulty. The conception, smuggled into sensitive experience as\nan \u2018impression,\u2019 can be extracted from it again as \u2018idea,\u2019 without\nostensible departure from the principle that the idea is only the\nweaker impression.\n\n[1] P. 338. [Book I, part II., sec. II.]\n\nHis ostensible explanation of it.\n\n260. \u2018The idea of time is not derived from a particular impression\nmixed up with others and plainly distinguishable from them, but\narises altogether from the manner in which impressions appear to the\nmind, without making one of the number. Five notes played on the\nflute give us the impression and idea of time, though time be not a\nsixth impression which presents itself to the hearing or any other of\nthe senses. Nor is it a sixth impression which the mind by reflection\nfinds in itself. These five sounds, making their appearance in this\nparticular manner, excite no emotion or affection in the mind, which\nbeing observed by it can give rise to a new idea. For _that_ is\nnecessary to produce a new idea of reflection; nor can the mind, by\nrevolving over a thousand times all its ideas of sensation, ever\nextract from them any new original idea, unless nature has so framed\nits faculties that it feels some new original impression arise from\nsuch a contemplation. But here it only takes notice of the _manner_\nin which the different sounds make their appearance, and that it may\nafterwards consider without considering these particular sounds, but\nmay conjoin it with any other objects. The ideas of some objects it\ncertainly must have, nor is it possible for it without these ever to\narrive at any conception of time; which, since it appears not as any\nprimary distinct impression, can plainly be nothing but different\nideas or impressions or objects disposed in a certain manner, _i.e._\nsucceeding each other. [1]\n\n[1] P. 343. [Book I, part II., sec. III.]\n\nIt turns upon equivocation between feeling and conception of\nrelations between felt things.\n\n261. In this passage the equivocation between \u2018impression\u2019 as\nfeeling, and \u2018impression\u2019 as conception of the manner in which\nfeelings occur, is less successfully disguised than is the like\nequivocation in the account of extension--not indeed from any failure\nin Hume\u2019s power of statement, but from the nature of the case. In\ntruth the mere reproduction of impressions can as little account for\nthe one conception as for the other. Just as, in order to account\nfor the \u2018impression\u2019 from which the abstract idea of space may\nbe derived, we have to suppose first that the feeling of colour,\nthrough being presented by the self-conscious subject to itself,\nbecomes a coloured thing, and next, that this thing is viewed as\na whole of parts limiting each other; so, in order to account for\nthe \u2018impression\u2019 from which the idea of time may be abstracted,\nwe have to suppose the presentation of the succession of feelings\nto a consciousness not in succession, and the consequent view of\nsuch presented succession as a sum of numerable parts. It is a\nrelation only possible for a thinking consciousness--a relation,\nin Hume\u2019s language, not depending on the nature of the impressions\nrelated--that has in each case to be introduced into experience in\norder to be extracted from it again by \u2018consideration:\u2019 but there is\nthis difference, that in one case the relation is not really between\nfeelings at all, but between things or parts of a thing; while in the\nother it is just that relation between feelings, the introduction\nof which excludes the possibility that any feeling should be the\nconsciousness of the relation. Thus to speak of a feeling of\nextension does not involve so direct a contradiction as to speak in\nthe same way of time. The reader gives Hume the benefit of a way of\nthinking which Hume\u2019s own theory excludes. Himself distinguishing\nbetween feeling and felt thing, and regarding extension as a relation\nbetween parts of a thing, he does not reflect that for Hume there is\nno such distinction; that a \u2018feeling of extension\u2019 means that feeling\nis extended, which again means that it has co-existent parts; and\nthat what is thus said of feeling as _extended_ is incompatible with\nwhat is said of it as _feeling_. But when it comes to a \u2018feeling\nof time\u2019--a feeling of the successiveness of all feelings--the\nincompatibility between what is said of feeling as the object and\nwhat is implied of it as the subject is less easy to disguise. In\nlike manner because we cannot really think of extension as being that\nwhich yet according to Hume it is, it does not strike us, when he\nspeaks of it as coloured or of colour as extended, that he is making\none feeling a quality of another. But it would be otherwise if any\nspecific feeling were taken as a quality of what is ostensibly a\nrelation between all feelings. There is thus no \u2018sensible quality\u2019\nwith which time can be said to be \u2018endowed,\u2019 as extension with\n\u2018colour and solidity;\u2019 none that can be made to do the same duty in\nregard to it as these do in regard to extension, \u2018giving the idea\u2019 of\nit without actually being it.\n\nHe fails to assign any impression or compound of impressions from\nwhich idea of time is copied.\n\n262. Hence, as the passage last quoted shows, in the case of time the\nalternative between ascribing it to a sixth sense, and confessing\nthat it is not an impression at all, is very hard to avoid. It would\nseem that there is an impression of \u2018the manner in which impressions\nappear to the mind,\u2019 which yet is no \u2018distinct impression.\u2019 What,\nthen, is it? It cannot be any one of the impressions of sense, for\nthen it would be a distinct impression. It cannot be a \u2018compound\nimpression,\u2019 for such composition is incompatible with that\nsuccessiveness of all feelings to each other which is the object of\nthe supposed impression. It cannot be any \u2018new original impression\u2019\narising from the contemplation of other impressions, for then,\naccording to Hume, it would be \u2018an affection or emotion.\u2019 But after\nthe exclusion of impressions of sense, compound impressions, and\nimpressions of reflection, Hume\u2019s inventory of the possible sources\nof ideas is exhausted. To have been consistent, he ought to have\ndealt with the relation of time as he afterwards does with that of\ncause and effect, and, in default of an impression from which it\ncould be derived, have reduced it to a figure of speech. But since\nthe possibility of accounting for the propensities to feign, which\nour language about cause and effect according to him represents,\nrequired the consciousness of relation in time, this course could\nnot be taken. Accordingly after the possibility of time being an\nimpression has been excluded as plainly as it can be by anything\nshort of a direct negation, by a device singularly _na\u00eff_ it is\nmade to appear as an impression after all. On being told that\nthe consciousness of time is not a \u2018new original impression of\nreflection,\u2019 since in that case it would be an emotion or affection,\nbut \u2018_only_ the notice which the mind takes of the manner in which\nimpressions appear to it,\u2019 the reader must be supposed to forget\nthe previous admission that it is no distinct impression at all,\nand to interpret this \u2018notice which the mind takes,\u2019 because it is\nnot an impression of reflection, as an impression of sense. To make\nsuch interpretation easier, the account given of time earlier in\nthe paragraph quoted is judiciously altered at its close, so that\ninstead of having to ascribe to feeling a consciousness of \u2018the\nmanner in which impressions appear to the mind,\u2019 we have only to\nascribe to it the impressions so appearing. But this alteration\nadmitted, what becomes of the \u2018abstractness\u2019 of the idea of time,\n_i.e._ of the possibility of its being \u2018conjoined with any objects\u2019\nindifferently? It is the essential condition of such indifferent\nconjunction, as Hume puts it, that time should be only the manner of\nappearance as distinct from the impressions themselves. If time _is_\nthe impressions, it must have the specific sensuous character which\nbelongs to these. It must be a multitude of sounds, a multitude of\ntastes, a multitude of smells--these one after the other in endless\nseries. How then can such a series of impressions become such an\nidea, _i.e._ so grow fainter as to be \u2018conjoined\u2019 indifferently \u2018with\nany impressions whatever\u2019?\n\nHow can he adjust the exact sciences to his theory of space and time?\n\n263. The case then between Hume and the conceptions which the exact\nsciences presuppose, as we have so far examined it, stands thus. Of\nthe idea of quantity, as such, he gives no account whatever. We are\ntold, indeed, that there are \u2018unites which can exist alone,\u2019 _i.e._\ncan be felt separately, and which are indivisible; but how such\nunites, being separate impressions, can form a sum or number, or what\nmeaning a unite can have except as one of a number--how again a sum\nformed of separate unites can be a continuous whole or magnitude--we\nare not told at all. Of the ideas of space and time we do find an\naccount. They are said to be given in impressions, but, to justify\nthis account of them, each impression has to be taken to be at the\nsame time a consciousness of the manner of its own existence, as\ndetermined by relation to other impressions not felt along with it\nand as interpreted in a way that presupposes the unexplained idea\nof quantity. With this supposed origin of the ideas the sciences\nresting on them have to be adjusted. They may take the relations of\nnumber and magnitude, time and space, for granted, as \u2018qualities of\nperceptions,\u2019 and no question will be asked as to how the perceptions\ncome to assume qualities confessed to be \u2018independent of their own\nnature.\u2019 It is only when they treat them in a way incompatible not\nmerely with their being feelings--that must always be the case--but\nwith their being relations between felt things, that they are\nsupposed to cross the line which separates experimental knowledge\nfrom metaphysical jargon. So long then as space is considered merely\nas the relation of externality between objects of the \u2018outer,\u2019 time\nas that of succession between objects of the \u2018inner,\u2019 sense--in\nother words, so long as they remain what they are to the earliest\nself-consciousness and do not become the subject matter of any\nscience of quantity--if we sink the difference between feelings and\nrelations of felt things, and ask no questions about the origin of\nthe distinction between outer and inner sense, they may be taken as\ndata of sensitive experience. It is otherwise when they are treated\nas quantities, and it is their susceptibility of being so treated\nthat, rightly understood, brings out their true character as the\nintelligible element in sensitive experience. But Hume contrives at\nonce to treat them as quantities, thus seeming to give the exact\nsciences their due, and yet to appeal to their supposed origin in\nsense as evidence of their not having properties which, if they are\nquantities, they certainly must have. Having thus seemingly disposed\nof the purely intelligible character of quantity in its application\nto space and time, he can more safely ignore what he could not so\nplausibly dispose of--its pure intelligibility as number.\n\nIn order to seem to do so, he must get rid of \u2018Infinite Divisibility\u2019.\n\n264. The condition of such a method being acquiesced in is, that\nquantity in all its forms should be found reducible to ultimate\nunites or indivisible parts in the shape of separate impressions.\nShould it be found so, the whole question indeed, how ideas of\nrelation are possible for a merely feeling consciousness, would\nstill remain, but mathematics would stand on the same footing\nwith the experimental sciences, as a science of relations between\nimpressions. Upon this reducibility, then, we find Hume constantly\ninsisting. In regard to number indeed he could not ignore the fact\nthat the science which deals with it recognizes no ultimate unite,\nbut only such a one as \u2018is itself a true number.\u2019 But he passes\nlightly over this difficulty with the remark that the divisible\nunite of actual arithmetic is a \u2018fictitious denomination\u2019--leaving\nhis reader to guess how the fiction can be possible if the real\nunite is a separate indivisible impression--and proceeds with the\nmore hopeful task of resolving space into such impressions. He is\nwell aware that the constitution of space by impressions and its\nconstitution by indivisible parts stand or fall together. If space is\na compound impression, it is made up of indivisible parts, for there\nis a \u2018minimum visibile\u2019 and by consequence a minimum of imagination;\nand conversely, if its parts are indivisible, they can be nothing\nbut impressions; for, being indivisible, they cannot be extended,\nand, not being extended, they must be either simple impressions or\nnothing. With that instinct of literary strategy which never fails\nhim, Hume feels that the case against infinite divisibility, from its\napparent implication of an infinite capacity in the mind, is more\neffective than that in favour of space being a compound impression,\nand accordingly puts that to the front in the Second Part of the\nTreatise, in order, having found credit for establishing it, to argue\nback to the constitution of space by impressions. In fact, however,\nit is on the supposed composition of all quantity from separate\nimpressions that his argument against its infinite divisibility rests.\n\nQuantity made up of impressions, and there must be a least possible\nimpression.\n\n265. The essence of his doctrine is contained in the following\npassages: \u2018\u2019Tis certain that the imagination reaches a _minimum_,\nand may raise up to itself an idea, of which it cannot conceive\nany subdivision, and which cannot be diminished without a total\nannihilation. When you tell me of the thousandth and ten thousandth\npart of a grain of sand, I have a distinct idea of these numbers\nand of their several proportions, but the images which I form in\nmy mind to represent the things themselves are nothing different\nfrom each other nor inferior to that image by which I represent\nthe grain of sand itself, which is supposed so vastly to exceed\nthem. What consists of parts is distinguishable into them, and what\nis distinguishable is separable. But whatever we may imagine of\nthe thing, the idea of a grain of sand is not distinguishable nor\nseparable into twenty, much less into a thousand, ten thousand, or\nan infinite number of different ideas. \u2019Tis the same case with the\nimpressions of the senses as with the ideas of the imagination. Put\na spot of ink upon paper, fix your eye upon that spot, and retire\nto such a distance that at last you lose sight of it; \u2019tis plain\nthat the moment before it vanished the image or impression was\nperfectly indivisible. \u2019Tis not for want of rays of light striking\non our eyes that the minute parts of distant bodies convey not\nany sensible impression; but because they are removed beyond that\ndistance at which their impressions were reduced to a _minimum_, and\nwere incapable of any further diminution. A microscope or telescope,\nwhich renders them visible, produces not any new rays of light,\nbut only spreads those which always flowed from them; and by that\nmeans both gives parts to impressions, which to the naked eye appear\nsimple and uncompounded, and advances to a minimum what was formerly\nimperceptible.\u2019 [1]\n\n[1] P. 335, Part II. \u00a7 1.\n\nYet it is admitted that there is an idea of number not made up of\nimpressions. A finite division into impressions no more possible than\nan infinite one.\n\n266. In this passage it will be seen that Hume virtually yields the\npoint as regards number. When he is told of the thousandth or ten\nthousandth part of a grain of sand he has \u2018a distinct idea of these\nnumbers and of their different proportions,\u2019 though to this idea no\ndistinct \u2018image\u2019 corresponds; in other words, though the idea is\nnot a copy of any impression. It is of such parts _as parts of the\ngrain of sand_--as parts of a \u2018compound impression\u2019--that he can\nform no idea, and for the reason given in the sequel, that they are\nless than any possible impression, less than the \u2018minimum visibile.\u2019\nThis, it would seem, is a fixed quantity. That which is the least\npossible impression once is so always. Telescopes and microscopes do\nnot alter it, but present it under conditions under which it could\nnot be presented to the naked eye. Their effect, according to Hume,\ncould not be to render that visible which existed unseen before, nor\nto reveal parts in that which previously had, though it seemed not\nto have, them--that would imply that an impression was \u2018an image of\nsomething distinct and external\u2019--but either to present a simple\nimpression of sight where previously there was none or to substitute\na compound impression for one that was simple. [1] It is then because\nall divisibility is supposed to be into impressions, _i.e._ into\nfeelings, and because there are conditions under which every feeling\ndisappears, that an infinite divisibility is pronounced impossible.\nBut the question is whether a finite divisibility into feelings is\nnot just as impossible as an infinite one. Just as for the reasons\nstated above [2] a \u2018compound feeling\u2019 is impossible, so is the\ndivision of a compound into feelings. Undoubtedly if the \u2018minimum\nvisibile\u2019 were a feeling it would not be divisible, but for the same\nreason it would not be a quantity. But if it is not a quantity,\nwith what meaning is it called a minimum, and how can a quantity be\nsupposed to be made up of such \u2018visibilia\u2019 as have themselves no\nquantity? In truth the \u2018minimum visibile\u2019 is not a feeling at all but\na felt thing, conceived under attributes of quantity; in particular,\nas the term \u2018minimum\u2019 implies, under a relation of proportion to\nother quantities of which, if expressed numerically, Hume himself,\naccording to the admission above noticed, would have to confess there\nwas an idea which was an image of no impression. That which thought\nthus presents to itself as a thing doubtless has been a feeling;\nbut, as thus presented, it is already other than and independent of\nfeeling. With a step backward or a turn of the head, the feeling\nmay cease, \u2018the spot of ink may vanish;\u2019 but the thing does not\ntherefore cease to be a thing or to have quantity, which implies the\npossibility of continuous division.\n\n[1] It will be noticed that in the last sentence of the passage\nquoted, Hume assumes the convenient privilege of \u2018speaking with the\nvulgar,\u2019 and treats the \u2018minimum visibile\u2019 presented by telescope\nor microscope as representing something other than itself, which\npreviously existed, though it was imperceptible.\n\n[2] See above, \u00a7\u00a7 241 & 246.\n\nIn Hume\u2019s instances it is not really a feeling, but a conceived\nthing, that appears as finitely divisible.\n\n267. It is thus the confusion between feeling and conception that\nis at the bottom of the difficulty about divisibility. For a\nconsciousness formed merely by the succession of feelings, as there\nwould be no _thing_ at all, so there would be no parts of a thing--no\naddibility or divisibility. But Hume is forced by the exigencies\nof his theory to hold together, as best he may, the reduction of\nall consciousness to feeling and the existence for it of divisible\nobjects. The consequence is his supposition of \u2018compound impressions\u2019\nor feelings having parts, divisible into separate impressions but\ndivisible no further when these separate impressions have been\nreached. We find, however, that in all the instances he gives it is\nnot really a feeling that is divided into feelings, but a thing into\nother things. It is the heap of sand, for instance, that is divided\ninto grains, not the feeling which, by intellectual interpretation,\nrepresents to me a heap of sand that is divided into lesser feelings.\nI may feel the heap and feel the grain, but it is not a feeling\nthat is the heap nor a feeling that is the grain. Hume would not\noffend common sense by saying that it was so, but his theory really\nrequired that he should, for the supposition that the grain is no\nfurther divisible when there are no separate impressions into which\nit may be divided, implies that in that case it is itself a separate\nimpression, even as the heap is a compound one. But what difference,\nit may be asked, does it make to say that the heap and the grain are\nnot feelings, but things conceived of, if it is admitted, as since\nBerkeley it must be, that the thing is nothing outside or independent\nof consciousness? Do we not by such a statement merely change names\nand invite the question how a thought can have parts, in place of the\nquestion how a feeling can have them?\n\nUpon true notion of quantity infinite divisibility follows of course.\n\n268. If thought were no more than Hume takes feeling to be, this\nobjection would be valid. But if by thought we understand the\nself-conscious principle which, present to all feelings, forms out\nof them a world of mutually related objects, permanent with its own\npermanence, we shall also understand that the relations by which\nthought qualifies its object are not qualities of itself--that,\nin thinking of its object as made up of parts, it does not become\nitself a quantum. We shall also be on the way to understand how\nthought, detaching that relation of simple distinctness by which it\nhas qualified its objects, finds before it a multitude of units of\nwhich each, as combining in itself distinctions from all the other\nunits, is at the same time itself a multitude; in other words,\nfinds a quantum of which each part, being the same in kind with\nthe whole and all other parts, is also a quantum; _i.e._ which is\ninfinitely divisible. When once it is understood, in short, that\nquantity is simply the most elementary of the relations by which\nthought constitutes the real world, as detached from this world and\npresented by thought to itself as a separate object, then infinite\ndivisibility becomes a matter of course. It is real just in so far\nas quantity, of which it is a necessary attribute, is real. If\nquantity, though not feeling, is yet real, that its parts should not\nbe feelings can be nothing against their reality. This once admitted,\nthe objections to infinite divisibility disappear; but so likewise\ndoes that mysterious dignity supposed to attach to it, or to its\ncorrelative, the infinitely addible, as implying an infinite capacity\nin the mind. From Hume\u2019s point of view, the mind being \u2018a bundle\nof impressions\u2019--though how impressions, being successive, should\nform a bundle is not explained--its capacity must mean the number of\nits impressions, and, all divisibility being into impressions, it\nfollows that infinite divisibility means an infinite capacity in the\nmind. This notion however arises, as we have shown, from a confusion\nbetween a _felt_ division of an impossible \u2018compound feeling,\u2019 and\nthat conceived divisibility of an object which constitutes but a\nsingle attribute of the object and represents a single relation of\nthe mind towards it. There may be a sense in which all conception\nimplies infinity in the conceiving mind, but so far from this\ndoing so in any special way, it arises, as we have seen, from the\npresentation of objects under that very condition of endless,\nunremoved, distinction which constitutes the true limitation of our\nthought.\n\nWhat are the ultimate elements of extension? If not extended, what\nare they?\n\n269. When, as with Hume, it is only in its application to space and\ntime that the question of infinite divisibility is treated, its true\nnature is more easily disguised, for the reason already indicated,\nthat space and time are not necessarily considered as quanta. When\nHume, indeed, speaks of space as a \u2018composition of parts\u2019 or \u2018made\nup of points,\u2019 he is of course treating it as a quantum; but we\nshall find that in seeking to avoid the necessary consequence of\nits being a quantum--the consequence, namely, that it is infinitely\ndivisible--he can take advantage of the possibility of treating\nit as the simple, unquantified, relation of externality. We have\nalready spoken of the dexterity with which, having shown that all\ndivisibility, because into impressions, is into simple parts, he\nturns this into an argument in favour of the composition of space\nby impressions. \u2018Our idea of space is compounded of parts which are\nindivisible.\u2019 Let us take one of these parts, then, and ask what sort\nof idea it is: \u2018let us form a judgment of its nature and qualities.\u2019\n\u2018\u2019Tis plain it is not an idea of extension: for the idea of extension\nconsists of parts; and this idea, according to the supposition, is\nperfectly simple and indivisible. Is it therefore nothing? That is\nimpossible,\u2019 for it would imply that a real idea was composed of\nnonentities. The way out of the difficulty is to \u2018endow the simple\nparts with colour and solidity.\u2019 In words already quoted, \u2018that\ncompound impression, which represents extension, consists of several\nlesser impressions, that are indivisible to the eye or feeling, and\nmay be called impressions of atoms or corpuscles endowed with colour\nand solidity.\u2019 (Part II. \u00a7 3, near the end.)\n\nColours or coloured points? What is the difference?\n\n270. It is very plain that in this passage Hume is riding two horses\nat once. He is trying so to combine the notion of the constitution\nof space by impressions with that of its constitution by points, as\nto disguise the real meaning of each. In what lies the difference\nbetween the feelings of colour, of which we have shown that they\ncannot without contradiction be supposed to \u2018make up extension,\u2019\nand \u2018coloured points or corpuscles\u2019? Unless the points, as points,\nmean something, the substitution of coloured points for colours\nmeans nothing. But according to Hume the point is nothing except as\nan impression of sight or touch. If then we refuse his words the\nbenefit of an interpretation which his doctrine excludes, we find\nthat there remains simply the impossible supposition that space\nconsists of feelings. This result cannot be avoided, unless in\nspeaking of space as composed of points, we understand by the point\nthat which is definitely other than an impression. Thus the question\nwhich Hume puts--If extension is made up of parts, and these, being\nindivisible, are unextended, what are they?--really remains untouched\nby his ostensible answer. Such a question indeed to a philosophy like\nLocke\u2019s, which, ignoring the constitution of reality by relations,\nsupposed real things to be first found and then relations to be\nsuperinduced by the mind--much more to one like Hume\u2019s, which left no\nmind to superinduce them--was necessarily unanswerable.\n\nTrue way of dealing with the question.\n\n271. In truth, extension is the relation of mutual externality. The\nconstituents of this relation have not, as such, any nature but what\nis given by the relation. If in Hume\u2019s language we \u2018separate each\nfrom the others and, considering it apart, form a judgment of its\nnature and qualities,\u2019 by the very way we put the problem we render\nit insoluble or, more properly, destroy it; for, thus separated,\nthey have no nature. It is this that we express by the proposition\nwhich would otherwise be tautological, that extension is a relation\nbetween extended points. The \u2018points\u2019 are the simplest expression\nfor those coefficients to the relation of mutual externality, which,\nas determined by that relation and no otherwise, have themselves the\nattribute of being extended and that only. If it is asked whether\nthe points, being extended, are therefore divisible, the answer must\nbe twofold. _Separately_ they are not divisible, for separately they\nare nothing. Whether, as determined by mutual relation, they are\ndivisible or no, depends on whether they are treated as forming a\nquantum or no. If they are not so treated, we cannot with propriety\npronounce them to be either further divisible or not so, for the\nquestion of divisibility has no application to them. But being\nperfectly homogeneous with each other and with that which together\nthey constitute, they are susceptible of being so treated, and are\nso treated when, with Hume in the passage before us, we speak of\nthem as the parts of which extended matter consists. Thus considered\nas parts of a quantum and therefore themselves quanta, the infinite\ndivisibility which belongs to all quantity belongs also to them.\n\n\u2018If the point were divisible, it would be no termination of a line.\u2019\nAnswer to this.\n\n272. In this lies the answer to the most really cogent argument which\nHume offers against infinite divisibility \u2018A surface terminates\na solid; a line terminates a surface; a point terminates a line:\nbut I assert that if the _ideas_ of a point, line, or surface were\nnot indivisible, \u2019tis impossible we should ever conceive these\nterminations. For let these ideas be supposed infinitely divisible,\nand then let the fancy endeavour to fix itself on the idea of the\nlast surface, line, or point, it immediately finds this idea to\nbreak into parts; and upon its seizing the last of these parts it\nloses its hold by a new division, and so on _ad infinitum_, without\nany possibility of its arriving at a concluding idea\u2019. [1] If\n\u2018point,\u2019 \u2018line,\u2019 or \u2018surface\u2019 were really names for \u2018ideas\u2019 either in\nHume\u2019s sense, as feelings grown fainter, or in Locke\u2019s, as definite\nimprints made by outward things, this passage would be perplexing.\nIn truth they represent objects determined by certain conceived\nrelations, and the relation under which the object is considered\nmay vary without a corresponding variation in the name. When a\n\u2018point\u2019 is considered simply as the \u2018termination of a line,\u2019 it is\nnot considered as a quantum. It represents the abstraction of the\nrelation of externality, as existing between _two lines_. It is these\nlines, not the point, that in this case are the constituents of the\nrelation, and thus it is they alone that are for the time considered\nas extended, therefore as quanta, therefore as divisible. So when the\nline in turn is considered as the \u2018termination of a surface.\u2019 It then\nrepresents the relation of externality _as between surfaces_, and\nfor the time it is the surfaces, not the line, that are considered\nto have extension and its consequences. The same applies to the view\nof a surface as the termination of a solid. Just as the line, though\nnot a quantum when considered simply as a relation between surfaces,\nbecomes so when considered in relation to another line, so the point,\nthough it \u2018has no magnitude\u2019 when considered as the termination of\na line, yet acquires parts, or becomes divisible, so soon as it is\nconsidered in relation to other points as a constituent of extended\nmatter; and it is thus that Hume considers it, \u1f11\u03ba\u1f7c\u03bd \u1f22 \u1f04\u03ba\u03c9\u03bd [2], when\nhe talks of extension as \u2018made up of coloured points.\u2019\n\n[1] P. 345. [Book I, part II., sec. IV.]\n\n[2] [Greek \u1f11\u03ba\u1f7c\u03bd \u1f22 \u1f04\u03ba\u03c9\u03bd (hekon e akon) = like it or not. Tr.]\n\nWhat becomes of the exactness of mathematics according to Hume?\n\n273. It is the necessity then, according to his theory, of making\nspace an impression that throughout underlies Hume\u2019s argument against\nits infinite divisibility; and, as we have seen, the same theory\nwhich excludes its infinite divisibility logically extinguishes it as\na quantity, divisible and measurable, altogether. He of course does\nnot recognize this consequence. He is obliged indeed to admit that\nin regard to the proportions of \u2018greater, equal and less,\u2019 and the\nrelations of different parts of space to each other, no judgments\nof universality or exactness are possible. We may judge of them,\nhowever, he holds, with various approximations to exactness, whereas\nupon the supposition of infinite divisibility, as he ingeniously\nmakes out, we could not judge of them at all. He \u2018asks the\nmathematicians, what they mean when they say that one line or surface\nis equal to, or greater or less than, another.\u2019 If they \u2018maintain the\ncomposition of extension by indivisible points,\u2019 their answer, he\nsupposes, will be that \u2018lines or surfaces are equal when the numbers\nof points in each are equal.\u2019 This answer he reckons \u2018just,\u2019 but the\nstandard of equality given is entirely useless. \u2018For as the points\nwhich enter into the composition of any line or surface, whether\nperceived by the sight or touch, are so minute and so confounded with\neach other that \u2019tis utterly impossible for the mind to compute their\nnumber, such a computation will never afford us a standard by which\nwe may judge of proportions.\u2019 The opposite sect of mathematicians,\nhowever, are in worse case, having no standard of equality whatever\nto assign. \u2018For since, according to their hypothesis, the least as\nwell as greatest figures contain an infinite number of parts, and\nsince infinite numbers, properly speaking, can neither be equal\nnor unequal with respect to each other, the equality or inequality\nof any portion of space can never depend on any proportion in the\nnumber of their parts.\u2019 His own doctrine is \u2018that the only useful\nnotion of equality or inequality is derived from the whole united\nappearance, and the comparison of, particular objects.\u2019 The judgments\nthus derived are in many cases certain and infallible. \u2018When the\nmeasure of a yard and that of a foot are presented, the mind can no\nmore question that the first is longer than the second than it can\ndoubt of those principles which are most clear and self-evident.\u2019\nSuch judgments, however, though \u2018sometimes infallible, are not always\nso.\u2019 Upon a \u2018review and reflection\u2019 we often \u2018pronounce those objects\nequal which at first we esteemed unequal,\u2019 and vice vers\u00e2. Often\nalso \u2018we discover our error by a juxtaposition of the objects; or,\nwhere that is impracticable, by the use of some common and invariable\nmeasure which, being successively applied to each, informs us of\ntheir different proportions. And even this correction is susceptible\nof a new correction, and of different degrees of exactness, according\nto the nature of the instrument by which we measure the bodies, and\nthe care which we employ in the comparison.\u2019 [1]\n\n[1] Pp. 351-53. [Book I, part II., sec. IV.]\n\nThe universal propositions of geometry either untrue or unmeaning.\n\n274. Such indefinite approach to exactness is all that Hume can allow\nto the mathematician. But it is undoubtedly another and an absolute\nsort of exactness that the mathematician himself supposes when he\npronounces all right angles equal. Such perfect equality \u2018beyond what\nwe have instruments and art\u2019 to ascertain, Hume boldly calls a \u2018mere\nfiction of the mind, useless as well as incomprehensible\u2019. [1] Thus\nwhen the mathematician talks of certain angles as always equal, of\ncertain lines as never meeting, he is either making statements that\nare untrue or speaking of nonentities. If his \u2018lines\u2019 and \u2018angles\u2019\nmean ideas that we can possibly have, his universal propositions are\nuntrue; if they do not, according to Hume they can mean nothing.\nHe says, for instance, that \u2018two right lines cannot have a common\nsegment;\u2019 but of such ideas of right lines as we can possibly have\nthis is only true \u2018where the right lines incline upon each other with\na sensible angle.\u2019 [2] It is not true when they \u2018approach at the rate\nof an inch in 20 leagues.\u2019 According to the \u2018original standard of a\nright line,\u2019 which is \u2018nothing but a certain general appearance, \u2019tis\nevident right lines may be made to concur with each other\u2019. [3] Any\nother standard is a \u2018useless and incomprehensible fiction.\u2019 Strictly\nspeaking, according to Hume, we have it not, but only a tendency to\nsuppose that we have it arising from the progressive correction of\nour actual measurements. [4]\n\n[1] P. 353. [Book I, part II., sec. IV.]\n\n[2] Cf. Aristotle, Metaph. 998a, on a corresponding view ascribed to\nProtagoras.\n\n[3] P. 356. [Book I, part II., sec. IV.]\n\n[4] P. 354. [Book I, part II., sec. IV.]\n\nDistinction between Hume\u2019s doctrine and that of the hypothetical\nnature of mathematics.\n\n275. Now it is obvious that what Hume accounts for by means of\nthis tendency to feign, even if the tendency did not presuppose\nconditions incompatible with his theory, is not mathematical\nscience as it exists. It has even less appearance of being so\nthan (to anticipate) has that which is accounted for by those\npropensities to feign, which he substitutes for the ideas of cause\nand substance, of being natural science as it exists. In the latter\ncase, when the idea of necessary connexion has been disposed of,\nan impression of reflection can with some plausibility be made\nto do duty instead; but there is no impression of reflection in\nHume\u2019s sense of the word, no \u2018propensity,\u2019 that can be the subject\nof mathematical reasoning. He speaks, indeed, of our _supposing_\nsome imaginary standard--of our having \u2018an obscure and implicit\nnotion\u2019--of perfect equality, but such language is only a way of\nsaving appearances; for according to him, a \u2018supposition\u2019 or \u2018notion\u2019\nwhich is neither impression nor idea, cannot be anything. A hasty\nreader, catching at the term \u2018supposition,\u2019 may find his statement\nplausible with all the plausibility of the modern doctrine, which\naccounts for the universality and exactness of mathematical truths\nas \u2018hypothetical\u2019--the doctrine that we suppose figures exactly\ncorresponding to our definitions, though such do not really exist.\nWith those who take this view, however, it is always understood that\nthe definitions represent ideas, though not ideas to which real\nobjects can be found exactly answering. Perhaps, if pressed about\ntheir distinction between idea and reality, they might find it hard\nconsistently to maintain it, but it is by this practically that they\nkeep their theory afloat. Hume can admit no such distinction. The\nreal with him is the impression, and the idea the fainter impression.\nThere can be no idea of a straight line, a curve, a circle, a right\nangle, a plane, other than the impression, other than the \u2018appearance\nto the eye,\u2019 and there are no appearances exactly answering to the\nmathematical definitions. If they do not _exactly_ answer, they might\nas well for the purposes of mathematical demonstration not answer\nat all. The Geometrician, having found that the angles at the base\nof _this_ isosceles triangle are equal to each other, at once takes\nthe equality to be true of all isosceles triangles, as being exactly\nlike the original one, and on the strength of this establishes many\nother propositions. But, according to Hume, no idea that we could\nhave would be one of which the sides were precisely equal. The Fifth\nProposition of Euclid then is not precisely true of the particular\nidea that we have before us when we follow the demonstration. Much\nless can it be true of the ideas, _i.e._ the several appearances\nof colour, indefinitely varying from this, which we have before us\nwhen we follow the other demonstrations in which the equality of the\nangles at the base of an isosceles is taken for granted.\n\nThe admission that no relations of quantity are data of sense removes\ndifficulty as to general propositions about them.\n\n276. Here, as elsewhere, what we have to lament is not that Hume\n\u2018pushed his doctrine too far,\u2019 so far as to exclude ideas of those\nexact proportions in space with which geometry purports to deal,\nbut that he did not carry it far enough to see that it excluded all\nideas of quantitative relations whatever. He thus pays the penalty\nfor his equivocation between a feeling of colour and a disposition\nof coloured points. Even alongside of his admission that \u2018relations\nof space and time\u2019 are independent of the nature of the ideas so\nrelated, which amounts to the admission that of space and time there\nare no ideas at all in his sense of the word, he allows himself to\ntreat \u2018proportions between spaces\u2019 as depending entirely on our\nideas of the spaces--depending on ideas which in the context he\nby implication admits that we have not. [1] If, instead of thus\nequivocating, he had asked himself how sensations of colour and touch\ncould be added or divided, how one could serve as a measure of the\nsize of another, he might have seen that only in virtue of that in\nthe \u2018general appearance\u2019 of objects which, in his own language, is\n\u2018independent of the nature of the ideas themselves\u2019--_i.e._ which\ndoes not belong to them as feelings, but is added by the comparing\nand combining thought--are the proportions of greater, less, and\nequal predicable of them at all; that what thought has thus added,\nviz. limitation by mutual externality, it can abstract; and that by\nsuch abstraction of the limit it obtains those several terminations,\nas Hume well calls them--the surface terminating bodies, the line\nterminating surfaces, the point terminating lines--from which it\nconstructs the world of pure space: that thus the same action of\nthought in sense, which alone renders appearances measurable, gives\nan object matter which, because the pure construction of thought, we\ncan measure exactly and with the certainty that the judgment based on\na comparison of magnitudes in a single case is true of all possible\ncases, because in none of these can any other conditions be present\nthan those which we have consciously put there.\n\n[1] Part III. \u00a7 1, sub init.\n\nHume does virtually admit this in regard to numbers.\n\n277. To have arrived at this conclusion Hume had only to extend to\nproportions in space the principle upon which the impossibility of\nsensualizing arithmetic compels him to deal with proportions in\nnumber. \u2018We are possessed,\u2019 he says, \u2018of a precise standard by which\nwe can judge of the equality and proportion of numbers; and according\nas they correspond or not to that standard we determine their\nrelations without any possibility of error. When two numbers are so\ncombined, as that the one has always an unite answering to every\nunite of the other, we pronounce them equal\u2019. [1] Now what are the\nunites here spoken of? If they were those single impressions which he\nelsewhere [2] seems to regard as alone properly unites, the point of\nthe passage would be gone, for combinations of such unites could at\nany rate only yield those \u2018general appearances\u2019 of whose proportions\nwe have been previously told there can be no precise standard. They\ncan be no other than those unites which, not being impressions, he\nhas to call \u2018fictitious denominations\u2019--unites which are nothing\nexcept in relation to each other and of which each, being in turn\ndivisible, is itself a true number. We can easily retort upon Hume,\nthen, when he argues that the supposition of infinite divisibility\nis incompatible with any comparison of quantities because with any\nunite of measurement, that, according to his own virtual admission,\nin the only case where such comparison is exact the ultimate unite\nof measurement is still itself divisible; which, indeed, is no\nmore than saying that whatever measures quantity must itself be a\nquantity, and that therefore quantity is infinitely divisible. If\nHume, instead of slurring over this characteristic of the science of\nnumber, had set himself to explain it, he would have found that the\nonly possible explanation of it was one equally applicable to the\nscience of space--that what is true of the unite, as the abstraction\nof distinctness, is true also of the abstraction of externality. As\nthe unite, because constituted by relation to other unites, so soon\nas considered breaks into multiplicity, and only for that reason is\na quantity by which other quantities can be measured; so is it also\nwith the limit in whatever form abstracted, whether as point, line,\nor surface. If the fact that number can have no least part since each\npart is itself a number or nothing, so far from being incompatible\nwith the finiteness of number, is the consequence of that finiteness,\nneither can the like attribute in spaces be incompatible with their\nbeing definite magnitudes, that can be compared with and measured\nby each other. The real difference, which is also the rationale of\nHume\u2019s different procedure in the two cases, is that the conception\nof space is more easily confused than that of number with the\nfeelings to which it is applied, and which through such application\nbecome sensible spaces. Hence the liability to the supposition,\nwhich is at bottom Hume\u2019s, that the last feeling in the process of\ndiminution before such sensible space disappears (being the \u2018minimum\nvisibile\u2019) is the least possible portion of space.\n\n[1] P. 374. [Book I, part III., sec. I.]\n\n[2] Above, par. 258.\n\nWith Hume idea of vacuum impossible, but logically not more so than\nthat of space.\n\n278. Just as that reduction of consciousness to feeling, which\nreally excludes the idea of quantity altogether, is by Hume only\nrecognised as incompatible with its infinite divisibility, so it is\nnot recognised as extinguishing space altogether, but only space as\na vacuum. If it be true, he says, \u2018that the idea of space is nothing\nbut the idea of visible or tangible points distributed in a certain\norder, it follows that we can form no idea of vacuum, or space where\nthere is nothing visible or tangible\u2019. [1] Here as elsewhere the\nacceptability of his statement lies in its being taken in a sense\nwhich according to his principles cannot properly belong to it. It\nis one doctrine that the ideas of space and body are essentially\ncorrelative, and quite another that the idea of space is equivalent\nto a feeling of sight or touch. It is of the latter doctrine that\nHume\u2019s denial of a vacuum is the corollary; but it is the former\nthat gains acceptance for this denial in the mind of his reader.\nSpace we have already spoken of as the relation of externality. If,\nabstracting this relation from the world of which it is the uniform\nbut most elementary determination, we regard it as a relation between\nobjects having no other determination, these become spaces and\nnothing but spaces--space pure and simple, _vacuum_. But we have\nknown the world in confused fulness before we detach its constituent\nrelations in the clearness of unreal abstraction. We have known\nbodies \u03c3\u03c5\u03b3\u03ba\u03b5\u03c7\u03c5\u03bc\u03ad\u03bd\u03bf\u03c2 [2], before we think their limits apart and out\nof these construct a world of pure space. It is thus in a sense true\nthat in the development of our consciousness an idea of body precedes\nthat of space, though the _abstraction_ of space--the detachment of\nthe relation so-called from the real complex of relations--precedes\nthat of body; and it is this fact that, in the face of geometry,\nstrengthens common sense in its position that an idea of vacuum is\nimpossible. It is not, however, the inseparability of space from body\nwhether in reality or for our consciousness, but its identity with a\ncertain sort of feeling, that is implied in Hume\u2019s exclusion of the\nidea of vacuum. \u2018Body,\u2019 as other than feeling, is with him as much a\nfiction as vacuum. That there can be no idea of vacuum, is thus in\nfact merely his negative way of putting that proposition of which the\npositive form is, that space is a compound impression of sight and\ntouch. Having examined that proposition in the positive, we need not\nexamine it again in the negative form. It will be more to the purpose\nto enquire whether the \u2018tendency to suppose\u2019 or \u2018propensity to feign\u2019\nby which, in the absence of any such idea, our language about \u2018pure\nspace\u2019 has to be accounted for, does not according to Hume\u2019s own\nshowing presuppose such an idea.\n\n[1] P. 358. [Book I, part II., sec. V.]\n\n[2] [Greek \u03c3\u03c5\u03b3\u03ba\u03b5\u03c7\u03c5\u03bc\u03ad\u03bd\u03bf\u03c2 (synkechymenos) = confused or jumbled-up. Tr.]\n\nHow it is that we talk as if we had idea of vacuum according to Hume.\n\n279. By vacuum he understands invisible and intangible extension.\nIf an idea of vacuum, then, is possible at all, he argues, it must\nbe possible for darkness and mere motion to convey it. That they\ncannot do so _alone_ is clear from the consideration that darkness\nis \u2018no positive idea\u2019 and that an \u2018invariable motion,\u2019 such as that\nof a \u2018man supported in the air and softly conveyed along by some\ninvisible power,\u2019 gives no idea at all. Neither can they do so when\n\u2018attended with visible and tangible objects.\u2019 \u2018When two bodies\npresent themselves where there was formerly an entire darkness, the\nonly change that is discoverable is in the appearance of these two\nobjects: all the rest continues to be, as before, a perfect negation\nof light and of every coloured or tangible object\u2019. [1] \u2018Such dark\nand indistinguishable distance between two bodies can never produce\nthe idea of extension,\u2019 any more than blindness can. Neither can a\nlike \u2018imaginary distance between tangible and solid bodies.\u2019 \u2018Suppose\ntwo cases, viz. that of a man supported in the air, and moving his\nlimbs to and fro without meeting anything tangible; and that of a\nman who, feeling something tangible, leaves it, and after a motion\nof which he is sensible perceives another tangible object. Wherein\nconsists the difference between these two cases? No one will scruple\nto affirm that it consists merely in the perceiving those objects,\nand that the sensation which arises from the motion is in both cases\nthe same; and as that sensation is not capable of conveying to us an\nidea of extension, when unaccompanied with some other perception,\nit can no more give us that idea, when mixed with the impressions\nof tangible objects, since that mixture produces no alteration upon\nit\u2019. [2] But though a \u2018distance not filled with any coloured or\nsolid object\u2019 cannot give us an idea of vacuum, it is the cause why\nwe falsely imagine that we can form such an idea. There are \u2018three\nrelations\u2019--_natural_ relations according to Hume\u2019s phraseology\n[3]--between it and that distance which really \u2018conveys the idea\nof extension.\u2019 \u2018The distant objects affect the senses in the same\nmanner, whether separated by the one distance or the other; the\nformer species of distance is found capable of receiving the latter;\nand they both equally diminish the force of every quality. These\nrelations betwixt the two kinds of distance will afford us an easy\nreason why the one has so often been taken for the other, and why we\nimagine we have an idea of extension without the idea of any object\neither of the sight or feeling\u2019. [4]\n\n[1] P. 362. [Book I, part II., sec. V.]\n\n[2] P. 363. [Book I, part II., sec. V.]\n\n[3] Above, \u00a7 206.\n\n[4] P. 364. [Book I, part II., sec. V.]\n\nHis explanation implies that we have an idea virtually the same.\n\n280. It appears then that we have an idea of \u2018distance unfilled\nwith any coloured or solid object.\u2019 To speak of this distance as\n\u2018imaginary\u2019 or fictitious can according to Hume\u2019s principles make no\ndifference, so long as he admits, which he is obliged to do, that we\nactually have an idea of it; for every idea, being derived from an\nimpression, is as much or as little imaginary as every other. And\nnot only have we such an idea, but Hume\u2019s account of the \u2018relations\u2019\nbetween it and the idea of extension implies that, _as ideas of\ndistance_, they do not differ at all. But the idea of \u2018distance\nunfilled with any coloured or solid object\u2019 _is_ the idea of vacuum.\nIt follows that the idea of extension does not differ from that of\nvacuum, except so far as it is other than the idea of distance. But\nit is from the consideration of distance that Hume himself expressly\nderives it; [1] and so derived, it can no more differ from distance\nthan an idea from a corresponding impression. Thus, after all, he has\nto all intents and purposes to admit the idea of vacuum, but saves\nappearances by refusing to call it extension--the sole reason for\nsuch refusal being the supposition that every idea, and therefore the\nidea of extension, must be a datum of sense, which the admission of\nan idea of \u2018invisible and intangible distance\u2019 already contradicts.\n\n[1] Part II. \u00a7 3, sub. inst.\n\nBy a like device that he is able to explain the appearance of our\nhaving such ideas as Causation and Identity.\n\n281. We now know the nature of that preliminary manipulation which\n\u2018impressions and ideas\u2019 have to undergo, if their association is to\nyield the result which Hume requires--if through it the succession\nof feelings is to become a knowledge of things and their relations.\nSuch a result was required as the only means of maintaining together\nthe two characteristic positions of Locke\u2019s philosophy; that,\nnamely, the only world we can know is the world of \u2018ideas,\u2019 and\nthat thought cannot originate ideas. Those relations, which Locke\nhad inconsistently treated at once as intellectual superinductions\nand as ultimate conditions of reality, must be dealt with by one\nof two methods. They must be reduced to impressions where that\ncould plausibly be done: where it could not, it must be admitted\nthat we have no ideas of them, but only \u2018tendencies to suppose\u2019\nthat we have such, arising from the association, through \u2018natural\nrelations,\u2019 of the ideas that we have. So dexterously does Hume work\nthe former method that, of all the \u2018philosophical relations\u2019 which\nhe recognizes, only Identity and Causation remain to be disposed of\nby the latter; and if the other relations--resemblance, time and\nspace, proportion in quantity and degree in quality--could really\nbe admitted as data of sense, there would at least be a possible\nbasis for those \u2018tendencies to suppose\u2019 which, in the absence of any\ncorresponding ideas, the terms \u2018Identity\u2019 and \u2018Causation\u2019 must be\ntaken to represent. But, as we have shown, they can only be claimed\nfor sense, if sense is so far one with thought--one not by conversion\nof thought into sense but by taking of sense into thought--as that\nHume\u2019s favourite appeals to sense against the reality of intelligible\nrelations become unmeaning. They may be \u2018impressions,\u2019 there may\nbe \u2018impressions of them,\u2019 but only if we deny of the impression\nwhat Hume asserts of it, and assert of it what he denies--only\nif we understand by \u2018impression\u2019 not an \u2018internal and perishing\nexistence;\u2019 not that which, if other than taste, colour, sound, smell\nor touch, must be a \u2018passion or emotion \u2018; _not_ that which carries\nno reference to an object other than itself, and which must _either_\nbe single _or_ compound; but something permanent and constituted\nby permanently coexisting parts; something that may \u2018be conjoined\nwith\u2019 any feeling, because it is none; that always carries with it a\nreference to a subject which it is not but of which it is a quality;\nand that is both many and one, since \u2018in its simplicity it contains\nmany different resemblances and relations.\u2019\n\n282. In the account just adduced of vacuum, the effect of that double\ndealing with \u2018impressions,\u2019 which we shall have to trace at large in\nHume\u2019s explanation of our language about Causation and Identity, is\nalready exhibited in little. Just as, after the idea of pure space\nhas been excluded because not a copy of any possible impression, we\nyet find an \u2018idea,\u2019 only differing from it in name, introduced as\nthe basis of that tendency to suppose which is to take the place of\nthe excluded idea, so we shall find ideas of relation in the way\nof Identity and Causation--ideas which according to Hume we have\nnot--presupposed as the source of those \u2018propensities to feign\u2019 which\nhe accounts for the appearance of our having them.\n\nKnowledge of relation in way of Identity and Causation excluded by\nLocke\u2019s definition of knowledge.\n\n283. The primary characteristic of these relations according to Hume,\nwhich they share with those of space and time, and which in fact\nvitiates that definition of \u2018philosophical relation,\u2019 as depending\non comparison, which he adopts, is that they \u2018depend not on the\nideas compared together, but may be changed without any change in\nthe ideas\u2019. [1] It follows that they are not objects of knowledge,\naccording to the definition of knowledge which Hume inherited,\nas \u2018the perception of agreement or disagreement between ideas.\u2019\nA partial recognition of this consequence in regard to cause and\neffect we found in Locke\u2019s suspicion that a science of nature was\nimpossible--impossible because, however often a certain \u2018idea of\nquality and substance\u2019 may have followed or accompanied another, such\nsequence or accompaniment never amounts to agreement or \u2018necessary\nconnexion\u2019 between the ideas, and therefore never can warrant a\ngeneral assertion, but only the particular one, that the ideas in\nquestion have so many times occurred in such an order. \u2018Matters of\nfact,\u2019 however, which no more consist in agreement of ideas than\ndoes causation, are by Locke treated without scruple as matter of\nknowledge when they can be regarded as relations between present\nsensations. Thus the \u2018particular experiment\u2019 in Physics constitutes\nknowledge--the knowledge, for instance, that a piece of gold is now\ndissolved in aqua regia; and when \u2018I myself see a man walk on the\nice, it is knowledge.\u2019 In such cases it does not occur to him to ask,\neither what are the ideas that agree or how much of the experiment is\na present sensation. [2] Nor does Hume commonly carry his analysis\nfurther. After admitting that the relations called \u2018identity and\nsituation in time and place\u2019 do not depend on the nature of the\nideas related, he proceeds: \u2018When both the objects are present to\nthe senses along with the relation, we call _this_ perception rather\nthan reasoning; nor is there in this case any exercise of the thought\nor any action, properly speaking, but a mere passive admission of\nthe impressions through the organs of sensation. According to this\nway of thinking, we ought not to receive as reasoning any of the\nobservations we may make concerning _identity_ and the _relations_ of\n_time_ and _place_; since in none of them the mind can go beyond what\nis immediately present to the senses, either to discover the real\nexistence or the relations of objects\u2019. [3]\n\n[1] P. 372. [Book I, part III., sec. I.]\n\n[2] Above, \u00a7\u00a7 122 & 123.\n\n[3] P. 376. [Book I, part III., sec. II.]\n\nInference a transition from an object perceived or remembered to one\nthat is not so.\n\n284. This passage points out the way which Hume\u2019s doctrine of\ncausation was to follow. That in any case \u2018the mind should go beyond\na present feeling, either to discover the real existence or the\nrelations of objects\u2019 other than present feelings, was what he could\nnot consistently admit. In the judgment of causation, however, it\nseems to do so. \u2018From the existence or action of one object,\u2019 seen\nor remembered, it seems to be assured of the existence or action\nof another, not seen or remembered, on the ground of a necessary\nconnection between the two. [1] It is such assurance that is reckoned\nto constitute reasoning in the distinctive sense of the term, as\ndifferent at once from the analysis of complex ideas and the simple\nsuccession of ideas--such reasoning as, in the language of a later\nphilosophy, can yield synthetic propositions. What Hume has to do,\nthen, is to explain this \u2018assurance\u2019 away by showing that it is\nnot essentially different from that judgment of relation in time\nand place which, because the related objects are \u2018present to the\nsenses along with the relation,\u2019 is called \u2018perception rather than\nreasoning,\u2019 and to which no \u2018exercise of the thought\u2019 is necessary,\nbut a \u2018mere passive admission of impressions through the organs of\nsensation.\u2019 Nor, for the assimilation of reasoning to perception,\nis anything further needed than a reference to the connection\nof ideas with impressions and of the ideas of imagination with\nthose of memory, as originally stated by Hume. When both of the\nobjects compared are present to the senses, we call the comparison\nperception; when neither, or only one, is so present, we call it\nreasoning. But the difference between the object that is present\nto sense, and that which is not, is merely the difference between\nimpression and idea, which again is merely the difference between\nthe more and the less lively feeling. [2] To feeling, whether with\nmore or with less vivacity, every object, whether of perception\nor reasoning, must alike be present. Is it then a sufficient\naccount of the matter, according to Hume, to say that when we are\nconscious of contiguity and succession between objects of which\nboth are impressions we call it perception; but that when both\nobjects are ideas, or one an impression and the other an idea, we\ncall it reasoning? Not quite so. Suppose that I \u2018have seen that\nspecies of object we call flame, and have afterwards felt that\nspecies of sensation we call heat.\u2019 If I afterwards remembered the\nsuccession of the feeling upon the sight, both objects (according\nto Hume\u2019s original usage of terms [3]) would be ideas as distinct\nfrom the impressions; or, if upon seeing the flame I remembered the\nprevious experience of heat, one object would be an idea; but we\nshould not reckon it a case of reasoning. \u2018In all cases wherein we\nreason concerning objects, there is only one either perceived or\n_remembered_, and the other is supplied in conformity to our past\nexperience\u2019--supplied by the only other faculty than memory that can\n\u2018supply an idea,\u2019 viz. imagination. [4]\n\n[1] Pp. 376, 384. [Book I, part III., secs. II. and IV.]\n\n[2] Pp. 327, 375. [Book I, part I., sec. VII. and part III., sec.\nIII.]\n\n[3] Above, par. 195.\n\n[4] Pp. 384, 388. [Book I, part III., secs. IV. and V.]\n\nRelation of cause and effect the same as this transition.\n\n285. This being the only account of \u2018inference from the known to\nthe unknown,\u2019 which Hume could consistently admit, his view of the\nrelation of cause and effect must be adjusted to it. It could not\nbe other than a relation either between impression and impression,\nor between impression and idea, or between idea and idea; and all\nthese relations are equally between feelings that we experience.\nThus, instead of being the \u2018objective basis\u2019 on which inference from\nthe known to the unknown rests, it is itself the inference; or, more\nproperly, it and the inference alike disappear into a particular\nsort of transition from feeling to feeling. The problem, then, is to\naccount for its seeming to be other than this. \u2018There is nothing in\nany objects to persuade us that they are always _remote_ or always\n_contiguous_; and when from experience and observation we discover\nthat the relation in this particular is invariable, we always\nconclude that there is some secret _cause_ which separates or unites\nthem\u2019. [1] It would seem, then, that the relation of cause and effect\nis something which we infer from experience, from the connection of\nimpressions and ideas, but which is not itself impression or idea.\nAnd it would _seem_ further, that, as we infer such an unexperienced\nrelation, so likewise we make inferences from it. In regard to\nidentity \u2018we readily suppose an object may continue individually the\nsame, though several times absent from and present to the senses;\nand ascribe to it an identity, notwithstanding the interruption of\nthe perception, whenever we conclude that if we had kept our hand\nor eye constantly upon it, it would have conveyed an invariable and\nuninterrupted perception. But this conclusion beyond the impressions\nof our senses can be founded only on the connection of _cause and\neffect_; nor can we otherwise have any security that the object is\nnot changed upon us, however much the new object may resemble that\nwhich was formerly present to the senses.\u2019\n\n[1] P. 376. [Book I, part III., sec. II.]\n\nYet seems other than this. How this appearance is to be explained.\n\n286. This relation which, going beyond our actual experience, we seem\nto infer as the explanation of invariable contiguity in place or time\nof certain impressions, and from which again we seem to infer the\nidentity of an object of which the perception has been interrupted,\nis what we call necessary connection. It is their supposed necessary\nconnection which distinguishes objects related as cause and effect\nfrom those related merely in the way of contiguity and succession,\n[1] and it is a like supposition that leads us to infer what we do\nnot see or remember from what we do. If then the reduction of thought\nand the intelligible world to feeling was to be made good, this\nsupposition, not being an impression of sense or a copy of such, must\nbe shown to be an \u2018impression of reflection,\u2019 according to Hume\u2019s\nsense of the term, _i.e._ a tendency of the soul, analogous to desire\nand aversion, hope and fear, derived from impressions of sense but\nnot copied from them; [2] and the inference which it determines\nmust be shown to be the work of imagination, as affected by such\nimpression of reflection. This in brief is the purport of Hume\u2019s\ndoctrine of causation.\n\n[1] P. 376. [Book I, part III., sec. II.]\n\n[2] Above, par. 195.\n\nInference, resting on supposition of necessary connection, to be\nexplained before that connection.\n\n287. After his manner, however, he will go about with his reader. The\nsupposed \u2018objective basis\u2019 of knowledge is to be made to disappear,\nbut in such a way that no one shall miss it. So dexterously, indeed,\nis this done, that perhaps to this day the ordinary student of Hume\nis scarcely conscious of the disappearance. Hume merely announces to\nbegin with that he will \u2018postpone the direct survey of this question\nconcerning the nature of necessary connection,\u2019 and deal first with\nthese other two questions, viz. (1) \u2018For what reason we pronounce it\n_necessary_ that everything whose existence has a beginning, should\nalso have a cause?\u2019 and (2) \u2018Why we conclude that such particular\ncauses must _necessarily_ have such particular effects; and what is\nthe nature of that _inference_ we draw from the one to the other, and\nof the _belief_ we repose in it?\u2019 That is to say, he will consider\nthe inference from cause or effect, before he considers cause and\neffect as a relation between objects, on which the inference is\nsupposed to depend. Meanwhile necessary connection, as a relation\nbetween objects, is naturally supposed in some sense or other to\nsurvive. In _what_ sense, the reader expects to find when these two\npreliminary questions have been answered. But when they have been\nanswered, necessary connection, as a relation between objects, turns\nout to have vanished.\n\nAccount of the inference given by Locke and Clarke rejected.\n\n288. With the first of the above questions Hume only concerns\nhimself so far as to show that we cannot know either intuitively or\ndemonstratively, in Locke\u2019s sense of the words, that \u2018everything\nwhose existence has a beginning also has a cause.\u2019 Locke\u2019s own\nargument for the necessity of causation--that \u2018something cannot be\nproduced by nothing\u2019--as well as Clarke\u2019s--that \u2018if anything wanted a\ncause it would produce itself, _i.e._ exist before it existed\u2019--are\nmerely different ways, as Hume shows, of assuming the point in\nquestion. \u2018If everything must have a cause, it follows that upon\nexclusion of other causes we must accept of the object itself, or\nof nothing, as causes. But \u2019tis the very point in question, whether\neverything must have a cause or not\u2019. [1] On that point, according\nto Locke\u2019s own showing, there can be no certainty, intuitive or\ndemonstrative; for between the idea of beginning to exist and the\nidea of cause there is clearly no agreement, mediate or immediate.\nThey are not similar feelings, they are not quantities that can be\nmeasured against each other, and to these alone can the definition\nof knowledge and reasoning, which Hume retained, apply. There thus\ndisappears that last remnant of \u2018knowledge\u2019 in regard to nature which\nLocke had allowed to survive--the knowledge that there is a necessary\nconnection, though one which we cannot find out. [2]\n\n[1] P. 382. [Book I, part III., sec. III.]\n\n[2] cf. Locke IV. 3, 29, and Introduc, par. 121.\n\nThree points to be explained in the inference according to Hume.\n\n289. Having thus shown, as he conceives, what the true answer to the\nfirst of the above questions is not, Hume proceeds to show what it\nis by answering the second. \u2018Since it is not from knowledge or any\nscientific reasoning that we derive the opinion of the necessity\nof a cause to every new production,\u2019 it must be from experience;\n[1] and every general opinion derived from experience is merely the\nsummary of a multitude of particular ones. Accordingly when it has\nbeen explained why we infer particular causes from particular effects\n(and _vice vers\u00e2_), the inference from every event to a cause will\nhave explained itself. Now \u2018all our arguments concerning causes\nand effects consist both of an impression of the memory or senses,\nand of the idea of that existence which produces the object of the\nimpression or is produced by it. Here, therefore, we have three\nthings to explain, viz. _first_, the original impression; _secondly_,\nthe transition to the idea of the connected cause or effect;\n_thirdly_, the nature and qualities of that idea.\u2019 [2]\n\n[1] P. 383. [Book I, part III., sec. III.]\n\n[2] P. 385. [Book I, part III., sec. V.]\n\n_a_. The original impression from which the transition is made, and\n_b_. The transition to inferred idea\n\n290. As to the original impression we must notice that there is a\ncertain inconsistency with Hume\u2019s previous usage of terms in speaking\nof an _impression_ of memory at all. [1] This, however, will be\nexcused when we reflect that according to him impression and idea\nonly differ in liveliness, and that he is consistent in claiming for\nthe ideas of memory, not indeed the maximum, but a high degree of\nvivacity, superior to that which belongs to ideas of imagination. All\nthat can be said, then, of that \u2018original impression,\u2019 whether of the\nmemory or senses, which is necessary to any \u2018reasoning from cause or\neffect,\u2019 is that it is highly vivacious. That the transition from it\nto the \u2018idea of the connected cause or effect\u2019 is not determined by\nreason, has already been settled. It could only be so determined,\naccording to the received account of reason, if there were some\nagreement in respect of quantity or quality between the idea of cause\nand that of the effect, to be ascertained by the interposition of\nother ideas. [2] But when we examine any particular objects that we\nhold to be related as cause and effect, _e.g._ the sight of flame and\nthe feeling of heat, we find no such agreement. What we _do_ find\nis their \u2018constant conjunction\u2019 in experience, and \u2018conjunction\u2019 is\nequivalent to that \u2018contiguity in time and place,\u2019 which has already\nbeen pointed out as one of those \u2018natural relations\u2019 which act as\n\u2018principles of union\u2019 between ideas. [3] Because the impression of\nflame has always been found to be followed by the impression of heat,\nthe idea of flame always suggests the idea of heat. It is simple\ncustom then that determines the transition from the one to the other,\nor renders \u2018necessary\u2019 the connection between them. In order that the\ntransition, however, may constitute an inference from cause to effect\n(or _vice vers\u00e2_), one of the two objects thus naturally related,\nbut not both, must be presented as an impression. If both were\nimpressions it would be a case of \u2018sensation, not reasoning;\u2019 if both\nwere ideas, no belief would attend the transition. This brings us to\nthe question as to the \u2018nature and qualities\u2019 of the inferred idea.\n\n[1] Above, par. 195.\n\n[2] Cf. Locke IV. 17, 2.\n\n[3] Above, par. 206.\n\n_c_. The qualities of this idea.\n\n291. \u2018\u2019Tis evident that all reasonings from causes or effects\nterminate in conclusions concerning matter of fact, _i.e._ concerning\nthe existence of objects or of their qualities\u2019; [1] in other words,\nin belief. If this meant a new idea, an idea that we have not\npreviously had, it would follow that inference could really carry us\nbeyond sense, that there could be an idea not copied from any prior\nimpression. But according to Hume it does not mean this. \u2018The idea\nof existence is the very same with the idea of what we conceive to\nbe existent;\u2019 [2] and not only so, \u2018the _belief_ of existence joins\nno new ideas to those which compose the idea of the object. When I\nthink of God, when I think of him as existent, and when I believe him\nto be existent, my idea of him neither increases nor diminishes\u2019.\n[3] In what then lies the difference between incredulity and belief;\nbetween an \u2018idea assented to,\u2019 or an object believed to exist, and\na fictitious object or idea from which we dissent? The answer is,\n\u2018not in the parts or composition of the idea, but in the manner\nof conceiving it,\u2019 which must be understood to mean the manner of\n\u2018feeling\u2019 it; and this difference is further explained to lie in \u2018the\nsuperior force, or vivacity, or steadiness\u2019 with which it is felt.\u2019\n[4] We are thus brought to the further question, how it is that this\n\u2018superior vivacity\u2019 belongs to the inferred idea when we \u2018reason\u2019\nfrom cause to effect or from effect to cause. The answer here is\nthat the \u2018impression of the memory or senses,\u2019 which in virtue of\na \u2018natural relation\u2019 suggests the idea, also \u2018communicates to it a\nshare of its force or vivacity.\u2019\n\n[1] P. 394. [Book I, part III., sec. VII.]\n\n[2] P. 370. [Book I, part II., sec. VI.]\n\n[3] P. 395. [Book I, part III., sec. II.]\n\n[4] P. 398 [Book I, part III., sec. VII.]. Cf. above, par. 170, for\nthe corresponding view in Berkeley.\n\nIt results that necessary connection is an impression of reflection,\n_i.e._, a propensity to the transition described.\n\n292. Thus it appears that in order to the conclusion that any\nparticular cause must have any particular effect, there is needed\nfirst the presence of an impression, and secondly the joint action\nof those two \u2018principles of union among ideas,\u2019 resemblance and\ncontiguity. In virtue of the former principle the given impression\ncalls up the image of a like impression previously experienced,\nwhich again in virtue of the latter calls up the image of its usual\nattendant, and the liveliness of the given impression so communicates\nitself to the recalled ideas as to constitute belief in their\nexistence. If this is the true account of the matter, the question\nas to the nature of necessary connexion has answered itself. \u2018The\nnecessary connexion betwixt causes and effects is the foundation of\nour inference from one to the other. The foundation of the inference\nis the transition arising from the accustomed union. These are\ntherefore the same\u2019. [1] We may thus understand how it is that there\nseems to be an idea of such connexion to which no impression of the\nsenses, or (to use an equivalent phrase of Hume\u2019s) no \u2018quality in\nobjects\u2019 corresponds. If the first presentation of two objects, of\nwhich one is cause, the other effect, (_i.e._ of which we afterwards\ncome to consider one the cause, the other the effect) gives no idea\nof a connexion between them, as it clearly does not, neither can\nit do so however often repeated. It would not do so, unless the\nrepetition \u2018either discovered or produced something new\u2019 in the\nobjects; and it does neither. But it does \u2018produce a new impression\nin the mind.\u2019 After observing a \u2018constant conjunction of the objects,\nand an uninterrupted resemblance of their relations of contiguity\nand succession, we immediately feel a determination of the mind to\npass from one of the objects to its usual attendant, and to conceive\nit in a stronger light on account of that relation.\u2019 It is of this\n\u2018internal impression,\u2019 this \u2018propensity which custom produces,\u2019 that\nthe idea of necessary connexion is the copy. [2]\n\n[1] P. 460. [Book I, part III., sec. XIV.]\n\n[2] Pp. 457-460. [Book I, part III., sec. XIV.]\n\nThe transition not to anything beyond sense.\n\n293. The sequence of ideas, which this propensity determines, clearly\ndoes not involve any inference \u2018beyond sense,\u2019 \u2018from the known to the\nunknown,\u2019 \u2018from instances of which we have had experience, to those\nof which we have had none,\u2019 any more than does any other \u2018recurrence\nof an idea\u2019--which, as we have seen, merely means, according to Hume,\nthe return of a feeling at a lower level of intensity after it has\nbeen felt at a higher. The idea which we speak of as an inferred\ncause or effect is only an \u2018instance of which we have no experience\u2019\nin the sense of being _numerically different_ from the similar\nideas, whose previous constant association with an impression like\nthe given one, determines the \u2018inference;\u2019 but in the same sense\nthe \u2018impression\u2019 which I now feel on putting my hand to the fire\nis different from the impressions previously felt under the same\ncircumstances, and I do not for that reason speak of this impression\nas an instance of which I have had no experience. Thus Hume, though\nretaining the received phraseology in reference to the \u2018conclusion\nfrom any particular cause to any particular effect\u2019--phraseology\nwhich implies that prior to the inference the object inferred is in\nsome sense unknown or unexperienced--yet deprives it of meaning by a\ndoctrine which makes inference, as he himself puts it, \u2018a species of\nsensation,\u2019 \u2018an unintelligible instinct of our souls,\u2019 \u2018more properly\nan act of the sensitive than of the cogitative part of our natures\u2019\n[1]--which in fact leaves no \u2018part of our natures\u2019 to be cogitative\nat all.\n\n[1] Pp. 404, 475, and 471. [Book I, part III., sec. VIII., part IV.\nsec. I. and part III., sec. XVI.]\n\nNor determined by any objective relation.\n\n294. We are not entitled then, it would seem, to say that any\ninference to matter of fact, any proof of an \u2018instructive\nproposition,\u2019--as distinct from the conclusion of a syllogism,\nwhich is simply derived from the analysis of a proposition already\nconceded,--rests on the relation of cause and effect. Such language\nimplies that the relation is other than the inference, whereas, in\nfact, they are one and the same, each being merely a particular\nsort of sequence of feeling upon feeling--that sort of which the\ncharacteristic is that, when the former feeling only has the maximum\nof vivacity, it still, owing to the frequency with which it has been\nattended by the other, imparts to it a large, though less, amount of\nvivacity. This is the naked result to which Hume\u2019s doctrine leads--a\nresult which, thus put, might have set men upon reconsidering the\nfirst principles of the Lockeian philosophy. But he wished to find\nacceptance, and would not so put it. A consideration of the points\nin which he had to sacrifice consistency to plausibility--since he\nwas always consistent where he decently could be--will lead us to the\ntrue \u03b1\u1f34\u03c4\u03b9\u03bf\u03bd \u03c4\u03bf\u1fe6 \u03c8\u03b5\u03c5\u03b4\u03bf\u1fe6\u03c2 [1], the impossibility on his principles of\nexplaining the world of knowledge.\n\n[1] [Greek \u03b1\u1f34\u03c4\u03b9\u03bf\u03bd \u03c4\u03bf\u1fe6 \u03c8\u03b5\u03c5\u03b4\u03bf\u1fe6\u03c2 (aition tou pseudous) = the cause of\nthe error. Tr.]\n\nDefinitions of cause: a. As a \u2018philosophical\u2019 relation.\n\n295. As the outcome of his doctrine, he submits two definitions\nof the relation of cause and effect. Considering it as \u2018a\n_philosophical_ relation or comparison of two ideas, we may define a\ncause to be an object precedent and contiguous to another, and where\nall objects resembling the former are placed in like relations of\nprecedency and contiguity to those objects that resemble the latter.\u2019\nConsidering the relation as \u2018a _natural_ one, or as an association\nbetween ideas,\u2019 we may say that \u2018a _cause_ is an object precedent and\ncontiguous to another, and so united with it that the idea of one\ndetermines the mind to form the idea of the other, and the impression\nof the one to form a more lively idea of the other\u2019. [1]\n\n[1] P. 464. [Book I, part III., sec. XIV.]\n\nIs Hume entitled to retain \u2018philosophical\u2019 relations as distinct from\n\u2018natural\u2019?\n\n296. Our first enquiry must be how far these definitions are really\nconsistent with the theory from which they are derived. At the\noutset, it is a surprise to find that the \u2018philosophical relation\u2019\nof cause and effect, as distinct from the natural one, should still\nappear to survive. Such a distinction has no meaning unless it\nimplies a conceived relation of objects other than the _de facto_\nsequence of feelings, of which one \u2018naturally\u2019 introduces the other.\nIt is the characteristic of Locke\u2019s doctrine of knowledge that in it\nthis distinction is still latent. His language constantly implies\nthat knowledge, as a perception of relations, is other than the\nsequence of feelings; but by confining his view chiefly to relation\nin the way of likeness and unlikeness--a relation that exists\nbetween feelings merely as felt, or as they are for the feeling\nconsciousness--he avoids the necessity of deciding what the \u2018ideas\u2019\nare in the connection of which knowledge and reasoning consist,\nwhether objects constituted by conceived relations or feelings\nsuggestive of each other. But when once attention had been fixed, as\nit was by Hume, on an ostensible relation between objects, like that\nof cause and effect, which, if it exist at all, is clearly not one in\nthe way of resemblance between feelings, the distinction spoken of\nbecomes patent. If the colour red had not the likeness and unlikeness\nwhich it has to the colour blue, the colours would be different\nfeelings from what they are; but if the flame of fire and its heat\nwere not regarded severally as cause and effect, it would make no\ndifference to them as feelings; or, to put it conversely, it is not\nupon any comparison of two feelings with each other that we regard\nthem as related in the way of cause and effect. In what sense then\ncan the relation between flame and heat be a philosophical relation,\nas defined by Hume--a relation in virtue of which we compare objects,\nor an idea that we acquire upon comparison?\n\nExamination of Hume\u2019s language about them.\n\n297. This definition, indeed, is not stated so exactly or so\nuniformly as might be wished. In different passages \u2018philosophical\nrelation\u2019 appears as that in respect of which we compare any two\nideas; as that of which we acquire the idea by comparing objects, [1]\nand finally (in the context of the passage last quoted) as itself\nthe comparison. [2] The real source of this ambiguity lies in that\nimpossibility of regarding an object as anything apart from its\nrelations, which compels any theory that does not recognize it to be\ninconsistent with itself. It is Locke\u2019s cardinal doctrine that real\n\u2018objects\u2019 are first given as simple ideas, and that their relations,\nunreal in contrast with the simple ideas, are superinduced by the\nmind--a doctrine which Hume completes by excluding all ideas that are\nnot either copies of simple feelings or compounds of these, and by\nconsequence ideas of relation altogether. The three statements of the\nnature of philosophical relation, given above, mark three stages of\ndeparture from, or approach to, consistency with this doctrine. The\nfirst, implying as it does that relation is not merely a subjective\nresult in our minds from the comparison of ideas, but belongs to the\nideas themselves, is most obviously inconsistent with it according to\nthe form in which it is presented by Locke; but the second is equally\nincompatible with Hume\u2019s completion of the doctrine, for it implies\nthat we so compare ideas as to acquire an idea of relation other than\nthe ideas put together--an idea at once open to Hume\u2019s own challenge,\n\u2018Is it a colour, sound, smell, &c.; or is it a passion or emotion?\u2019\n\n[1] Cf. Part I. 5.\n\n[2] P. 464. [Book I, part III., sec. XIV.]\n\nPhilosophical relation consists in a comparison, but no comparison\nbetween cause and effect.\n\n298. We are thus brought to the third statement, according to which\nphilosophical relation, instead of being an idea acquired upon\ncomparison, is itself the comparison. A comparison of ideas may seem\nnot far removed from the simple sequence of resembling ideas; but\nif we examine the definition of cause, as stated above, which with\nHume corresponds to the view of the relation of cause and effect as\na \u2018_philosophical_\u2019 one, we find that the relation in question is\nneither a comparison of the related objects nor an idea which arises\nupon such comparison. According to his statement a comparison is\nindeed necessary to give us an idea of the relation--a comparison,\nhowever, not of the objects which we reckon severally cause and\neffect with each other, but _(a)_ of each of the two objects with\nother like objects, and _(b)_ of the relation of precedency and\ncontiguity between the two objects with that previously observed\nbetween the like objects. Now, unless the idea of relation between\nobjects in the way of cause and effect is one that consists in, or\nis acquired by, comparison _of those objects_, the fact that another\nsort of comparison is necessary to constitute it does not touch the\nquestion of its possibility. However we come to have it, however\nreducible to impressions the objects may be, it is not only other\nthan the idea of either object taken singly; it is not, as an idea\nof resemblance might be supposed to be, constituted by the joint\npresence or immediate sequence upon each other of the objects. Here,\nthen, is an idea which is not taken either from an impression or from\na compound of impressions (if such composition be possible), and this\nidea is \u2018the source of all our reasonings concerning matters of fact.\u2019\n\nThe comparison is between present and past experience of succession\nof objects.\n\n299. The modern followers of Hume may perhaps seek refuge in the\nconsideration that though the relation of cause and effect between\nobjects is not one in the way of resemblance or one of which the\nidea is given by comparison of the objects, it yet results from\ncomparisons, which may be supposed to act like chemical substances\nwhose combination produces a substance with properties quite\ndifferent from those of the combined substances, whether taken\nseparately or together. Some anticipation of such a solution,\nit may be said, we find in Hume himself, who is aware that from\nthe repetition of impressions of sense and their ideas new,\nheterogeneous, impressions--those of \u2018reflection\u2019--are formed. Of\nthis more will be said when we come to Hume\u2019s treatment of cause\nand effect as a \u2018natural relation.\u2019 For the present we have to\nenquire what exactly is implied in the comparisons from which this\nheterogeneous idea of relation is derived. If we look closely we\nshall find that they presuppose a consciousness of relations as\nlittle reducible to resemblance, _i.e._ as little the result of\ncomparison, as that of cause and effect itself. It has been already\nnoticed how Hume treats the judgment of proportion between figures\nas a mere affair of sense, because such relation depends entirely\non the ideas compared, without reflecting that the existence of the\nfigures presupposes those relations of space to which, because (as\nhe admits) they do not depend on the comparison of ideas, the only\nexcuse for reckoning any relation sensible does not apply. In the\nsame way he contents himself with the fact that the judgment of cause\nand effect implies a comparison of present with past experience, and\nmay thus be brought under his definition of \u2018philosophical relation,\u2019\nwithout observing that the experiences compared are themselves by\nno means reducible to comparison. We judge that an object, which we\nnow find to be precedent and contiguous to another, is its cause\nwhen, comparing present experience with past, we find that it always\nhas been so. That in effect is Hume\u2019s account of the relation,\n\u2018considered as a philosophical one:\u2019 and it implies that the\nconstitution of the several experiences compared involves two sorts\nof relation which Hume admits not to be derived from comparison,\n_(a)_ relation in time and place, _(b)_ relation in the way of\nidentity.\n\nObservation of succession already goes beyond sense.\n\n300. As to relations in time and space, we have already traced out\nthe inconsistencies which attend Hume\u2019s attempt to represent them\nas compound ideas. The statement at the beginning of Part III.,\nthat they are relations not dependent on the nature of compared\nideas, is itself a confession that such representation is erroneous.\nIf the difficulty about the synthesis of successive feelings in\na consciousness that consists merely of the succession could be\novercome, we might admit that the putting together of ideas might\nconstitute such an idea of relation as depends on the nature of the\ncombined ideas. But no combination of ideas can yield a relation\nwhich remains the same while the ideas change, and changes while\nthey remain the same. Thus, when Hume tells us that \u2018in none of the\nobservations we may make concerning relations of time and place can\nthe mind go beyond what is immediately present to the senses, to\ndiscover the relations of objects\u2019 [1] the statement contradicts\nitself. Either we can make no observation concerning relation in\ntime and place at all, or in making it we already \u2018go beyond what\nis immediately present to the senses,\u2019 since we observe what is\nneither a feeling nor several feelings put together. If then Hume\nhad succeeded in his reduction of reasoning from cause or effect to\nobservation of this kind, as modified in a certain way by habit,\nthe purpose for which the reduction is attempted would not have\nbeen attained. The separation between perception and inference,\nbetween \u2018intuition\u2019 and \u2018discourse,\u2019 would have been got rid of,\nbut inference and discourse would not therefore have been brought\nnearer to the mere succession of feelings, for the separation between\nfeeling and perception would remain complete; and that being so, the\nquestion would inevitably recur--If the \u2018observation\u2019 of objects\nas related in space and time already involves a transition from\nthe felt to the unfelt, what greater difficulty is there about the\ninterpretation of a feeling as a change to be accounted for (which is\nwhat is meant by inference to a cause), that we should do violence to\nthe sciences by reducing it to repeated observation lest it should\nseem that in it we \u2018go beyond\u2019 present feeling?\n\n[1] P. 376. [Book I, part III., sec. II.]\n\nAs also does the \u2018observation concerning identity,\u2019 which the\ncomparison involves.\n\n301. Relation in the way of identity is treated by Hume in the third\npart of the Treatise [1] pretty much as he treats contiguity and\ndistance. He admits that it does not depend on the nature of any\nideas so related--in other words, that it is not constituted by\nfeelings as they would be for a merely feeling consciousness--yet\nhe denies that the mind \u2018in any observations we may make concerning\nit\u2019 can go beyond what is immediately present to the senses.\nDirectly afterwards, however, we find that there _is_ a judgment of\nidentity which involves a \u2018conclusion beyond the impressions of our\nsenses\u2019--the judgment, namely, that an object of which the perception\nis interrupted continues individually the same notwithstanding the\ninterruption. Such a judgment, we are told, is a supposition founded\nonly on the connection of cause and effect. How any \u2018observation\nconcerning identity\u2019 can be made without it is not there explained,\nand, pending such explanation, observations concerning identity\nare freely taken for granted as elements given by sense in the\nexperience from which the judgment of cause and effect is derived.\nIn the second chapter of Part IV., however, where \u2018belief in an\nexternal world\u2019 first comes to be explicitly discussed by Hume, we\nfind that \u2018propensities to feign\u2019 are as necessary to account for\nthe judgment of identity as for that of necessary connection. If\nthat chapter had preceded, instead of following, the theory of cause\nand effect as given in Part III., the latter would have seemed much\nless plain sailing than to most readers it has done. It is probably\nbecause nothing corresponding to it appears in that later redaction\nof his theory by which Hume sought popular acceptance, that the true\nsuggestiveness of his speculation was ignored, and the scepticism,\nwhich awakened Kant, reduced to the commonplaces of inductive logic.\nTo examine its purport is the next step to be taken in the process\nof testing the possibility of a \u2018natural history\u2019 of knowledge. Its\nbearing on the doctrine of cause will appear as we proceed.\n\n[1] P. 376. [Book I, part III., sec. II.]\n\nIdentity of objects an unavoidable crux for Hume. His account of it.\n\n302. The problem of identity necessarily arises from the fusion of\nreality and feeling. We must once again recall the propositions in\nwhich Hume represents this fusion--that \u2018everything which enters\nthe mind is both in reality and appearance as the perception;\u2019\nthat \u2018so far as the senses are judges, all perceptions are the\nsame in the manner of their existence;\u2019 that \u2018perceptions\u2019 are\neither impressions, or ideas which are \u2018fainter impressions;\u2019 and\n\u2018impressions are internal and perishing existences, and appear\nas such.\u2019 If these propositions are true--and the \u2018new way of\nideas\u2019 inevitably leads to them--how is it that we _believe_ in \u2018a\n_continued_ existence of objects even when they are not present\nto the senses,\u2019 and an existence \u2018distinct from the mind and\nperception\u2019? They are the same questions from which Berkeley derived\nhis demonstration of an eternal mind--a demonstration premature\nbecause, till the doctrine of \u2018ideas,\u2019 and of mind as their subject,\nhad been definitely altered in a way that Berkeley did not attempt,\nit was explaining a belief difficult to account for by one wholly\nunaccountable. Before Theism could be exhibited with the necessity\nwhich Locke claimed for it, it was requisite to try what could be\ndone with association of ideas and \u2018propensities to feign\u2019 in the\nway of accounting for the world of knowledge, in order that upon\ntheir failure another point of departure than Locke\u2019s might be found\nnecessary. The experiment was made by Hume. He has the merit, to\nbegin with, of stating the nature of identity with a precision which\nwe found wanting in Locke. \u2018In that proposition, _an object is the\nsame with itself_, if the idea expressed by the word _object_ were\nno ways distinguished from that meant by _itself_, we really should\nmean nothing.\u2019 \u2018On the other hand, a multiplicity of objects can\nnever convey the idea of identity, however resembling they may be\nsupposed. ... Since then both number and unity are incompatible with\nthe relation of identity, it must lie in something that is neither\nof them. But at first sight this seems impossible.\u2019 The explanation\nis that when \u2018we say that an object is the same with itself, we mean\nthat the object existent at one time is the same with itself existent\nat another. By this means we make a difference betwixt the idea meant\nby the word _object_ and that meant by _itself_ without going the\nlength of number, and at the same time without restraining ourselves\nto a strict and absolute unity.\u2019 In other words, identity means the\nunity of a thing through a multiplicity of times; or, as Hume puts\nit, \u2018the invariableness and uninterruptedness of any object through a\nsupposed variation of time\u2019. [1]\n\n[1] Pp. 489, 490. [Book I, part IV., sec. II.]\n\nProperly with him it is a fiction, in the sense that we have no\nsuch idea. Yet he implies that we have such idea, in saying that we\nmistake something else for it.\n\n303. Now that \u2018an object exists\u2019 can with Hume mean no more than\nthat an \u2018impression\u2019 is felt, and without succession of feelings\naccording to him there is no time. [1] It follows that unity in the\nexistence of the object, being incompatible with _succession_ of\nfeelings, is incompatible also with existence in time. Either then\nthe unity of the object or its existence at manifold times--both\nbeing involved in the conception of identity--must be a fiction; and\nsince \u2018all impressions are perishing existences,\u2019 perishing with a\nturn of the head or the eyes, it cannot be doubted which it is that\nis the fiction. That the existence of an object, which we call the\nsame with itself, is broken by as many intervals of time as there are\nsuccessive and different, however resembling, \u2018perceptions,\u2019 must be\nthe fact; that it should yet be one throughout the intervals is a\nfiction to be accounted for, Hume accounts for it by supposing that\nwhen the separate \u2018perceptions\u2019 have a strong \u2018natural relation\u2019 to\neach other in the way of resemblance, the transition from one to\nthe other is so \u2018smooth and easy\u2019 that we are apt to take it for\nthe \u2018same disposition of mind with which we consider one constant\nand uninterrupted perception;\u2019 and that, as a consequence of this\nmistake, we make the further one of taking the successive resembling\nperceptions for an identical, _i.e._ uninterrupted as well as\ninvariable object. [2] But we cannot mistake one object for another\nunless we have an idea of that other object. If then we \u2018mistake the\nsuccession of our interrupted perceptions for an identical object,\u2019\nit follows that we have an idea of such an object--of a thing one\nwith itself throughout the succession of impressions--an idea which\ncan be a copy neither of any one of the impressions nor, even if\nsuccessive impressions could put themselves together, of all so\nput together. Such an idea being according to Hume\u2019s principles\nimpossible, the appearance of our having it was the fiction he had\nto account for; and he accounts for it, as we find, by a \u2018habit of\nmind\u2019 which already presupposes it. His procedure here is just the\nsame as in dealing with the idea of vacuum. In that case, as we saw,\nhaving to account for the appearance of there being the impossible\nidea of pure space, he does so by showing, that having \u2018an idea of\ndistance not filled with any coloured or tangible object,\u2019 we mistake\nthis for an idea of extension, and hence suppose that the latter\nmay be invisible and intangible. He thus admits an idea, virtually\nthe same with the one excluded, as the source of the \u2018tendency to\nsuppose\u2019 which is to replace the excluded idea. So in his account of\nidentity. Either the habit, in virtue of which we convert resembling\nperceptions into an identical object, is what Hume admits to be a\ncontradiction, \u2018a habit acquired by what was never present to the\nmind\u2019; [3] or the idea of identity must be present to the mind in\norder to render the habit possible.\n\n[1] \u2018Wherever we have no successive perceptions, we have no notion of\ntime.\u2019 (p. 342) [Book I, part II., sec. III.].\n\n[2] P. 492. [Book I, part IV., sec. II.]\n\n[3] P. 487. [Book I, part IV., sec. II.]\n\nSuccession of like feelings mistaken for an identical object: but the\nfeelings, as described, are already such objects.\n\n304. The device by which this _petitio principii_ is covered is one\nalready familiar to us in Hume. In this case it is so palpable that\nit is difficult to believe he was unconscious of it. As he has \u2018to\naccount for the belief of the vulgar with regard to the existence\nof body,\u2019 he will \u2018entirely conform himself to their manner of\nthinking and expressing themselves;\u2019 in other words, he will assume\nthe fiction in question as the beginning of a process by which its\nformation is to be accounted for. The vulgar make no distinction\nbetween thing and appearance. \u2018Those very sensations which enter by\nthe eye or ear are with them the true objects, nor can they readily\nconceive that this pen or this paper, which is immediately perceived,\nrepresents another which is different from, but resembling it. In\norder therefore to accommodate myself to their notions, I shall at\nfirst suppose that there is only a single existence, which I shall\ncall indifferently _object_ and _perception_, according as it shall\nseem best to suit my purpose, understanding by both of them what\nany common man may mean by a hat, or shoe, or stone, or any other\nimpression conveyed to him by his senses\u2019. [1] Now it is of course\ntrue that the vulgar are innocent of the doctrine of representative\nideas. They do not suppose that this pen or this paper, which is\nimmediately perceived, represents another which is different from,\nbut resembling, it; but neither do they suppose that this pen or\nthis paper is a sensation. It is the intellectual transition from\nthis, that, and the other successive sensations to this pen or\nthis paper, as the identical object to which the sensations are\nreferred as qualities, that is unaccountable if, according to Hume\u2019s\ndoctrine, the succession of feelings constitutes our consciousness.\nIn the passage quoted he quietly ignores it, covering his own\nreduction of felt thing to feeling under the popular identification\nof the real thing with the perceived. With \u2018the vulgar\u2019 that which\nis \u2018immediately perceived\u2019 is the real thing, just because it is\nnot the mere feeling which with Hume it is. But under pretence of\nprovisionally adopting the vulgar view, he entitles himself to treat\nthe mere feeling, because according to him it is that which is\nimmediately perceived, as if it were the permanent identical thing,\nwhich according to the vulgar is what is immediately perceived.\n\n[1] P. 491. [Book I, part IV., sec. II.]\n\nFiction of identity thus implied as source of the propensity which is\nto account for it.\n\n305. Thus without professedly admitting into consciousness anything\nbut the succession of feelings he gets such individual objects as\nLocke would have called objects of \u2018actual present sensation.\u2019 When\n\u2018I survey the furniture of my chamber,\u2019 according to him, I see\nsundry \u2018identical objects\u2019--this chair, this table, this inkstand,\n&c. [1] So far there is no fiction to be accounted for. It is only\nwhen, having left my chamber for an interval and returned to it, I\nsuppose the objects which I see to be identical with those I saw\nbefore, that the \u2018propensity to feign\u2019 comes into play, which has\nto be explained as above. But in fact the original \u2018survey\u2019 during\nwhich, seeing the objects, I suppose them to continue the same with\nthemselves, involves precisely the same fiction. In that case, says\nHume, I \u2018suppose the change\u2019 (which is necessary to constitute the\nidea of identity) \u2018to lie only in the time.\u2019 But without \u2018succession\nof perceptions,\u2019 different however resembling, there could according\nto him be no change of time. The continuous survey of this table, or\nthis chair, then, involves the notion of its remaining the same with\nitself throughout a succession of different perceptions--_i.e._ the\nfull-grown fiction of identity--just as much as does the supposition\nthat the table I see now is identical with the one I saw before. The\n\u2018reality,\u2019 confusion with which of \u2018a smooth passage along resembling\nideas\u2019 is supposed to constitute the \u2018fiction,\u2019 is already itself\nthe fiction--the fiction of an object which must be other than our\nfeelings, since it is permanent while they are successive, yet so\nrelated to them that in virtue of reference to it, instead of being\nmerely different from each other, they become changes of a thing.\n\n[1] P. 493. [Book I, part IV., sec. II.]\n\nWith Hume continued existence of perceptions a fiction different from\ntheir identity. Can perceptions exist when not perceived?\n\n306. Having thus in effect imported all three \u2018fictions of\nimagination\u2019--identity, continued existence, and existence distinct\nfrom perception--into the original \u2018perception,\u2019 Hume, we may think,\nmight have saved himself the trouble of treating them as separate and\nsuccessive formations. Unless he had so treated them, however, his\n\u2018natural history\u2019 of consciousness would have been far less imposing\nthan it is. The device, by which he represents the \u2018vulgar\u2019 belief in\nthe reality of the felt thing as a belief that the mere feeling is\nthe real object, enables him also to represent the identity, which\na smooth transition along closely resembling sensations leads us\nto suppose, as still merely identity of a _perception_. \u2018The very\nimage which is present to the senses is with us the real body; and\n\u2019tis to these interrupted images we ascribe a perfect identity\u2019.\n[1] The identity lying thus in the images or appearances, not in\nanything to which they are referred, a further fiction seems to be\nrequired by which we may overcome the contradiction between the\ninterruption of the appearances and their identity--the fiction\nof \u2018a continued being which may fill the intervals\u2019 between the\nappearances. [2] That a \u2018propension\u2019 towards such a fiction would\nnaturally arise from the uneasiness caused by such a contradiction,\nwe may readily admit. The question is how the propension can be\nsatisfied by a supposition which is merely another expression for\none of the contradictory beliefs. What difference is there between\nthe appearance of a perception and its existence, that interruption\nof the perception, though incompatible with uninterruptedness in its\nappearance, should not be so with uninterruptedness in its existence?\nIt may be answered that there is just the difference between relation\nto a feeling subject and relation to a thinking one--between relation\nto a consciousness which is in time, or successive, and relation to\na thinking subject which, not being itself in time, is the source of\nthat determination by permanent conditions, which is what is meant by\nthe real existence of a perceived thing. But to Hume, who expressly\nexcludes such a subject--with whom \u2018it exists\u2019 = \u2018it is felt\u2019--such\nan answer is inadmissible. He can, in fact, only meet the difficulty\nby supposing the existence of unfelt feelings, of unperceived\nperceptions. The appearance of a perception is its presence to\n\u2018what we call a mind,\u2019 which \u2018is nothing but a heap or collection\nof different perceptions, united together by certain relations, and\nsupposed, though falsely, to be endowed with a perfect simplicity\nand identity\u2019. [3] To consider a perception, then, as existing\nthough not appearing is merely to consider it as detached from\nthis \u2018heap\u2019 of other perceptions, which, on Hume\u2019s principle that\nwhatever is distinguishable is separable, is no more impossible than\nto distinguish one perception from all others. [4] In fact, however,\nit is obvious that the supposed detachment is the very opposite of\nsuch distinction. A perception distinguished from all others is\ndetermined by that distinction in the fullest possible measure. A\nperception _detached_ from all others, left out of the \u2018heap which\nwe call a mind,\u2019 being out of all relation, has no qualities--is\nsimply nothing. We can no more \u2018consider\u2019 it than we can see vacancy.\nYet it is by the consideration of such nonentity, by supposing a\nworld of unperceived perceptions, of \u2018existences\u2019 without relation\nor quality, that the mind, according to Hume--itself only \u2018a heap of\nperceptions\u2019--arrives at that fiction of a continued being which,\nas involved in the supposition of identity, is the condition of our\nbelieving in a world of real things at all.\n\n[1] P. 493. [Book I, part IV., sec. II.]\n\n[2] Pp. 494, 495. [Book I, part IV., sec. II.]\n\n[3] P. 495. [Book I, part IV., sec. II.]\n\n[4] Ibid.\n\nExistence of objects, distinct from perceptions, a further fiction\nstill.\n\n307. It is implied, then, in the process by which, according to\nHume, the fiction of a continued being is arrived at, that this\nbeing is supposed to be not only continued but \u2018distinct from the\nmind\u2019 and \u2018independent\u2019 of it. With Hume, however, the supposition\nof a distinct and \u2018independent\u2019 existence of the _perception_ is\nquite different from that of a distinct and independent object\nother than the perception. The former is the \u2018vulgar hypothesis,\u2019\nand though a fiction, it is also a universal belief: the latter\nis the \u2018philosophical hypothesis,\u2019 which, if it has a tendency to\nobtain belief at all, at any rate derives that tendency, in other\nwords \u2018acquires all its influence over the imagination,\u2019 from the\nvulgar one. [1] Just as the belief in the independent and continued\nexistence of perceptions results from an instinctive effort to escape\nthe uneasiness, caused by the contradiction between the interruption\nof resembling perceptions and their imagined identity, so the\ncontradiction between this belief and the evident dependence of all\nperceptions \u2018on our organs and the disposition of our nerves and\nanimal spirits\u2019 leads to the doctrine of representative ideas or \u2018the\ndouble existence of perceptions and objects.\u2019 \u2018This philosophical\nsystem, therefore, is the monstrous offspring of two principles\nwhich are contrary to each other, which are both at once embraced by\nthe mind and which are unable mutually to destroy each other. The\nimagination tells us that our resembling perceptions have a continued\nand uninterrupted existence, and are not annihilated by their\nabsence. Reflection tells us that even our resembling perceptions\nare interrupted in their existence and different from each other.\nThe contradiction betwixt these opinions we elude by a new fiction\nwhich is conformable to the hypotheses both of reflection and fancy,\nby ascribing these contrary qualities to different existences; the\ninterruption to _perceptions_, and the continuance to _objects_\u2019. [2]\n\n[1] P. 500. [Book I, part IV., sec. II.]\n\n[2] P. 502. [Book I, part IV., sec. II.]\n\nAre these several \u2018fictions\u2019 really different from each other?\n\n308. Here, again, we find that the contradictory announcements,\nwhich it is the object of this new fiction to elude, are virtually\nthe same as those implied in that judgment of identity which is\nnecessary to the \u2018perception\u2019 of this pen or this paper. That\n\u2018interruption of our resembling perceptions,\u2019 of which \u2018reflection\u2019\n(in the immediate context \u2018Reason\u2019) is here said to \u2018tell us,\u2019 is\nmerely that difference in time, or succession, which Hume everywhere\nelse treats as a datum of sense, and which, as he points out, is as\nnecessary a factor in the idea of identity, as is the imagination\nof an existence continued throughout the succession. Thus the\ncontradiction, which suggests this philosophical fiction of double\nexistence, has been already present and overcome in every perception\nof a qualified object. Nor does the fiction itself, by which the\ncontradiction is eluded, differ except verbally from that suggested\nby the contradiction between the interruption and the identity of\nperceptions. What power is there in the word \u2018object\u2019 that the\nsupposition of an unperceived existence of perceptions, continued\nwhile their appearance is broken, should be an unavoidable fiction of\nthe imagination, while that of \u2018the double existence of perceptions\nand objects\u2019 is a gratuitous fiction of philosophers, of which\n\u2018vulgar\u2019 thinking is entirely innocent?\n\nAre they not all involved in the simplest perception?\n\n309. That it is gratuitous we may readily admit, but only because a\nrecognition of the function of the Ego in the primary constitution\nof the qualified individual object--this pen or this paper--renders\nit superfluous. To the philosophy, however, in which Hume was\nbred, the perception of a qualified object was simply a feeling.\nNo intellectual synthesis of successive feelings was recognized as\ninvolved in it. It was only so far as the dependence of the feeling\non our organs, in the absence of any clear distinction between\nfeeling and felt thing, seemed to imply a dependent and broken\nexistence of the thing, that any difficulty arose--a difficulty met\nby the supposition that the felt thing, whose existence was thus\nbroken and dependent, represented an unfelt and permanent thing of\nwhich it is a copy or effect. To the Berkeleian objections, already\nfatal to this supposition, Hume has his own to add, viz. that we can\nhave no idea of relation in the way of cause and effect except as\nbetween objects which we have observed, and therefore can have no\nidea of it as existing between a perception and an object of which\nwe can only say that it is not a perception. Is all existence then\n\u2018broken and dependent\u2019? That is the \u2018sceptical\u2019 conclusion which\nHume professes to adopt--subject, however, to the condition of\naccounting for the contrary supposition (without which, as he has to\nadmit, we could not think or speak, and which alone gives a meaning\nto his own phraseology about impressions and ideas) as a fiction of\nthe imagination. He does this, as we have seen, by tracing a series\nof contradictions, with corresponding hypotheses invented, either\ninstinctively or upon reflection, in order to escape the uneasiness\nwhich they cause, all ultimately due to our mistaking similar\nsuccessive feelings for an identical object. Of such an object, then,\nwe must have an idea to begin with, and it is an object permanent\nthroughout a variation of time, which means a succession of feelings;\nin other words, it is a felt thing, as distinct from feelings but to\nwhich feelings are referred as its qualities. Thus the most primary\nperception--that in default of which Hume would have no reality\nto oppose to fiction, nor any point of departure for the supposed\nconstruction of fictions--already implies that transformation of\nfeelings into changing relations of a thing which, preventing any\nincompatibility between the perpetual brokenness of the feeling\nand the permanence of the thing, \u2018eludes\u2019 by anticipation all the\ncontradictions which, according to Hume, we only \u2018elude\u2019 by speaking\nas if we had ideas that we have not.\n\nYet they are not possible ideas, because copied from no impressions.\n\n310. \u2018Ideas that we _have not_;\u2019 for no one of the fictions by\nwhich we elude the contradictions, nor indeed any one of the\ncontradictory judgments themselves, can be taken to represent an\n\u2018idea\u2019 according to Hume\u2019s account of ideas. He allows himself indeed\nto speak of our having ideas of identical objects, such as _this\ntable while I see or touch it_--though in this case, as has been\nshown, either the object is not identical or the idea of it cannot\nbe copied from an impression--and of our transferring this idea to\nresembling but interrupted perceptions. But the supposition to which\nthe contradiction involved in this transference gives rise--the\nsupposition that the perception continues to exist when it is not\nperceived--is shown by the very statement of it to be no possible\ncopy of an impression. Yet according to Hume it is a \u2018belief,\u2019 and\na belief is \u2018a lively idea associated with a present impression.\u2019\nWhat then is the impression and what the associated idea? \u2018As the\npropensity to feign the continued existence of sensible objects\narises from some lively impressions of the memory, it bestows a\nvivacity on that fiction; or, in other words, makes us believe the\ncontinued existence of body\u2019. [1] Well and good: but this only\nanswers the first part of our question. It tells us what are the\nimpressions in the supposed case of belief, but not what is the\nassociated idea to which their liveliness is communicated. To say\nthat it arises from a propensity to feign, strong in proportion to\nthe liveliness of the supposed impressions of memory, does not tell\nus of what impression it is a copy. Such a propensity indeed would be\nan \u2018impression of reflection,\u2019 but the fiction itself is neither the\npropensity nor a copy of it. The only possible supposition left for\nHume would be that it is a \u2018compound idea;\u2019 but what combination of\n\u2018perceptions\u2019 can amount to the existence of perceptions when they\nare not perceived?\n\n[1] P. 496. [Book I, part IV., sec. II.]\n\nComparison of present experience with past, which yields relation of\ncause and effect, pre-supposes judgment of identity;\n\n311. From this long excursion into Hume\u2019s doctrine of relation in the\nway of identity--having found him admitting explicitly that it is\nonly by a \u2018fiction of the imagination\u2019 that we identify this table\nas now seen with this table as seen an hour ago, and implicitly that\nthe same fiction is involved in the perception of this table as an\nidentical object even when hand or eye is kept upon it, while yet\nhe says not a word to vindicate the possibility of such a fiction\nfor a faculty which can merely reproduce and combine \u2018perishing\nimpressions\u2019--we return to consider its bearing upon his doctrine of\nrelation in the way of cause and effect. According to him, as we saw,\n[1] that relation, \u2018considered as a philosophical\u2019 one, is founded\non a comparison of present experience with past, in the sense that\nwe regard an object, precedent and contiguous to another, as its\ncause when all like objects have been found similarly related. The\nquestion then arises whether the experiences compared--the present\nand the past alike--do not involve the fiction of identity along with\nthe whole family of other fictions which Hume affiliates to it? Does\nthe relation of precedence and sequence, which, if constant, amounts\nto that of cause and effect, merely mean precedence and sequence\nof two feelings, indefinitely like an indefinite number of other\nfeelings that have thus the one preceded and the other followed;\nor is it a relation between one qualified thing or definite fact\nalways the same with itself, and another such thing or fact always\nthe same with itself? The question carries its own answer. If in the\ndefinition quoted Hume used the phrase \u2018all like objects\u2019 instead of\nthe \u2018same object,\u2019 in order to avoid the appearance of introducing\nthe \u2018fiction\u2019 of identity into the definition of cause, the device\ndoes not avail him much. The effect of the \u2018like\u2019 is neutralized by\nthe \u2018all.\u2019 A _uniform_ relation is impossible except between objects\nof which each has a definite identity.\n\n[1] Above, pars. 298 and 299.\n\n... without which there could be no recognition of an object as one\nobserved before.\n\n312. When Hume has to describe the experience which gives the\nidea of cause and effect, he virtually admits this. \u2018The nature\nof experience,\u2019 he tells us, \u2018is this. We remember to have had\nfrequent instances of the existence of one species of objects, and\nalso remember that the individuals of another species of objects\nhave always attended them, and have existed in a regular order of\ncontiguity and succession with regard to them. Thus we remember\nto have seen that species of object we call _flame_, and to have\nfelt that species of sensation we call _heat_. We likewise call to\nmind their constant conjunction in all past instances. Without any\nfarther ceremony we call the one cause, and the other effect, and\ninfer the existence of the one from the other\u2019. [1] It appears, then,\nthat upon experiencing certain sensations of sight and touch, we\nrecognize each as \u2018one of a species of objects\u2019 which we remember to\nhave observed in certain constant relations before. In virtue of the\nrecognition the sensations become severally this _flame_ and this\n_heat_; and in virtue of the remembrance the objects thus recognized\nare held to be related in the way of cause and effect. Now it is\nclear that though the recognition takes place upon occasion of a\nfeeling, the object recognized--this flame or this heat--is by no\nmeans the feeling as a \u2018perishing existence.\u2019 Unless the feeling\nwere taken to represent a thing, conceived as permanently existing\nunder certain relations and attributes--in other words, unless it\nwere _identified_ by thought--it would be no definite object, not\nthis _flame_ or this _heat_, at all. The moment it is named, it\nhas ceased to be a feeling and become a felt thing, or, in Hume\u2019s\nlanguage, an \u2018individual of _a species of objects_.\u2019 And just as\nthe present \u2018perception\u2019 is the recognition of such an individual,\nso the remembrance which determines the recognition is one wholly\ndifferent from the return with lessened liveliness of a feeling more\nstrongly felt before. According to Hume\u2019s own statement, it consists\nin recalling \u2018frequent instances of the existence of _a species of\nobjects_.\u2019 It is remembrance of an experience in which every feeling,\nthat has been attended to, has been interpreted as a fresh appearance\nof some qualified object that \u2018exists\u2019 throughout its appearances--an\nexperience which for that reason forms a connected whole. If it were\nnot so, there could be no such comparison of the relations in which\ntwo objects are now presented with those in which they have always\nbeen presented, as that which according to Hume determines us to\nregard them as cause and effect. The condition of our so regarding\nthem is that we suppose the objects now presented to be _the same_\nwith those of which we have had previous experience. It is only on\nsupposition that a certain sensation of sight is not merely like a\nmultitude of others, but represents the same object as that which I\nhave previously known as flame, that I infer the sequence of heat\nand, when it does follow, regard it as an effect. If I thought that\nthe sensation of sight, however like those previously referred to\nflame, did not represent the same object, I should not infer heat\nas effect; and conversely, if, having identified the sensation\nof sight as representative of flame, I found that the inferred\nheat was not actually felt, I should judge that I was mistaken in\nthe identification. It follows that it is only an experience of\nidentical, and by consequence related and qualified, objects, of\nwhich the memory can so determine a sequence of feelings as to\nconstitute it an experience of cause and effect. Thus the perception\nand remembrance upon which, according to Hume, we judge one object to\nbe the cause of another, alike rest on the \u2018fictions of identity and\ncontinued existence.\u2019 Without these no present experience would, in\nhis language, be an instance of an individual of a certain species\nexisting in a certain relation, nor would there be a past experience\nof individuals of the same species, by comparison with which the\nconstancy of the relation might be ascertained.\n\n[1] P. 388. [Book I, part III., sec. VI.]\n\nHume makes conceptions of identity and cause each come before the\nother. Their true correlativity.\n\n313. Against this derivation of the conception of cause and effect,\nas implying that of identity, may be urged the fact that when\nwe would ascertain the truth of any identification we do so by\nreference to causes and effects. As Hume himself puts it at the\noutset of his discussion of causation, an inference of identity\n\u2018beyond the impressions of our senses can be founded only on the\nconnexion of cause and effect.\u2019 ... \u2018Whenever we discover a perfect\nresemblance between a new object and one which was formerly present\nto the senses, we consider whether it be common in that species of\nobjects; whether possibly or probably any cause could operate in\nproducing the change and resemblance; and according as we determine\nconcerning these causes and effects, we form our judgment concerning\nthe identity of the object\u2019. [1] This admission, it may be said,\nthough it tells against Hume\u2019s own subsequent explanation of identity\nas a fiction of the imagination, is equally inconsistent with any\ndoctrine that would treat identity as the presupposition of inference\nto cause or effect. Now undoubtedly if the identity of interrupted\nperceptions is one fiction of the imagination and the relation of\ncause and effect another, each resulting from \u2018custom,\u2019 to say with\nHume, that we must have the idea of cause in order to arrive at the\nsupposition of identity, is logically to exclude any derivation\nof that idea from an experience which involves the supposition of\nidentity. The \u2018custom\u2019 which generates the idea of cause must have\ndone its work before that which generates the supposition of identity\ncan begin. Hume therefore, after the admission just quoted, was\nnot entitled to treat the inference to cause or effect as a habit\nderived from experience of identical things. But it is otherwise\nif the conceptions of causation and identity are correlative--not\nresults of experience of which one must be formed before the\nother, but co-ordinate expressions of one and the same synthetic\nprinciple, which renders experience possible. And this is the real\nstate of the case. It is true, as Hume points out, that when we\nwant to know whether a certain sensation, precisely resembling one\nthat we have previously experienced, represents the same object,\nwe do so by asking how otherwise it can be accounted for. If no\ndifference appears in its antecedents or sequents, we identify\nit--refer it to the same thing--as that previously experienced; for\nits relations (which, since it is an event in time, take the form\nof antecedence and sequence) _are_ the thing. The conceptions of\nidentity and of relation in the way of cause and effect are thus as\nstrictly correlative and inseparable as those of the thing and of\nits relations. Without the conception of identity experience would\nwant a centre, without that of cause and effect it would want a\ncircumference. Without the supposition of objects which \u2018existing at\none time are the same with themselves as existing at other times\u2019--a\nsupposition which at last, when through acquaintance with the\nendlessness of orderly change we have learnt that there is but one\nobject for which such identity can be claimed without qualification,\nbecomes the conception of nature as a uniform whole--there could\nbe no such comparison of the relations in which an object is now\npresented with those in which it has been before presented, as\ndetermines us to reckon it the cause or effect of another; but it is\nequally true that it is only by such comparison of relations that the\nidentity of any particular object can be ascertained.\n\n[1] P. 376. [Book I, part III., sec. II.]\n\nHume quite right in saying that we do not go _more_ beyond sense in\nreasoning than in perception.\n\n314. Thus, though we may concede to Hume that neither in the\ninference to the relation of cause and effect nor in the conclusions\nwe draw from it do we go \u2018beyond experience,\u2019 [1] this will merely\nbe, if his account of it as a \u2018philosophical relation\u2019 be true,\nbecause in experience we already go beyond sense. \u2018There is nothing,\u2019\nsays Hume, \u2018in any object considered in itself that can afford\nus a reason for drawing a conclusion beyond it\u2019 [2]--a statement\nwhich to him means that, if the mind really passes from it to\nanother, this is only because as a matter of fact another feeling\nfollows on the first. But, in truth, if each feeling were merely\n\u2018considered in itself,\u2019 the fact that one follows on another would\nbe no fact _for the subject of the feelings_, no starting-point of\nintelligent experience at all; for the fact is the relation between\nthe feelings--a relation which only exists for a subject that\nconsiders neither feeling \u2018in itself,\u2019 as a \u2018separate and perishing\nexistence,\u2019 but finds a reality in the determination of each by the\nother which, as it is not either or both of them, so survives, while\nthey pass, as a permanent factor of experience. Thus in order that\nany definite \u2018object\u2019 of experience may exist for us, our feelings\nmust have ceased to be what according to Hume they are in themselves.\nThey cease to be so in virtue of the presence to them of the Ego,\nin common relation to which they become related to each other as\nmutually qualified members of a permanent system--a system which at\nfirst for the individual consciousness exists only as a forecast\nor in outline, and is gradually realized and filled up with the\naccession of experience. It is quite true that nothing more than\nthe reference to such a system, already necessary to constitute the\nsimplest object of experience, is involved in that interpretation\nof every event as a changed appearance of an unchanging order,\nand therefore to be accounted for, which we call inference to a\ncause or the inference of necessary connection; or, again, in the\nidentification of the event, the determination of its particular\nnature by the discovery of its particular cause.\n\n[1] Above, pars. 285 & 286.\n\n[2] P. 436 and elsewhere. [Book I, part III., sec. XII.]\n\nHow his doctrine might have been developed. Its actual outcome.\n\n315. The supposed difference then between immediate and mediate\ncognition is no absolute difference. It is not a difference between\nexperience and a process that goes beyond experience, or between an\nexperience unregulated by a conception of a permanent system and\none that is so regulated. It lies merely in the degree of fullness\nand articulation which that conception has attained. If this had\nbeen what Hume meant to convey in his assimilation of inference to\nperception, he would have gone far to anticipate the result of the\nenquiry which Kant started. And this is what he might have come\nto mean if, instead of playing fast and loose with \u2018impression\u2019\nand \u2018object,\u2019 using each as plausibility required on the principle\nof accommodation to the \u2018vulgar,\u2019 he had faced the consequence of\nhis own implicit admission, that every perception of an object as\nidentical is a \u2018fiction\u2019 in which we go beyond present feeling. As\nit is, his \u2018scepticism with regard to the senses\u2019 goes far enough\nto empty their \u2018reports\u2019 of the content which the \u2018vulgar\u2019 ascribe\nto them, and thus to put a breach between sense and the processes\nof knowledge, but not far enough to replace the \u2018sensible thing\u2019 by\na function of reason. In default of such replacement, there was no\nway of filling the breach but to bring back the vulgar theory under\nthe cover of habits and \u2018tendencies to feign,\u2019 which all suppose a\nready-made knowledge of the sensible thing as their starting-point.\nHence the constant contradiction, which it is our thankless task\nto trace, between his solution of the real world into a succession\nof feelings and the devices by which he sought to make room in his\nsystem for the actual procedure of the physical sciences. Conspicuous\namong these is his allowance of that view of relation in the way of\ncause and effect as an objective reality, which is represented by his\ndefinition of it as a \u2018philosophical relation.\u2019 It is in the sense\nrepresented by that definition that his doctrine has been understood\nand retained by subsequent formulators of inductive logic; but on\nexamining it in the light of his own statements we have found that\nthe relation, as thus defined, is not that which his theory required,\nand as which to represent it is the whole motive of his disquisition\non the subject. It is not a sequence of impression upon impression,\ndistinguished merely by its constancy; nor a sequence of idea upon\nimpression, distinguished merely by that transfer of liveliness to\nthe idea which arises from the constancy of its sequence upon the\nimpression. It is a relation between \u2018objects\u2019 of which each is what\nit is only as \u2018an instance of a species\u2019 that exists continuously,\nand therefore in distinction from our \u2018perishing impressions,\u2019\naccording to a regular order of \u2018contiguity and succession.\u2019 As\nsuch existence and order are by Hume\u2019s own showing no possible\nimpressions, and by consequence no possible ideas, so neither are the\n\u2018objects\u2019 which derive their whole character from them.\n\nNo philosophical relation admissible with Hume that is not derived\nfrom a natural one.\n\n316. It may be said, however, that wherever Hume admits a definition\npurporting to be of a \u2018philosophical relation,\u2019 he does so only as\nan accommodation, and under warning that every such relation is\n\u2018fictitious\u2019 except so far as it is equivalent to a natural one; that\naccording to his express statement \u2018it is only so far as causation\nis a _natural_ relation, and produces an union among our ideas, that\nwe are able to reason upon it or draw any inference from it\u2019; [1]\nand that therefore it is only by his definition of it as a \u2018natural\nrelation\u2019 that he is to be judged. Such a vindication of Hume would\nbe more true than effective. That with him the \u2018philosophical\u2019\nrelation of cause and effect is \u2018fictitious,\u2019 with all the\nfictitiousness of a \u2018continued existence distinct from perceptions,\u2019\nis what it has been the object of the preceding paragraphs to show.\nBut the fictitiousness of a relation can with him mean nothing else\nthan that, instead of having an idea of it, we have only a \u2018tendency\nto suppose\u2019 that we have such an idea. Thus the designation of the\nphilosophical relation of cause and effect carries with it two\nconditions, one negative, the other positive, on the observance of\nwhich the logical value of the designation depends. The \u2018tendency\nto suppose\u2019 must _not_ after all be itself translated into the idea\nwhich it is to replace; and it _must_ be accounted for as derived\nfrom a \u2018natural relation\u2019 which is not fictitious. That the negative\ncondition is violated by Hume, we have sufficiently seen. He treats\nthe \u2018philosophical relation\u2019 of cause and effect, in spite of the\n\u2018fictions\u2019 which it involves, not as a name for a tendency to suppose\nthat we have an idea which we have not, but as itself a definite\nidea on which he founds various \u2018rules for judging what objects are\nreally so related and what are not\u2019. [2] That the positive condition\nis violated also--that the \u2018natural relation\u2019 of cause and effect,\naccording to the sense in which his definition of it is meant to be\nunderstood, already itself involves \u2018fictions,\u2019 and only for that\nreason is a possible source of the \u2018philosophical\u2019--is what we have\nnext to show.\n\n[1] P. 394. [Book I, part III., sec. VII.]\n\n[2] Part III. \u00a7 15.\n\nExamination of his account of cause and effect as \u2018natural relation\u2019.\n\n317. That definition, it will be remembered, runs as follows: \u2018A\ncause is an object precedent and contiguous to another, and so united\nwith it in the imagination that the idea of the one determines the\nmind to form the idea of the other, and the impression of the one to\nform a more lively idea of the other.\u2019 Now, as has been sufficiently\nshown, the object of an idea with Hume can properly mean nothing but\nthe impression from which the idea is derived, which again is only\nthe livelier idea, even as the idea is the fainter impression. The\nidea and the object of it, then, only differ as different stages\nin the vivacity of a feeling. [1] It must be remembered, further,\nin regard to the \u2018determination of the mind\u2019 spoken of in the\ndefinition, that the \u2018mind\u2019 according to Hume is merely a succession\nof impressions and ideas, and that its \u2018determination\u2019 means no more\nthan a certain habitualness in this succession. Deprived of the\nbenefit of ambiguous phraseology, then, the definition would run\nthus: \u2018A cause is a lively feeling immediately precedent to another,\n[2] and so united with it that when either of the two more faintly\nrecurs, the other follows with like faintness, and when either occurs\nwith the maximum of liveliness the other follows with less, but still\ngreat, liveliness.\u2019 Thus stated, the definition would correspond\nwell enough to the process by which Hume arrives at it, of which the\nwhole drift, as we have seen, is to merge the so-called objective\nrelation of cause and effect, with the so-called inference from it,\nin the mere habitual transition from one feeling to another. But it\nis only because not thus stated, and because the actual statement\nis understood to carry a meaning of which Hume\u2019s doctrine does not\nconsistently admit, that it has a chance of finding acceptance. Its\nplausibility depends on \u2018object\u2019 and \u2018mind\u2019 and \u2018determination\u2019 being\nunderstood precisely in the sense in which, according to Hume, they\nought not to be understood, so that it shall express not a sequence\nof feeling upon feeling, as this might be for a merely feeling\nsubject, but that permanent relation or law of nature which to a\nsubject that thinks upon its feelings, and only to such a subject,\ntheir sequence constitutes or on which it depends.\n\n[1] See above, paragraphs 195 and 208. Cf. also, among other\npassages, one in the chapter now under consideration (p. 451) [Book\nI, part III., sec. XIV.]--\u2018Ideas always represent their _objects or\nimpressions_.\u2019\n\n[2] The phrase \u2018immediately precedent\u2019 would seem to convey Hume\u2019s\nmeaning better than his own phrase \u2018precedent and contiguous.\u2019\nContiguity _in space_ (which is what we naturally understand by\n\u2018contiguity,\u2019 when used absolutely) he could not have deliberately\ntaken to be necessary to constitute the relation of cause and effect,\nsince the impressions so related, as he elsewhere shows, may often\nnot be in space at all.\n\nDouble meaning of natural relation. How Hume turns it to account.\n\n318. It is this essential distinction between the sequence of\nfeeling upon feeling for a sentient subject and the relation which\nto a thinking subject this sequence constitutes--a distinction not\nless essential than that between the conditions, through which a\nman passes in sleep, as they are for the sleeping subject himself,\nand as they are for another thinking upon them--which it is the\ncharacteristic of Hume\u2019s doctrine of natural relation in all its\nforms to disguise. Only in virtue of the presence to feelings of\na subject, which distinguishes itself from them, do they become\nrelated objects. Thus, with Hume\u2019s exclusion of such a subject,\nwith his reduction of mind and world alike to the succession of\nfeelings, relations and ideas of relation logically disappear. But\nby help of the phrase \u2018natural relation,\u2019 covering, as it does, two\nwholly different things--the involuntary sequence of one feeling\nupon another, and that determination of each by the other which can\nonly take place for a synthetic self-consciousness--he is able on\nthe one hand to deny that the relations which form the framework\nof knowledge are more than sequences of feeling, and on the other\nto clothe them with so much of the real character of relations as\nqualifies them for \u2018principles of union among ideas.\u2019 Thus the mere\noccurrence of similar feelings is with him already that relation in\nthe way of resemblance, which in truth only exists for a subject\nthat can contemplate them as permanent objects. In like manner the\nsuccession of feelings, which can only constitute time for a subject\nthat contrasts the succession with its own unity, and which, if ideas\nwere feelings, would exclude the possibility of an idea of time, is\nyet with him indifferently time and the idea of time, though ideas\nare feelings and there is no \u2018mind\u2019 but their succession.\n\nIf an effect is merely a constantly observed sequence, how can an\nevent be an effect the first time it is observed? Hume evades this\nquestion;\n\n319. The fallacy of Hume\u2019s doctrine of causation is merely an\naggravated form of that which has generally passed muster in his\ndoctrine of time. If time, because a relation between feelings, can\nbe supposed to survive the exclusion of a thinking self and the\nreduction of the world and mind to a succession of feelings, the\nrelation of cause and effect has only to be assimilated to that of\ntime in order that its incompatibility with the desired reduction\nmay disappear, The great obstacle to such assimilation lies in that\nopposition to the mere sequence of feelings which causation as\n\u2018matter of fact\u2019--as that in discovering which we \u2018discover the real\nexistence and relations of objects\u2019--purports to carry with it. Why\ndo we set aside our usual experience as delusive in contrast with the\nexceptional experience of the laboratory--why do we decide that an\nevent which has seemed to happen cannot really have happened, because\nunder the given conditions no adequate cause of it could have been\noperative--if the relation of cause and effect is itself merely a\nsuccession of seemings, repeated so often as to leave behind it a\nlively expectation of its recurrence? This question, once fairly put,\ncannot be answered: it can only be evaded. It is Hume\u2019s method of\nevasion that we have now more particularly to notice.\n\nStill, he is a long way off the Inductive Logic, which supposes an\nobjective sequence.\n\n320. In its detailed statement it is very different from the method\nadopted in those modern treatises of Logic which, beginning with\nthe doctrine that facts are merely feelings in the constitution\nof which thought has no share, still contrive to make free use in\ntheir logical canon of the antithesis between the real and apparent.\nThe key to this modern method is to be found in its ambiguous use\nof the term \u2018phenomenon,\u2019 alike for the feeling as it is felt,\n\u2018perishing\u2019 when it ceases to be felt, and for the feeling as it\nis for a thinking subject--a qualifying and qualified element in a\npermanent world. Only if facts were \u2018phenomena\u2019 in the former sense\nwould the antithesis between facts and conceptions be valid; only if\n\u2018phenomena\u2019 are understood in the latter sense can causation be said\nto be a law of phenomena. So strong, however, is the charm which this\nambiguous term has exercised, that to the ordinary modern logician\nthe question above put may probably seem unmeaning. \u2018The appearance,\u2019\nhe will say, \u2018which we set aside as delusive does not consist in any\nof the reports of the senses--these are always true--but in some\nfalse supposition in regard to them due to an insufficient analysis\nof experience, in some reference of an actual sensation to a group\nof supposed possibilities of sensation, called a \u201cthing,\u201d which\nare either unreal or with which it is not really connected. The\ncorrection of the false appearance by a discovery of causation is\nthe replacement of a false supposition, as to the possibility of the\nantecedence or sequence of one feeling to another, by the discovery,\nthrough analysis of experience, of what feelings do actually precede\nand follow each other. It implies no transition from feelings to\nthings, but only from a supposed sequence of feelings to the actual\none. Science in its farthest range leaves us among appearances still.\nIt only teaches us what really appears.\u2019\n\nCan the principle of uniformity of nature be derived from sequence of\nfeelings?\n\n321. Now the presupposition of this answer is the existence of just\nthat necessary connexion as between appearances, just that objective\norder, for which, because it is not a possible \u2018impression or idea,\u2019\nHume has to substitute a blind propensity produced by habit. Those\nwho make it, indeed, would repel the imputation of believing in any\n\u2018necessary connexion,\u2019 which to them represents that \u2018mysterious\ntie\u2019 in which they vaguely suppose \u2018metaphysicians\u2019 to believe. They\nwould say that necessary connexion is no more than uniformity of\nsequence. But sequence of what? Not of feelings as the individual\nfeels them, for then there would be no perfect uniformities, but\nonly various degrees of approximation to uniformity, and the\nmeasure of approximation in each case would be the amount of the\nindividual\u2019s experience in that particular direction. The procedure\nof the inductive logician shows that his belief in the uniformity\nof a sequence is irrespective of the number of instances in which\nit has been experienced. A single instance in which one feeling is\nfelt after another, if it satisfy the requirements of the \u2018method\nof difference,\u2019 _i.e._ if it show exactly what it is that precedes\nand what it is that follows in that instance, suffices to establish\na uniformity of sequence, on the principle that what is fact once\nis fact always. Now a uniformity that can be thus established is in\nthe proper sense necessary. Its existence is not contingent on its\nbeing felt by anyone or everyone. It does not come into being with\nthe experiment that shows it. It is felt because it is real, not real\nbecause it is felt. It may be objected indeed that the principle of\nthe \u2018uniformity of nature,\u2019 the principle that what is fact once is\nfact always, itself gradually results from the observation of facts\nwhich are feelings, and that thus the principle which enables us to\ndispense with the repetition of a sensible experience is itself due\nto such repetition. The answer is, that feelings which are conceived\nas facts are already conceived as constituents of a nature. The same\npresence of the thinking subject to, and distinction of itself from,\nthe feelings, which renders them knowable _facts_, renders them\nmembers of a world which is one throughout its changes. In other\nwords, the presence of facts from which the uniformity of nature, as\nan abstract rule, is to be inferred, is already the consciousness of\nthat uniformity _in concreto_.\n\nWith Hume the only uniformity is in expectation, as determined by\nhabit; but strength of such expectation must vary indefinitely.\n\n322. Hume himself makes a much more thorough attempt to avoid\nthat pre-determination of feelings by the conception of a world,\nof things and relations, which is implied in the view of them as\npermanent facts. He will not, if he can help it, so openly depart\nfrom the original doctrine that thought is merely weaker sense. Such\nconceptions as those of the uniformity of nature and of reality,\nbeing no possible \u2018impressions or ideas,\u2019 he only professes to admit\nin a character wholly different from that in which they actually\ngovern inductive philosophy. Just as by reality he understands not\nsomething to which liveliness of feeling may be an index, but simply\nthat liveliness itself, and by an inferred or believed reality a\nfeeling to which this liveliness has been communicated from one that\nalready has it; so he is careful to tell us \u2018that the supposition\nthat the future resembles the past is derived entirely from habit,\nby which we are determined to expect for the future the same train\nof objects to which we have been accustomed\u2019. [1] The supposition\nthen _is_ this \u2018determination,\u2019 this \u2018propensity,\u2019 to expect. Any\n\u2018idea\u2019 derived from the propensity can only be the propensity itself\nat a fainter stage; and between such a propensity and the conception\nof \u2018nature,\u2019 whether as uniform or otherwise, there is a difference\nwhich only the most hasty reader can be liable to ignore. But if\nby any confusion an expectation of future feelings, determined by\nthe remembrance of past feelings, could be made equivalent to any\nconception of nature, it would not be of nature as uniform. As is\nthe \u2018habit\u2019 which determines the expectation, such must be the\nexpectation itself; and as have been the sequences of feeling in\neach man\u2019s past, such must be the habit which results from them. Now\nno one\u2019s feelings have always occurred to him in the same relative\norder. There may be some pairs of feelings of which one has always\nbeen felt before the other and never after it, and between which\nthere has never been an intervention of a third--although (to take\nHume\u2019s favourite instance) even the feeling of heat may sometimes\nprecede the sight of the flame--and in these cases upon occurrence of\none there will be nothing to qualify the expectation of the other.\nBut just so far as there are exceptions in our past experience to\nthe immediate sequence of one feeling upon another, must there be a\nqualification of our expectation of the future, if it be undetermined\nby extraneous conceptions, with reference to those particular\nfeelings.\n\n[1] P. 431. [Book I, part III., sec. XII.]\n\nIt could not serve the same purpose as the conception of uniformity\nof nature.\n\n323. Thus the expectation that \u2018the future will resemble the past,\u2019\nif the past means to each man (and Hume could not allow of its\nmeaning more) merely the succession of his own feelings, must be made\nup of a multitude of different expectations--some few of these being\nof that absolute and unqualified sort which alone, it would seem,\ncan regulate the transition that we are pleased to call \u2018necessary\nconnexion;\u2019 the rest as various in their strength and liveliness\nas there are possible differences between cases where the chances\nare evenly balanced and where they are all on one side. From Hume\u2019s\npoint of view, as he himself says, \u2018every past experiment,\u2019 _i.e._\nevery instance in which feeling _(a)_ has been found to follow\nfeeling _(b)_, \u2018may be considered a kind of chance\u2019. [1] As are the\ninstances of this kind to the instances in which some other feeling\nhas followed _(b)_, such are the chances or \u2018probability\u2019 that _(a)_\nwill follow _(b)_ again, and such upon the occurrence of _(b)_ will\nbe that liveliness in the expectation of _(a)_, which alone with Hume\nis the reality of the connexion between them. In such an expectation,\nin an expectation made up of such expectations, there would be\nnothing to serve the purpose which the conception of the uniformity\nof nature actually serves in inductive science. It could never make\nus believe that a feeling felt before another--as when the motion\nof a bell is seen before the sound of it has been heard--represents\nthe real antecedent. It could never set us upon that analysis of our\nexperience by which we seek to get beyond sequences that are merely\nusual, and admit of indefinite exceptions, to such as are invariable;\nupon that \u2018interrogation of nature\u2019 by which, on the faith that\nthere is a uniformity if only we could find it out, we wrest from\nher that confession of a law which she does not spontaneously offer.\nThe fact that some sequences of feeling have been so uniform as to\nresult in unqualified expectations (if it be so) could of itself\nafford no motive for trying to compass other expectations of a like\ncharacter which do not naturally present themselves. Nor could there\nbe anything in the appearance of an exception to a sequence, hitherto\nfound uniform, to lead us to change our previous expectation for\none which shall not be liable to such modification. The previous\nexpectation would be so far weakened, but there is nothing in the\nmere weakening of our expectations that should lead to the effort\nto place them beyond the possibility of being weakened. Much less\ncould the bundle of expectations come to conceive themselves as one\nsystem so as that, through the interpretation of each exception to\na supposed uniformity of sequence as an instance of a real one, the\nchanges of the parts should prove the unchangeableness of the whole.\n\n[1] P. 433. [Book I, part III., sec. XII.]\n\nHume changes the meaning of this expectation by his account of\nthe \u2018remembrance\u2019 which determines it. Bearing of his doctrine\nof necessary connexion upon his argument against miracles. This\nremembrance, as he describes it, supposes conception of a system of\nnature.\n\n324. That a doctrine which reduces the order of nature to strength\nof expectation, and exactly reverses the positions severally given\nto belief and reality in the actual procedure of science, [1] should\nhave been ostensibly adopted by scientific men as their own--with\nevery allowance for Hume\u2019s literary skill and for the charm which the\nprospect of overcoming the separation between reason and instinct\nexercises over naturalists--would have been unaccountable if the\ndoctrine had been thus nakedly put or consistently maintained.\nBut it was not so. Hume\u2019s sense of consistency was satisfied when\nexpectation determined by remembrance had been put in the place\nof necessary connexion, as the basis of \u2018inference to matters of\nfact.\u2019 It does not lead him to adjust his view of the fact inferred\nto his view of the basis on which the inference rests. Expectation\nis an \u2018impression of reflection,\u2019 and if the relation of cause and\neffect is no more than expectation, that which seemed most strongly\nto resist reduction to feeling has yet been so reduced. But if the\nexpectation is to be no more than an impression of reflection, the\nobject expected must itself be no more than an impression of some\nkind or other. The expectation must be expectation of a feeling,\npure and simple. Nor does Hume in so many words allow that it is\notherwise, but meanwhile though the expectation itself is not openly\ntampered with, the remembrance that determines it is so. This is\nbeing taken to be that, which it cannot be unless ideas unborrowed\nfrom impressions are operative in and upon it. It is being regarded,\nnot as the recurrence of a multitude of feelings with a liveliness\nindefinitely less than that in virtue of which they are called\nimpressions of sense, and indefinitely greater than that in virtue of\nwhich they are called ideas of imagination, but as the recognition\nof a world of experience, one, real and abiding. An expectation\ndetermined by such remembrance is governed by the same \u2018fictions\u2019 of\nidentity and continued existence which are the formative conditions\nof the remembrance. Expectation and remembrance, in fact, are one and\nthe same intellectual act, one and the same reference of feelings,\ngiven in time, to an order that is not in time, distinguished\naccording to the two faces which, its \u2018matter\u2019 being in time, it\nhas to present severally to past and future. The remembrance is the\nmeasure of the expectation, but as the remembrance carries with it\nthe notion of a world whose existence does not depend on its being\nremembered, and whose laws do not vary according to the regularity\nor looseness with which our ideas are associated, so too does the\nexpectation, and only as so doing becomes the mover and regulator of\n\u2018inference from the known to the unknown.\u2019\n\n[1] It is by a curious fate that Hume should have been remembered,\nat any rate in the \u2018religious\u2019 world, chiefly by the argument\nagainst miracles which appears in the \u2018Essays\u2019--an argument which,\nhowever irrefragable in itself, turns wholly upon that conception of\nnature as other than our instinctive expectations and imaginations,\nwhich has no proper place in his system (see Vol. IV. page 89). If\n\u2018necessary connexion\u2019 were really no more than the transition of\nimagination, as determined by constant association, from an idea to\nits usual attendant--if there were no conception of an objective\norder to determine belief other than the belief itself--the fact that\nsuch an event, as the revival of one four-days-dead at the command of\na person, had been believed, since it would show that the imagination\nwas at liberty to pass from the idea of the revival to that of the\ncommand (or _vice versa_) with that liveliness which constitutes\nreality, would show also that no necessary connexion, no law of\nnature in the only sense in which Hume entitles himself to speak of\nsuch, was violated by the sequence of the revival on the command.\nAt the same time there would be nothing \u2018miraculous,\u2019 according to\nhis definition of the miraculous as distinct from the extraordinary,\nin the case. Taken strictly, indeed, his doctrine implies that a\nbelief in a miracle is a contradiction in terms. An event is not\nregarded as miraculous unless it is regarded as a \u2018transgression\nof a law of nature by a particular volition of the Deity or by\nthe interposition of some invisible agent\u2019 (page 93, note 1); but\nit could not transgress a law of nature in Hume\u2019s sense unless it\nwere so inconsistent with the habitual association of ideas as that\nit could not be believed. Hume\u2019s only consistent way of attacking\nmiracles, then, would have been to show that the events in question,\nas _miraculous_, had never been believed. Having been obliged to\nrecognize the belief in their having happened, he is open to the\nretort \u2018ad hominem\u2019 that according to his own showing the belief\nin the events constitutes their reality. Such a retort, however,\nwould be of no avail in the theological interest, which requires not\nmerely that the events should have happened but that they should have\nbeen _miraculous_, _i.e._ \u2018transgressions of a law of nature by a\nparticular volition of the Deity.\u2019\n\n325. In the passage already quoted, where Hume is speaking of the\nexpectation in question as depending simply on habit, he yet speaks\nof it as an expectation \u2018of the _same train of objects_ to which\nwe have been accustomed.\u2019 These words in effect imply that it is\n_not_ habit, as constituted simply by the repetition of separate\nsequences of feelings, that governs the expectation--in which case,\nas we have seen, the expectation would be made up of expectations\nas many and as various in strength as have been the sequences and\ntheir several degrees of regularity--but, if habit in any sense,\nhabit as itself governed by conceptions of \u2018identity and distinct\ncontinued existence,\u2019 in virtue of which, as past experience is\nnot an indefinite series of perishing impressions of separate men\nbut represents one world, so all fresh experience becomes part \u2018of\nthe same train of objects;\u2019 part of a system of which, as a whole,\n\u2018the change lies only in the time\u2019. [1] If now we look back to the\naccount given of the relation of memory to belief we shall find that\nit is just so far as, without distinct avowal, and in violation of\nhis principles, he makes \u2018impressions of memory\u2019 carry with them the\nconception of a real system, other than the consciousness of their\nown liveliness, that he gains a meaning for belief which makes it in\nany respect equivalent to the judgment, based on inference, of actual\nscience.\n\n[1] P. 492. [Book I, part IV., sec. II.]\n\nThis explains his occasional inconsistent ascription of an objective\ncharacter to causation.\n\n326. Any one who has carefully read the chapters on inference and\nbelief will have found himself frequently doubting whether he has\ncaught the author\u2019s meaning correctly. A clear line of thought may\nbe traced throughout, as we have already tried to trace it [1]--one\nperfectly consistent with itself and leading properly to the\nconclusion that \u2018all reasonings are nothing but the effect of custom,\nand that custom has no influence but by enlivening the imagination\u2019\n[2]--but its even tenour is disturbed by the exigency of showing\nthat proven fact, after turning out to be no more than enlivened\nimagination, is still what common sense and physical science take it\nto be. According to the consistent theory, ideas of memory are needed\nfor inference to cause or effect, simply because they are lively.\nSuch inference is inference to a \u2018real existence,\u2019 that is to an\n\u2018idea assented to,\u2019 that is to a feeling having such liveliness as,\nnot being itself one of sense or memory, it can only derive from one\nof sense or memory through association with it. That the inferred\nidea is a cause or effect and, as such, has \u2018real existence,\u2019 merely\nmeans that it has this derived liveliness or is believed; just as\nthe reality ascribed to the impression of memory lies merely in its\nhaving this abundant liveliness from which to communicate to its\n\u2018usual attendant.\u2019 But while the title of an idea to be reckoned a\ncause or effect is thus made to depend on its having the derived\nliveliness which constitutes belief, [3] on the other hand we find\nHume from time to time making belief depend on causation, as on a\nrelation of objects distinct from the lively suggestion of one by\nthe others. \u2018Belief arises only from causation, and we can draw no\ninference from one object to another except they be connected by\nthis relation.\u2019 \u2018The relation of cause and effect is requisite to\npersuade us of any real existence\u2019. [4] In the context of these\ndisturbing admissions we find a reconsideration of the doctrine of\nmemory which explains them, but only throws back on that doctrine the\ninconsistency which they exhibit in the doctrine of belief.\n\n[1] Above, paragraphs 289 and ff.\n\n[2] P. 445. [Book I, part III., sec. XIII.]\n\n[3] It may be as well here to point out the inconsistency in Hume\u2019s\nuse of \u2018belief.\u2019 At the end of sec. 5 (Part III.) the term is\nextended to \u2018impressions of the senses and memory.\u2019 We are said to\nbelieve when \u2018we feel an _immediate impression_ of the senses, or a\nrepetition of that impression in the memory. But in the following\nsection the characteristic of belief is placed in the _derived_\nliveliness of an _idea_ as distinct from the immediate liveliness of\nimpression.\n\n[4] Pp. 407 & 409. [Book I, part III., sec. IX.]\n\nReality of remembered \u2018system\u2019 transferred to \u2018system of judgment\u2019.\n\n327. This reconsideration arises out of an objection to his doctrine\nwhich Hume anticipates, to the effect that since, according to it,\nbelief is a lively idea associated \u2018to a present impression,\u2019 any\nsuggestion of an idea by a resembling or contiguous impression\nshould constitute belief. How is it then that \u2018belief arises only\nfrom causation\u2019? His answer, which must be quoted at length, is as\nfollows:--\u2018\u2019Tis evident that whatever is present to the memory,\nstriking upon the mind with a vivacity which resembles an immediate\nimpression, must become of considerable moment in all the operations\nof the mind and must easily distinguish itself above the mere\nfictions of the imagination. Of these impressions or ideas of the\nmemory we form a kind of system, comprehending whatever we remember\nto have been present either to our internal perception or senses, and\nevery particular of that system, joined to the present impressions,\nwe are pleased to call a _reality_. But the mind stops not here.\nFor finding that with this system of perceptions there is another\nconnected by custom or, if you will, by the relation of cause and\neffect, it proceeds to the consideration of their ideas; and as it\nfeels that \u2019tis in a manner necessarily determined to view these\nparticular ideas, and that the custom or relation by which it is\ndetermined admits not of the least change, it forms them into a new\nsystem, which it likewise dignifies with the title of _realities_.\nThe first of these systems is the object of the memory and senses;\nthe second of the judgment. \u2019Tis this latter principle which peoples\nthe world, and brings us acquainted which such existences as, by\ntheir removal in time and place, lie beyond the reach of the senses\nand memory\u2019. [1]\n\n[1] P. 408. [Book I, part III., sec. IX.]\n\nReality of the former \u2018system\u2019 other than vivacity of impressions.\n\n328. From this it appears that \u2018what we are pleased to call\nreality\u2019 belongs, not merely to a \u2018present impression,\u2019 but to\n\u2018every particular of a system joined to the present impression\u2019 and\n\u2018comprehending whatever we remember to have been present either\nto our internal perception or senses.\u2019 This admission already\namounts to an abandonment of the doctrine that reality consists in\nliveliness of feeling. It cannot be that every particular of the\nsystem comprehending all remembered facts, which is joined with the\npresent impression, can have the vivacity of that impression either\nalong with it or by successive communication. We can only feel one\nthing at a time, and by the time the vivacity had spread far from the\npresent impression along the particulars of the system, it must have\ndeclined from that indefinite degree which marks an impression of\nsense. It is not, then, the derivation of vivacity from the present\nimpression, to which it is joined, that renders the \u2018remembered\nsystem\u2019 real; and what other vivacity can it be? It may be said\nindeed that each particular of the system had once the required\nvivacity, was once a present impression; but if in ceasing to be so,\nit did not cease to be real--if, on the contrary, it could not become\na \u2018particular of the system,\u2019 counted real, without becoming other\nthan the \u2018perishing existence\u2019 which an impression is--it is clear\nthat there is a reality which lively feeling does not constitute\nand which involves the \u2018fiction\u2019 of an existence continued in the\nabsence, not only of lively feeling, but of all feelings whatsoever.\nSo soon, in short, as reality is ascribed to a system, which cannot\nbe an \u2018impression\u2019 and of which consequently there cannot be an\n\u2018idea,\u2019 the first principle of Hume\u2019s speculation is abandoned. The\ntruth is implicitly recognized that the reality of an individual\nobject consists in that system of its relations which only exists for\na conceiving, as distinct from a feeling, subject, even as the unreal\nhas no meaning except as a confused or inadequate conception of such\nrelations; and that thus the \u2018present impression\u2019 is neither real\nnor unreal in itself, but may be equally one or the other according\nas the relations, under which it is conceived by the subject of\nit, correspond to those by which it is determined for a perfect\nintelligence. [1]\n\n[1] See above, paragraphs 184 & 183.\n\nIt is constituted by relations, which are not impressions at all;\nand in this lies explanation of the inference from it to \u2018system of\njudgment\u2019.\n\n329. A clear recognition of this truth can alone explain the nature\nof belief as a result of inference from the known to the unknown,\nwhich is, at the same time, inference to a matter of fact. The\npopular notion, of course, is that certain facts are given by feeling\nwithout inference and then other facts inferred from them. But what\nis \u2018fact\u2019 taken to mean? If a feeling, then an inferred fact is a\ncontradiction, for it is an unfelt feeling. If (as should be the\ncase) it is taken to mean the relation of a feeling to something,\nthen it already involves inference--the interpretation of the feeling\nby means of the conception of a universal, self or world, brought\nto it--an inference which is all inference _in posse_, for it\nimplies that a universe of relations is there, which I must know if\nI would know the full reality of the individual object: so that no\nfact can be even partially known without compelling an inference to\nthe unknown, nor can there be any inference to the unknown without\nmodification of what already purports to be known. Hume, trying to\ncarry out the equivalence of fact and feeling, and having clearer\nsight than his masters, finds himself in the presence of this\ndifficulty about inference. Unless the inferred object is other than\none of sense (outer or inner) or of memory, there is no reasoning,\nbut only perception; [1] but if it is other, how can it be real\nor even an object of consciousness at all, since consciousness is\nonly of impressions, stronger or fainter? The only consistent way\nout of the difficulty, as we have seen, is to explain inference as\nthe expectation of the recurrence of a feeling felt before, through\nwhich the unknown becomes known merely in the sense that from the\nrepetition of the recurrence the expectation has come to amount\nto the fullest assurance. But according to this explanation the\ndifference between the inferences of the savage and those of the man\nof science will lie, not in the objects inferred, but in the strength\nof the expectation that constitutes the inference. Meanwhile, if a\nsemblance of explanation has been given for the inference from cause\nto effect, that from effect to cause remains quite in the dark. How\ncan there be inference from a given feeling to that felt immediately\nbefore it?\n\n[1] Pp. 376 & 388. [Book I, part III., secs. II. and VI.]\n\nNot seeing this, Hume has to explain inference to latter system as\nsomething forced upon us by habit.\n\n330. From the avowal of such paradoxical results, Hume only saved\nhimself by reverting, as in the passage before us, to the popular\nview--to the distinction between two \u2018systems of reality,\u2019 one\nperceived, the other inferred; one \u2018the object of the senses and\nmemory,\u2019 the other \u2018of the judgment.\u2019 He sees that if the educated\nman erased from his knowledge upon us by of the world all \u2018facts\u2019 but\nthose for which he has \u2018the evidence of his senses and memory,\u2019 his\nworld would be unpeopled; but he has not the key to the true identity\nbetween the two systems. Not recognizing the inference already\ninvolved in a fact of sense or memory, he does not see that it is\nonly a further articulation of this inference which gives the fact of\njudgment; that as the simplest fact for which we have the \u2018evidence\nof sense\u2019 is already not a feeling but an explanation of a feeling,\nwhich connects it by relations, that are not feelings, with an unfelt\nuniverse, so inferred causes and effects are explanations of these\nexplanations, by which they are connected as mutually determinant\nin the one world whose presence the simplest fact, the most primary\nexplanation of feeling, supposes no less than the most complete. Not\nseeing this, what is he to make of the system of merely inferred\nrealities? He will represent the relation of cause and effect, which\nconnects it with the \u2018system of memory,\u2019 as a habit derived from the\nconstant _de facto_ sequence of this or that \u2018inferred\u2019 upon this\nor that remembered idea. The mind, \u2018feeling\u2019 the unchangeableness\nof this habit, regards the idea, which in virtue of it follows upon\nthe impression of memory, as equally real with that impression.\nIn this he finds an answer to the two questions which he himself\nraises: _(a)_ \u2018Why is it that we draw no inference from one object\nto another, except they be connected by the relation of cause and\neffect;\u2019 or (which is the same, since inference to an object implies\nthe ascription of reality to it), \u2018Why is this relation requisite to\npersuade us of any real existence?\u2019 and _(b)_, \u2018How is it that the\nrelations of resemblance and contiguity have not the same effect?\u2019\nThe answer to the first is, that we do not ascribe reality to an\nidea recalled by an impression, unless we find that, owing to its\ncustomary sequence upon the impression, we cannot help passing from\nthe one to the other. The answer to the second corresponds. The\ncontiguity of an idea to an impression, if it has been repeated\noften enough and without any \u2018arbitrary\u2019 action on our part, is\nthe relation of cause and effect, and thus does \u2018persuade us of\nreal existence.\u2019 A \u2018feigned\u2019 contiguity, on the other hand, because\nwe are conscious that it is \u2018of our mere good-will and pleasure\u2019\nthat we give the idea that relation to the impression, can produce\nno belief. \u2018There is no reason why, upon the return of the same\nimpression, we should be determined to place the same object in the\nsame relation to it\u2019. [1] In like manner we must suppose (though this\nis not so clearly stated) that when an impression--such as the sight\nof a picture--calls up a resembling idea (that of the man depicted)\nwith much vivacity, it does not \u2018persuade us of his real existence\u2019\nbecause we are conscious that it is by the \u2018mere good-will and\npleasure\u2019 of some one that the likeness has been produced.\n\n[1] P. 409. [Book I, part III., sec. IX.]\n\nBut if so, \u2018system of judgment\u2019 must consist of feelings constantly\nexperienced;\n\n331. Now this account has the fault of being inconsistent with Hume\u2019s\nprimary doctrine, inasmuch as it makes the real an object of thought\nin distinction from feeling, without the merit of explaining the\nextension of knowledge beyond the objects of sense and memory. It\nturns upon a conception of the real, as the unchangeable, which the\nsuccession of feelings, in endless variety, neither is nor could\nsuggest. It implies that not in themselves, but as representing\nsuch an unchangeable, are the feelings which \u2018return on us whether\nwe will or no,\u2019 regarded as real. The peculiar sequence of one idea\non another, which is supposed to constitute the relation of cause\nand effect, is not, according to this description of it, a sequence\nof feelings simply; it is a sequence reflected on, found to be\nunchangeable, and thus to entitle the sequent idea to the prerogative\nof reality previously awarded (but only by the admission as real\nof the \u2018fiction\u2019 of distinct continued existence) to the system of\nmemory. But while the identification of the real with feeling is\nthus in effect abandoned, in saving the appearance of retaining it,\nHume makes his explanation of the \u2018system of judgment\u2019 futile for\nits purpose. He saves the appearance by intimating that the relation\nof cause and effect, by which the inferred idea is connected with\nthe idea of memory and derives reality from it, is only the repeated\nsequence of the one idea upon the other, of the less lively feelings\nupon the more lively, or a habit that results from such repetition.\nBut if the sequence of the inferred idea upon the other must have\nbeen so often repeated in order to the existence of the relation\nwhich renders the inference possible, the inferred idea can be no\nnew one, but must itself be an idea of memory, and the question, how\nany one\u2019s knowledge comes to extend beyond the range of his memory,\nremains unanswered.\n\n... which only differ from remembered feelings inasmuch as their\nliveliness has faded. But how can it have faded, if they have been\nconstantly repeated?\n\n332. What Hume himself seems to mean us to understand is, that the\ninferred idea is one of imagination, as distinct from memory; and\nthat the characteristic of the relation of cause and effect is that\nthrough it ideas of imagination acquire the reality that would\notherwise be confined to impressions of sense and memory. But,\naccording to him, ideas of imagination only differ from those of\nmemory in respect of their less liveliness, and of the freedom with\nwhich we can combine ideas in imagination that have not been given\ntogether as impressions. [1] Now the latter difference is in this\ncase out of the question. A compound idea of imagination, in which\nsimple ideas are put together that have never been felt together, can\nclearly never be connected with an impression of sense or memory by\na relation derived from constant experience of the sequence of one\nupon the other, and specially opposed to the creations of \u2018caprice\u2019.\n[2] We are left, then, to the supposition that the inferred idea,\nas idea of imagination, is one originally given as an impression\nof sense, but of which the liveliness has faded and requires to be\nrevived by association in the way of cause and effect with one that\nhas retained the liveliness proper to an idea of memory. Then the\nquestion recurs, how the restoration of its liveliness by association\nwith an impression, on which it must have been constantly sequent in\norder that the association may be possible, is compatible with the\nfact that its liveliness has faded. And however this question may be\ndealt with, if the relation of cause and effect is merely custom, the\nextension of knowledge by means of it remains unaccounted for; the\nbreach between the expectation of the recurrence of familiar feelings\nand inductive science remains unfilled; Locke\u2019s \u2018suspicion\u2019 that\n\u2018a science of nature is impossible,\u2019 instead of being overcome, is\nelaborated into a system.\n\n[1] Part I., sec. 3; cf. note on p. 416 [Book I, part III., sec. IX.].\n\n[2] P. 409. [Book I, part III., sec. IX.]\n\nInference then can give no new knowledge.\n\n333. Thus inference, according to Hume\u2019s account of it as originating\nin habit, suffers from a weakness quite as fatal as that which he\nsupposes to attach to it if accounted for as the work of reason.\n\u2018The work of reason\u2019 to a follower of Locke meant either the mediate\nperception of likeness between ideas, which the discovery of cause or\neffect cannot be; or else syllogism, of which Locke had shown once\nfor all that it could yield no \u2018instructive propositions.\u2019 But if an\nidea arrived at by that process could be neither new nor real--not\nnew, because we must have been familiar with it before we put it\ninto the compound idea from which we \u2018deduce\u2019 it; not real, because\nit has not the liveliness either of sensation or of memory--the\nidea inferred according to Hume\u2019s process, however real with the\nreality of liveliness, is certainly not new. \u2018If this means\u2019 (the\nmodern logician may perhaps reply), \u2018that according to Hume no new\nphenomenon can be given by inference, he was quite right in thinking\nso. If the object of inference were a separate phenomenon, it would\nbe quite true that it must have been repeatedly perceived before it\ncould be inferred, and that thus inference would be nugatory. But\ninference is in fact not to such an object, but to a uniform relation\nof certain phenomena in the way of co-existence and sequence; and\nwhat Hume may be presumed to mean is not that every such relation\nmust have been perceived before it can be inferred, much less that\nit must have been perceived so constantly that an appearance of the\none phenomenon causes instinctive expectation of the other, but _(a)_\nthat the phenomena themselves must have been given by immediate\nperception, and _(b)_ that the conception of a law of causation, in\nvirtue of which a uniformity of relation between them is inferred\nfrom a single instance of it, is itself the result of an \u201cinductio\nper enumerationem simplicem,\u201d of the accumulated experience of\ngenerations that the same sequents follow the same antecedents.\u2019\n\nNor does this merely mean that it cannot constitute new phenomena,\nwhile it can prove relations, previously unknown, between phenomena.\nSuch a distinction inadmissible with Hume.\n\n334. At the point which our discussion has reached, few words should\nbe wanted to show that thus to interpret Hume is to read into him\nan essentially alien theory, which has doubtless grown out of his,\nbut only by a process of adaptation which it needs a principle the\nopposite of his to justify. Hume, according to his own profession,\nknows of no objects but impressions and ideas--feelings stronger or\nmore faint--of no reality which it needs thought, as distinct from\nfeeling, to constitute. But a uniform relation between phenomena\nis neither impression nor idea, and can only exist for thought.\nHe could not therefore admit inference to such relation as to a\nreal existence, without a double contradiction, nor does he ever\nexplicitly do so. He never allows that inference is other than a\ntransition to a certain sort of feeling, or that it is other than\nthe work of imagination, the weakened sense, as enlivened by custom\nto a degree that puts it _almost_ on a level with sense; which\nimplies that in every case of inference the inferred object is\n_not_ a uniform relation--for how can there be an image of uniform\nrelation?--and that it _is_ something which has been repeatedly\nand without exception perceived to follow another before it can be\ninferred. Even when in violation of his principle he has admitted\na \u2018system of memory\u2019--a system of things which have been felt, but\nwhich are not feelings, stronger or fainter, and which are what they\nare only through relation--he still in effect, as we have seen, makes\nthe \u2018system of judgment,\u2019 which he speaks of as inferred from it,\nonly the double of it. To suppose that, on the strength of a general\ninference, itself the result of habit, in regard to the uniformity\nof nature, particular inferences may be made which shall be other\nthan repetitions of a sequence already habitually repeated, is, if\nthere can be degrees of contradiction, even more incompatible with\nHume\u2019s principles than to suppose such inferences without it. If\na uniformity of relation between particular phenomena is neither\nimpression nor idea, even less so is the system of all relations.\n\nHis distinction of probability of causes from that of chances might\nseem to imply conception of nature, as determining inference.\n\n335. There is language, however, in the chapters on \u2018Probability of\nChances and of Causes,\u2019 which at first sight might seem to warrant\nthe ascription of such a supposition to Hume. According to the\ndistinction which he inherited from Locke all inference to or from\ncauses or effects, since it does not consist in any comparison of the\nrelated ideas, should be merely probable. And as such he often speaks\nof it. His originality lies in his effort to explain what Locke had\nnamed; in his treating that \u2018something not joined on both sides\nto, and so not showing the agreement or disagreement of, the ideas\nunder consideration\u2019 which yet \u2018makes me believe\u2019, [1] definitely\nas Habit. But \u2018in common discourse,\u2019 as he remarks, \u2018we readily\naffirm that many arguments from causation exceed probability\u2019; [2]\nthe explanation being that in these cases the habit which determines\nthe transition from impression to idea is \u2018full and perfect.\u2019 There\nhas been enough past experience of the immediate sequence of the\none \u2018perception\u2019 on the other to form the habit, and there has been\nno exception to it. In these cases the \u2018assurance,\u2019 though distinct\nfrom knowledge, may be fitly styled \u2018proof,\u2019 the term \u2018probability\u2019\nbeing confined to those in which the assurance is not complete. Hume\nthus comes to use \u2018probability\u2019 as equivalent to incompleteness of\nassurance, and in this sense speaks of it as \u2018derived either from\nimperfect experience, or from contrary causes, or from analogy\u2019.\n[3] It is derived from analogy when the present impression, which\nis needed to give vivacity to the \u2018related idea,\u2019 is not perfectly\nlike the impressions with which the idea has been previously found\nunited; \u2018from contrary causes,\u2019 when there have been exceptions to\nthe immediate sequence or antecedence of the one perception to the\nother; \u2018from imperfect experience\u2019 when, though there have been no\nexceptions, there has not been enough experience of the sequence to\nform a \u2018full and perfect habit of transition.\u2019 Of this last \u2018species\nof probability,\u2019 Hume says that it is a kind which, \u2018though it\nnaturally takes place before any entire proof can exist, yet no one\nwho is arrived at the age of maturity can any longer be acquainted\nwith. \u2019Tis true, nothing is more common than for people of the most\nadvanced knowledge to have attained only an imperfect experience of\nmany particular events; which naturally produces only an imperfect\nhabit and transition; but then we must consider that the mind, having\nformed another observation concerning the connexion of causes and\neffects, gives new force to its reasoning from that observation;\nand by means of it can build an argument on one single experiment,\nwhen duly prepared and examined. What we have found once to follow\nfrom any object we conclude will for ever follow from it; and if\nthis maxim be not always built upon as certain, \u2019tis not for want of\na sufficient number of experiments, but because we frequently meet\nwith instances to the contrary\u2019--which give rise to the other sort of\nweakened assurance or probability, that from \u2018contrary causes\u2019. [4]\n\n[1] Locke, 4, 15, 3.\n\n[2] P. 423. [Book I, part III., sec. XI.]\n\n[3] P. 439. [Book I, part III., sec. XIII.]\n\n[4] Pp. 429 & 430. [Book I, part III., sec. XII.]\n\nBut this distinction he only professes to adopt in order to explain\nit away.\n\n336. There is a great difference between the meaning which the above\npassage conveys when read in the light of the accepted logic of\nscience, and that which it conveys when interpreted consistently with\nthe theory in the statement of which it occurs. Whether Hume, in\nwriting as he does of that conclusion from a single experiment, which\nour observation concerning the connexion of cause and effect enables\nus to draw, understood himself to be expressing his own theory or\nmerely using the received language provisionally, one cannot be\nsure; but it is certain that such language can only be justified by\nthose \u2018maxims of philosophers\u2019 which it is the purpose or effect of\nhis doctrine to explain away--in particular the maxims that \u2018the\nconnexion between all causes and effects is equally necessary and\nthat its seeming uncertainty in some instances proceeds from the\nsecret opposition of contrary causes;\u2019 and that \u2018what the vulgar\ncall chance is but a concealed cause\u2019. [1] These maxims represent\nthe notion that the law of causation is objective and universal;\nthat all seeming limitations to it, all \u2018probable and contingent\nmatter,\u2019 are the reflections of our ignorance, and exist merely _ex\nparte nostr\u00e2_. In other words, they represent the notion of that\n\u2018continued existence distinct from our perceptions,\u2019 which with Hume\nis a phrase generated by \u2018propensities to feign.\u2019 Yet he does not\nprofess to reject them; nay, he handles them as if they were his own,\nbut after a very little of his manipulation they are so \u2018translated\u2019\nthat they would not know themselves. Because philosophers \u2018allow\nthat what the vulgar call chance is nothing but a concealed cause,\u2019\n\u2018probability of causes\u2019 and \u2018probability of chances\u2019 may be taken as\nequivalent. But chance, as \u2018merely negation of a cause,\u2019 has been\npreviously explained, on the supposition that causation means a\n\u2018perfect habit of imagination,\u2019 to be the absence of such habit--the\nstate in which imagination is perfectly indifferent in regard to the\ntransition from a given impression to an idea, because the transition\nhas not been repeated often enough to form even the beginning of a\nhabit. Such being mere chance, \u2018probability of chances\u2019 means a state\nof imagination between the perfect indifference and that perfect\nhabit of transition, which is \u2018necessary connexion.\u2019 \u2018Probability\nof causes\u2019 is the same thing. Its strength or weakness depends\nsimply on the proportion between the number of experiments (\u2018each\nexperiment being a kind of chance\u2019) in which A has been found to\nimmediately follow B, and the number of those in which it has not.\n[2] Mere chance, probability, and causation then are equally states\nof imagination. The \u2018equal necessity of the connexion between all\ncauses and effects\u2019 means not that any \u2018law of causation pervades\nthe universe,\u2019 but that, unless the habit of transition between any\nfeelings is \u2018full and perfect,\u2019 we do not speak of these feelings as\nrelated in the way of cause and effect.\n\n[1] Pp. 429 & 430. [Book I, part III., sec. XII.]\n\n[2] Pp. 424-428, 432-434. [Book I, part III., secs. XI. and XII.]\n\nLaws of nature are unqualified habits of expectation.\n\n337. Interpreted consistently with this doctrine, the passage quoted\nin the last paragraph but one can only mean that, when a man has\narrived at maturity, his experience of the sequence of feelings\ncannot fail in quantity. He must have had experience _enough_ to form\nnot only a perfect habit of transition from any impression to the\nidea of its usual attendant, but a habit which would act upon us even\nin the case of novel events, and lead us after a single experiment\nor a sequence confidently to expect its recurrence, if only the\nexperience had been _uniform_. It is because it has not been so,\nthat in many cases the habit of transition is still imperfect, and\nthe sequence of A on B not \u2018proven,\u2019 but \u2018probable.\u2019 The probability\nthen which affects the imagination of the matured man is of the sort\nthat arises from \u2018contrary causes,\u2019 as distinct from \u2018imperfect\nexperience.\u2019 This is all that the passage in question can fairly\nmean. Such \u2018probability\u2019 cannot become \u2018proof,\u2019 or the \u2018imperfect\nhabit,\u2019 perfect, by _discovery_ of any necessary connexion or law\nof causation, for the perfect habit of transition, the imagination\nenlivened to the maximum by custom, _is_ the law of causation. The\nformation of the habit constitutes the law: to discover it would be\nto discover what does not yet exist. The incompleteness of the habit\nin certain directions, the limitation of our assurance to certain\nsequences as distinct from others, must be equally a limitation\nto the universality of the law. It is impossible then that on the\nfaith of the universality of the law we should seek to extend the\nrange of that assurance which is identical with it. Our \u2018observation\nconcerning the connexion of causes and effects\u2019 merely means the sum\nof our assured expectations, founded on habit, at any given time,\nand that on the strength of this we should \u2018prepare an experiment,\u2019\nwith a view to assuring ourselves of a universal sequence from a\nsingle instance, is as unaccountable as that, given the instance, the\nassurance should follow.\n\nExperience, according to his account of it, cannot be a parent of\nknowledge.\n\n338. The case then stands thus. In order to make the required\ndistinction between inference to real existence and the lively\nsuggestion of an idea, Hume has to graft on his theory the alien\nnotion of an objective system, an order of nature, represented by\nideas of memory, and on the strength of such a notion to interpret\na transition from these ideas to others, because we cannot help\nmaking it, as an objective necessity. Of such alien notion and\ninterpretation he avails himself in his definition (understood as he\nmeans it to be understood) of cause as a \u2018natural relation\u2019. [1] But\nhe had not the boldness of his later disciples. Though he could be\ninconsistent so far, he could not be inconsistent far enough to make\nhis theory of inference fit the practice of natural philosophers.\nBound by his doctrine of ideas as copied from impressions, he can\ngive no account of inferred ideas that shall explain the extension\nof knowledge beyond the expectation that we shall feel again what we\nhave felt already. It was not till another theory of experience was\nforthcoming than that given by the philosophers who were most fond of\ndeclaring their devotion to it, that the procedure of science could\nbe justified. The old philosophy, we are often truly told, had been\nbarren for want of contact with fact. It sought truth by a process\nwhich really consisted in evolving the \u2018connotation\u2019 of general\nnames. The new birth came when the mind had learnt to leave the idols\nof the tribe and cave, and to cleave solely to experience. If the\nold philosophy, however, was superseded by science, science itself\nrequired a new philosophy to answer the question. What constitutes\nexperience? It was in effect to answer this question that Locke\nand Hume wrote, and it is the condemnation of their doctrine that,\naccording to it, experience is not a possible parent of science. It\nis not those, we know, who cry \u2018Lord, Lord!\u2019 the loudest, that enter\ninto the kingdom of heaven, nor does the strongest assertion of our\ndependence on experience imply a true insight into its nature. Hume\nhas found acceptance with men of science as the great exponent of the\ndoctrine that there can be no new knowledge without new experience.\nIt has not been noticed that with him such \u2018new experience\u2019 could\nonly mean a further repetition of familiar feelings, and that if it\nmeans more to his followers, it is only because they have been less\nfaithful than he was to that antithesis between thought and reality\nwhich they are not less loud in asserting.\n\n[1] See above, paragraph 317.\n\nHis attitude towards doctrine of thinking substance.\n\n339. From the point that our enquiry has reached, we can anticipate\nthe line which Hume could not but take in regard to Self and God.\nHis scepticism lay ready to his hand in the incompatibility between\nthe principles of Locke and that doctrine of \u2018thinking substance,\u2019\nwhich Locke and Berkeley alike maintained. If the reader will revert\nto the previous part of this introduction, in which that doctrine\nwas discussed, [1] he will find it equally a commentary upon\nthose sections of the \u2018Treatise on Human Nature\u2019 which deal with\n\u2018immateriality of the soul\u2019 and \u2018personal identity.\u2019 Substance, we\nsaw, alike as \u2018extended\u2019 and as \u2018thinking,\u2019 was a \u2018creation of the\nmind,\u2019 yet real; something of which there was an \u2018idea,\u2019 but of which\nnothing could be said but that it was not an \u2018idea.\u2019 The \u2018thinking\u2019\nsubstance, moreover, was at a special disadvantage in contrast with\nthe \u2018extended,\u2019 because, in the first place, it could not, like body,\nbe represented as given to consciousness in the feeling of solidity,\nand secondly it was not wanted. It was a mere double of the extended\nsubstance to which, as the \u2018something wherein they do subsist and\nfrom which they do result\u2019 our ideas had already been referred.\nHaving no conception, then, of Spirit or Self before him but that\nof the thinking substance, of which Berkeley had confessed that\nit was not a possible idea or object of an idea, Hume had only to\napply the method, by which Berkeley himself had disposed of extended\nsubstance, to get rid of Spirit likewise. This could be done in a\nsentence, [2] but having done it, Hume is at further pains to show\nthat immateriality, simplicity, and identity cannot be ascribed to\nthe soul; as if there were a soul left to which anything could be\nascribed.\n\n[1] Above, paragraphs 127-135, 144-146, & 192.\n\n[2] P. 517. [Book I, part IV., sec. V.]\n\nAs to Immateriality of the Soul, he plays off Locke and Berkeley\nagainst each other, and proves Berkeley a Spinozist.\n\n340. There were two ways of conceiving the soul as immaterial, of\nwhich Hume was cognizant. One, current among the theologians and\nordinary Cartesians and adopted by Locke, distinguishing extension\nand thought as severally divisible and indivisible, supposed\nseparate substances--matter and the soul--to which these attributes,\nincapable of \u2018local conjunction,\u2019 severally belonged. The other,\nBerkeley\u2019s, having ostensibly reduced extended matter to a succession\nof feelings, took the exclusion of all \u2018matter\u2019 to which thought\ncould be \u2018joined\u2019 as a proof that the soul was immaterial. Hume,\nwith cool ingenuity, turns each doctrine to account against the\nother. From Berkeley he accepts the reduction of sensible things to\nsensations. Our feelings do not represent extended objects other\nthan themselves; but we cannot admit this without acknowledging the\nconsequence, as Berkeley himself implicitly did, [1] that certain of\nour impressions--those of sight and touch--are themselves extended.\nWhat then becomes of the doctrine, that the soul must be immaterial\nbecause thought is not extended, and cannot be joined to what is\nso? Thought means the succession of impressions. Of these some,\nthough the smaller number, are actually extended; and those that\nare not so are united to those that are by the \u2018natural relations\u2019\nof resemblance and of contiguity in time of appearance, and by the\nconsequent relation of cause and effect. [2] The relation of local\nconjunction, it is true, can only obtain between impressions which\nare alike extended. The ascription of it to such as are unextended\narises from the \u2018propensity in human nature, when objects are united\nby any relation, to add some new relation in order to complete the\nunion\u2019. [3] This admission, however, can yield no triumph to those\nwho hold that thought can only be joined to a \u2018simple and indivisible\nsubstance.\u2019 If the existence of unextended impressions requires the\nsupposition of a thinking substance \u2018simple and indivisible,\u2019 the\nexistence of extended ones must equally imply a thinking substance\nthat has all the properties of extended objects. If it is absurd\nto suppose that perceptions which are unextended can belong to a\nsubstance which is extended, it is equally absurd to suppose that\nperceptions which are extended can belong to a substance that is\nnot so. Thus Berkeley\u2019s criticism has indeed prevailed against the\nvulgar notion of a material substance as opposed to a thinking one,\nbut meanwhile he is himself \u2018hoist with his own petard.\u2019 If that\nthinking substance, the survival of which was the condition of his\ntheory serving its theological purpose, [4] is to survive at all,\nit can only be as equivalent to Spinoza\u2019s substance, in which \u2018both\nmatter and thought were supposed to inhere.\u2019 The universe of our\nexperience--\u2018the sun, moon, and stars; the earth, seas, plants,\nanimals, men, ships, houses, and other productions, either of art\nor nature\u2019--is the same universe when it is called \u2018the universe of\nobjects or of body,\u2019 and when it is called \u2018the universe of thought,\nor of impressions and ideas;\u2019 but to hold, according to Spinoza\u2019s\n\u2018hideous hypothesis,\u2019 that \u2018the universe of objects or of body\u2019\ninheres in one simple uncompounded substance, is to rouse \u2018a hundred\nvoices of scorn and detestation;\u2019 while the same hypothesis in regard\nto the \u2018universe of impressions and ideas\u2019 is treated \u2018with applause\nand veneration.\u2019 It was to save God and Immortality that the \u2018great\nphilosopher,\u2019 who had found the true way out of the scholastic\nabsurdity of abstract ideas, [5] had yet clung to the \u2018unintelligible\nchimaera\u2019 of thinking substance; and after all, in doing so, he fell\ninto a \u2018true atheism,\u2019 indistinguishable from that which had rendered\nthe unbelieving Jew \u2018so universally infamous\u2019. [6]\n\n[1] See above, par. 177.\n\n[2] Pp. 520-521. [Book I, part IV., sec. V.]\n\n[3] P. 521. [Book I, part IV., sec. V.]\n\n[4] See page 325. [Book I, part I., sec. VII.]\n\n[5] See above, paragraphs 191 and foll.\n\n[6] Pp. 523-526. [Book I, part IV., sec. V.]\n\nCausality of spirit treated in the same way.\n\n341. The supposition of spiritual substance being thus at once\nabsurd, and of a tendency the very opposite of the purpose it was\nmeant to serve, can anything better be said for the supposition of\na spiritual cause? It was to the representation of spirit as cause\nrather than as substance, it will be remembered, that both Locke\nand Berkeley trusted for the establishment of a Theism which should\nnot be Pantheism. [1] Locke, in his demonstration of the being of\nGod, trusted for proof of a first cause to the inference from that\nwhich begins to exist to something having power to produce it, and\nto the principle of necessary connexion--connexion in the way of\nagreement of ideas--between cause and effect for proof that this\nfirst cause must be immaterial, even as its effect, viz. our thought,\nis. Hume\u2019s doctrine of causation, of course, renders both sides of\nthe demonstration unmeaning. Inference being only the suggestion by\na feeling of the image of its \u2018usual attendant,\u2019 there can be no\ninference to that which is not a possible image of an impression.\nNor, since causation merely means the constant conjunction of\nimpressions, and there is no such contrariety between the impression\nwe call \u2018motion of matter\u2019 and that we call \u2018thought,\u2019 anymore than\nbetween any other impressions, [2] as is incompatible with their\nconstant conjunction, is there any reason why we should set aside the\nhourly experience, which tells us that bodily motions are the cause\nof thoughts and sentiments. If, however, there were that necessary\nconnexion between effect and cause, by which Locke sought to show the\nspirituality of the first cause, it would really go to show just the\nreverse of infinite power in such cause. It is from our impressions\nand ideas that we are supposed to infer this cause; but in these--as\nBerkeley had shown, and shown as his way of proving the existence of\nGod--there is no efficacy whatever. They are \u2018inert.\u2019 If then the\ncause must agree with the effect, the Supreme Being, as the cause\nof our impressions and ideas, must be \u2018inert\u2019 likewise. If, on the\nother hand, with Berkeley we cling to the notion that there must be\nefficient power somewhere, and having excluded it from the relation\nof ideas to each other or of matter to ideas, find it in the direct\nrelation of God to ideas, we fall \u2018into the grossest impieties;\u2019\nfor it will follow that God \u2018is the author of all our volitions and\nimpressions.\u2019 [3]\n\n[1] See above, \u00a7\u00a7 147, 171, 193.\n\n[2] There is no contrariety, according to Hume, except between\nexistence and non-existence (p. 323) [Book I, part I., sec. V.] and\nas all impressions and ideas equally exist (p. 394) [Book I, part\nIII., sec. VII.], there can be no contrariety between any of them.\nHe does indeed in certain leading passages allow himself to speak\nof contrariety between ideas (_e.g._ pp. 494 and 535 [Book I, part\nIV., secs. II. and VI.]), which is incidental evidence that the ideas\nthere treated of are not so, according to his account of ideas, at\nall.\n\n[3] Pp. 529-531 [Book I, part IV., sec. V.], a commentary on the\nargument here given has been in effect supplied in paragraphs\n148-152, and 194.\n\nDisposes of \u2018personal\u2019 identity by showing contradictions in Locke\u2019s\naccount of it.\n\n342. Against the doctrine of a real \u2018identity of the self or person\u2019\nHume had merely to exhibit the contradictions which Locke\u2019s own\nstatement of it involves. [1] To have transferred this identity\ndefinitely from \u2018matter\u2019 to consciousness was in itself a great\nmerit, but, so transferred, in the absence of any other theory of\nconsciousness than Locke\u2019s, it only becomes more obviously a fiction.\nIf there is nothing real but the succession of feelings, identity of\nbody, it is true, disappears as inevitably as identity of mind; and\nso we have already found it to do in Hume. [2] But whereas the notion\nof a unity of body throughout the succession of perceptions only\nbecomes contradictory through the medium of a reduction of body to\na succession of perceptions, the identity of a mind, which has been\nalready defined as a succession of perceptions, is a contradiction\nin terms. There can be \u2018properly no simplicity in it at one time,\nnor identity at different; it is a kind of theatre where several\nperceptions successively make their appearance.\u2019 But this comparison\nmust not mislead us. \u2018They are the successive perceptions only, that\nconstitute the mind; nor have we the most distant notion of the place\nwhere these scenes are represented, or of the materials of which it\nis composed.\u2019 The problem for Hume then in regard to personal, as\nit had been in regard to bodily, identity is to account for that\n\u2018natural propension to imagine\u2019 it which language implies.\n\n[1] See above, \u00a7\u00a7 134 and foll.\n\n[2] See above, \u00a7\u00a7 306 and foll.\n\nYet can only account for it as a \u2018fiction\u2019 by supposing ideas which\nwith him are impossible.\n\n343. The method of explanation in each case is the same. He starts\nwith two suppositions, to neither of which he is logically entitled.\nOne is that we have a \u2018distinct idea of identity or sameness,\u2019 _i.e._\nof an object that remains invariable and uninterrupted through a\nsupposed variation of time\u2019--a supposition which, as we have seen,\nupon his principles must mean that a feeling, which is one in a\nsuccession of feelings, is yet all the successive feelings at once.\nThe other is that we have an idea \u2018of several different objects\nexisting in succession, and connected together by a close\u2019 (natural)\n\u2018relation\u2019--which in like manner implies that a feeling, which is one\namong a succession of feelings, is at the same time a consciousness\nof these feelings as successive and under that qualification by\nmutual relation which implies their equal presence to it. These\ntwo ideas, which in truth are \u2018distinct and even contrary\u2019 [1] we\nyet come to confuse with each other, because \u2018that action of the\nimagination, by which we consider the uninterrupted and invisible\nobject, and that by which we reflect on the succession of related\nobjects, are almost the same to the feeling.\u2019 Thus, though what we\ncall our mind is really a \u2018succession of related objects,\u2019 we have a\nstrong propensity to mistake it for an \u2018invariable and uninterrupted\nobject.\u2019 To this propensity we at last so far yield as to assert our\nsuccessive perceptions to be in effect the same, however interrupted\nand variable; and then, by way of \u2018justifying to ourselves this\nabsurdity, feign the continued existence of the perceptions of our\nsenses, to remove the interruption; and run into the notion of a\n_soul_ and _self_, and _substance_, to disguise the variation\u2019. [2]\n\n[1] See note to \u00a7 341.\n\n[2] Pp. 535-536. [Book I, part IV., sec. VI.].\n\nIn origin this \u2018fiction\u2019 the same as that of \u2018Body\u2019.\n\n344. It will be seen that the theory, which we have just summarised,\nwould merely be a briefer version of that given in the section\non \u2018Scepticism with regard to the Senses,\u2019 if in the sentence,\nwhich states its conclusion, for \u2018the notion of a soul and self\nand substance\u2019 were written \u2018the notion of a double existence of\nperceptions and objects\u2019. [1] To a reader who has not thoroughly\nentered into the fusion of being and feeling, which belongs to the\n\u2018new way of ideas,\u2019 it may seem strange that one and the same process\nof so-called confusion has to account for such apparently disparate\nresults, as the notion of a permanently identical self and that of\nthe distinct existence of body. If he bears in mind, however, that\nwith Hume the universe of our experience is the same when it is\ncalled \u2018the universe of objects or of body\u2019 and when it is called\nthe \u2018universe of thought or my impressions and ideas\u2019, [2] he will\nsee that on the score of consistency Hume is to be blamed, not for\napplying the same method to account for the fictions of material\nand spiritual identity, but for allowing himself, in his preference\nfor physical, as against theological, pretension, to write as if\nthe supposition of spiritual were really distinct from that of\nmaterial identity, and might be more contemptuously disposed of. The\noriginal \u2018mistake,\u2019 out of which according to him the two fictitious\nsuppositions arise, is one and the same; and though it is a \u2018mistake\u2019\nwithout which, as we have found [3] from Hume\u2019s own admissions, we\ncould not speak even in singular propositions of the most ordinary\n\u2018objects of sense\u2019--this pen, this table, this chair--it is yet one\nthat on his principles is logically impossible, since it consists\nin a confusion between ideas that we cannot have. Of this original\n\u2018mistake\u2019 the fictions of body and of its \u2018continued and distinct\nexistence\u2019 are but altered expressions. They represent in truth the\nsame logical category of substance and relation. And of the Self\naccording to Locke\u2019s notion of it [4] (which was the only one that\nHume had in view), as a \u2018thinking thing\u2019 within each man among a\nmultitude of other thinking things, the same would have to be said.\nBut in order to account for the \u2018mistake,\u2019 of which the suppositions\nof thinking and material substance are the correlative expressions,\nand which it is the net result of Hume\u2019s speculation to exhibit at\nonce as necessary and as impossible, we have found another notion\nof the self forced upon us--not as a double of body, but as the\nsource of that \u2018familiar theory\u2019 which body in truth is, and without\nwhich there would be no universe of objects, whether \u2018bodies\u2019 or\n\u2018impressions and ideas,\u2019 at all.\n\n[1] Above, \u00a7\u00a7 306-310.\n\n[2] Above, \u00a7 340.\n\n[3] Above, \u00a7\u00a7 303 & 304.\n\n[4] Above, \u00a7\u00a7 129-132.\n\nPossibility of such fictitious ideas implies refutation of Hume\u2019s\ndoctrine.\n\n345. Thus the more strongly Hume insists that \u2018the identity which\nwe ascribe to the mind of man is only a fictitious one\u2019, [1] the\nmore completely does his doctrine refute itself. If he had really\nsucceeded in reducing those \u2018invented\u2019 relations, which Locke had\nimplicitly recognised as the framework of the universe, to what\nhe calls \u2018natural\u2019 ones--to mere sequences of feeling--the case\nwould have been different. With the disappearance of the conception\nof the world as a system of related elements, the necessity of a\nthinking subject, without whose presence to feelings they could\nnot become such elements, would have disappeared likewise. But he\ncannot so reduce them. In all his attempts to do so we find that\nthe relation, which has to be explained away, is pre-supposed under\nsome other expression, and that it is \u2018fictitious\u2019 not in the sense\nwhich Hume\u2019s theory requires--the sense, namely, that there is no\nsuch thing either really or in imagination, either as impression or\nidea--but in the sense that it would not exist if we did not think\nabout our feelings. Thus, whereas identity ought for Hume\u2019s purpose\nto be either a \u2018natural relation,\u2019 or a propensity arising from\nsuch relation, or nothing, we find that according to his account,\nthough neither natural relation nor propensity, it yet exists both as\nidea and as reality. He saves appearances indeed by saying [2] that\nnatural relations of ideas \u2018produce it,\u2019 but they do so, according\nto his detailed account of the matter, in the sense that, the idea\nof an identical object being given, we mistake our successive and\nresembling feelings for such an object. In other words, the existence\nof numerically identical things is a \u2018fiction,\u2019 not as if there\nwere no such things, but because it implies a certain operation of\nthought upon our feelings, a certain interpretation of impressions\nunder direction of an idea not derived from impressions. By a\nlike equivocal use of \u2018fiction\u2019 Hume covers the admission of real\nidentity in its more complex forms--the identity of a mass, whose\nparts undergo perpetual change of distribution; of a body whose form\nsurvives not merely the redistribution of its materials, but the\nsubstitution of others; of animals and vegetables, in which nothing\nbut the \u2018common end\u2019 of the changing members remains the same. The\nreality of such identity of mass, of form, of organism, he quietly\ntakes for granted. [3] He calls it \u2018fictitious\u2019 indeed, but only\neither in the sense above given or in the sense that it is mistaken\nfor mere numerical identity.\n\n[1] P. 540. [Book I, part IV., sec. VI.]\n\n[2] P. 543 [Book I, part IV., sec. VI.]. \u2018Identity depends on the\nrelations of ideas; and these relations produce identity by means\nof that easy transition they occasion.\u2019 Strictly it should be \u2018that\neasy transition in which they consist;\u2019 since, according to Hume, the\n\u2018easiness of transition\u2019 is not an effect of natural relation, but\nconstitutes it. Cf. pp. 322 & 497 [Book I, part I., sec. V. and part\nIV., sec. II.], and above, \u00a7 318.\n\n[3] Pp. 536-538. [Book I, part IV., sec. VI.]\n\n346. After he has thus admitted, as constituents of the \u2018universe\nof objects,\u2019 a whole hierarchy of ideas of which the simplest must\nvanish before the demand to \u2018point out the impression from which\nit is derived,\u2019 we are the less surprised to find him pronouncing\nin conclusion \u2018that the true idea of the human mind is to consider\nit as a system of different perceptions or different existences,\nwhich are linked together by the relation of cause and effect, and\nmutually produce, destroy, influence and modify each other\u2019. [1] A\nbetter definition than this, as a _definition of nature_, or one more\ncharged with \u2018fictions of thought,\u2019 could scarcely be desired. If\nthe idea of such a system is a true idea at all, which we are only\nwrong in confusing with mere numerical identity, we need be the less\nconcerned that it should be adduced as the true idea not of nature\nbut of the \u2018human mind.\u2019 Having learnt, through the discipline which\nHume himself furnishes, that the recognition of a system of nature\nlogically carries with it that of a self-conscious subject, in\nrelation to which alone \u2018different perceptions\u2019 become a system of\nnature, we know that we cannot naturalise the \u2018human mind\u2019 without\npresupposing that which is neither nature nor natural, though apart\nfrom it nature would not be--that of which the designation as \u2018mind,\u2019\nas \u2018human,\u2019 as \u2018personal,\u2019 is of secondary importance, but which is\neternal, self-determined, and thinks.\n\n[1] P. 541. [Book I, part IV., sec. VI.]\n\nT.H. Green\n\n\n\n\nGENERAL INTRODUCTION TO VOL. II.\n\nHume\u2019s doctrine of morals parallel to his doctrine of nature.\n\n1. In his speculation on morals, no less than on knowledge, Hume\nfollows the lines laid down by Locke. With each there is a precise\ncorrespondence between the doctrine of nature and the doctrine of\nthe good. Each gives an account of reason consistent at least in\nthis that, as it allows reason no place in the constitution of real\nobjects, so it allows it none in the constitution of objects that\ndetermine desire and, through it, the will. With each, consequently,\nthe \u2018moral faculty,\u2019 whether regarded as the source of the judgments\n\u2018ought and ought not,\u2019 or of acts to which these judgments are\nappropriate, can only be a certain faculty of feeling, a particular\nsusceptibility of pleasure and pain. The originality of Hume lies in\nhis systematic effort to account for those objects, apparently other\nthan pleasure and pain, which determine desire, and which Locke had\ntaken for granted without troubling himself about their adjustment to\nhis theory, as resulting from the modification of primary feelings\nby \u2018associated ideas.\u2019 \u2018Natural relation,\u2019 the close and uniform\nsequence of certain impressions and ideas upon each other, is the\nsolvent by which in the moral world, as in the world of knowledge, he\ndisposes of those ostensibly necessary ideas that seem to regulate\nimpressions without being copied from them; and in regard to the one\napplication of it as much as to the other, the question is whether\nthe efficiency of the solvent does not depend on its secretly\nincluding the very ideas of which it seems to get rid.\n\nIts relation to Locke: Locke\u2019s account of freedom, will, and desire.\n\n2. The place held by the \u2018Essay concerning Human Understanding,\u2019 as\na sort of philosopher\u2019s Bible in the last century, is strikingly\nillustrated by the effect of doctrines that only appear in it\nincidentally. It does not profess to be ethical treatise at all, yet\nthe moral psychology contained in the chapter \u2018of Power\u2019 (II. 21),\nand the account of moral good and evil contained In the chapter \u2018of\nother Relations\u2019 (II. 28) furnished the text for most of the ethical\nspeculation that prevailed in England, France, and Scotland for a\ncentury later. If Locke\u2019s theory was essentially a reproduction of\nHobbes\u2019, it was yet in the form he gave it that it survived while\nHobbes was decried and forgotten. The chapter on Power is in effect\nan account of determination by motives. More, perhaps, than any other\npart of the essay it bears the marks of having been written \u2018currente\ncalamo.\u2019 In the second edition a summary was annexed which differs\nsomewhat in the use of terms, but not otherwise, from the original\ndraught. The main course of thought, however, is clear throughout.\nWill and freedom are at first defined in all but identical terms\nas each a \u2018power to begin or forbear action barely by a preference\nof the mind\u2019 (\u00a7\u00a7 5, 8, 71). Nor is this identification departed\nfrom, except that the term \u2018will\u2019 is afterwards restricted to the\n\u2018preference\u2019 or \u2018power of preference,\u2019 while freedom is confined to\nthe power of acting upon preference; in which sense it is pointed\nout that though there cannot be freedom without will, there may be\nwill without freedom, as when, through the breaking of a bridge, a\nman cannot help falling into the water, though he prefers not to do\nso. \u2018Freedom\u2019 and \u2018will\u2019 being thus alike powers, if not the same\npower, it is as improper to ask whether the will is free as whether\none power has another power. The proper question is whether man is\nfree (\u00a7\u00a7 14, 21), and the answer to this question, according to\nLocke, is that within certain limits he is free to act, but that he\nis not free to will. When in any case he has the option of acting\nor forbearing to act, he cannot help preferring, _i.e._ willing,\none or other alternative. If it is further asked, What determines\nthe will or preference? the answer is that \u2018nothing sets us upon\nany new action but some uneasiness\u2019 (\u00a7 29), viz., the \u2018most urgent\nuneasiness we at any time feel\u2019 (\u00a7 40), which again is always \u2018the\nuneasiness of desire fixed on some absent good, either negative, as\nindolence to one in pain, or positive, as enjoyment of pleasure.\u2019 In\none sense, indeed, it may be said that the will often runs counter\nto desire, but this merely means that we \u2018being in this world beset\nwith sundry uneasinesses, distressed with different desires,\u2019 the\ndetermination of the will by the most pressing desire often implies\nthe counteraction of other desires which would, indeed, under other\ncircumstances, be the most pressing, but at the particular time of\nthe supposed action are not so.\n\nTwo questions: Does man always act from the strongest motive? and,\nWhat constitutes his motive? The latter the important question:\nDistinction between desires that are, and those that are not,\ndetermined by the conception of self.\n\n3. So far Locke\u2019s doctrine amounts to no more than this, that action\nis always determined by the strongest motive; and only those who\nstrangely hold that human freedom is to be vindicated by disputing\nthat truism will care to question it. To admit that the strongest\ndesire always moves action (there being, in fact, no test of its\nstrength but its effect on action) and that, since every desire\ncauses uneasiness till it is satisfied, the strongest desire is\nalso the most pressing uneasiness, [1] is compatible with the most\nopposite views as to the constitution of the objects which determine\ndesire. To understand that it is this constitution of the desired\nobject, not any possible intervention of unmotived willing between\nthe presentation of a strongest motive and action, which forms the\ncentral question of ethics, is the condition of all clear thinking\non the subject. It is a question, however, which Locke ignores, and\npopular philosophy, to its great confusion, has not only continued\nto do the same, but would probably resent as pedantic any attempt\nat more accurate analysis. When we hear of the strongest \u2018desire\u2019\nbeing the uniform motive to action, we have to ask, in the first\nplace, whether the term is confined to impulses determined by a prior\nconsciousness, or is taken to include those impulses, commonly called\n\u2018mere appetites,\u2019 which are not so determined, but depend directly\nand solely on the \u2018constitution of our bodily organs.\u2019 The _appetite_\nof hunger is obviously quite independent of any remembrance of the\npleasure of eating, yet nothing is commoner than to identify with\nsuch simple appetite the desire determined by consciousness of some\nsort, as when we say of a drunkard, who never drinks merely because\nhe is thirsty, that he is governed by his appetite. Upon this\ndistinction, however, since it is recognised by current psychology,\nit is less important to insist than on that between the kinds of\nprior consciousness which may determine desire proper. Does this\nprior consciousness consist simply in the return of an image of\npast pleasure with consequent hope of its renewal, or is it a\nconception--the thought of an object under relations to self or of\nself in relation to certain objects--in a word, self-consciousness as\ndistinct from simple feeling?\n\n[1] Locke\u2019s language in regard to \u2018the most pressing uneasiness\u2019\nwill not be found uniformly consistent. His usual doctrine is that\nthe strength of a desire, as evinced by the resulting action, and\nthe uneasiness which it causes are in exact proportion to each\nother. According to this view, desire for future happiness can only\nbecome a prevalent motive when the uneasiness which it causes has\ncome to outweigh every other (Cf. Chap, xxi., Secs. 43 and 45). On\nthe other hand, he sometimes seems to distinguish the desire for\nfuture pleasure from present uneasiness, while at the same time\nimplying that it may be a strongest motive (Cf. sec. 65). But if so,\nit follows that there may be a strongest desire which is not the\nmost pressing uneasiness. (See below, sec. 13.) Hume, distinguishing\nstrong from violent desires, and restricting \u2018uneasiness\u2019 to the\nlatter, is able to hold that it is not alone the present uneasiness\nwhich determines action. (Book II., part 3, sec. 3, sub fin.)\n\nEffect of this conception on the objects of human desire.\n\n4. Of desire determined in the former way we have experience, if at\nall, in those motives which actuate us, as we say, \u2018unconsciously\u2019;\nwhich means, without our attending to them--feelings which we do\nnot fix even momentarily by reference to self or to a thing. As we\ncannot set ourselves to recall such feelings without thinking them,\nwithout determining them by that reference to self which we suppose\nthem to exclude, they cannot be described; but some of our actions\n(such as the instinctive recurrence to a sweet smell), seem only to\nbe thus accounted for, and probably those actions of animals which\ndo not proceed from appetite proper are to be accounted for in the\nsame way. But whether such actions are facts in human experience or\nno, those which make us what we are as men are not so determined.\nThe man whom we call the slave of his appetite, the enlightened\npleasure-hunter, the man who lives for his family, the artist, the\nenthusiast for humanity, are alike in this, that the desire which\nmoves their action is itself determined not by the recurring image\nof a past pleasure, but by the conception of self. The self may be\nconceived of simply as a subject to be pleased, or may be a subject\nof interests, which, indeed, when gratified, produce pleasure but\nare not produced by it--interests in persons, in beautiful things,\nin the order of nature and society--but self is still not less the\n\u2018punctum stans\u2019 whose presence to each passing pleasure renders it\na constituent of a happiness which is to be permanently pursued,\nthan it is the focus in which the influences of that world which\nonly self-conscious reason could constitute--the world of science,\nof art, of human society--must be regathered in order to become the\npersonal interests which move the actions of individuals. It is in\nthis self-consciousness involved in our motives, in that conversion\ninto a conception by reference to self, which the image even of the\nmerest animal pleasure must undergo before it can become an element\nin the formation of character, that the possibility of freedom lies.\nWithout it we should be as sinless and as unprogressive, as free from\nremorse and aspiration, as incapable of selfishness and self-denial\nas the animals. Each pleasure would be taken as it came. We should\nhave \u2018the greatest happiness of which our nature is capable,\u2019 without\npossibility of asking ourselves whether we might not have had more.\nIt is only the conception of himself as a permanent subject to be\npleased that can set man upon the invention of new pleasures, and\nthen, making each pleasure a disappointment when it comes, produce\nthe \u2018vicious\u2019 temper; only this that can suggest the reflection\nhow much more pleasure he might have had than he has had, and thus\nproduce what the moralists know as \u2018cool selfishness\u2019; only this,\non the other hand, which, as \u2018enlightened self-love,\u2019 perpetually\nbalances the attraction of imagined pleasure by the calculation\nwhether it will be good for one as a whole. Nor less is it the\nconception of self, with a \u2018matter\u2019 more adequate to its \u2018form,\u2019\ntaking its content not from imagined pleasure, but from the work of\nreason in the world of nature and humanity, which determines that\npersonal devotion to a work or a cause, to a state, a church, or\nmankind, which we call self-sacrifice.\n\nObjects so constituted Locke should consistently exclude: But he\nfinds room for them by treating every desire for an object, of which\nthe attainment gives pleasure, as a desire for pleasure.\n\n5. If, now, we ask ourselves whether Locke recognised this function\nof reason, as self-consciousness, in the determination of the will,\nthe answer must be yes and no. His cardinal doctrine, as we have\nsufficiently seen, forbade him to admit that reason or thought could\noriginate an object. The only possible objects with him are either\nsimple ideas or resoluble into these, and the simple idea, as that\nwhich we receive in pure passivity, is virtually feeling. Now no\ncombination of feelings (supposing it possible [1]) can yield the\nconception of self as a permanent subject even of pleasure, much\nless as a subject of social claims. It cannot, therefore, yield the\nobjects, ranging from sensual happiness to the moral law, humanity,\nand God, of which this conception is the correlative condition.\nThus, strictly taken, Locke\u2019s doctrine excludes every motive to\naction, but appetite proper and such desire as is determined by the\nimagination of animal pleasure or pain, and in doing so renders vice\nas well as virtue unaccountable--the excessive pursuit of pleasure as\nwell as that dissatisfaction with it which affords the possibility\nof ordinary reform. On the other hand, the same happy intellectual\nunscrupulousness, which we have traced in his theory of knowledge,\nattends him also here. Just as he is ready on occasion to treat\nany conceived object that determines sense as if it were itself a\nsensation, so he is ready to treat any object that determines desire,\nwithout reference to the work of thought in its construction, as if\nit were itself the feeling of pleasure, or of uneasiness removed,\nwhich arises upon satisfaction of the desire. In this way, without\nprofessedly admitting any motive but remembered pleasure--a motive\nwhich, if it were our only one, would leave \u2018man\u2019s life as cheap as\nbeasts\u2019\u2019--he can take for granted any objects of recognised interest\nas accounting for the movement of human life, and as constituents of\nan utmost possible pleasure which it is his own fault if every one\ndoes not pursue.\n\n[1] Cf. Introduction to Vol. I., \u00a7\u00a7 215 and 247.\n\nConfusion covered by calling \u2018happiness\u2019 the general object of desire.\n\n6. The term \u2018happiness\u2019 is the familiar cover for confusion between\nthe animal imagination of pleasure and the conception of personal\nwell-being. It is so when--having raised the question. What moves\ndesire?--Locke answers, \u2018happiness, and that alone.\u2019 What, then,\nis happiness? \u2018Good and evil are nothing but pleasure and pain,\u2019\nand \u2018happiness in its full extent is the utmost pleasure we are\ncapable of.\u2019 [1] This is \u2018the proper object of desire in general,\u2019\nbut Locke is careful to explain that the happiness which \u2018moves\nevery particular man\u2019s desire\u2019 is not the full extent of it, but\n\u2019so much of it as is considered and taken to make a necessary part\nof his happiness.\u2019 It is that \u2018wherewith he in his present thoughts\ncan satisfy himself.\u2019 Happiness in this sense \u2018every one constantly\npursues,\u2019 and without possibility of error; for \u2018as to present\npleasure the mind never mistakes that which is really good or\nevil.\u2019 Every one \u2018knows what best pleases him, and that he actually\nprefers.\u2019 That which is the greater pleasure or the greater pain\nis really just as it appears (Ibid. \u00a7\u00a7 43, 58, 63). Now in these\nstatements, if we look closely, we shall find that four different\nmeanings of happiness are mixed up, which we will take leave to\ndistinguish by letters--_(a)_ happiness as an abstract conception,\nthe sum of possible pleasure; _(b)_ happiness as equivalent to the\npleasure which at any time survives most strongly in imagination;\n_(c)_ happiness as the object of the self-conscious pleasure-seeker;\n_(d)_ happiness as equivalent to any object at any time most\nstrongly desired, not really a pleasure, but by Locke identified\nwith happiness in sense _(b)_ through the fallacy of supposing that\nthe pleasure which arises on satisfaction of any desire, great in\nproportion to the strength of the desire, is itself the object which\nexcites desire.\n\n[1] Ibid., sec. 42, and cap. 28, sec. 5.\n\n\u2018Greatest sum of pleasure\u2019 and \u2018Pleasure in general\u2019 unmeaning\nexpressions.\n\n7. Happiness \u2018in its full extent,\u2019 as \u2018the utmost pleasure we are\ncapable of,\u2019 is an unreal abstraction if ever there was one. It\nis curious that those who are most forward to deny the reality of\nuniversals, in that sense in which they are the condition of all\nreality, viz., as relations, should yet, having pronounced these\nto be mere names, be found ascribing reality to a universal, which\ncannot without contradiction be supposed more than a name. Does\nthis \u2018happiness in its full extent\u2019 mean the \u2018aggregate of possible\nenjoyments,\u2019 of which modern utilitarians tell us? Such a phrase\nsimply represents the vain attempt to get a definite by addition\nof indefinites. It has no more meaning than \u2018the greatest possible\nquantity of time\u2019 would have. Pleasant feelings are not quantities\nthat can be added. Each is over before the next begins, and the man\nwho has been pleased a million times is not really better off--has\nno more of the supposed chief good in possession--than the man who\nhas only been pleased a thousand times. When we speak of pleasures,\nthen, as forming a possible whole, we cannot mean pleasures as\nfeelings, and what else do we mean? Are we, then, by the \u2018happiness\u2019\nin question to understand pleasure _in general_, as might be inferred\nfrom Locke\u2019s speaking of it as the \u2018object of desire _in general_\u2019?\nBut it is in its mere particularity that each pleasure has its being.\nIt is a simple idea, and therefore, as Locke and Hume have themselves\ntaught us, momentary, indefinable, in \u2018perpetual flux,\u2019 changing\nevery moment upon us. Pleasure _in general_, therefore, is not\npleasure, and it is nothing else. It is not a conceived reality, as a\nrelation, or a thing determined by relations, is, since pleasure as\nfeeling, in distinction from its conditions which are not feelings,\nfor the same reason that it cannot be defined, cannot be conceived.\nIt is a mere name which utilitarian philosophy has mistaken for\na thing; but for which--since no one, whatever his theory of the\ndesirable, can actually desire either the abstraction of pleasure\nin general or the aggregate of possible pleasures--a practical\nsubstitute is apt to be found in any lust of the flesh that may for\nthe time be the strongest.\n\nIn what sense of happiness is it true that it \u2018is really just as it\nappears?\u2019 In what sense that it is every one\u2019s object?\n\n8. Having begun by making this fiction \u2018the proper object of\ndesire in general,\u2019 Locke saves the appearance of consistency by\nrepresenting the particular pleasure or removal of uneasiness, which\nhe in fact believed to be the object of every desire, as if it were a\ncertain part of the \u2018full extent of happiness\u2019 which the individual,\nhaving this full extent before him, picked out as being what \u2018in\nhis present thoughts would satisfy him.\u2019 Nor does he ever give up\nthe notion of a \u2018happiness in general,\u2019 in distinction from the\nhappiness of each man\u2019s actual choice, as a possible motive, which\na man who finds himself wretched in consequence of his actions may\nbe told that he ought to have adopted. His real notion, however, of\nthe happiness which is motive to action is a confused result of the\nthree other notions of happiness, distinguished above as _(b)_, _(c)_\nand _(d)_. As that about which no one can be mistaken, \u2018happiness\u2019\ncan only be so in sense _(b)_, as the \u2018pleasure which survives most\nstrongly in imagination.\u2019 Of this it can be said truly, and of this\nonly, that \u2018it really is just as it appears,\u2019 and that \u2018a man never\nchooses amiss\u2019 since he must \u2018know what best pleases him.\u2019 But with\nthis, almost in the same breath, Locke confuses \u2018happiness\u2019 in senses\n_(c)_ and _(d)_. So soon as it is said of an object that it is\n\u2018taken by the individual to make a necessary part of his happiness,\u2019\nit is implied that it is determined by his conception of self. It\nis something which, as the result of the action of this conception\non his past experience, he has come to present to himself as a\nconstituent of his personal good. Unless he were conscious of himself\nas a permanent subject, he could have no conception of happiness as a\nwhole from relation to which each present object takes its character\nas a part. Nor of the objects determined by this relation is it true,\nas Locke says, that they are always pleasures, or that they \u2018are\nreally just as they appear.\u2019 Our readiness to accept his statements\nto this effect, is at bottom due to a confusion between the pleasure,\nor removal of uneasiness, incidental to the satisfaction of a desire\nand the object which excites the desire. If having explained desire,\nas Locke does, by reference to the good, we then allow ourselves to\nexplain the good by reference to desire, it will indeed be true that\nno man can be mistaken as to his present good, but only in the sense\nof the identical proposition that every man most desires what he does\nmost desire; and true also, that every attained good is pleasure,\nbut only in the sense that what satisfies desire does satisfy it.\nThe man of whom it could be truly said, in any other sense than\nthat of the above identical proposition, that his only objects of\ndesire--the only objects which he \u2018takes to make a necessary part\nof his happiness\u2019--were pleasures, would be a man, as we say, of no\ninterests. He would be a man who either lived simply for pleasures\nincidental to the satisfaction of animal appetite, or one who, having\nbeen interested in certain objects in which reason alone enables us\nto be interested--_e.g._, persons, pursuits, or works of art--and\nhaving found consequent pleasure, afterwards vainly tries to get the\npleasure without the interests. To the former type of character, of\ncourse, the approximations are numerous enough, though it may be\ndoubted whether such an ideal of sensuality is often fully realised.\nThe latter in its completeness, which would mean a perfect misery\nthat could only issue in suicide, would seem to be an impossibility,\nthough it is constantly being approached in proportion to the\nunworthiness and fleetingness of the interests by which men allow\nthemselves to be governed, and which, after stimulating an indefinite\nhunger for good, leave it without an object to satisfy it; in\nproportion, too, to the modern habit of hugging and poring over the\npleasures which our higher interests cause us till these interests\nare vitiated, and we find ourselves in restless and hopeless pursuit\nof the pleasure when the interest which might alone produce it is\ngone.\n\nNo real object of human desire can ever be just as it appears.\n\n9. Just as it is untrue, then, of the object of desire, as \u2018taken\nto be part of one\u2019s happiness\u2019 or determined by the conception of\nself, that it is always a pleasure, so it is untrue that it is always\nreally just as it appears, except in the trifling sense that what\nis most strongly desired is most strongly desired. Rather it is\nnever really what it appears. It is least of all so to the professed\npleasure-seeker. Obviously, to the man who seeks the pleasure\nincidental to interests which he has lost, there is a contradiction\nin his quest which for ever prevents what seems to him desirable\nfrom satisfying his desire. And even the man who lives for merely\nanimal pleasure, just because he seeks it as part of a happiness,\nnever finds it to be that which he sought. There is no mistake about\nthe pleasure, but he seeks it as that which shall satisfy him, and\nsatisfy him, since he is not an animal, it cannot. Nor are our higher\nobjects of desire ever what they seem. That is too old a topic with\npoets and moralizers to need enforcing. Each in its turn, we know,\npromises happiness when it shall have been attained, but when it\nis attained the happiness has not come. The craving for an object\nadequate to oneself, which is the source of the desire, is still\nnot quenched; and because it is not, nor can be, even \u2018the joy of\nsuccess\u2019 has its own bitterness.\n\nCan Locke consistently allow the distinction between true happiness\nand false? Or responsibility?\n\n10. The case, then, stands thus. Locke, having too much \u2018common\nsense\u2019 to reduce all objects of desire to the pleasures incidental to\nsatisfactions of appetite, takes for granted any number of objects\nwhich only reason can constitute (or, in other words, which can only\nexist for a self-conscious subject) without any question as to their\norigin. It is enough for him that they are not conscious inventions\nof the individual, and that they are related to feeling--though\nrelated as determining it. This being so, they are to him no more\nthe work of thought than are the satisfactions of appetite. The\nconception of them is of a kind with the simple remembrance or\nimagination of pleasures caused by such satisfactions. The question\nhow, if only pleasure is the object of desire, they came to be\ndesired before there had been experience of the pleasures incidental\nto their attainment, is virtually shelved by treating these latter\npleasures as if they were themselves the objects originally desired.\nSo far consistency at least is saved. No object but feeling, present\nor remembered, is ostensibly admitted within human experience. But\nmeanwhile, alongside of this view, comes the account of the strongest\nmotive as determined by the conception of self--as something which\na man \u2018takes to be a necessary part of his happiness,\u2019 and which he\nis \u2018answerable to himself\u2019 for so taking. The inconsistency of such\nlanguage with the view that every desired object must needs be a\npleasure, would have been less noticeable if Locke himself had not\nfrankly admitted, as the corollary of this view that the desired good\n\u2018is really just as it appears.\u2019 The necessity of this admission has\nalways been the rock on which consistent Hedonism has broken. Locke\nhimself has scarcely made it when he becomes aware of its dangerous\nconsequences, and great part of the chapter on Power is taken up by\nawkward attempts to reconcile it with the distinction between true\nhappiness and false, and with the existence of moral responsibility.\nIf greatest pleasure is the only possible object, and the production\nof such pleasure the only possible criterion of action, and if \u2018as\nto present pleasure and pain the mind never mistakes that which is\nreally good or evil,\u2019 with what propriety can any one be told that\nhe might or that he ought to have chosen otherwise than he has done?\n\u2018He has missed the true good,\u2019 we say, \u2018which he might and should\nhave found\u2019; but \u2018good,\u2019 according to Locke, is only pleasure, and\npleasure, as Locke in any other connexion would be eager to tell\nus, must mean either some actual present pleasure or a series of\npleasures of which each in turn is present. If every one without\npossibility of mistake has on each occasion chosen the greatest\npresent pleasure, how can the result for him at any time be other\nthan the true good, _i.e._, the series of greatest pleasures, each in\nits turn present, that have been hitherto possible for him?\n\nObjections to the Utilitarian answer to these questions.\n\n11. A modern utilitarian, if faithful to the principle which excludes\nany test of pleasure but pleasure itself, will probably answer that\nevery one does attain the maximum of pleasure possible for him,\nhis character and circumstances being what they are; but that with\na change in these his choice would be different. He would still\nchoose on each occasion the greatest pleasure of which he was then\ncapable, but this pleasure would be one \u2018truer\u2019--in the sense of\nbeing more intense, more durable, and compatible with a greater\nquantity of other pleasures--than is that which he actually chooses.\nBut admitting that this answer justifies us in speaking of any sort\nof pleasure as \u2018truer\u2019 than that at any time chosen by any one--which\nis a very large admission, for of the intensity of any pleasure we\nhave no test but its being actually preferred, and of durability\nand compatibility with other pleasures the tests are so vague that\na healthy and unrepentant voluptuary would always have the best of\nit in an attempt to strike the balance between the pleasures he has\nactually chosen and any truer sort--it still only throws us back on\na further question. With a better character, it is said, such as\nbetter education and improved circumstances might have produced, the\nactually greatest happiness of the individual--_i.e._, the series of\npleasures which, because he has chosen them, we know to have been\nthe greatest possible for him--might have been greater or \u2018truer.\u2019\nBut the man\u2019s character is the result of his previous preferences;\nand if every one has always chosen the greatest pleasure of which\nhe was at the time capable, and if no other motive is possible,\nhow could any other than his actual character have been produced?\nHow could that conception of a happiness truer than the actual, of\nsomething that should be most pleasant, and therefore preferred,\nthough it is not--a conception which all education implies--have been\na possible motive among mankind? To say that the individual is, to\nbegin with, destitute of such a conception, but acquires it through\neducation from others, does not remove the difficulty. How do the\neducators come by it? Common sense assumes them to have found out\nthat more happiness might have been got by another than the merely\nnatural course of living, and to wish to give others the benefit\nof their experience. But such experience implies that each has a\nconception of himself as other than the subject of a succession of\npleasures, of which each has been the greatest possible at the time\nof its occurrence; and the wish to give another the benefit of the\nexperience implies that this conception, which is no possible image\nof a feeling, can originate action. The assumption of common sense,\nthen, contradicts the two cardinal principles of the Hedonistic\nphilosophy; yet, however disguised in the terminology of development\nand evolution, it, or some equivalent supposition, is involved in\nevery theory of the progress of mankind.\n\nAccording to Locke present pleasures may be compared with future, and\ndesire suspended till comparison has been made.\n\n12. Such difficulties do not suggest themselves to Locke, because he\nis always ready to fall back on the language of common sense without\nasking whether it is reconcilable with his theory. Having asserted,\nwithout qualification, that the will in every case is determined\nby the strongest desire, that the strongest desire is desire for\nthe greatest pleasure, and that \u2018pleasure is just so great, and\nno greater, than it is felt,\u2019 he finds a place for moral freedom\nand responsibility in the \u2018power a man has to suspend his desires\nand stop them from determining his will to any action till he has\nexamined whether it be really of a nature in itself and consequences\nto make him happy or no.\u2019 [1] But how does it happen that there\nis any need for such suspense, if as to pleasure and pain \u2018a man\nnever chooses amiss,\u2019 and pleasure is the same with happiness or\nthe good? To this Locke answers that it is only present pleasure\nwhich is just as it appears, and that in \u2018comparing present pleasure\nor pain with future we often make wrong judgments of them;\u2019 again,\nthat not only present pleasure and pain, but \u2018things that draw after\nthem pleasure and pain, are considered as good and evil,\u2019 and that\nof these consequences under the influence of present pleasure or\npain we may judge amiss. [2] By these wrong judgments, it will be\nobserved, Locke does not mean mistakes in discovering the proper\nmeans to a desired end (Aristotle\u2019s \u1f00\u03b3\u03bd\u03bf\u03af\u03b1 \u1f21 \u03ba\u03b1\u03b8\u2019 \u1f15\u03ba\u03b1\u03c3\u03c4\u03b1) [3], which\nit is agreed are not a ground for blame or punishment, but wrong\ndesires--desires for certain pleasures as being the greater, which\nare not really the greater. Regarding such desires as involving\ncomparisons of one good with another, he counts them judgments, and\n(the comparison being incorrectly made) _wrong_ judgments. A certain\npresent pleasure, and a certain future one, are compared, and though\nthe future would really be the greater, the present is preferred;\nor a present pleasure, \u2018drawing after it\u2019 a certain amount of pain,\nis compared with a less amount of present pain, drawing after it a\ngreater pleasure, and the present pleasure preferred. In such cases\nthe man \u2018may justly incur punishment\u2019 for the wrong preference,\nbecause having \u2018the power to suspend his desire\u2019 for the present\npleasure, he has not done so, but \u2018by too hasty choice of his own\nmaking has imposed on himself wrong measures of good and evil.\u2019\n\u2018When he has once chosen it,\u2019 indeed, \u2018and thereby it is become part\nof his happiness, it raises desire, and that proportionately gives\nhim uneasiness, which determines his will.\u2019 But the original wrong\nchoice, having the \u2018power of suspending his desires,\u2019 he might have\nprevented. In not doing so he \u2018vitiated his own palate,\u2019 and must be\n\u2018answerable to himself\u2019 for the consequences. [4]\n\n[1] II. 21, Sec. 51 and 56.\n\n[2] Ibid., Sec. 61, 63, 67.\n\n[3] [Greek \u1f00\u03b3\u03bd\u03bf\u03af\u03b1 \u1f21 \u03ba\u03b1\u03b8\u2019 \u1f15\u03ba\u03b1\u03c3\u03c4\u03b1 (agnoia he kath\u2019 hekasta) =\nunawareness of the particular circumstances. Tr.]\n\n[4] Ibid., Sec. 56.\n\nWhat is meant by \u2018present\u2019 and \u2018future\u2019 pleasure? By the supposed\ncomparison Locke ought only to have meant the competition of\npleasures equally present in imagination ...\n\n13. Responsibility for evil, then (with its conditions, blame,\npunishment, and remorse) supposes that a man has gone wrong in\nthe comparison of present with future pleasure or pain, having\nhad the chance of going right. Upon this we must remark that as\nmoving desire--and it is the determination of desire that is here\nin question--NO pleasure can be present in the sense of actual\nenjoyment, or (in Hume\u2019s language) as \u2018impression,\u2019 but only in\nmemory or imagination, as \u2018idea.\u2019 Otherwise desire would not be\ndesire. It would not be that uneasiness which, according to Locke,\nimplies the absence of good, and alone moves action. On the other\nhand, to imagination EVERY pleasure must be present that is to act as\nmotive at all. In whatever sense, then, pleasure, as pleasure, _i.e._\nas undetermined by conceptions, can properly be said to move desire,\nevery pleasure is equally present and equally future. [1] For man, if\nhe only felt and retained his feelings in memory, or recalled them in\nimagination, the only difference among the imagined pleasures which\nsolicit his desires, other than difference of intensity, would lie in\nthe imagined pains with which each may have become associated. One\npleasure might be imagined in association with a greater amount of\nthe pain of waiting than another. In that sense, and only in that,\ncould one be distinguished from the other as a future pleasure from\na present one. According as the greater imagined intensity of the\nfuture pleasure did or did not outweigh the imagined pain of waiting\nfor it, the scale of desire would turn one way or the other. Or\nwith one pleasure, imagined as more intense than another, might be\nassociated an expectation of a greater amount of pain to be \u2018drawn\nafter it.\u2019 Here, again, the question would be whether the greater\nimagined intensity of pleasure would have the more effect in exciting\ndesire, or the greater amount of imagined sequent pain in quenching\nit--a question only to be settled by the action which results. In\nwhatever sense it is true of the \u2018present pleasure or pain,\u2019 that\nit is really just as it appears, it is equally true of the future.\nWhenever the determination of desire is in question, the statement\nthat present pleasure is just as it appears must mean that the\npleasure _present in imagination_ is so, and in this sense all motive\npleasures are equally so present. Undoubtedly the pleasure associated\nwith the pain of prolonged expectancy might turn out greater, and\nthat associated with sequent pain less, than was imagined; but so\nmight a pleasure not thus associated. Of every pleasure alike it is\nas true, that while it is imagined it is just as it is imagined, as\nthat while felt it is just as it is felt; and if man only felt and\nimagined, there would be no more reason why he should hold himself\naccountable for his imaginations than for his feelings. Whatever\npleasure was most attractive in imagination would determine desire,\nand, through it, action, which would be the only measure of the\namount of the attraction. It would not indeed follow because an\naction was determined by the pleasure most attractive in imagination,\nthat the ensuing pleasure in actual enjoyment would be greater than\nmight have been attained by a different action--though it would be\nvery hard to show the contrary--but it would follow that the man\nattained the greatest pleasure of which his nature was capable. There\nwould be no reason why he should blame himself, or be blamed by\nothers, for the result.\n\n[1] It is noticeable that when Locke takes to distinguishing the\npleasures that move desire into present and future, he speaks as if\nthe future pleasure alone were an absent good, in contradiction to\nhis previous view that every object of desire is an absent good. (Cf.\nsec. 65 with sec. 57 of cap. 21.)\n\n... and this could give no ground for responsibility. In order to do\nso, it must be understood as implying determination by conception of\nself.\n\n14. Thus on Locke\u2019s supposition, that desire is only moved by\npleasure--which must mean _imagined_ pleasure, since pleasure,\ndetermined by conceptions, is excluded by the supposition that\npleasure alone is the ultimate motive, and pleasure in actual\nenjoyment is no longer desired--the \u2018suspense of desire,\u2019 that he\nspeaks of, can only mean an interval, during which a competition of\nimagined pleasures (one associated with more, another with less, of\nsequent or antecedent pain) is still going on, and none has become\nfinally the strongest motive. Of such suspense it is unmeaning to\nsay that a man has \u2018the power of it,\u2019 or that, when it terminates in\nan action which does not produce so much pleasure as another might\nhave done, it is because the man \u2018has vitiated his palate,\u2019 and that\ntherefore he must be \u2018answerable to himself\u2019 for the consequences.\nThis language really implies that pleasures, instead of being\nultimate ends, are determined to be ends through reference to an\nobject beyond them which the man himself constitutes; that it is\nonly through his conception of self that every pleasure--not indeed\nbest pleases him, or is most attractive in imagination--but becomes\nhis personal good. It may be that he identifies his personal good\nwith the pleasure most attractive in imagination; but a pleasure\nso identified is quite a different motive from a pleasure simply\nas imagined. It is no longer mere pleasure that the man seeks, but\nself-satisfaction through the pleasure. The same consciousness of\nself, which sets him on the act, continues through the act and its\nconsequences, carrying with it the knowledge (commonly called the\n\u2018voice of conscience\u2019) that it is to himself, as the ultimate motive,\nthat the act and its consequences, whether in the shape of natural\npains or civil penalties, are due--a knowledge which breeds remorse,\nand, through it, the possibility of a better mind. Thus, when Locke\nfinds the ground of responsibility in a man\u2019s power of suspending\nhis desire till he has considered whether the act, to which it\ninclines him, is of a kind to make him happy or no, the value of the\nexplanation lies in the distinction which it may be taken to imply,\nbut which Locke could not consistently admit, between the imagination\nof pleasure and the conception of self as a permanent subject of\nhappiness, by reference to which an imagined pleasure becomes a\nstrongest motive. It is not really as involving a comparison between\nimagined pleasures, but as involving the consideration whether the\ngreatest imagined pleasure will be the best for one in the long\nrun, that the suspense of desire establishes the responsibility of\nman. Even if we admitted with Locke that nothing entered into the\nconsideration but an estimate of \u2018future pleasures\u2019--and Locke, it\nwill be observed, by supposing the estimate to include \u2018pleasures\nof a sort we are unacquainted with,\u2019 [1] which is as much of a\ncontradiction as to suppose a man influenced by unfelt feelings,\nrenders this restriction unmeaning--still to be determined by the\nconsideration whether something is good for me on the whole is to be\ndetermined, not by the imagination of pleasure, but by the conception\nof self, though it be of self only as a subject to be pleased.\n\n[1] Cap. 21, sec. 65. He has specially in view the pleasures of\n\u2018another life\u2019, which \u2018being intended for a state of happiness, must\ncertainly be agreeable to every one\u2019s wish and desire: could we\nsuppose their relishes as different there as they are here, yet the\nmanna in heaven will suit every one\u2019s palate.\u2019\n\nLocke finds moral freedom in necessity of pursuing happiness.\n\n15. The mischief is that, though his language implies this\ndistinction, he does not himself understand it. \u2018The care of\nourselves,\u2019 he tells us, \u2018that we mistake not imaginary for real\nhappiness, is the necessary foundation of our liberty. The stronger\nties we have to an unalterable pursuit of happiness in general, which\nis our greatest good, and which, as such, our desires always follow,\nthe more are we free from any necessary determination of our will\nto any particular action, till we have examined whether it has a\ntendency to, or is inconsistent with, our real happiness.\u2019 [1] But he\ndoes not see that the _rationale_ of the freedom, thus paradoxically,\nthough truly, placed in the strength of a tie, lies in that\ndetermination by the conception of self to which the \u2018unalterable\npursuit of happiness\u2019 is really equivalent. To him it is not as one\nmode among others in which that self-determination appears, but\nsimply in itself, that the consideration of what is for our real\nhappiness is the \u2018foundation of our liberty,\u2019 and the consideration\nitself is no more than a comparison between imagined pleasures\nand pains. Hence to a reader who refuses to read into Locke an\ninterpretation which he does not himself supply, the range of moral\nliberty must seem as narrow as its nature is ambiguous. As to its\nrange, the greater part of our actions, and among them those which\nwe are apt to think our best, are not and could not be preceded by\nany consideration whether they are for our real happiness or no. In\ntruth, they result from a character which the conception of self has\nrendered possible, or express an interest in objects of which this\nconception is the condition, and for that reason they represent a\nwill self-determined and free; but they do not rest on the foundation\nwhich Locke calls \u2018the necessary foundation of our liberty.\u2019 As\nto the nature of this liberty, the reader, who takes Locke at his\nword, would find himself left to choose between the view of it as\nthe condition of a mind \u2018suspended\u2019 between rival presentations of\nthe pleasant, and the equally untenable view of it as that \u2018liberty\nof indifference,\u2019 which Locke himself is quite ready to deride--as\nconsisting in a choice prior to desire, which determines what the\ndesire shall be. [2]\n\n[1] Cap. 21, sec. 51.\n\n[2] Cf. the passage in sec. 56: \u2018When he has once chosen it, and\nthereby it is become part of his happiness, it raises desire,\u2019 &c.\n(Cf. also sec. 43 sub fin.)\n\nIf an action is moved by desire for an object, Locke asks no\nquestions about origin of the object: But what is to be said of\nactions, which we only do because we ought?\n\n16. This ambiguous deliverance about moral freedom, it must be\nobserved, is the necessary result on a mind, having too strong a\npractical hold on life to tamper with human responsibility, of\na doctrine which denies the originativeness of thought, and in\nconsequence cannot consistently allow any motive to desire, but the\nimage of a past pleasure or pain. The full logical effect of the\ndoctrine, however, does not appear in Locke, because, with his way\nof taking any desire of which the satisfaction produces pleasure to\nhave pleasure for its object, he never comes in sight of the question\nhow the manifold objects of actual human interest are possible for\na being who only feels and retains, or combines, his feelings. An\naction moved by love of country, love of fame, love of a friend, love\nof the beautiful, would cause him no more difficulty than one moved\nby desire for the renewal of some sensual enjoyment, or for that\nmaintenance of health which is the condition of such enjoyment in the\nfuture. If pressed about them, we may suppose that--availing himself\nof the language probably current in the philosophic society in which\nhe lived, though it first became generally current in England through\nthe writings of his quasi-pupil, Shaftesbury--he would have said\nthat he found in his breast affections for public good, as well as\nfor self-good, the satisfaction of which gave pleasure, and to which\nhis doctrine, that pleasure is the \u2018object of desire in general,\u2019\nwas accordingly applicable. The question--of what feelings or\ncombinations of feelings are the objects which excite these several\ndesires copies?--it does not occur to him to ask. It is only when a\nclass of actions presents itself for which a motive in the way of\ndesire or aversion is not readily assignable that any difficulty\narises, and then it is a difficulty which the assignment of such a\nmotive, without any question asked as to its possibility for a merely\nfeeling and imagining subject, is thought sufficiently to dispose\nof. Such a class of actions is that of which we say that we \u2018ought\u2019\nto do them, even when we are not compelled and had rather not. We\nought, it is generally admitted, to keep our promises, even when it\nis inconvenient to us to do so and no punishment could overtake us if\nwe did not. We ought to be just even in ways that the law does not\nprescribe, and when we are beyond its ken; and that, too, in dealing\nwith men towards whom we have no inclination to be generous. We ought\neven--so at least Locke \u2018on the authority of Revelation\u2019 would have\nsaid--to forgive injuries which we cannot forget, and if not \u2018to love\nour enemies\u2019 in the literal sense, which may be an impossibility, yet\nto act as if we did. To what motive are such actions to be assigned?\n\nTheir object is pleasure, but pleasure given not by nature but by law.\n\n17. \u2018To desire for pleasure or aversion from pain,\u2019 Locke would\nanswer, \u2018but a pleasure and pain other than the natural consequences\nof acts and attached to them by some law.\u2019 This is the result of\nhis enquiry into \u2018Moral Relations\u2019 (Book II., chap. 28). Good and\nevil, he tells us, being \u2018nothing but pleasure and pain, moral good\nor evil is only the conformity or disagreement of our actions to\nsome law, whereby good or evil, _i.e._, pleasure or pain, is drawn\non us by the will and power of the law-maker.\u2019 All law according\nto its \u2018true nature\u2019 is a rule set to the actions of others by an\nintelligent being, having \u2018power to reward the compliance with,\nand punish deviation from, his rule by some good and evil that is\nnot the natural product and consequence of the action itself; for\nthat, being a natural convenience or inconvenience, would operate\nof itself without a law.\u2019 Of such law there are three sorts. 1.\nDivine Law, \u2018promulgated to men by the light of nature or voice of\nrevelation, by comparing their actions to which they judge whether,\nas duties or sins, they are like to procure them happiness or misery\nfrom the hands of the Almighty.\u2019 2. Civil Law, \u2018the rule set by the\nCommonwealth to the actions of those who belong to it,\u2019 reference\nto which decides \u2018whether they be criminal or no.\u2019 3. \u2018The law of\nopinion or reputation,\u2019 according to agreement or disagreement with\nwhich actions are reckoned \u2018virtues or vices.\u2019 This law may or may\nnot coincide with the divine law. So far as it does, virtues and\nvices are really, what they are always supposed to be, actions \u2018in\ntheir own nature \u2018severally right or wrong. It is not as really right\nor wrong, however, but only as esteemed so, that an act is virtuous\nor vicious, and thus \u2018the common measure of virtue and vice is the\napprobation or dislike, praise or blame, which by a tacit consent\nestablishes itself in the several societies, tribes, and clubs of men\nin the world, whereby several actions come to find credit or disgrace\namong them, according to the judgment, maxims, or fashions of the\nplace.\u2019 Each sort of law has its own \u2018enforcement in the way of good\nand evil.\u2019 That of the civil law is obvious. That of the Divine Law\nlies in the pleasures and pains of \u2018another world,\u2019 which (we have to\nsuppose) render actions \u2018in their own nature good and evil.\u2019 That of\nthe third sort of law lies in those consequences of social reputation\nand dislike which are stronger motives to most men than are the\nrewards and punishments either of God or the magistrate (chap. 28, \u00a7\u00a7\n5-12).\n\nConformity to law not the moral good, but a means to it.\n\n18. \u2018Moral goodness or evil,\u2019 Locke concludes, \u2018is the conformity or\nnon-conformity of any action\u2019 to one or other of the above rules (\u00a7\n14). But such conformity or non-conformity is not a feeling, pleasant\nor painful, at all. If, then, the account of the good as consisting\nin pleasure, of which the morally good is a particular form, is to\nbe adhered to, we must suppose that, when moral goodness is said\nto be conformity to law, it is so called merely with reference\nto the specific means of attaining that pleasure in which moral\ngood consists. Not the conception of conformity to law, but the\nimagination of a certain pleasure, will determine the desire that\nmoves the moral act, as every other desire. The distinction between\nthe moral act and an act judiciously done for the sake, let us say,\nof some pleasure of the palate, will lie only in the channel through\nwhich comes the pleasure that each is calculated to obtain. If the\nmotive of an act done for the sake of the pleasure of eating differs\nfrom the motive of an act done for the sake of sexual pleasure on\naccount of the difference of the channels through which the pleasures\nare severally obtained, in that sense only can the motive of either\nof these acts, upon Locke\u2019s principles, be taken to differ from the\nmotive of an act morally done. The explanation, then, of the acts\nnot readily assignable to desire or aversion, of which we say that\nwe only do them because we \u2018ought,\u2019 has been found. They are so far\nof a kind with all actions done to obtain or avoid what Locke calls\n\u2018future\u2019 pleasures or pains that the difficulty of assigning a motive\nfor them only arises from the fact that their immediate result is not\nan end but a means. They differ from these, however, inasmuch as the\npleasure they draw after them is not their \u2018natural consequence,\u2019 any\nmore than the pain attaching to a contrary act would be, but is only\npossible through the action of God, the magistrate, or society in\nsome of its forms.\n\nHume has to derive from \u2018impressions\u2019 the objects which Locke took\nfor granted.\n\n19. After the above examination we can easily anticipate the points\non which a candid and clear-headed man, who accepted the principles\nof Locke\u2019s doctrine, would see that it needed explanation and\ndevelopment. If all action is determined by impulse to remove the\nmost pressing uneasiness, as consisting in desire for the greatest\npleasure of which the agent is at the time capable; if this, again,\nmeans desire for the renewal of some \u2018impression\u2019 previously\nexperienced, and all impressions are either those of sense or derived\nfrom them, how are we to account for those actual objects of human\ninterest and pursuit which seem far removed from any combination\nof animal pleasures or of the means thereto, and specially for\nthat class of actions determined, as Locke says, by expectation of\npain or pleasure other than the \u2018natural consequence\u2019 of the act,\nto which the term \u2018moral\u2019 is properly applied? Hume, as we have\nseen, [1] in accepting Locke\u2019s principles, clothes them in a more\nprecise terminology, marking the distinction between the feeling as\noriginally felt and the same as returning in memory or imagination as\nthat between \u2018impression and idea,\u2019 and excluding _original_ ideas\nof reflection. \u2018An impression first strikes upon the senses, and\nmakes us perceive heat or cold, thirst or hunger, pleasure or pain,\nof some kind or other. Of this impression there is a copy taken by\nthe mind, which remains after the impression ceases; and this we call\nan idea. This idea of pleasure or pain, when it returns upon the\nsoul, produces the new impressions of desire and aversion, hope and\nfear, which may properly be called impressions of reflection, because\nderived from it\u2019 _(a)_. \u2018These, again, are copied by the memory and\nimagination, and become ideas; which perhaps in their turn give rise\nto other impressions\u2019 _(b)_. Thus the impressions of reflection,\nmarked _(a)_, will be determined by ideas copied from impressions\nof sense. If desires, they will be desires for the renewal either\nof a pleasure incidental to the satisfaction of appetite, or of\na pleasant sight or sound, a sweet taste or smell. These desires\nand their satisfactions will again be copied in ideas, but how can\nthe impressions _(b)_ to which these ideas give rise be other than\ndesires for the renewal of the original animal pleasures? How do they\ncome to be desires as unlike these as are the motives which actuate\nnot merely the saint or the philanthropist, but the ordinary good\nneighbour or honest citizen or head of a family?\n\n[1] General Introd., Vol. I, par. 195\n\nQuestions which he found at issue, a. Is virtue interested? b. What\nis conscience?\n\n20. During the interval between the publication of Locke\u2019s essay\nand the \u2018Treatise on Human Nature\u2019 there had been much writing on\nethical questions in English. The effect of this on Hume is plain\nenough. He writes with reference to current controversy, and in\nthe moral part of the treatise probably had the views of Clarke,\nShaftesbury, Butler, and Hutcheson more consciously before him than\nLocke\u2019s. This does not interfere, however, with the propriety of\naffiliating him in respect of his views on morals, no less than on\nknowledge, directly to Locke, whose principles and method were in the\nmain accepted by all the moralists of that age. His characteristic\nlies in his more consistent application of these, and the effect of\ncurrent controversy upon him was chiefly to show him the line which\nthis application must take. It was a controversy which turned almost\nwholly on two points; _(a)_ the distinction between \u2018interested and\ndisinterested,\u2019 selfish and unselfish affections; _(b)_ the origin\nand nature of that \u2018law,\u2019 relation to which, according to Locke,\nconstitutes our action \u2018virtuous or vicious.\u2019 In the absence of any\nnotion of thought but as a faculty which puts together simple ideas\ninto complex ones, of reason but as a faculty which calculates means\nand perceives the agreement of ideas mediately, it could have but one\nend.\n\nHobbes\u2019 answer to first question,\n\n21. By the generation in which Hume was bred the issue as to the\npossible disinterestedness of action was supposed to lie between\nthe view of Hobbes and that of Shaftesbury. Hobbes\u2019 moral doctrine\nhad not been essentially different from Locke\u2019s, but he had been\noffensively explicit on questions which Locke left open to more\ngenial views than his doctrine logically justified. Each started from\nthe position that the ultimate motive to every action can only be\nthe imagination of one\u2019s own pleasure or pain, and neither properly\nleft room for the determination of desire by a conceived object as\ndistinct from remembered pleasure. But while Locke, as we have seen,\nillogically took for granted desires so determined, and thus made it\npossible for a disciple to admit any benevolent desires as motives\non the strength of the pleasure which they produce when satisfied,\nHobbes had been more severe in his method, and had explained every\ndesire, of which the direct motive could not be taken to be the\nrenewal of some animal pleasure, as desire either for the power\nin oneself to command such pleasure at will or for the pleasure\nincidental to the contemplation of the signs of such power. Hence his\npeculiar treatment of compassion and the other \u2018social affections,\u2019\nwhich it is easier to show to be untrue to the facts of the case\nthan to be other than the proper consequence of principles which\nLocke had rendered orthodox. [1] The counter-doctrine of Shaftesbury\nholds water just so far as it involves the rejection of the doctrine\nthat pleasure is the sole ultimate motive. It becomes confused just\nbecause its author had no definite theory of reason, as constitutive\nof objects, that could justify this rejection.\n\n[1] See \u2018Leviathan,\u2019 part 1, chap. 6.\n\nCounter-doctrine of Shaftesbury. Vice is selfishness: But no clear\naccount of selfishness.\n\n22. He begins with a doctrine that directly contradicts Locke\u2019s\nidentification of the good with pleasure, and of the morally good\nwith pleasure occurring in a particular way, \u2018In a sensible creature\nthat which is not done through any affection at all makes neither\ngood nor ill in the nature of that creature; who then only is\nsupposed good, when the good or ill of the system to which he has\nrelation is the immediate object of some passion or affection moving\nhim.\u2019 [1] This, it will be seen, as against Locke, implies that the\ngood of a man\u2019s action lies not in any pleasure sequent upon it to\nhim, but in the nature of the affection from which it proceeds; and\nthat the goodness of this affection depends on its being determined\nby an object wholly different from imagined pleasure--the _conceived_\ngood of a system to which the man has relation, _i.e._, of human\nsociety, which in Shaftesbury\u2019s language is the \u2018public\u2019 as distinct\nfrom the \u2018private\u2019 system. It is not enough that an action should\nresult in good to this system; it must proceed from affection for it.\n\u2018Whatever is done which happens to be advantageous to the species\nthrough an affection merely towards self-good does not imply any\nmore goodness in the creature than as the affection itself is good.\nLet him in any particular act ever so well; if at the bottom it\nbe that selfish affection alone which moves him, he is in himself\nstill vicious.\u2019 [2] Here, then, we seem to have a clear theory of\nmoral evil as consisting in selfish, of moral good as consisting\nin unselfish affections. But what exactly constitutes a selfish\naffection, according to Shaftesbury? The answer that first suggests\nitself, is that as the unselfish affection is an affection for\npublic good, so a selfish one is an affection for \u2018self-good,\u2019 the\ngood of the \u2018private system.\u2019 Shaftesbury, however, does not give\nthis answer. \u2018Affection for private good\u2019 with him is not, as such,\nselfish; it is so only when \u2018excessive\u2019 and \u2018inconsistent with the\ninterest of the species or public.\u2019 [3] This qualification seems\nat once to efface the clear line of distinction previously drawn.\nIt puts \u2018self-affection\u2019 on a level with public affection which,\naccording to Shaftesbury, may equally err on the side of excess. It\nimplies that an affection for self-good, if only it be advantageous\nto the species, may be good; which is just what had been previously\ndenied. And not only so; although, when the self-affections are under\nview, they are only allowed a qualified goodness in virtue of their\nindirect contribution to the good of the species, yet conversely, the\nsuperiority of the affections, which have this latter good for their\nobject, is urged specially on the ground of the greater amount of\nhappiness or \u2018self-good\u2019 which they produce.\n\n[1] \u2018Inquiry concerning Virtue,\u2019 Book I. part 2, sec. 1.\n\n[2] Ibid., Book I., part 2, sec. 2.\n\n[3] Ibid., Book II., part 1, sec. 3.\n\nConfusion in his notions of self-good and public good: Is all living\nfor pleasure, or only too much of it, selfish?\n\n23. The truth is that the notions which Shaftesbury attached to the\nterms \u2018affection for self-good\u2019 and \u2018affection for public good\u2019 were\nnot such as allowed of a consistent opposition between them. They can\nonly be so opposed if, on the one hand, self-good is identified with\npleasure; and on the other, affection for public good is carefully\ndistinguished from desire for that sort of pleasure of which the\ngratification of others is a condition. But with Shaftesbury,\naffections for self-good do not represent merely those desires for\npleasure determined by self-consciousness--for pleasure presented as\none\u2019s personal good--which can alone be properly reckoned sources\nof moral evil. They include equally mere natural appetites--hunger,\nthe sexual impulse, &c.--which are morally neutral, and they do not\nclearly exclude any desire for an object which a man has so \u2018made\nhis own\u2019 as to find his happiness--\u2018self-enjoyment\u2019 or \u2018self-good,\u2019\naccording to Shaftesbury\u2019s language--in attaining it, though it\nbe as remote from imagined pleasure as possible. [1] On the other\nhand, \u2018affections for public good,\u2019 as he describes them, are not\nrestricted to such desires for the good of others as are irrespective\nof pleasure to self. They include not only such natural instincts as\n\u2018parental kindness and concern for the nurture and propagation of\nthe young,\u2019 which, morally, at any rate, are not to be distinguished\nfrom the appetites reckoned as affections for self-good, but also\ndesires for sympathetic pleasure--the pleasure to oneself which\narises on consciousness that another is pleased. Shaftesbury\u2019s\nspecial antipathy, indeed, is the doctrine that benevolent affections\nare interested in the sense of having for their object a pleasure\nto oneself, apart from and beyond the pleasure of the person whom\nthey move us to please; but unless he regards them as desires for\nthe pleasure which the subject of them experiences in the pleasure\nof another, there is no purpose in enlarging, as he does with much\nunction, on the special pleasantness of the pleasures which they\nproduce. With such vagueness in his notions of what he meant by\naffections for \u2018self-good\u2019 and for \u2018public good,\u2019 it is not strange\nthat he should have failed to give any tenable account of the\nselfishness in which he conceived moral evil to consist. He could not\napply such a term of reproach to the \u2018self-affections\u2019 in general,\nwithout condemning as selfish the man who \u2018finds his own happiness\nin doing good,\u2019 and who is in truth indistinguishable from one to\nwhom \u2018affection for public good\u2019 has become, as we say, the law of\nhis being. Nor could he identify selfishness, as he should have\ndone, with all living for pleasure without a more complete rupture\nthan he was capable of with the received doctrine of his time and\nwithout bringing affection for public good, in the form in which it\nwas most generally conceived, and which was, at any rate, one of the\nforms under which he presented it to himself--as desire, namely, for\nsympathetic pleasure--into the same condemnation. His way out of the\ndifficulty is, as we have seen, in violation of his own principle\nto find the characteristic of selfishness not in the motive of any\naffection but in its result; not in the fact that a man\u2019s desire\nhas his own good for its object, which is true of one to whom his\nneighbour\u2019s good is as his own, nor in the fact that it has pleasure\nfor its object, which Shaftesbury, as the child of his age, could\nscarcely help thinking was the case with every desire, but in the\nfact that it is stronger than is \u2018consistent with the interest of the\nspecies or public.\u2019\n\n[1] Book II., part 2, sec. 2.\n\nWhat have Butler and Hutcheson to say about it? Chiefly that\naffections terminate upon their objects. But this does not exclude\nthe view that all desire is for pleasure.\n\n24. Neither Butler nor Hutcheson [1] can claim to have carried the\nethical controversy much beyond the point at which Shaftesbury left\nit. Each took for granted that the object of the \u2018self-affection\u2019 was\nnecessarily one\u2019s own happiness, and neither made any distinction\nbetween living for happiness and living for pleasure. They could\nnot then identify selfishness with the living for pleasure without\ncondemning the self-affection, and with, it the best man\u2019s pursuit\nof his own highest good in the service of others, altogether as\nevil. Nor in the absence of any better theory of the object of\nthe self-affection could the social affections, which, according\nto Butler, are subject in the developed man to the direction of\nself-love, escape the suggestion that they are one mode of the\ngeneral desire for pleasure. Butler and Hutcheson, indeed, are quite\nclear that they are \u2018disinterested\u2019 in the sense of \u2018terminating upon\ntheir objects.\u2019 [2] This means, what is sufficiently obvious when\nonce pointed out, _(a)_ that a benevolent desire is not a desire\nfor that particular pleasure, or rather \u2018removal of uneasiness,\u2019\nwhich shall ensue when it is satisfied, and _(b)_ that it cannot\noriginally arise from the general desire for happiness, since\nthis creates no pleasures but merely directs us to the pursuit of\nobjects found pleasant independently of it, and thus, if it directs\nus to benevolent acts, presupposes a pleasure previously found in\nthem. This, however, as Butler points out, is equally true of all\nparticular desires whatever--of those styled self-regarding, no less\nthan of the social--and if it is not incompatible with the former\nbeing desires for pleasure, no more is it with the latter being so.\nMuch confusion on the matter, it may be truly said, arises from the\nloose way in which the words \u2018affection\u2019 and \u2018passion\u2019 are used by\nButler and his contemporaries, not excluding Hume himself, alike\nfor appetite, desire, and emotion. In every case a pleasure other\nthan satisfaction of desire must have been experienced before desire\ncan be excited by the imagination of it. A pleasure incidental to\nthe satisfaction of _appetite_ must have been experienced before\nimagination of it could excite the _desire_ of the glutton. In like\nmanner, social affection, as _desire_, cannot be first excited by the\npleasure which shall arise when it is satisfied; it must previously\nexist as the condition of that pleasure being experienced; but it\ndoes not follow that it is other than a desire for an imagined\npleasure, for that sympathetic pleasure in the pleasure of another\nin which the social affection as _emotion_ consists. Now though\nButler and Hutcheson sufficiently showed that it is no other pleasure\nthan this which is the original object of benevolent desires, they\ndid not attempt to show that it is not this; and failing such an\nattempt, the received doctrine that the object of all desire, social\nand self-regarding alike, is pleasure of one sort or another, would\nnaturally be taken to stand. This admitted, there can be nothing in\nthe fact that a certain pleasure depends on the pleasure of another,\nand that a certain other does not, to entitle an action moved by\ndesire for the former sort of pleasure to be called unselfish in the\nway of praise, and one moved by desire for the latter sort selfish\nin the way of reproach. The motive--desire for his own pleasure--is\nthe same to the doer in both cases. The distinction between the acts\ncan only lie in that which Shaftesbury had said could not constitute\nmoral good or ill--in the consequences by which society judges of\nthem, but which do not form the motive of the agent. In other words,\nit will be a distinction fixed by that law of opinion or reputation,\nin which Locke had found the common measure of virtue and vice,\nthough he had not entered on the question of the considerations by\nwhich that law is formed.\n\n[1] The works of Hutcheson, published before Hume\u2019s treatise was\nwritten, and which strongly affected it, were the \u2018Enquiry into the\nOriginal of our Ideas of Beauty and Virtue\u2019 (1725), and the \u2018Essay\non the Nature and Conduct of the Passions and Affections\u2019 (1728). In\nwhat follows I wrote with direct reference to his posthumous work,\nnot published till after Hume\u2019s treatise, but which only reproduces\nmore systematically his earlier views.\n\n[2] See in Preface to Butler\u2019s Sermons, the part relating to Sermon\nXI., \u2018Besides, the only idea of an interested pursuit\u2019 &c.; also the\nearly part of Sermon XI., \u2018Every man hath a general desire,\u2019 &c.\n\nOf moral goodness Butler\u2019s account circular: Hutcheson\u2019s inconsistent\nwith his doctrine that reason gives no end.\n\n25. Such a conclusion would lie ready to hand for such a reader of\nButler and Hutcheson as we may suppose Hume to have been, but it\nis needless to say that it is not that at which they themselves\narrive. Butler, indeed, distinctly refuses to identify moral good\nand evil respectively with disinterested and interested action, [1]\nbut neither does he admit that desire for pleasure or aversion from\npain is the uniform motive of action in such a way as to compel the\nconclusion that moral good and ill represent a distinction, not of\nmotives, but of consequences of action contemplated by the onlooker.\nAn act is morally good, according to him, when it is approved by the\n\u2018reflex faculty of approbation,\u2019 bad when it is disapproved, but\nwhat it is that this \u2018faculty\u2019 approves he never distinctly tells\nus. The good is what \u2018conscience\u2019 approves, and conscience is what\napproves the good--that is the circle out of which he never escapes.\nIf we insist on extracting from him any more satisfactory conclusion\nas to the object of moral approbation, it must be that it is the\nobject which \u2018self-love\u2019 pursues, _i.e._, the greatest happiness\nof the individual, a conclusion which in some places he certainly\nadopts. [2] Hutcheson, on the other hand, gives a plain definition of\nthe object which this faculty approves. It consists in \u2018affections\ntending to the happiness of others and the moral perfection of the\nmind possessing them.\u2019 If in this definition by \u2018tending to\u2019 may be\nunderstood \u2018of which the motive is\u2019--an interpretation which the\ngeneral tenor of Hutcheson\u2019s view would justify--it implies in effect\nthat the morally good lies in desires of which the object is not\npleasure. That desire for moral perfection, if there is such a thing,\nis not desire for pleasure is obvious enough; nor could desire for\nthe happiness of others be taken to be so except through confusion\nbetween determination by the conception of another\u2019s good, to which\nhis apparent pleasure is rightly or wrongly taken as a guide, and\nby the imagination of a pleasure to be experienced by oneself in\nsympathy with the pleasure of another. Nor is it doubtful that\nHutcheson himself, though he might have hesitated to identify moral\nevil, as selfishness, with the living for pleasure, yet understood\nby the morally good the living for objects wholly different from\npleasure. The question is whether the recognition of such motives is\nlogically compatible with his doctrine that reason gives no ends,\nbut is only a \u2018subservient power\u2019 of calculating means. If feeling,\nundetermined by thought or reason, can alone supply motives, and\nof feeling, thus undetermined, nothing can be said but that it is\npleasant or painful, what motive can there be but imagination of\none\u2019s own pleasure or pain--_one\u2019s own_, for if imagination is\nmerely the return of feeling in fainter form, no one can imagine any\nfeeling, any more than he can originally feel it, except as his own?\n\n[1] See preface to Sermons (about four pages from the end in most\neditions):-\u2018The goodness or badness of actions does not arise hence,\u2019\n&c. The conclusion he there arrives at is that a good action is one\nwhich \u2018becomes such creatures as we are\u2019; and this, read in the\nlight of the second sermon, must be understood to mean an action\n\u2018suitable to our whole nature,\u2019 as containing a principle of \u2018reflex\napprobation.\u2019 In other words, the good action is so because approved\nby conscience.\n\n[2] See a passage towards the end of Sermon III., \u2018Reasonable\nself-love and conscience are the chief,\u2019 &c. &c.; also a passage\ntowards the end of Sermon XI., \u2018Let it be allowed though virtue,\u2019 &c.\n&c.\n\nSource of the moral judgment: Received notion of reason incompatible\nwith true view. Shaftesbury\u2019s doctrine of rational affection; spoilt\nby doctrine of \u2018moral sense\u2019\n\n26. The work of reason in constituting the moral judgment (\u2018I\nought\u2019), as well as the moral motive (\u2018I must, because I ought\u2019),\ncould not find due recognition in an age which took its notion of\nreason from Locke. The only theory then known which found the source\nof moral distinctions in reason was Clarke\u2019s, and Clarke\u2019s notion\nof reason was essentially the same as that which appears in Locke\u2019s\naccount of demonstrative knowledge. [1] It was in truth derived from\nthe procedure of mathematics, and only applicable to the comparison\nof quantities. Clarke talks loftily about the Eternal Reason of\nthings, but by this he means nothing definite except the laws of\nproportion, and when he finds the virtue of an act to consist in\nconformity to this Eternal Reason, the inevitable rejoinder is the\nquestion--Between what quantities is this virtue a proportion? [2]\nIn Shaftesbury first appears a doctrine of moral sense. Over and\nabove the social and self-regarding affections proper to a \u2018sensible\u2019\ncreature, the characteristic of man is a \u2018rational affection\u2019 for\ngoodness as consisting in the proper adjustment of the two orders\nof \u2018sensible\u2019 affection. This rational affection is not only a\npossible motive to action--it is the only motive that can make that\ncharacter good of which human action is the expression; for with\nShaftesbury, though a balance of the social and self-affections\nconstitutes the goodness of those affections, yet the man is only\ngood as actuated by affection for this goodness, and \u2018should the\n_sensible_ affections stand ever so much amiss, yet if they prevail\nnot because of those other rational affections spoken of, the person\nis esteemed virtuous.\u2019 [3] Such a notion, it is clear, if it had\nmet with a psychology answering it, had only to be worked out in\norder to become Kant\u2019s doctrine of the rational will as determined\nby reverence for law; but Shaftesbury had no such psychology, nor,\nwith his aristocratic indifference to completeness of system, does\nhe seem ever to have felt the want of it. He never asked himself\nwhat precisely was the theory of reason implied in the admission of\nan affection \u2018rational\u2019 in the sense, not that reason calculates the\nmeans to its satisfaction, but that it is determined by an object\nonly possible for a rational as distinct from a \u2018sensible\u2019 creature;\nand just because he did not do so, he slipped into adaptations to the\ncurrent view of the good as pleasure and of desire as determined by\nthe pleasure incidental to its own satisfaction. Thus, to a disciple,\nwho wished to extract from Shaftesbury a more definite system than\nShaftesbury had himself formed, the \u2018rational affection\u2019 would become\ndesire for a specific feeling of pleasure supposed to arise on the\nview of good actions as exhibiting a proper balance between social\nand self-regarding affections. This pleasure is the \u2018moral sense,\u2019\n[4] with which Shaftesbury\u2019s name has become specially associated,\nwhile the doctrine of rational affection, with which he certainly\nhimself connected it, but which it essentially vitiates, has been\nforgotten.\n\n[1] See Clarke\u2019s Boyle Lectures, Vol. II., proposition 1. The germ\nof Clarke\u2019s doctrine of morals is to be found in Locke\u2019s occasional\nassimilation of moral to mathematical truth and certainty. (Cf.\nEssay, Book IV, ch. 4, sec. 7, and ch. 12, sec. 8).\n\n[2] Cf. Hume, Vol. II., p. 238. [Book III., part I., sec. I.]\n\n[3] \u2018Inq. concerning Virtue,\u2019 Book I., pt. 2. sec. 4. Cf. Sec. 3 sub\ninit.\n\n[4] In using the term \u2018moral sense,\u2019 Shaftesbury himself, no doubt,\nmeant to convey the notion that the moral faculty was one of\n\u2018intuition,\u2019 in Locke\u2019s sense of the word, as opposed to reason, the\nfaculty of demonstration, rather than that it was a susceptibility of\npleasure and pain.\n\nConsequences of the latter.\n\n27. That doctrine is of value as maintaining that those actions only\nare morally good of which the rational affection is the motive, in\nthe sense that they spring from a character which this affection has\nfashioned. But if the rational affection is desire for the pleasure\nof moral sense, we find ourselves in the contradiction of supposing\nthat the only motive which can produce good acts is one that cannot\noperate till after the good acts have been done. It is desire for a\npleasure which yet can only have been experienced as a consequence of\nthe previous existence of the desire. Shaftesbury himself, indeed,\ntreats the moral sense of pleasure in the contemplation of good\nactions as a pleasure in the view of the right adjustment between the\nsocial and self-affections. If, however, on the strength of this, we\nsuppose that certain actions are first done, not from the rational\naffection, but yet good, and that then remembrance of the pleasure\nfound in the view of their goodness, exciting desire, becomes motive\nto another set of acts which are thus done from rational affection,\nwe contradict his statement that only the rational affection forms\nthe goodness of man, and are none the nearer to an account of what\ndoes form it. To say that it is the \u2018right adjustment\u2019 of the two\norders of affection tells us nothing. Except as suggesting an analogy\nfrom the world of art, really inapplicable, but by which Shaftesbury\nwas much influenced, this expression means no more than that goodness\nis a good state of the affections. From such a circle the outlet\nmost consistent with the spirit of that philosophy, which had led\nShaftesbury himself to bring down the rational affection to the level\nof a desire for pleasure, would lie in the notion that a state of the\naffections is good in proportion as it is productive of pleasure;\nwhich again would suggest the question whether the specific pleasure\nof moral sense itself, the supposed object of rational affection, is\nmore than pleasure in that indefinite anticipation of pleasure which\nthe view of affections so ordered tends to raise in us.\n\nIs an act done for \u2018virtue\u2019s sake\u2019 done for pleasure of moral sense?\n\n28. Here, again, neither Butler nor Hutcheson, while they avoid the\nmost obvious inconsistency of Shaftesbury\u2019s doctrine, do much for its\npositive development. With each the \u2018moral faculty,\u2019 though it is\nsaid to approve and disapprove, is still a \u2018sense\u2019 or \u2018sentiment,\u2019 a\nspecific susceptibility of pleasure in the contemplation of goodness;\nand each again recognises a \u2018reflex affection\u2019 for--a desire to\nhave--the goodness of which the view conveys this pleasure. But they\nneither have the merit of stating so explicitly as Shaftesbury does\nthat this rational affection alone constitutes the goodness of man,\nas man; nor, on the other hand, do they lapse, as he does, into the\nrepresentation of it as a desire for the pleasure which the view of\ngoodness causes. Butler, indeed, having no account to give of the\ngoodness which is approved or morally pleasing, but the fact that it\nis so pleasing, could logically have nothing to say against the view\nthat this reflex affection is merely a desire for this particular\nsort of pleasure; but by representing it as equivalent in its highest\nform to the love of God, to the longing of the soul after Him as the\nperfectly good, he in effect gives it a wholly different character.\nHutcheson, by his definition of the object of moral approbation, [1]\nwhich is also a definition of the object of the reflex affection, is\nfairly entitled to exclude, as he does, along with the notion that\nthe goodness which we morally approve is the quality of exciting the\npleasure of such approval, the notion that \u2018affection for goodness\u2019\nmeans desire for this or any other pleasure. But, in spite of his\nexpress rejection of this view, the question will still return, how\neither a faculty of consciousness of which we only know that it is \u2018a\nkind of taste or relish,\u2019 or a desire from the determination of which\nreason is expressly excluded, can have any other object than pleasure\nor pain.\n\n[1] See above, sec. 25.\n\nHume excludes every object of desire but pleasure.\n\n29. In contrast with these well-meant efforts to derive that\ndistinction between the selfish and unselfish, between the pleasant\nand the morally good, which the Christian conscience requires, from\nprinciples that do not admit of it, Hume\u2019s system has the merit\nof relative consistency. He sees that the two sides of Locke\u2019s\ndoctrine--one that thought originates nothing, but takes its objects\nas given in feeling, the other that the good which is object of\ndesire is pleasant feeling--are inseparable. Hence he decisively\nrejects every notion of rational or unselfish affections, which\nwould imply that they are other than desires for pleasure; of\nvirtue, which would imply that it antecedently determines, rather\nthan is constituted by, the specific pleasure of moral sense; and\nof this pleasure itself, which would imply that anything but the\nview of tendencies to produce pleasure can excite it. But here\nhis consistency stops. The principle which forbade him to admit\nany object of desire but pleasure is practically forgotten in his\naccount of the sources of pleasure, and its being so forgotten is\nthe condition of the desire for pleasure being made plausibly to\nserve as a foundation for morals. It is the assumption of pleasures\ndetermined by objects only possible for reason, made in the treatise\non the Passions, that prepares the way for the rejection of reason,\nas supplying either moral motive or moral standard, in the treatise\non Morals.\n\nHis account of \u2018direct passions\u2019: All desire is for pleasure.\n\n30. \u2018The passions\u2019 is Hume\u2019s generic term for \u2018impressions of\nreflection\u2019--appetites, desires, and emotions alike. He divides them\ninto two main orders, \u2018direct and indirect,\u2019 both \u2018founded on pain\nand pleasure.\u2019 The _direct_ passions are enumerated as \u2018desire and\naversion, grief and joy, hope and fear, along with volition\u2019 or will.\nThese \u2018arise from good and evil\u2019 (which are the same as pleasure and\npain) \u2018most naturally and with least preparation.\u2019 \u2018Desire arises\nfrom good, aversion from evil, considered simply.\u2019 They become will\nor volition, \u2018when the good may be attained or evil avoided by\nany action of the mind or body\u2019--will being simply \u2018the internal\nimpression we feel and are conscious of when we knowingly give rise\nto any new motion of our body or new perception of our mind.\u2019 \u2018When\ngood is certain or probable it produces joy\u2019 (which is described\nalso as a pleasure produced by pleasure or by the imagination of\npleasure); \u2018when it is uncertain, it gives rise to hope.\u2019 To these\nthe corresponding opposites are grief and fear. We must suppose\nthem to be distinguished from desire and aversion as being what\nhe elsewhere calls \u2018pure emotions\u2019; such as do not, like desires,\n\u2018immediately excite us to action.\u2019 Given such an immediate impression\nof pleasure or pain as excites a \u2018distinct passion\u2019 of one or other\nof these kinds, and supposing it to \u2018arise from an object related to\nourselves or others,\u2019 it excites mediately, through this relation,\nthe new impressions of pride or humility, love or hatred--pride when\nthe object is related to oneself, love when it is related to another\nperson. These are _indirect_ passions. They do not tend to displace\nthe immediate impression which is the condition of their excitement,\nbut being themselves agreeable give it additional force. \u2018Thus a\nsuit of fine clothes produces pleasure from their beauty; and this\npleasure produces the direct passions, or the impressions of volition\nand desire. Again, when these clothes are considered as belonging to\noneself, the double relation conveys to us the sentiment of pride,\nwhich is an indirect passion; and the pleasure which attends that\npassion returns back to the direct affections, and gives new force to\nour desire or volition, joy or hope.\u2019 [1]\n\n[1] Vol. II., pp. 214, 215. Cf. pp. 76, 90, 153 and 203. [Book II.,\npart III., sec. IX.; part 1, sec. I; part I., sec. VI.; part II.,\nsec. VI.; part III., sec. VI.]\n\nYet he admits \u2018passions\u2019 which produce pleasure, but proceed not from\nit\n\n31. Alongside of the unqualified statement that \u2018the passions, both\ndirect and indirect, are founded on pain and pleasure,\u2019 and the\nconsequent theory of them, we find the curiously cool admission that\n\u2018beside pain and pleasure, the direct passions frequently arise from\na natural impulse or instinct, which is perfectly unaccountable.\nOf this kind is the desire of punishment to our enemies, and of\nhappiness to our friends; hunger and lust, and a few other bodily\nappetites. These passions, properly speaking, produce good and\nevil, and proceed not from them like the other affections.\u2019 [1] In\nthis casual way appears the recognition of that difference of the\ndesire for imagined pleasure from appetite proper on the one side,\nand on the other from desire determined by reason, which it is the\npoint of Hume\u2019s system to ignore. The question is, how many of the\npleasures in which he finds the springs of human conduct are other\nthan products of a desire which is not itself moved by pleasure, or\nemotions excited by objects which reason constitutes.\n\n[1] P. 215. [Book II., part III., sec. IX.] The passage in the\n\u2018Dissertation on the Passions\u2019 (Vol. IV., \u2018Dissertation on the\nPassions,\u2019 sub init.), which corresponds to the one here quoted,\nthrows light on the relation in which Hume\u2019s later redaction of his\ntheory stands to the earlier, as occasionally disguising, but never\nremoving, its inconsistencies. \u2018Some objects, by being naturally\nconformable or contrary to passion, excite an agreeable or painful\nsensation, and are thence called _good_ or _evil_. The punishment\nof an adversary, by gratifying revenge, is good: the sickness of a\ncompanion, by affecting friendship, is evil.\u2019 Here he avoids the\ninconsistency of admitting in so many words a \u2018desire\u2019 which is not\nfor a pleasure. But the inconsistency really remains. What is the\npassion, the \u2018conformability\u2019 to which of an object in the supposed\ncases constitutes pleasure? Since it is neither an appetite (such\nas hunger), nor an emotion (such as pride), it remains that it is a\ndesire, and a desire which, though the \u2018gratification\u2019 of it is a\npleasure, cannot be a desire for that or any other pleasure.\n\nDesire for objects, as he understands it, excluded by his theory of\nimpressions and ideas.\n\n32. In what sense, we have first to ask, do Hume\u2019s principles justify\nhim in speaking of desire _for an object_ at all. \u2018The appearance of\nan object to the senses\u2019 is the same thing as \u2018an impression becoming\npresent to the mind,\u2019 [1] and if this is true of impressions of sense\nit cannot be less true of impressions of reflection. If sense \u2018offers\nnot its object as anything distinct from itself,\u2019 neither can desire.\nIts object, according to Hume, is an idea of a past impression; but\nthis, if we take him at his word, can merely mean that a feeling\nwhich, when at its liveliest, was pleasant, has passed into a fainter\nstage, which, in contrast with the livelier, is pain--the pain of\nwant, which is also a wish for the renewal of the original pleasure.\nIn fact, however, when Hume or anyone else (whether he admit the\npossibility of desiring an object not previously found pleasant, or\nno), speaks of desire for an object, he means something different\nfrom this. He means either desire for an object that causes pleasure,\nwhich is impossible except so far as the original pleasure has\nbeen--consciously to the subject feeling it--pleasure caused by an\nobject, _i.e._, a feeling determined by the conception of a thing\nunder relations to self; or else desire for pleasure as an object,\n_i.e._, not merely desire for the revival of some feeling which,\nhaving been pleasant as \u2018impression,\u2019 survives without being pleasant\nas \u2018idea,\u2019 but desire determined by the consciousness of self as a\npermanent subject that has been pleased, and is to be pleased again.\nIt is here, then, as in the case of the attempted derivation of\nspace, or of identity and substance, from impressions of sense. In\norder to give rise to such an impression of reflection as desire for\nan object is, either the original impression of sense, or the idea\nof this, must be other than Hume could allow it to be. Either the\noriginal impression must be other than a satisfaction of appetite,\nother than a sight, smell, sound, &c., or the idea must be other\nthan a copy of the impression. One or other must be determined by\nconceptions not derived from feeling, the correlative conceptions of\nself and thing. Thus, in order to be able to interpret his primary\nclass of impressions of reflection [2] as desires for objects, or\nfor pleasures as good, Hume has already made the assumption that is\nneeded for the transition to that secondary class of impressions\nthrough which he has to account for morality. He has assumed that\nthought determines feeling, and not merely reproduces it. Even if the\nmaterials out of which it constructs the determining object be merely\nremembered pleasures, the object is no more to be identified with\nthese materials than the living body with its chemical constituents.\n\n[1] See General Introduction, paragraph 208.\n\n[2] See above, sec. 19.\n\nPride determined by reference to self.\n\n33. In the account of the \u2018indirect passions\u2019 the term _object_ is no\nlonger applied, as in the account of the direct ones, to the pleasure\nor pain which excites desire or aversion. It is expressly transferred\nto the self or other person, to whom the \u2018exciting causes\u2019 of pride\nand love must be severally related. \u2018Pride and humility, though\ndirectly contrary, have yet the same object,\u2019 viz., self; but since\nthey are contrary, \u2018\u2019tis impossible this object can be their cause,\nor sufficient alone to excite them ... We must therefore make a\ndistinction betwixt that idea which excites them, and that to which\nthey direct their view when excited.... The first idea that is\npresented to the mind is that of the cause or productive principle.\nThis excites the passion connected with it; and that passion, when\nexcited, turns our view to another idea, which is that of self....\nThe first idea represents the _cause_, the second the _object_ of\nthe passion.\u2019 [1] Again a further distinction must be made \u2018in the\ncauses of the passion betwixt that _quality_ which operates, and\nthe _subject_ on which it is placed. A man, for instance, is vain\nof a beautiful house which belongs to him, or which he has himself\nbuilt or contrived. Here the object of the passion is himself, and\nthe cause is the beautiful house; which cause again is subdivided\ninto two parts, viz., the quality which operates upon the passion,\nand the subject in which the quality inheres. The quality is the\nbeauty, and the subject is the house, considered as his property or\ncontrivance.\u2019 [2] It is next found that the operative qualities which\nproduce pride, however various, agree in this, that they produce\npleasure--a \u2018separate pleasure,\u2019 independent of the resulting pride.\nIn all cases, again, \u2018the subjects to which these qualities adhere\nare either parts of ourselves or something nearly related to us.\u2019\nThe conclusion is that \u2018the cause, which excites the passion, is\nrelated to the object which nature has attributed to the passion;\nthe sensation, which the cause separately produces, is related to\nthe sensation of the passion: from this double relation of ideas\nand impressions the passion is derived.\u2019 [3] The ideas, it will be\nobserved, are severally those of the exciting \u2018subject\u2019 (in the\nillustrative case quoted, the beautiful house) and of the \u2018object\u2019\nself; the impressions are severally the pleasure immediately caused\nby the \u2018subject\u2019 (in the case given, the pleasure of feeling beauty)\nand the pleasure of pride. The relation between the ideas may be\nany of the \u2018natural ones\u2019 that regulate association. [4] In the\nsupposed case it is that of cause and effect, since a man\u2019s property\n\u2018produces effects on him and he on it.\u2019 The relation between the\nimpressions must be that of resemblance--this, as we are told by the\nway (somewhat strangely, if impressions are only stronger ideas),\nbeing the only possible relation between impressions--the resemblance\nof one pleasure to another.\n\n[1] Vol. II., pp. 77 and 78. [Book II., part I., sec. II.]\n\n[2] Ibid., p. 79. [Book II., part I., sec. II.]\n\n[3] Vol. II., pp. 84, 85. [Book II., part I., sec. V.]\n\n[4] Book I., part 1, secs. 4 and 5.\n\nThis means that it takes its character from that which is not a\npossible \u2018impression\u2019.\n\n34. Pride, then, is a special sort of pleasure excited by another\nspecial sort of pleasure, and the distinction of the two sorts of\npleasure from each other depends on the character which each derives\nfrom an idea--one from the idea of self, the other from the idea of\nsome \u2018quality in a subject,\u2019 which may be the beauty of a picture, or\nthe achievement of an ancestor, or any other quality as unlike these\nas these are unlike each other, so long as the idea of it is capable\nof association with the idea of self. Apart from such determination\nby ideas, the pleasure of pride itself and the pleasure which excites\nit, on the separateness of which from each other Hume insists, could\nonly be separate in time and degree of liveliness--a separation\nwhich might equally obtain between successive feelings of pride. Of\nneither could anything be said but that it was pleasant--more or less\npleasant than the other, before or after it, as the case might be.\nIs the idea, then, that gives each impression its character, itself\nan impression grown fainter? It should be so, of course, if Hume\u2019s\ntheory of consciousness is to hold good, either in its general form,\nor in its application to morals, according to which all actions,\nthose moved by pride among the rest, have pleasure for their ultimate\nmotive; and no doubt he would have said that it was so. The idea of\nthe beauty of a picture, for instance, is the original impression\nwhich it \u2018makes on the senses\u2019 as more faintly retained by the mind.\nBut is the original impression _merely_ an impression--an impression\nundetermined by conceptions, and of which, therefore, as it is to the\nsubject of it, nothing can be said, but simply that it is pleasant?\nThis, too, in the particular instance of beauty, Hume seems to hold;\n[1] but if it is so, the idea of beauty, as determined by reference\nto the impression, is determined by reference to the indeterminate,\nand we know no more of the separate pleasure that excites the\npleasure of pride, when we are told that its source is an impression\nof beauty, than we did before. Apart from any other reference, we\nonly know that pride is a pleasure excited by a pleasure which is\nitself excited by a pleasure grown fainter. Of effect, proximate\ncause, and ultimate cause, only one and the same thing can be said,\nviz., that each feels pleasant. Meanwhile in regard to that other\nrelation from which the pleasure of pride, on its part, is supposed\nto take its character, the same question arises. This pleasure \u2018has\nself for its object.\u2019 Is self, then, an impression stronger or\nfainter? Can one feeling be said without nonsense to have another\nfeeling for its object? If it can, what specification is gained for\na pleasure or pain by reference to an object of which, as a mere\nfeeling, nothing more can be said than that it is a pleasure or pain?\nIf, on the other hand, the idea of self, relation to which makes the\nfeeling of pride what it is, and through it determines action, is\nnot a copy of any impression of sense or reflection--not a copy of\nany sight or sound, any passion or emotion [2]--how can it be true\nthat the ultimate determination of action in all cases arises from\npleasure or pain?\n\n[1] Vol. II., p. 96; IV [Book II., part I., sec. VIII.],\n\u2018Dissertation on the Passions,\u2019 II. 7.\n\n[2] Intr. to Vol. I., paragraph 208.\n\nHume\u2019s attempt to represent idea of self as derived from impression.\n\n35. From the pressure of such questions as these Hume offers us two\nmain subterfuges. One is furnished by his account of the self, as\n\u2018that succession of related ideas and impressions of which we have\nan intimate memory and consciousness\u2019 [1]--an account which, to an\nincurious reader, conveys the notion that \u2018self,\u2019 if not exactly an\nimpression, is something in the nature of an impression, while yet it\nseems to give the required determination to the impression which has\nthis for its \u2018object.\u2019 It is evident, however, that its plausibility\ndepends entirely on the qualification of the \u2018succession, &c.,\u2019 as\nthat of which we have an \u2018intimate consciousness.\u2019 The succession\nof impressions, simply as such, and in the absence of relation to\na single subject, is nothing intelligible at all. Hume, indeed,\nelsewhere represents it as constituting time, which, as we have\npreviously shown, [2] by itself it could not properly be said to do;\nbut if it could, the characterisation of pleasure as having time\nfor its object would not be much to the purpose. The successive\nimpressions and ideas are further said to be \u2018related,\u2019 _i.e._,\n_naturally_ related, according to Hume\u2019s sense of the term; but\nthis we have found means no more than that when two feelings have\nbeen often felt to be either like each other or \u2018contiguous,\u2019 the\nrecurrence of one is apt to be followed by the recurrence in fainter\nform of the other. This characteristic of the succession brings it\nno nearer to the intelligible unity which it must have, in order\nto be an object of which the idea makes the pleasure of pride what\nit is. The notion of its having such unity is really conveyed by\nthe statement that we have an \u2018intimate consciousness\u2019 of it. It is\nthrough these words, so to speak, that we read into the definition\nof self that conception of it which we carry with us, but of which\nit states the reverse. Now, however difficult it may be to say what\nthis intimate consciousness is, it is clear that it cannot be one of\nthe feelings, stronger or fainter--impressions or ideas--which the\nfirst part of the definition tells us form a succession, for this\nwould imply that one of them was at the same time all the rest. Nor\nyet can it be a compound of them all, for the fact that they are a\nsuccession is incompatible with their forming a compound. Here, then,\nis a consciousness, which is not an impression, and which we can\nonly take to be derived from impressions by supposing these to be\nwhat they first become in relation to this consciousness. In saying\nthat we have such a consciousness of the succession of impressions,\nwe say in effect that we are other than the succession. How, then,\nwithout contradiction, can our self be said to _be_ the succession of\nimpressions, &c.--a succession which in the very next word has to be\nqualified in a way that implies we are other than it? This question,\nonce put, will save us from surprise at finding that in one place,\namong frequent repetitions of the account of self already given, the\n\u2018succession &c.\u2019 is dropped, and for it substituted \u2018_the individual\nperson_ of whose actions and sentiments each of us is intimately\nconscious.\u2019 [3]\n\n[1] Vol, II., p. 77, &c. [Book II., part I., sec. II.]\n\n[2] Intr. to Vol. I., sec. 261.\n\n[3] Vol. II., p. 84. [Book II., part I., sec. V.]\n\nAnother device is to suggest a physiological account of pride.\n\n36. The other way of gaining an apparent determination for the\nimpression, pride, without making it depend on relation to that\nwhich is not an impression at all, corresponds to that appeal to\nthe \u2018anatomist\u2019 by the suggestion of which, it will be remembered,\nHume avoids the troublesome question, how the simple impressions of\nsense, undetermined by relation, can have that definite character\nwhich they must have if they are to serve as the elements of\nknowledge. The question in that case being really one that concerns\nthe simple impression, as it is for the consciousness of the\nsubject of it, Hume\u2019s answer is in effect a reference to what it\nis for the physiologist. So in regard to pride; the question being\nwhat character it can have, for the conscious subject of it, to\ndistinguish it from any other pleasant feeling, except such as is\nderived from a conception which is not an impression, Hume is ready\non occasion to suggest that it has the distinctive character which\nfor the physiologist it would derive from the nerves organic to it,\nif such nerves could be traced. \u2018We must suppose that nature has\ngiven to the organs of the human mind a certain disposition fitted\nto produce a peculiar impression or emotion, which we call PRIDE: to\nthis emotion she has assigned a certain idea, viz., that of SELF,\nwhich it never fails to produce. This contrivance of nature is easily\nconceived. We have many instances of such a situation of affairs.\nThe nerves of the nose and palate are so disposed, as in certain\ncircumstances to convey such peculiar sensations to the mind; the\nsensations of lust and hunger always produce in us the idea of those\npeculiar objects, which are suitable to each appetite. These two\ncircumstances are united in pride. The organs are so disposed as to\nproduce the passion; and the passion, after its production, naturally\nproduces a certain idea.\u2019 [1]\n\n[1] Vol. II., p. 85. [Book II., part I., sec. V.]\n\nFallacy of this: it does not tell us what pride is to the subject of\nit.\n\n37. Here, it will be noticed, the doctrine, that the pleasant emotion\nof pride derives its specific character from relation to the idea\nof self, is dropped. The emotion we call pride is supposed to be\nfirst produced, and then, in virtue of its specific character as\npride, to _produce_ the idea of self. [1] If the idea of self, then,\ndoes not give the pleasure its specific character, what does? \u2018That\ndisposition fitted to produce it,\u2019 Hume answers, which belongs to\nthe \u2018organs of the human mind.\u2019 Now either this is the old story of\nexplaining the soporific qualities of opium by its _vis soporifica_,\nor it means that the distinction of the pleasure of pride from other\npleasures, like the distinction of a smell from a taste, is due to\na particular kind of nervous irritation that conditions it, and\nmay presumably be ascertained by the physiologist. Whether such a\nphysical condition of pride can be discovered or no, it is not to\nthe purpose to dispute. The point to observe is that, if discovered,\nit would not afford an answer to the question to which an answer is\nbeing sought--to the question, namely, what the emotion of pride is\nto the conscious subject of it. If it were found to be conditioned\nby as specific a nervous irritation as the sensations of smell and\ntaste to which Hume assimilates it, it would yet be no more the\nconsciousness of such irritation than is the smell of a rose to the\nperson smelling it. In the one case as in the other, the feeling,\nas it is to the subject of it, can only be determined by relation\nto other feelings or other modes of consciousness. It is by such a\nrelation that, according to Hume\u2019s general account of it, pride is\ndetermined, but the relation is to the consciousness of an object\nwhich, not being any form of feeling, has no proper place in his\npsychology. Hence in the passage before us he tries to substitute for\nit a physical determination of the emotion, which for the subject\nof it is no determination at all; and, having gained an apparent\nspecification for it in this way, to represent as its product that\nidea of a distinctive object which he had previously treated as\nnecessary to constitute it. Pride produces the idea of self, just as\n\u2018the sensations of hunger and lust always produce in us the idea of\nthose peculiar objects, which are suitable to each appetite.\u2019 Now\nit is a large assumption in regard to animals other than men, that,\nbecause hunger and lust move them to eat and generate, they so move\nthem through the intervention of any ideas _of objects_ whatever--an\nassumption which in the absence of language on the part of the\nanimals it is impossible to verify--and one still more questionable,\nthat the ideas of objects which these appetites (if it be so) produce\nin the animals, except as determined by self-consciousness, are ideas\nin the same sense as the idea of self. But at any rate, if such\nfeelings produce ideas of peculiar objects, it must be in virtue\nof the distinctive character which, as feelings, they have for the\nsubjects of them. The withdrawal, however, of determination by the\nidea of self from the emotion of pride, leaves it with no distinctive\ncharacter whatever, and therefore with nothing by which we may\nexplain its production of that idea as analogous to the production by\nhunger, if we admit such to take place, of the \u2018idea of the peculiar\nobject suited to it.\u2019\n\n[1] Cf. Vol. IV., \u2018Dissertation on the Passions,\u2019 II. 2.\n\nAccount of love involves the same difficulties; and a further one as\nto nature of sympathy.\n\n38. If, in Hume\u2019s account of pride, for _pleasure_, wherever it\noccurs, is substituted _pain_, it becomes his account of humility.\nA criticism of one account is equally a criticism of the other; and\nwith him every passion that \u2018has self for its object,\u2019 according as\nit is pleasant or painful, is included under one or other of these\ndesignations. In like manner, every passion that has \u2018some other\nthinking being\u2019 for its object, according as it is pleasant or\npainful, is either love or hatred. To these the key is to be found in\nthe same \u2018double relation of impressions and ideas\u2019 by which pride\nand humility are explained. If beautiful pictures, for instance,\nbelong not to oneself but to another person, they tend to excite\nnot pride but esteem, which is a form of love. The idea of them is\n\u2018naturally related\u2019 to the idea of the person to whom they belong,\nand they cause a separate pleasure which naturally excites the\nresembling impression of which this other person is the object. Write\n\u2018other person,\u2019 in short, where before was written \u2018self,\u2019 and the\naccount of pride and humility becomes the account of love and hatred.\nOf this pleasure determined by the idea of another person, or of\nwhich such a person \u2018is the object,\u2019 Hume gives no _rationale_, and,\nfailing this, it must be taken to imply the same power of determining\nfeeling on the part of a conception not derived from feeling, which\nwe have found to be implied in the pleasure of which self is the\nobject. All his pains and ingenuity in the second part of the book\n\u2018on the Passions,\u2019 are spent on illustrating the \u2018double relation\nof impressions and ideas\u2019--on characterising the separate pleasures\nwhich excite the pleasure of love, and showing how the idea of the\nobject of the exciting pleasure is related to the idea of the beloved\nperson. The objection to this part of his theory, which most readily\nsuggests itself to a reader, arises from the essential discrepancy\nwhich in many cases seems to lie between the exciting and the excited\npleasure. The drinking of fine wine, and the feeling of love, are\ndoubtless \u2018resembling impressions,\u2019 so far as each is pleasant,\nand from the idea of the wine the transition is natural to that of\nthe person who gives it; but is there really anything, it will be\nasked, in my enjoyment of a rich man\u2019s wine, that tends to make me\nlove him, even in the wide sense of \u2018love\u2019 which Hume admits? This\nobjection, it will be found, is so far anticipated by Hume, that in\nmost cases he treats the exciting pleasure as taking its character\nfrom sympathy. Thus it is not chiefly the pleasure of ear, sight,\nand palate, caused by the rich man\u2019s music, and gardens, and wine,\nthat excites our love for him, but the pleasure we experience through\nsympathy with his pleasure in them. [1] The explanation of love being\nthus thrown back on sympathy (which had previously served to explain\nthat form of pride which is called \u2018love of fame\u2019), we have to ask\nwhether sympathy is any less dependent than we have found pride to be\non an originative, as distinct from a merely reproductive, reason.\n\n[1] Vol. II., p. 147. [Book II., part II., sec. V.]\n\nHume\u2019s account of sympathy.\n\n39. \u2018When any affection is infused by sympathy, it is at first known\nonly by its effects, and by those external signs in the countenance\nand conversation which convey an idea of it.\u2019 By inference from\neffect to cause, \u2018we are convinced of the reality of the passion,\u2019\nconceiving it \u2018to belong to another person, as we conceive any other\nmatter of fact.\u2019 This idea of another\u2019s affection \u2018is presently\nconverted into an impression, and acquires such a degree of force and\nvivacity as to become the very passion itself, and produce an equal\nemotion as any original affection.\u2019 The conversion is not difficult\nto account for when we reflect that \u2018all ideas are borrowed from\nimpressions, and that these two kinds of perceptions differ only\nin the degrees of force and vivacity with which they strike upon\nthe soul.... As this difference may be removed in some measure by a\nrelation between the impressions and ideas\u2019--in the case before us,\nthe relation between the impression of one\u2019s own person and the idea\nof another\u2019s, by which the vivacity of the former may be conveyed to\nthe latter--\u2018\u2019tis no wonder an idea of a sentiment or passion may\nby this means be so enlivened as to become the very sentiment or\npassion.\u2019 [1]\n\n[1] Vol. II., pp. 111-114. [Book II., part I., sec. IX.]\n\nIt implies a self-consciousness not reducible to impressions.\n\n40. Upon this it must be remarked that the inference from the\nexternal signs of an affection, according to Hume\u2019s doctrine of\ninference, can only mean that certain impressions of the other\nperson\u2019s words and gestures call up the ideas of their \u2018usual\nattendants\u2019; which, again, must mean either that they convey the\nbelief in certain exciting circumstances experienced by the other\nman, and the expectation of certain acts to follow upon his words\nand gestures; or else that they suggest to the spectator the memory\nof certain like manifestations on his own part and through these of\nthe emotion which in his own case was their antecedent. Either way,\nthe spectator\u2019s idea of the other person\u2019s affection is in no sense a\ncopy of it, or that affection in a fainter form. If it is an idea of\nan impression _of reflection_ at all, it is of such an impression as\nexperienced by the spectator himself, and determined, as Hume admits,\nby his consciousness of himself; nor could any conveyance of vivacity\nto the idea make it other than that impression. How it should\nbecome to the spectator consciously at once another\u2019s impression\nand his own, remains unexplained. Hume only seems to explain it by\nmeans of the equivocation lurking in the phrase, \u2018idea of another\u2019s\naffection.\u2019 The reader, not reflecting that, according to the copying\ntheory, so far as the idea is a copy of anything _in the other_, it\ncan only be a copy of certain \u2018external signs, &c.,\u2019 and so far as\nit is a copy _of an affection_, only of an affection experienced by\nthe man who has the idea, thinks of it as being to the spectator\nthe other\u2019s affection minus a certain amount of vivacity--the\nrestoration of which will render it an impression at once his own\nand the other\u2019s. It can in truth only be so in virtue _(a)_ of an\ninterpretation of words and gestures, as related to a person, which\nno suggestion by impressions of their usual attendants can account\nfor, and in virtue _(b)_ of there being such a conceived identity,\nor unity in difference, between the spectator\u2019s own person and the\nperson of the other that the same impression, in being determined by\nhis consciousness of himself, is determined also by his consciousness\nof the other as an \u2018alter ego.\u2019 Thus sympathy, according to Hume\u2019s\naccount of it, so soon as that account is rationalized, is found\nto involve the determination of pleasure and pain, not merely by\nself-consciousness, but by a self-consciousness which is also\nself-identification with another. If self-consciousness cannot in\nany of its functions be reduced to an impression or succession of\nimpressions, least of all can it in this. On the other hand, if it\nis only through its constitutive action, its reflection of itself,\nupon successive impressions of sense that these become the permanent\nobjects which we know, we can understand how by a like action on\ncertain impressions of reflection, certain emotions and desires, it\nconstitutes those objects of interest which we love as ourselves.\n\nAmbiguity in his account of benevolence: It is a desire and therefore\nhas pleasure for its object. What pleasure?\n\n41. Pride, love, and sympathy, then, are the motives which Hume\nmust have granted him, if his moral theory is to march. Sympathy is\nnot only necessary to his explanation of that most important form\nof pride which is the motive to a man in maintaining a character\nwith his neighbours when \u2018nothing is to be gained by it\u2019--nothing,\nthat is, beyond the immediate pleasure it gives--and of all forms\nof \u2018love,\u2019 except those of which the exciting cause lies in the\npleasures of beauty and sexual appetite: he finds in it also the\nground of benevolence. Where he first treats of benevolence, indeed,\nthis does not appear. Unlike pride and humility, we are told, which\n\u2018are pure emotions of the soul, unattended with any desire, and\nnot immediately exciting us to action, love and hatred are not\ncompleted within themselves ... Love is always followed by a desire\nof the happiness of the person beloved, and an aversion to his\nmisery; as hatred produces a desire of the misery, and an aversion\nto the happiness, of the person hated.\u2019 [1] This actual sequence of\n\u2018benevolence\u2019 and \u2018anger\u2019 severally upon love and hatred is due, it\nappears, to \u2018an original constitution of the mind\u2019 which cannot be\nfurther accounted for. That benevolence is no essential part of love\nis clear from the fact that the latter passion \u2018may express itself\nin a hundred ways, and may subsist a considerable time, without\nour reflecting on the happiness of its object.\u2019 Doubtless, when we\ndo reflect on it, we desire the happiness; but, \u2018if nature had so\npleased, love might have been unattended with any such desire.\u2019 [2]\nSo far, the view given tallies with what we have already quoted from\nthe summary account of the direct and indirect passions, where the\n\u2018desire of punishment to our enemies and happiness to our friends\u2019\nis expressly left outside the general theory of the passions as a\n\u2018natural impulse wholly unaccountable,\u2019 a \u2018direct passion\u2019 which yet\ndoes not proceed from pleasure.\u2019 With his instinct for consistency,\nhowever, Hume could scarcely help seeking to assimilate this alien\nelement to his definition of desire as universally for pleasure;\nand accordingly, while the above view of benevolence is never in so\nmany words given up, an essentially different one appears a little\nfurther on, which by help of the doctrine of sympathy at once makes\nthe connection of benevolence with love more accountable, and brings\nit under the general definition of desire. \u2018Benevolence,\u2019 we are\nthere told, \u2018is an original pleasure arising from the pleasure of\nthe person beloved, and a pain proceeding from his pain, from which\ncorrespondence of impressions there arises a subsequent desire of his\npleasure and aversion to his pain.\u2019 [3]\n\n[1] Vol. II., p. 153. [Book II., part II., sec. VI.]\n\n[2] Vol. II., p. 154. [Book II., part II., sec. VI.]\n\n[3] Vol. II., p. 170. [Book II., part II., sec. IX.] Compare Vol.\nII., \u2018Inquiry concerning the Principles of Morals,\u2019 Appendix II.,\n_note_ 3, where \u2018general benevolence,\u2019 also called \u2018humanity,\u2019 is\nidentified with \u2018sympathy.\u2019 \u2018Benevolence is naturally divided into\ntwo kinds, the _general_ and the _particular_. The first is, where we\nhave no friendship, or connection, or esteem for the person, but feel\nonly a general sympathy with him, or a compassion for his pains, and\na congratulation with his pleasures,\u2019 &c. &c.\n\nPleasure of sympathy with the pleasure of another.\n\n42. Now, strictly construed, this passage seems to efface the one\nclear distinction of benevolence that had been previously insisted\non--that it is a desire, namely, as opposed to a pure emotion. If\nbenevolence _is_ an \u2018original pleasure arising from the pleasure of\nthe person beloved,\u2019 it is identical with love, so far as sympathy\nis an exciting cause of love, instead of being distinguished from it\nas desire from emotion. We must suppose, however, that the sentence\nwas carelessly put together, and that Hume did not really mean to\nidentify benevolence with the pleasure spoken of in the former part\nof it (for which his proper term is simply sympathy), but with the\ndesire for that pleasure, spoken of in the latter part. In that case\nwe find that benevolence forms no exception to the general definition\nof desire. It is desire for one\u2019s own pleasure, but for a pleasure\nreceived through the communication by sympathy of the pleasure of\nanother. In like manner, the sequence of benevolence upon love,\ninstead of being an unaccountable \u2018disposition of nature,\u2019 would seem\nexplicable, as merely the ordinary sequence upon a pleasant emotion\nof a desire for its renewal. Though it be not strictly the pleasant\nemotion of love, but that of sympathy, for which benevolence is the\ndesire, yet if sympathy is necessary to the excitement of love,\nit will equally follow that benevolence attends on love. Pleasure\nsympathised with, we may suppose, first excites the secondary emotion\nof love, and afterwards, when reflected on, that desire for its\ncontinuance or renewal, which is benevolence. That love \u2018should\nexpress itself in a hundred ways, and subsist a considerable time\u2019\nwithout any consciousness of benevolence, will merely be the natural\nrelation of emotion to desire. When a pleasure is in full enjoyment,\nit cannot be so reflected on as to excite desire; and thus, if\nbenevolence is desire for that pleasure in the pleasure of another,\nwhich is an exciting cause of love, the latter emotion must naturally\nsubsist and express itself for some time before it reaches the stage\nin which reflection on its cause, and with it benevolent desire,\nensues.\n\nAll \u2018passions\u2019 equally interested or disinterested: Confusion arises\nfrom use of \u2018passion\u2019 alike for desire and emotion. Of this Hume\navails himself in his account of active pity.\n\n43. This _rationale_, however, of the relation between love and\nbenevolence is not explicitly given by Hume himself. He nowhere\nexpressly withdraws the exception, made in favour of benevolence, to\nthe rule that all desire is for pleasure--an exception which, once\nadmitted, undermines his whole system--or tells us in so many words\nthat benevolence is desire for pleasure to oneself in the pleasure\nof another. In an important note to the Essays, [1] indeed, he\ndistinctly puts benevolence on the same footing with such desires\nas avarice or ambition. \u2018A man is no more interested when he seeks\nhis own glory, than when the happiness of his friend is the object\nof his wishes; nor is he any more disinterested when he sacrifices\nhis own ease and quiet to public good, than when he labours for the\ngratification of avarice or ambition.\u2019 ... \u2018Though the satisfaction\nof these latter passions gives us enjoyment, yet the prospect of this\nenjoyment is not the cause of the passion, but, on the contrary, the\npassion is antecedent to the enjoyment, and without the former the\nlatter could not possibly exist.\u2019 In other words, if \u2018passion\u2019 means\n_desire_--and, as applied to _emotion_, the designation \u2018interested\u2019\nor \u2018disinterested\u2019 has no meaning--every passion is equally\ndisinterested in the sense of presupposing an \u2018enjoyment\u2019 a pleasant\nemotion, antecedent to that which consists in its satisfaction; but\nat the same time equally interested in the sense of being a desire\nfor such enjoyment. Whether from a wish to find acceptance, however,\nor because forms of man\u2019s good-will to man forced themselves on\nhis notice which forbade the consistent development of his theory,\nHume is always much more explicit about the disinterestedness of\nbenevolence in the former sense than about its interestedness in\nthe latter. [2] Accordingly he does not avail himself of such an\nexplanation of its relation to love as that above indicated, which\nby avowedly reducing benevolence to a desire for pleasure, while it\nsimplified his system, might have revolted the \u2018common sense\u2019 even of\nthe eighteenth century. He prefers--as his manner is, when he comes\nupon a question which he cannot face--to fall back on a \u2018disposition\nof nature\u2019 as the ground of the \u2018conjunction\u2019 of benevolence with\nlove. There is a form of benevolence, however, which would seem as\nlittle explicable by such natural conjunction as by reduction to a\ndesire for sympathetic pleasure. How is it that active good-will is\nshown towards those whom, according to Hume\u2019s theory of love, it\nshould be impossible to love--towards those with whom intercourse\nis impossible, or from whom, if intercourse is possible, we can\nderive no such pleasure as is supposed necessary to excite that\npleasant emotion, but rather such pain, in sympathy with their pain,\nas according to the theory should excite hatred? To this question\nHume in effect finds an answer in the simple device of using the\nsame terms, \u2018pity\u2019 and \u2018compassion,\u2019 alike for the painful _emotion_\nproduced by the spectacle of another\u2019s pain and for \u2018desire for the\nhappiness of another and aversion to his misery.\u2019 [3] According\nto the latter account of it, pity is already \u2018the same desire\u2019\nas benevolence, though \u2018proceeding from a different principle,\u2019\nand thus has a resemblance to the love with which benevolence is\nconjoined--a \u2018resemblance not of feeling or sentiment but of tendency\nor direction.\u2019 [4] Hence, whereas \u2018pity\u2019 in the former sense would\nmake us hate those whose pain gives us pain, by understanding it in\nthe latter sense we can explain how it leads us to love them, on the\nprinciple that one resembling passion excites another.\n\n[1] \u2018Inquiry concerning Human Understanding,\u2019 note to sec. 1. In the\neditions after the second, this note was omitted.\n\n[2] Attention should be called to a passage at the end of the account\nof \u2018self-love\u2019 in the Essays, where he seems to revert to the view of\nbenevolence as a desire not _originally_ produced by pleasure, but\nproductive of it, and thus passing into a secondary stage in which\nit is combined with desire for pleasure. He suggests tentatively\nthat \u2018from the original frame of our temper we may feel a desire\nfor another\u2019s happiness or good, which, by means of that affection,\nbecomes our own good, and is afterwards pursued from the combined\nmotives of benevolence and self-enjoyment.\u2019 The passage might have\nbeen written by Butler. (Vol. IV., \u2018Inquiry concerning Principles of\nMorals,\u2019 Appendix II.)\n\n[3] Book II., part 2, secs. 7 and 9. Within a few lines of each\nother will be found the statements _(a)_ that \u2018pity is an uneasiness\narising from the misery of others,\u2019 and _(b)_ that \u2018pity is desire\nfor the happiness of another,\u2019 &c.\n\n[4] \u2018Dissertation on the Passions\u2019 (in the Essays), sec. 3, sub-sec.\n5.\n\nExplanation of apparent conflict between reason and passion.\n\n44. We are now in a position to review the possible motives of human\naction according to Hume. Reason, constituting no objects, affords no\nmotives. \u2018It is only the slave of the passions, and can never pretend\nto any other office than to serve and obey them.\u2019 [1] To any logical\nthinker who accepted Locke\u2019s doctrine of reason, as having no other\nfunction but to \u2018lay in order intermediate ideas,\u2019 this followed of\nnecessity. It is the clearness with which Hume points out that, as it\ncannot move, so neither can it restrain, action, that in this regard\nchiefly distinguishes him from Locke. The check to any passion, he\npoints out, can only proceed from some counter-motive, and such\na motive reason, \u2018having no original influence,\u2019 cannot give.\nStrictly speaking, then, a passion can only be called unreasonable,\nas accompanied by some false judgment, which on its part must\nconsist in \u2018disagreement of ideas, considered as copies, with those\nobjects which they represent;\u2019 and \u2018even then it is not the passion,\nproperly speaking, which is unreasonable, but the judgment.\u2019 It is\nnothing against reason--not, as Locke had inadvertently said, a\nwrong judgment--\u2018to prefer my own acknowledged lesser good to my\ngreater.\u2019 The only unreasonableness would lie in supposing that \u2018my\nown acknowledged lesser good,\u2019 being preferred, could be attained by\nmeans that would not really lead to it. Hence \u2018we speak not strictly\nwhen we talk of the combat of reason and passion.\u2019 They can in truth\nnever oppose each other. The supposition. that they do so arises\nfrom a confusion between \u2018calm passions\u2019 and reason--a confusion\nfounded on the fact that the former \u2018produce little emotion in the\nmind, while the operation of reason produces none at all.\u2019 [2] Calm\npassions, undoubtedly, do often conflict with the violent ones and\neven prevail over them, and thus, as the violent passion causes most\nuneasiness, it is untrue to say with Locke [3] that it is the most\npressing uneasiness which always determines action. The calmness of a\npassion is not to be confounded with weakness, nor its violence with\nstrength. A desire may be calm either because its object is remote,\nor because it is customary. In the former case, it is true, the\ndesire is likely to be relatively weak; but in the latter case, the\ncalmer the desire, the greater is likely to be its strength, since\nthe repetition of a desire has the twofold effect, on the one hand\nof diminishing the \u2018sensible emotion\u2019 that accompanies it, on the\nother hand of \u2018bestowing a facility in the performance of the action\u2019\ncorresponding to the desire, which in turn creates a new inclination\nor tendency that combines with the original desire. [4]\n\n[1] Vol. II., p. 195. [Book II., part III., sec. III.]\n\n[2] Vol. II., pp. 195, 196. [Book II., part III., sec. III.]\n\n[3] Above, sec. 3.\n\n[4] Vol. II. pp. 198-200. [Book II., part III., sec. IV.] It will be\nfound that here Hume might have stated his case much more succinctly\nby avoiding the equivocal use of \u2018passion\u2019 at once for \u2018desire\u2019 and\n\u2018emotion.\u2019 When a \u2018passion\u2019 is designated as \u2018calm\u2019 or \u2018violent,\u2019\n\u2018passion\u2019 means emotion. When the terms \u2018strong\u2019 and \u2018weak\u2019 are\napplied to it, it means \u2018desire.\u2019 Since of the strength of any\ndesire there is in truth no test but the resulting action, and habit\nfacilitates action, if we will persist in asking the idle question\nabout the relative strength of desires, we must suppose that the most\nhabitual is the strongest.\n\nA \u2018reasonable\u2019 desire means one that excites little emotion:\nEnumeration of possible motives.\n\n45. The distinction, then, between \u2018reasonable\u2019 and \u2018unreasonable\u2019\ndesires--and it is only _desires_ that can be referred to when will,\nor the determination to action, is in question--in the only sense\nin which Hume can admit it, is a distinction not of objects but\nof our situation in regard to them. The object of desire in every\ncase--whether near or remote, whether either by its novelty or by\nits contrariety to other passions it excites more or less \u2018sensible\nemotion\u2019--is still \u2018good,\u2019 _i.e._ pleasure. The greater the pleasure\nin prospect, the stronger the desire. [1] The only proper question,\nthen, according to Hume, as to the pleasure which in any particular\ncase is an object of desire will be whether it is _(a)_ an immediate\nimpression of sense, or _(b)_ a pleasure of pride, or _(c)_ one\nof sympathy. Under the first head, apparently, he would include\npleasures incidental to the satisfaction of appetite, and pleasures\ncorresponding to the several senses--not only the smells and tastes\nwe call \u2018sweet,\u2019 but the sights and sounds we call \u2018beautiful.\u2019 [2]\nPleasures of this sort, we must suppose, are the _ultimate_ \u2018exciting\ncauses\u2019 [3] of all those secondary ones, which are distinguished\nfrom their \u2018exciting causes\u2019 as determined by the ideas either of\nself or of another thinking person--the pleasures, namely, of pride\nand sympathy. Sympathetic pleasure, again, will be of two kinds,\naccording as the pleasure in the pleasure of another does or does\nnot excite the further pleasure of love for the other person. If the\nobject desired is none of these pleasures, nor the means to them, it\nonly remains for the follower of Hume to suppose that it is \u2018pleasure\nin general\u2019--the object of \u2018self love.\u2019\n\n[1] Cf p. 198. [Book II., part III., sec. IV.] \u2018The same good, when\nnear, will cause a violent passion, which, when remote, produces\nonly a calm one.\u2019 The expression, here, is obviously inaccurate. It\ncannot be the _same good_ in Hume\u2019s sense, _i.e._ equally pleasant in\nprospect, when remote as when near.\n\n[2] No other account of pleasure in beauty can be extracted from Hume\nthan this--that it is either a \u2018primary impression of sense,\u2019 so far\nco-ordinate with any pleasant taste or smell that but for an accident\nof language the term \u2018beautiful\u2019 might be equally applicable to\nthese, or else a pleasure in that indefinite anticipation of pleasure\nwhich is called the contemplation of utility.\n\n[3] _Ultimate_ because according to Hume the _immediate_ exciting\ncause of a pleasure of pride may be one of love, and vice versa. In\nthat case, however, a more remote \u2018exciting cause\u2019 of the exciting\npleasure must be found in some impressions of sense, if the doctrine\nthat these are the sole \u2018original impressions\u2019 is to be maintained.\n\nIf pleasure sole motive, what is the distinction of self-love?\nIts opposition to disinterested desires, as commonly understood,\ndisappears: it is desire for pleasure in general.\n\n46. Anyone reading the \u2018Treatise on Human Nature\u2019 alongside of\nShaftesbury or Butler would be surprised to find that while sympathy\nand benevolence fill a very large place in it, self-love \u2018eo nomine\u2019\nhas a comparatively small one. At first, perhaps, he would please\nhimself with thinking that he had come upon a more \u2018genial\u2019 system\nof morals. The true account of the matter, however, he will find to\nbe that, whereas with Shaftesbury and his followers the notion of\nself-love was really determined by opposition to those desires for\nother objects than pleasure, in the existence of which they really\nbelieved, however much the current psychology may have embarrassed\ntheir belief, on the other hand with Hume\u2019s explicit reduction of all\ndesire to desire for pleasure self-love loses the significance which\nthis opposition gave it, and can have no meaning except as desire for\n\u2018pleasure in general\u2019 in distinction from this or that particular\npleasure. Passages from the Essays may be adduced, it is true,\nwhere self-love is spoken of under the same opposition under which\nShaftesbury and Hutcheson conceived of it, but in these, it will be\nfound, advantage is taken of the ambiguity between \u2018emotion\u2019 and\n\u2018desire,\u2019 covered by the term \u2018passion.\u2019 That there are sympathetic\n_emotions_--pleasures occasioned by the pleasure of others--is, no\ndoubt, as cardinal a point in Hume\u2019s system as that all _desire_\nis for pleasure to self; but between such emotions and self-love\nthere is no co-ordination. No emotion, as he points out, determines\naction directly, but only by exciting desire; which with him can\nonly mean that the image of the pleasant emotion excites desire for\nits renewal. In other words, no emotion amounts to volition or will.\nSelf-love, on the other hand, if it means anything, means desire\nand a possibly strongest desire, or will. It can thus be no more\ndetermined by opposition to generous or sympathetic _emotions_ than\ncan these by opposition to hunger and thirst. Hume, however, when\nhe insists on the existence of generous \u2018passions\u2019 as showing that\nself-love is not our uniform motive, though he cannot consistently\nmean more than that desire for \u2018pleasure in general,\u2019 or desire for\nthe satisfaction of desire, is not the uniform motive--which might\nequally be shown (as he admits) by pointing to such self-regarding\n\u2018passions\u2019 as love of fame, or such appetites as hunger--is yet\napt, through the reader\u2019s interpretation of \u2018generous passions\u2019 as\n_desires_ for something other than pleasure, to gain credit for\nrecognising a possibility of living for others, in distinction from\nliving for pleasure, which was in truth as completely excluded by his\ntheory as by that of Hobbes. If he himself meant to convey any other\ndistinction between self-love and the generous passions than one\nwhich would hold no less between it and every emotion whatever, it\nwas through a fresh intrusion upon him of that notion of benevolence,\nas a \u2018desire not founded on pleasure,\u2019 which was in too direct\ncontradiction to the first principles of his theory to be acquiesced\nin. [1]\n\n[1] Cf. II. p. 197 [Book II., part III., sec. III.], where, speaking\nof \u2018calm desires,\u2019 he says they \u2018are of two kinds; either certain\ninstincts originally implanted in our natures, such as benevolence\nand resentment, the love of life, and kindness to children; or the\ngeneral appetite to good and aversion to evil, considered merely as\nsuch.\u2019 This seems to imply a twofold distinction of the \u2018general\nappetite to good\u2019 _(a)_ from desires for particular pleasures, which\nare commonly not calm, and _(b)_ from certain desires, which resemble\nthe \u2018general appetite\u2019 in being calm but are not for pleasure at all.\nSee above, sec. 31. In that section of the Essays where \u2018self-love\u2019\nis expressly treated of, there is a still clearer appearance of the\ndoctrine, that there are desires (in that instance called \u2018mental\npassions\u2019) which have not pleasure for their object any more than\nhave such \u2018bodily wants\u2019 as hunger and thirst. From these self-love,\nas desire for pleasure, is distinguished, though, when the pleasure\nincidental to their satisfaction is discovered and reflected on, it\nis supposed to combine with them. (Vol. IV. Appendix on Self-love,\nnear the end. See above, sec. 43 and note.)\n\nThis amounts, in fact, to a complete withdrawal from Hume\u2019s original\nposition and the adoption of one which is most clearly stated in\nHutcheson\u2019s posthumous treatise--the position, namely, that we begin\nwith a multitude of \u2018particular\u2019 or \u2018violent\u2019 desires, severally\n\u2018terminating upon objects\u2019 which are not pleasures at all, and that,\nas reason developes, these gradually blend with, or are superseded\nby, the \u2018calm\u2019 desire for pleasure; so that moral growth means the\naccess of conscious pleasure-seeking. This in effect seems to be\nButler\u2019s view, and Hutcheson reckons it \u2018a lovely representation\nof human nature,\u2019 though he himself holds that benevolence may\nexist, not merely as one of the \u2018particular desires\u2019 controlled\nby self-love, but as itself a \u2018calm\u2019 and controlling principle,\nco-ordinate with self-love. (\u2018System of Moral Philosophy,\u2019 Vol. I. p.\n51, &c.)\n\nHow Hume gives meaning to this otherwise unmeaning definition:\n\u2018Interest,\u2019 like other motives described, implies determination by\nreason.\n\n47. Such desire, then, being excluded, what other motive than\n\u2018interest\u2019 remains, by contrast with which the latter may be defined?\nIt has been explained above (\u00a77) that since pleasure as such, or\nas a feeling, does not admit of generality, \u2018pleasure in general\u2019\nis an impossible object. When the motive of an action is said to\nbe \u2018pleasure in general,\u2019 what is really meant is that the action\nis determined by the conception of pleasure, or, more properly,\nof self as a subject to be pleased. Such determination, again, is\ndistinguished by opposition to two other kinds--_(a)_ to that sort\nof determination which is not by conception, but either by animal\nwant, or by the animal _imagination_ of pleasure, and _(b)_ to\ndetermination by the conception of other objects than pleasure.\nBy an author, however, who expressly excluded the latter sort of\ndetermination, and who did not recognise any distinction between the\nthinking and the animal subject, the motive in question could not\nthus be defined. Hence the difficulty of extracting from Hume himself\nany clear and consistent account of that which he variously describes\nas the \u2018general appetite for good, considered merely as such,\u2019 as\n\u2018interest,\u2019 and as \u2018self-love.\u2019 To say that he understood by it a\ndesire for pleasure which is yet not a desire for any pleasure in\nparticular, may seem a strange interpretation to put on one who\nregarded himself as a great liberator from abstractions, but there\nis no other which his statements, taken together, would justify.\nThis desire for nothing, however, he converts into a desire for\nsomething by identifying it on occasion, (1) with any desire for a\npleasure of which the attainment is regarded as sufficiently remote\nto allow of calmness in the desire, and (2) with desire for the means\nof having all pleasures indifferently at command. It is in one or\nother of these senses--either as desire for some particular pleasure\ndistinguished only by its calmness, or as desire for power--that he\nalways understands \u2018interest\u2019 or \u2018self-love,\u2019 except where he gains\na more precise meaning for it by the admission of desires, not for\npleasure at all, to which it may be opposed. Now taken in the former\nsense, its difference from the desires for the several pleasures of\n\u2018sense,\u2019 \u2018pride,\u2019 and \u2018sympathy,\u2019 of which Hume\u2019s account has already\nbeen examined, cannot lie in the object, but--as he himself says of\nthe distinction, which he regarded as an equivalent one, between\n\u2018reasonable and unreasonable\u2019 desires--in our situation with regard\nto it. If then the object of each of these desires, as we have shown\nto be implied in Hume\u2019s account of them, is one which only reason,\nas self-consciousness, can constitute, it cannot be less so when the\ndesire is calm enough to be called self-love. Still more plainly is\nthe desire in question determined by reason--by the conception of\nself as a permanent susceptibility of pleasure--if it is understood\nto be desire for power.\n\nThus Hume, having degraded morality for the sake of consistency,\nafter all is not consistent.\n\n48. Having now before us a complete view of the possible motives to\nhuman action which Hume admits, we find that while he has carried to\nits furthest limit, and with the least verbal inconsistency possible,\nthe effort to make thought deny its own originativeness in action,\nhe has yet not succeeded. He has made abstraction of everything in\nthe objects of human interest but their relation to our nervous\nirritability--he has left nothing of the beautiful in nature or art\nbut that which it has in common with a sweetmeat, nothing of that\nwhich is lovely and of good report to the saint or statesman but\nwhat they share with the dandy or diner-out--yet he cannot present\neven this poor residuum of an object, by which all action is to be\nexplained, except under the character it derives from the thinking\nsoul, which looks before and after, and determines everything by\nrelation to itself. Thus if, as he says, the distinction between\nreasonable and unreasonable desires does not lie in the object,\nthis will not be because reason has never anything to do with the\nconstitution of the object, but because it has always so much to\ndo with it as renders selfishness--the self-conscious pursuit of\npleasure--possible. Sensuality then will have been vindicated, the\ndistinction between the \u2018higher\u2019 and \u2018lower\u2019 modes of life will have\nbeen erased, and after all the theoretic consistency--for the sake\nof which, and not, of course, to gratify any sinister interest, Hume\nmade his philosophic venture--will not have been attained. Man will\nstill not be ultimately passive, nor human action natural. Reason\nmay be the \u2018slave of the passions,\u2019 but it will be a self-imposed\nsubjection.\n\nIf all good is pleasure, what is _moral_ good? Ambiguity in Locke\u2019s\nview.\n\n49. We have still, however, to explain how Hume himself completes the\nassimilation of the moral to the natural; how, on the supposition\nthat the \u2018good\u2019 can only mean the \u2018pleasant,\u2019 he accounts for the\napparent distinction between moral and other good, for the intrusion\nof the \u2018ought and ought not\u2019 of ethical propositions upon the\n\u2018is and is not\u2019 of truth concerning nature. [1] Here again he is\nfaithful to his _r\u00f4le_ as the expander and expurgator of Locke. With\nLocke, it will be remembered, the distinction of _moral_ good lay\nin the channel through which the pleasure, that constitutes it, is\nderived. It was pleasure accruing through the intervention of law,\nas opposed to the operation of nature: and from the pleasure thus\naccruing the term \u2018morally good\u2019 was transferred to the act which, as\n\u2018conformable to some law,\u2019 occasions it. [2] This view Hume retains,\nmerely remedying Locke\u2019s omissions and inconsistencies. Locke, as we\nsaw, not only neglected to derive the existence of the laws, whose\nintervention he counted necessary to constitute the morally good,\nfrom the operation of that desire for pleasure which he pronounced\nthe only motive of man; in speaking of moral goodness as consisting\nin conformity to law, he might, if taken at his word, be held to\nadmit something quite different from pleasure alike as the standard\nand the motive of morality. Hume then had, in the first place, to\naccount for the laws in question, and so account for them as to\nremove that absolute opposition between them and the operation of\nnature which Locke had taken for granted; secondly, to exhibit that\nconformity to law, in which the moral goodness of an act was held\nto consist, as itself a mode of pleasure--pleasure, namely, to the\ncontemplator of the act; and thirdly, to show that not the moral\ngoodness of the act, even thus understood, but pleasure to himself\nwas the motive to the doer of it. [3]\n\n[1] Vol. II, p. 245. [Book III., part I., sec. I.]\n\n[2] Above, secs. 16-18.\n\n[3] Of the three problems here specified, Hume\u2019s treatment of the\n_second_ is discussed in the following secs. 50-54; of the _first_ in\nsecs. 55-58; of the _third_ in secs. 60 to the end.\n\nDevelopment of it by Clarke, which breaks down for want of true view\nof reason.\n\n50. It was a necessary incident of this process that Locke\u2019s notion\nof a Law of God, conformity to which rendered actions \u2018in their own\nnature right and wrong,\u2019 should disappear. The existence of such\na law cannot be explained as a result of any desire for pleasure,\nnor conformity to it as a mode of pleasure. Locke, indeed, tries\nto bring the goodness, consisting in such conformity, under his\ngeneral definition by treating it as equivalent to the production of\npleasure in another world. This, however, is to seek refuge from the\ncontradictory in the unmeaning. The question--Is it the pleasure it\nproduces, or its conformity to law, that constitutes the goodness of\nan act?--remains unanswered, while the further one is suggested--What\nmeaning has pleasure except as the pleasure we experience? [1]\nBetween pleasure, then, and a \u2018conformity\u2019 irreducible to pleasure,\nas the moral standard, the reader of Locke had to choose. Clarke,\nsupported by Locke\u2019s occasional assimilation of moral to mathematical\ntruth, had elaborated the notion of conformity. To him an action\nwas \u2018in its own nature right\u2019 when it conformed to the \u2018reason of\nthings\u2019--_i.e._ to certain \u2018eternal proportions,\u2019 by which God, \u2018qui\nomnia numero, ordine, mensur\u00e2 posuit,\u2019 obliges Himself to govern\nthe world, and of which reason in us is \u2018the appearance.\u2019 [2] Thus\nreason, as an eternal \u2018agreement or disagreement of ideas,\u2019 was the\nstandard to which action ought to conform, and, as our consciousness\nof such agreement, at once the judge of and motive to conformity.\nTo this Hume\u2019s reply is in effect the challenge to instance any\nact, of which the morality consists either in any of those four\nrelations, \u2018depending on the nature of the ideas related,\u2019 which he\nregarded as alone admitting of demonstration, or in any other of\nthose relations (contiguity, identity, and cause and effect) which,\nas \u2018matters of fact,\u2019 can be \u2018discovered by the understanding.\u2019 [3]\nSuch a challenge admits of no reply, and no other function but the\nperception of such relations being allowed to reason or understanding\nin the school of Locke, it follows that it is not this faculty which\neither constitutes, or gives the consciousness of, the morally good.\nReason excluded, feeling remains. No action, then, can be called\n\u2018right in its own nature,\u2019 if that is taken to imply (as \u2018conformity\nto divine law\u2019 must be), relation to something else than our feeling.\nIt could only be so called with propriety in the sense of exciting\nsome pleasure _immediately_, as distinct from an act which may be a\ncondition of the attainment of pleasure, but does not directly convey\nit.\n\n[1] Above, sec. 14.\n\n[2] Boyle Lectures, Vol. II, prop. 1. secs. 1-4.\n\n[3] Book III. part 1, sec. 1. (Cf. Book I part 3, sec. 1, and\nIntroduction to Vol. I, secs. 283 and ff.) It will be observed\nthat throughout the polemic against Clarke and his congeners Hume\nwrites as if there were a difference between objects of reason and\nfeeling, which he could not consistently admit. He begins by putting\nthe question thus (page 234), \u2018whether \u2018tis by means of our ideas\nor impressions we distinguish betwixt vice and virtue:\u2019 but if, as\nhe tells us, \u2018the idea is merely the weaker impression, and the\nimpression the stronger idea,\u2019 such a question has no meaning. In\nlike manner he concludes by saying (page 245) that \u2018vice and virtue\nmay be compared to sounds, colours, heat and cold, which are not\nqualities in objects, but perceptions in the mind.\u2019 But, since the\nwhole drift of Book I. is to show that all \u2018objective relations\u2019 are\nsuch \u2018perceptions\u2019 or their succession, this still leaves us without\nany distinction between science and morality that shall be tenable\naccording to his own doctrine.\n\nWith Hume, moral good is pleasure excited in a particular way, viz.:\nin the _spectator_ of the \u2018good\u2019 act and by the view of its tendency\nto produce pleasure.\n\n51. So far, however, there is nothing to distinguish the moral\nact either from any \u2018inanimate object,\u2019 which may equally excite\nimmediate pleasure, or from actions which have no character, as\nvirtuous or vicious, at all. Some further limitation, then, must\nbe found for the immediate pleasure which constitutes the goodness\ncalled \u2018moral,\u2019 and of which praise is the expression. This Hume\nfinds in the exciting object which must be _(a)_ \u2018considered in\ngeneral and without reference to our particular interest,\u2019 and _(b)_\nan object so \u2018related\u2019 (in the sense above [1] explained) to oneself\nor to another as that the pleasure which it excites shall cause the\nfurther pleasure either of pride or love. [2] The precise effect of\nsuch limitation he does not explain in detail. A man\u2019s pictures,\ngardens, and clothes, we have been told, tend to excite pride in\nhimself and love in others. If then we can \u2018consider them in general\nand without reference to our particular interest,\u2019 and in such \u2018mere\nsurvey\u2019 find pleasure, this pleasure, according to Hume\u2019s showing,\nwill constitute them morally good. [3] He usually takes for granted,\nhowever, a further limitation of the pleasure in question, as excited\nonly by \u2018actions, sentiments, and characters,\u2019 and thus finds virtue\nto consist in the \u2018satisfaction produced to the spectator of an\nact or character by the mere view of it.\u2019 [4] Virtues and vices\nthen mean, as Locke well said, the usual likes and dislikes of\nsociety. If we choose with him to call that virtue of an act, which\nreally consists in the pleasure experienced by the spectator of it,\n\u2018conformity to the law of their opinion,\u2019 we may do so, provided we\ndo not suppose that there is some other law, which this imperfectly\nreflects, and that the virtue is something other than the pleasure,\nbut to be inferred from it. \u2018We do not infer a character to be\nvirtuous, because it pleases; but in feeling that it pleases after\nsuch a particular manner, we in effect feel that it is virtuous.\u2019 [5]\n\n[1] Sec. 33.\n\n[2] Vol. II. pp. 247 and 248. [Book III., part I., sec. II.]\n\n[3] Hume treats them as such in Book III. part 3, sec. 5.\n\n[4] Vol. II. p. 251. Cf. p. 225. [Book III., part I., sec. II.; Book\nII., part III., sec. X.]\n\n[5] Vol. II. p. 247. [Book III., part I., sec. II.]\n\n52. Some further explanation, however, of the \u2018particular manner\u2019\nof this pleasure was clearly needed in order at once to adjust it\nto the doctrine previously given of the passions (of which this, as\na pleasant emotion, must be one), and to account for our speaking\nof the actions which excite it--at least of some of them--as\nactions which we _ought_ to do. If we revert to the account of the\npassions, we can have no difficulty in fixing on that of which this\npeculiar pleasure, excited by the \u2018mere survey\u2019 of an action without\nreference to the spectator\u2019s \u2018particular interest,\u2019 must be a mode.\nIt must be a kind of sympathy--pleasure felt by the spectator in\nthe pleasure of another, as distinct from what might be felt in the\nprospect of pleasure to himself. [1] On the other hand, there seem\nto be certain discrepancies between pleasure and moral sentiment.\nWe sympathise where we neither approve nor disapprove; and,\nconversely, we express approbation where it would seem there was no\npleasure to sympathise with, _e.g._, in regard to an act of simple\njustice, or where the person experiencing it was one with whom we\ncould have no fellow-feeling--an enemy, a stranger, a character in\nhistory--or where the experience, being one not of pleasure but of\npain (say, that of a martyr at the stake), should excite the reverse\nof approbation in the spectator, if approbation means pleasure\nsympathised with. Our sympathies, moreover, are highly variable, but\nour moral sentiments on the whole constant. How must \u2018sympathy\u2019 be\nqualified, in order that, when we identify moral sentiment with it,\nthese objections may be avoided?\n\n[1] Vol. II. pp. 335-337. [Book III., part III., sec. I.]\n\nMoral sense is thus sympathy with pleasure qualified by consideration\nof general tendencies.\n\n53. Hume\u2019s answer, in brief, is that the sympathy, which constitutes\nmoral sentiment, is sympathy qualified by the consideration of\n\u2018general tendencies.\u2019 Thus we sympathise with the pleasure arising\nfrom any casual action, but the sympathy does not become moral\napprobation unless the act is regarded as a sign of some quality\nor character, generally permanently agreeable or useful (_sc._ and\nproductive of pleasure directly or indirectly) to the agent or\nothers. An act of justice may not be productive of any immediate\npleasure with which we can sympathise; nay, taken singly, it may\ncause pain both in itself and in its results, as when a judge \u2018takes\nfrom the poor to give to the rich, or bestows on the dissolute the\nlabour of the industrious; \u2018but we sympathise with the general\nsatisfaction resulting to society from \u2018the whole scheme of law\nand justice,\u2019 to which the act in question belongs, and approve it\naccordingly. The constancy which leads to a dungeon is a painful\ncommodity to its possessor, but sympathy with his pain need not\nincapacitate a spectator for that other sympathy with the general\npleasure caused by such a character to others, which constitutes it\nvirtuous. Again, though remote situation or the state of one\u2019s temper\nmay at any time modify or suppress sympathy with the pleasure caused\nby the good qualities of any particular person, we may still apply to\nhim terms expressive of our liking. \u2018External beauty is determined\nmerely by pleasure; and \u2018tis evident a beautiful countenance cannot\ngive so much pleasure, when seen at a distance of twenty paces, as\nwhen it is brought nearer to us. We say not, however, that it appears\nto us less beautiful; because we know what effect it will have in\nsuch a position, and by that reflection we correct its momentary\nappearance.\u2019 As with the beautiful, so with the morally good. \u2018In\norder to correct the continual contradictions\u2019 in our judgment of\nit, that would arise from changes in personal temper or situation,\n\u2018we fix on some steady and general points of view, and always in\nour thoughts place ourselves in them, whatever may be our present\nsituation.\u2019 Such a point of view is furnished by the consideration\nof \u2018the interest or pleasure of the person himself whose character\nis examined, and of the persons who have a connection with him,\u2019 as\ndistinct from the spectator\u2019s own. The imagination in time learns to\n\u2018adhere to these general views, and distinguishes the feelings they\nproduce from those which arise from our particular and momentary\nsituation.\u2019 Thus a certain constancy is introduced into sentiments of\nblame and praise, and the variations, to which they continue subject,\ndo not appear in language, which \u2018experience teaches us to correct,\neven where our sentiments are more stubborn and unalterable.\u2019 [1]\n\n[1] Book III. Vol. II. part 3, sec. 1. Specially pp. 339, 342, 346,\n349.\n\nIn order to account for the facts it has to become sympathy with\nunfelt feelings.\n\n54. It thus appears that though the virtue of an act means the\npleasure which it causes to a spectator, and though this again arises\nfrom sympathy with imagined pleasure of the doer or others, yet\nthe former may be a pleasure which no particular spectator at any\ngiven time does actually feel--he need only know that under other\nconditions on his part he would feel it--and the latter pleasure\nmay be one either not felt at all by any existing person, or only\nfelt as the opposite of the uneasiness with which society witnesses\na departure from its general rules. Of the essential distinction\nbetween a feeling of pleasure or pain and a knowledge of the\nconditions under which a pleasure or pain is generally felt, Hume\nshows no suspicion; nor, while he admits that without substitution\nof the knowledge for the feeling there could be no general standard\nof praise or blame, does he ask himself what the quest for such a\nstandard implies. As little does he trouble himself to explain how\nthere can be such sympathy with an unfelt feeling--with a pleasure\nwhich no one actually feels but which is possible for posterity--as\nwill explain our approval of the virtue which defies the world,\nand which is only assumed, for the credit of a theory, to bring\npleasure to its possessor, because it certainly brings pleasure\nto no one else. For the \u2018artificial\u2019 virtue, however, of acts\ndone in conformity with the \u2018general scheme of justice,\u2019 or other\nsocial conventions, he accounts at length in part II. of his Second\nBook--that entitled \u2018Of Justice and Injustice.\u2019\n\nCan the distinction between the \u2018moral\u2019 and \u2018natural\u2019 be maintained\nby Hume? What is \u2018artificial virtue\u2019?\n\n55. To a generation which has sufficiently freed itself from all\n\u2018mystical\u2019 views of law--which is aware that \u2018natural right,\u2019 if it\nmeans a right that existed in a \u2018state of nature,\u2019 is a contradiction\nin terms; that, since contracts could not be made, or property\nexist apart from social convention, any question about a primitive\nobligation to respect them is unmeaning--the negative side of this\npart of the treatise can have little interest. That all rights and\nobligations are in some sense \u2018artificial,\u2019 we are as much agreed as\nthat without experience there can be no knowledge. The question is,\nhow the artifice, which constitutes them, is to be understood, and\nwhat are its conditions. If we ask what Hume understood by it, we can\nget no other answer than that the artificial is the opposite of the\nnatural. If we go on to ask for the meaning of the natural, we only\nlearn that we must distinguish the senses in which it is opposed to\nthe miraculous and to the unusual from that in which it is opposed\nto the artificial, [1] but not what the latter sense is. The truth\nis that, if the first book of Hume\u2019s treatise has fulfilled its\npurpose, the only conception of the natural, which can give meaning\nto the doctrine that the obligation to observe contracts and respect\nproperty is artificial, must disappear. There are, we shall find,\ntwo different negations which in different contexts this doctrine\nconveys. Sometimes it means that such an obligation did not exist\nfor man in a \u2018state of nature,\u2019 _i.e._, as man was to begin with.\nBut in that sense the law of cause and effect, without which there\nwould be no nature at all, is, according to Hume, not natural, for\nit--not merely our recognition of it, but the law itself--is a habit\nof imagination, gradually formed. Sometimes it conveys an opposition\nto Clarke\u2019s doctrine of obligation as constituted by certain \u2018eternal\nrelations and proportions,\u2019 which also form the order of nature, and\nare other than, though regulative of, the succession of our feelings.\nNature, however, having been reduced by Hume to the succession of\nour feelings, the \u2018artifice,\u2019 by which he supposes obligations to\nbe formed, cannot be determined by opposition to it, unless the\noperation of motives, which explains the artifice, is something else\nthan a succession of feelings. But that it is nothing else is just\nwhat it is one great object of the moral part of his treatise to show.\n\n[1] Book II. part 1, sec. 2.\n\nNo ground for such distinction in relation between motive and act.\n\n56. He is nowhere more happy than in exposing the fallacies by\nwhich \u2018liberty of indifferency\u2019--the liberty supposed to consist\nin a possibility of unmotived action--was defended. [1] Every act,\nhe shows, is determined by a strongest motive, and the relation\nbetween motive and act is no other than that between any cause and\neffect in nature. In one case, as in the other, \u2018necessity\u2019 lies\nnot in an \u2018esse\u2019 but in a \u2018percipi.\u2019 It is the \u2018determination of\nthe thought of any intelligent being, who considers \u2018an act or\nevent,\u2019 to infer its existence from some preceding objects;\u2019 [2]\nand such determination is a habit formed by, and having a strength\nproportionate to, the frequency with which certain phenomena--actions\nor events--have followed certain others. The weakness in this part of\nHume\u2019s doctrine lies, not in the assumption of an equal uniformity\nin the sequence of act upon motive with that which obtains in\nnature, but in his inability consistently to justify the assumption\nof an absolute uniformity in either case. When there is an apparent\nirregularity in the consequences of a given motive--when according to\none \u2018experiment\u2019 action _(a)_ follows upon it, according to another\naction _(b)_, and so on--although \u2018these contrary experiments are\nentirely equal, we remove not the notion of causes and necessity;\nbut, supposing that the usual contrariety proceeds from the operation\nof contrary and concealed causes, we conclude that the chance or\nindifference lies only in our judgment on account of our imperfect\nknowledge, not in the things themselves, which are in every case\nequally necessary, though to appearance not equally constant or\nuniform.\u2019 [3] But we have already seen that, if necessary connection\nwere in truth only a habit arising from the frequency with which\ncertain phenomena follow certain others, the cases of exception to a\nusual sequence, or in which the balance of chances did not incline\none way more than another, could only so far weaken the habit. The\nexplanation of them by the \u2018operation of concealed causes\u2019 implies,\nas he here says, an opposition of real necessity to apparent\ninconstancy, which, if necessity were such a habit as he says it is,\nwould be impossible. [4] This difficulty, however, applying equally\nto moral and natural sequences, can constitute no difference between\nthem. It cannot therefore be in the relation between motive and act\nthat the followers of Hume can find any ground for a distinction\nbetween the process by which the conventions of society are formed,\nand that succession of feelings which he calls nature. May he\nthen find it in the character of the motive itself by which the\n\u2018invention\u2019 of justice is to be accounted for? Is this other than a\nfeeling determined by a previous, and determining a sequent, one?\nNot, we must answer, as Hume himself understood his own account of\nit, which is as follows:-\n\n[1] Book II. part 3, secs. 1 and 2.\n\n[2] Vol. II. p, 189. [Book II., part III., sec. II.]\n\n[3] Ibid., p. 185. [Book II., part III., sec. I.]\n\n[4] See Introduction to Vol. I. secs. 323 and 336.\n\nMotive to artificial virtues.\n\n57. He will examine, he says, \u2018two questions, viz., concerning the\nmanner in which the rules of justice are established by the artifice\nof men; and concerning the reasons which determine us to attribute\nto the observance or neglect of these rules a moral beauty and\ndeformity.\u2019 [1] Of the motives which he recognises (\u00a7 45) it is\nclear that only two--\u2018benevolence\u2019 and \u2018interest\u2019--can be thought of\nin this connection, and a little reflection suffices to show that\nbenevolence cannot account for the artifice in question. Benevolence\nwith Hume means either sympathy with pleasure--and this (though Hume\ncould forget it on occasion) [2] must be a particular pleasure of\nsome particular person--or desire for the pleasure of such sympathy.\nEven if a benevolence may be admitted, which is not a desire for\npleasure at all but an impulse to please, still this can only be an\nimpulse to please some particular person, and the only effect of\nthought upon it, which Hume recognises, is not to widen its object\nbut to render it \u2018interested.\u2019 [3] \u2018There is no such passion in\nhuman minds as the love of mankind, merely as such, independent of\npersonal qualities, of services, or of relation to ourself.\u2019 [4] The\nmotive, then, to the institution of rules of justice cannot be found\nin general benevolence. [5] As little can it be found in private\nbenevolence, for the person to whom I am obliged to be just may be\nan object of merited hatred. It is true that, \u2018though it be rare\nto meet with one who loves any single person better than himself,\nyet \u2018tis as rare to meet with one in whom all the kind affections,\ntaken together, do not overbalance all the selfish\u2019; but they are\naffections to his kinsfolk and acquaintance, and the generosity which\nthey prompt will constantly conflict with justice. [6] \u2018Interest,\u2019\nthen, must be the motive we are in quest of. Of the \u2018three species of\ngoods which we are possessed of--the satisfaction of our minds, the\nadvantages of our body, and the enjoyment of such possessions as we\nhave acquired by our industry and good fortune\u2019--the last only \u2018may\nbe transferred without suffering any loss or alteration; while at the\nsame time there is not sufficient quantity of them to supply every\none\u2019s desires and necessities.\u2019 Hence a special instability in their\npossession. Reflection on the general loss caused by such instability\nleads to a \u2018tacit convention, entered into by all the members of a\nsociety, to abstain from each other\u2019s possessions;\u2019 and thereupon\n\u2018immediately arise the ideas of justice and injustice; as also those\nof property, right, and obligation.\u2019 It is not to be supposed,\nhowever, that the \u2018convention\u2019 is of the nature of a promise, for\nall promises presuppose it. \u2018It is only a general sense of common\ninterest; which sense all the members of the society express to\none another, and which induces them to regulate their conduct by\ncertain rules;\u2019 and this \u2018general sense of common interest,\u2019 it need\nscarcely be said, is every man\u2019s sense of his own interest, as in\nfact coinciding with that of his neighbours. In short, \u2018\u2019tis only\nfrom the selfishness and confined generosity of man, along with the\nscanty provision nature has made for his wants, that justice derives\nits origin.\u2019 [7]\n\n[1] Book III. part 2, sec. 2.\n\n[2] Cf. sec. 54.\n\n[3] Cf. secs. 42, 43, and 46.\n\n[4] Vol. II. p. 255. [Book III., part II., sec. I.]\n\n[5] For the sense in which Hume did admit a \u2018general benevolence,\u2019\nsee sec. 41, note.\n\n[6] Vol. II. pp. 256 and 260. [Book III., part II., sec. II.]\n\n[7] Vol. II. pp. 261, 263, 268. [Book III., part II., sec. II.]\n\nHow artificial virtues become moral.\n\n58. Thus the origin of rules of justice is explained, but the\nobligation to observe them so far appears only as \u2018interested,\u2019 not\nas \u2018moral.\u2019 In order that it may become \u2018moral,\u2019 a pleasure must be\ngenerally experienced in the spectacle of their observance, and a\npain in that of their breach, apart from reference to any gain or\nloss likely to arise to the spectator himself from that observance\nor breach. In accounting for this experience Hume answers the second\nof the questions, proposed above. \u2018To the imposition and observance\nof these rules, both in general and in every particular instance,\nmen are at first induced only by a regard to interest; and this\nmotive, on the first formation of society, is sufficiently strong and\nforcible. But when society has become numerous, and has increased\nto a tribe or nation, this interest is more remote; nor do men so\nreadily perceive that disorder and confusion follow upon each breach\nof these rules, as in a more narrow and contracted society. But\nthough, in our own actions, we may frequently lose sight of that\ninterest which we have in maintaining order, and may follow a lesser\nand more present interest, we never fail to observe the prejudice\nwe receive, either mediately or immediately, from the injustice of\nothers.... Nay, when the injustice is so distant from us, as no way\nto affect our interest, it still displeases us, because we consider\nit as prejudicial to human society, and pernicious to every one that\napproaches the person guilty of it. We partake of their uneasiness\nby _sympathy_; and as everything which gives uneasiness in human\nactions, upon the general survey, is called vice, and whatever\nproduces satisfaction, in the same manner, denominated virtue, this\nis the reason why the sense of moral good and evil follows upon\njustice and injustice. And though this sense, in the present case, be\nderived only from contemplating the actions of others, yet we fail\nnot to extend it even to our own actions. The _general rule_ reaches\nbeyond those instances from which it arose, while at the same time we\nnaturally _sympathise_ with others in the sentiments they entertain\nof us.\u2019 [1]\n\n[1] Vol. II. p. 271. [Book III., part II., sec. II.]\n\nInterest and sympathy account for all obligations civil and moral.\n\n59. To this account of the process by which rules of justice have not\nonly come into being, but come to bind our \u2018conscience\u2019 as they do,\nthe modern critic will be prompt to object that it is still affected\nby the \u2018unhistorical\u2019 delusions of the systems against which it was\ndirected. In expression, at any rate, it bears the marks of descent\nfrom Hobbes, and, if read without due allowance, might convey the\nnotion that society first existed without any sort of justice, and\nthat afterwards its members, finding universal war inconvenient,\nsaid to themselves, \u2018Go to; let us abstain from each other\u2019s goods.\u2019\nIt would be hard, however, to expect from Hume the full-blown\nterminology of development. He would probably have been the first to\nadmit that rules of justice, as well as our feelings towards them,\nwere not made but grew; and in his view of the \u2018passions\u2019 whose\noperation this growth exhibits, he does not seriously differ from\nthe ordinary exponents of the \u2018natural history\u2019 of ethics. These\npassions, we have seen, are \u2018Interest\u2019 and \u2018Sympathy,\u2019 which with\nHume only differ from the pleasures and desires we call \u2018animal\u2019 as\nany one of these differs from another--the pleasure of eating, for\ninstance, from that of drinking, or desire for the former pleasure\nfrom desire for the latter. Nor do their effects in the regulation of\nsociety, and in the growth of \u2018artificial\u2019 virtues and vices, differ\naccording to his account of them from sentiments which, because they\n\u2018occur to us whether we will or no,\u2019 he reckons purely natural, save\nin respect of the further extent to which the modifying influence of\nimagination--itself reacted on by language--must have been carried\nin order to their existence; and since this in his view is a merely\n\u2018natural\u2019 influence, there can only be a relative difference between\nthe \u2018artificiality\u2019 of its more complex, and the \u2018naturalness\u2019 of\nits simpler, products. Locke\u2019s opposition, then, of \u2018moral\u2019 to\nother good, on the ground that other than natural instrumentality\nis implied in its attainment, will not hold even in regard to that\ngood which, it is admitted, would not be what it is, _i.e._, not a\npleasure, but for the intervention of civil law.\n\nWhat is meant by an action which _ought_ to be done.\n\n60. The doctrine, which we have now traversed, of \u2018interested\u2019\nand \u2018moral\u2019 obligation, implicitly answers the question as to the\norigin and significance of the ethical copula \u2018ought.\u2019 It originally\nexpresses, we must suppose, obligation by positive law, or rather\nby that authoritative custom in which (as Hume would probably have\nbeen ready to admit) the \u2018general sense of common interest\u2019 first\nembodies itself. In this primitive meaning it already implies an\nopposition between the \u2018interest which each man has in maintaining\norder\u2019 and his \u2018lesser and more present interests.\u2019 Its meaning will\nbe modified in proportion as the direct interest in maintaining order\nis reinforced or superseded by sympathy with the general uneasiness\nwhich any departure from the rules of justice causes. And as this\nuneasiness is not confined to cases where the law is directly or\nin the letter violated, the judgment, that an act _ought_ to be\ndone, not only need not imply a belief that the person, so judging,\nwill himself gain anything by its being done or lose anything by\nits omission; it need not imply that any positive law requires it.\nWhether it is applicable to every act \u2018causing pleasure on the mere\nsurvey\u2019--whether the range of \u2018imperfect obligation\u2019 is as wide\nas that of moral sentiment--Hume does not make clear. That every\naction representing a quality \u2018fitted to give immediate pleasure to\nits possessor\u2019 should be virtuous--as according to Hume\u2019s account\nof the exciting cause of moral sentiment it must be--seems strange\nenough, but it would be stranger that we should judge of it as an\nact which _ought_ to be done. It is less difficult, for instance, to\nsuppose that it is virtuous to be witty, than that one ought to be\nso. Perhaps it would be open to a disciple of Hume to hold that as,\naccording to his master\u2019s showing, an opposition between permanent\nand present interest is implied in the judgment of obligation as\nat first formed, so it is when the pleasure to be produced by an\nact, which gratifies moral sense, is remote rather than near, and a\npleasure to others rather than to the doer, that the term \u2018ought\u2019 is\nappropriate to it.\n\nSense of morality no motive: When it seems so the motive is really\npride.\n\n61. But though Hume leaves some doubt on this point, he leaves none\nin regard to the sense in which alone any one can be said to do an\naction _because he ought_. This must mean that he does it to avoid\neither a legal penalty or that pain of shame which would arise upon\nthe communication through sympathy of such uneasiness as a contrary\nact would excite in others upon the survey. So far from its being\ntrue that an act, in order to be thoroughly virtuous, must be done\nfor virtue\u2019s sake, \u2018no action can be virtuous or morally good unless\nthere is some motive to produce it, distinct from the sense of its\nmorality.\u2019 [1] An act is virtuous on account of the pleasure which\nsupervenes when it is contemplated as proceeding from a motive fitted\nto produce pleasure to the agent or to others. The presence of this\nmotive, then, being the antecedent condition of the act\u2019s being\nregarded as virtuous, the motive cannot itself have been a regard\nto the virtue. It may be replied, indeed, that though this shows\n\u2018regard to virtue\u2019 or \u2018sense of morality\u2019 to be not the primary or\nonly virtuous motive, it does not follow that it cannot be a motive\nat all. An action cannot be prompted for the first time by desire\nfor a pleasure which can only be felt as a consequence of the action\nhaving been done, but it may be repeated, after experience of this\npleasure, from desire for its renewal. In like manner, since with\nHume the \u2018sense of morality\u2019 is not a desire at all but an emotion,\nand an emotion which cannot be felt till an act of a certain kind\nhas been done, it cannot be the original motive to such an action;\nbut why may not desire for so pleasant an emotion, when once it has\nbeen experienced, lead to a repetition of the act? The answer to this\nquestion is that the pleasure of moral sentiment, as Hume thinks of\nit, is essentially a pleasure experienced by a spectator of an act\nwho is other than the doer of it. If the doer and spectator were\nregarded as one person, there would be no meaning in the rule that\nthe tendency to produce pleasure, which excites the sentiment of\napprobation, must be a tendency to produce it to the doer himself or\nothers, as distinct from the spectator himself. Thus pleasure, in\nthe specific form in which Hume would call it \u2018moral sentiment,\u2019 is\nnot what any one could attain by his own action, and consequently\ncannot be a motive to action. Transferred by sympathy to the\nconsciousness of the man whose act is approved, \u2018moral sentiment\u2019\nbecomes \u2018pride,\u2019 and desire for the pleasure of pride--otherwise\ncalled \u2018love of fame\u2019--is one of the \u2018virtuous\u2019 motives on which\nHume dwells most. When an action, however, is done for the sake of\nany such positive pleasure, he would not allow apparently that the\nagent does it \u2018from a sense of duty\u2019 or \u2018because he ought.\u2019 He would\nconfine this description to cases where the object was rather the\navoidance of humiliation. \u2018I ought\u2019 means \u2018it is expected of me.\u2019\n\u2018When any virtuous motive or principle is common in human nature, a\nperson who feels his heart devoid of that motive may hate himself\u2019\n(strictly, according to Hume\u2019s usage of terms, \u2018despise himself\u2019) \u2018on\nthat account, and may perform the action without the motive from a\ncertain sense of duty, in order to acquire by practice that virtuous\nprinciple, or at least to disguise to himself as much as possible his\nwant of it.\u2019 [2]\n\n[1] Vol. II., p. 253. [Book III., part II., sec. I.]\n\n[2] Vol. II., p. 253. [Book III., part II., sec. I.]\n\nDistinction between virtuous and vicious motive does not exist for\nperson moved.\n\n62. What difference, then, we have finally to ask, does Hume leave\nbetween one motive and another, which can give any significance\nto the assertion that an act, to be virtuous, must proceed from a\nvirtuous motive? When a writer has so far distinguished between\nmotive and action as to tell us that the moral value of an action\ndepends on its motive--which is what Hume is on occasion ready to\ntell us--we naturally suppose that any predicate, which he proceeds\nto apply to the motive, is meant to represent what it is in relation\nto the subject of it. It cannot be so, however, when Hume calls a\nmotive virtuous. This predicate, as he explains, refers not to an\n\u2018esse\u2019 but to a \u2018percipi;\u2019 which means that it does not represent\nwhat the motive is to the person whom it moves, but a pleasant\nfeeling excited in the spectator of the act. To the excitement of\nthis feeling it is necessary that the action should not merely from\nsome temporary combination of circumstances produce pleasure for\nthat time and turn, but that the desire, to which the spectator\nascribes it, should be one according to his expectation \u2018fitted to\nproduce pleasure to the agent or to others.\u2019 In this sense only can\nHume consistently mean that virtue in the motive is the condition\nof virtue in the act, and in this sense the qualification has not\nmuch significance for the spectator of the act, and none at all in\nrelation to the doer. It has not much for the spectator, because,\naccording to it, no supposed desire will excite his displeasure and\nconsequently be vicious unless in its general operation it produces\na distinct overbalance of pain to the subject of it _and_ to others;\n[1] and by this test it would be more difficult to show that an\nunseasonable passion for reforming mankind was _not_ vicious than\nthat moderate lechery was so. It has no significance at all for the\nperson to whom vice or virtue is imputed, because a difference in\nthe results, which others anticipate from any desire that moves him\nto action, makes no difference in that desire, as he feels and is\nmoved by it. To him, according to Hume, it is simply desire for the\npleasure of which the idea is for the time most lively, and, being\nmost lively, cannot but excite the strongest desire. In this--in the\ncharacter which they severally bear for the subjects of them--the\nvirtuous motive and the vicious are alike. Hume, it is true, allows\nthat the subject of a vicious desire may become conscious through\nsympathy of the uneasiness which the contemplation of it causes to\nothers, but if this sympathy were strong enough to neutralize the\nimagination which excites the desire, the desire would not move him\nto act. That predominance of anticipated pain over pleasure in the\neffects of a motive, which renders it vicious to the spectator,\ncannot be transferred to the imagination of the subject of it without\nmaking it cease to be his motive because no longer his strongest\ndesire. A vicious motive, in short, would be a contradiction in\nterms, if that productivity of pain, which belongs to the motive\nin the imagination of the spectator, belonged to it also in the\nimagination of the agent.\n\n[1] I write \u2018AND to others,\u2019 not \u2018OR,\u2019 because according to Hume the\nproduction of pleasure to the agent alone is enough to render an\naction virtuous, if it proceeds from some permanent quality. Thus an\naction could not be unmistakably vicious unless it tended to produce\npain _both_ to the doer and to others. If, though tending to bring\npain to others, it had a contrary tendency for the agent himself,\nthere would be nothing to decide whether the viciousness of the\nformer tendency was, or was not, balanced by the virtuousness of the\nlatter.\n\n\u2018Consciousness of sin\u2019 disappears.\n\n63. Thus the consequence, which we found to be involved in Locke\u2019s\ndoctrine of motives, is virtually admitted by its most logical\nexponent. Locke\u2019s confusions began when he tried to reconcile his\ndoctrine with the fact of self-condemnation, with the individual\u2019s\nconsciousness of vice as a condition of himself; or, in his own\nwords, to explain how the vicious man could be \u2018answerable to\nhimself\u2019 for his vice. Consciousness of vice could only mean\nconsciousness of pleasure wilfully foregone, and since pleasure could\nnot be wilfully foregone, there could be no such consciousness. Hume,\nas we have seen, cuts the knot by disposing of the consciousness\nof vice, as a relation in which the individual stands to himself,\naltogether. A man\u2019s vice is someone else\u2019s displeasure with him, and,\nif we wish to be precise, we must not speak of self-condemnation\nor desire for excellence as influencing human conduct, but of\naversion from the pain of humiliation and desire for the pleasure\nof pride--humiliation and pride of that sort of which each man\u2019s\nsympathy with the feeling of others about him is the condition.\n\nOnly respectability remains; and even this not consistently accounted\nfor.\n\n64. That such a doctrine leaves large fields of human experience\nunexplained, few will now dispute. Wesley, Wordsworth, Fichte,\nMazzini, and the German theologians, lie between us and the\ngeneration in which, to so healthy a nature as Hume\u2019s, and in so\nexplicit a form, it could be possible. Enthusiasm--religious,\npolitical, and poetic--if it has not attained higher forms, has been\nforced to understand itself better since the time when Shaftesbury\u2019s\nthin and stilted rhapsody was its most intelligent expression.\nIt is now generally agreed that the saint is not explained by\nbeing called a fanatic, that there is a patriotism which is not\n\u2018the last refuge of a scoundrel,\u2019 and that we know no more about\nthe poet, when we have been told that he seeks the beautiful,\nand that what is beautiful is pleasant, than we did before. This\nadmitted, Hume\u2019s Hedonism needs only to be clearly stated to be\nfound \u2018unsatisfactory.\u2019 If it ever tends to find acceptance with\nserious people, it is through confusion with that hybrid, though\nbeneficent, utilitarianism which finds the moral good in the\n\u2018greatest happiness of the greatest number\u2019 without reflecting that\ndesire for such an object, not being for a feeling of pleasure to be\nexperienced by the subject of the desire, is with Hume impossible.\nUnderstood as he himself understood his doctrine, it is only\n\u2018respectability\u2019--the temper of the man who \u2018naturally,\u2019 _i.e._,\nwithout definite expectation of ulterior gain, seeks to stand well\nwith his neighbours--that it will explain; and this it can only\ntreat as a fixed quantity. Taking for granted the heroic virtue,\nfor which it cannot account, it still must leave it a mystery how\nthe heroic virtue of an earlier age can become the respectability\nof a later one. Recent literary fashion has led us perhaps unduly\nto depreciate respectability, but the avowed insufficiency of a\nmoral theory to explain anything beyond it may fairly entitle us\nto enquire whether it can consistently explain even that. The\nreason, as we have sufficiently seen, why Hume\u2019s ethical speculation\nhas such an issue is that he does not recognize the constitutive\naction of self-conscious thought. Misunderstanding our passivity in\nexperience--unaware that it has no meaning except in relation to\nan object which thought itself projects, yet too clear-sighted to\nacquiesce in the vulgar notion of either laws of matter or laws of\naction, as simply thrust upon us from an unaccountable without--he\nseeks in the mere abstraction of passivity, of feeling which is\na feeling of nothing, the explanation of the natural and moral\nworld. Nature is a sequence of sensations, morality a succession of\npleasures and pains. It is under the pressure of this abstraction\nthat he so empties morality of its actual content as to leave only\nthe residuum we have described. Yet to account even for this he\nhas to admit such motives as \u2018pride,\u2019 \u2018love,\u2019 and \u2018interest;\u2019 and\neach of these, as we have shown, implies that very constitutive\naction of reason, by ignoring which he compels himself to reduce all\nmorality to that of the average man in his least exalted moments.\nThe formative power of thought, as exhibited in such motives, only\ndiffers in respect of the lower degree, to which it has fashioned\nits matter, from the same power as the source of the \u2018desire for\nexcellence,\u2019 of the will autonomous in the service of mankind, of\nthe forever (to us) unfilled ideal of a perfect society. It is\nbecause Hume de-rationalizes respectability, that he can find no\n_rationale_, and therefore no room, for the higher morality. This\nmight warn us that an \u2018ideal\u2019 theory of ethics tampers with its only\nsure foundation when it depreciates respectability; and if it were\nour business to extract a practical lesson from him, it would be that\nthere is no other genuine \u2018enthusiasm of humanity\u2019 than one which\nhas travelled the common highway of reason--the life of the good\nneighbour and honest citizen--and can never forget that it is still\nonly on a further stage of the same journey. Our business, however,\nhas not been to moralise, but to show that the philosophy based on\nthe abstraction of feeling, in regard to morals no less than to\nnature, was with Hume played out, and that the next step forward in\nspeculation could only be an effort to re-think the process of nature\nand human action from its true beginning in thought. If this object\nhas been in any way attained, so that the attention of Englishmen\n\u2018under five-and-twenty\u2019 may be diverted from the anachronistic\nsystems hitherto prevalent among us to the study of Kant and Hegel,\nan irksome labour will not have been in vain.\n\nT. H. Green.","title":"A Treatise of Human Nature: Being an Attempt to Introduce the Experimental Method Into Moral Subjects; and Dialogues Concerning Natural Religion","tradition":"philosophy-ethics","translated_text":null,"word_count":169128}