{"collection":"philosophy-ethics","full_text_available":true,"graph_url":"https://unchartedknowledge.5gfusion.net/v2/api/page/gutenberg_2529/graph","html_url":"https://unchartedknowledge.5gfusion.net/v2/page/gutenberg_2529","id":"gutenberg_2529","license":"Public Domain (US)","rights_basis":"gutenberg_metadata_copyright_false","source":"gutenberg","source_url":"https://www.gutenberg.org/ebooks/2529","summary":null,"tags":["Category: Philosophy & Ethics","Category: Psychiatry/Psychology","Consciousness","Knowledge, Theory of","Nobel Prizes in Literature","Philosophy","Psychology"],"text":"Produced by Dianne Bean\n\n\n\n\n\nTHE ANALYSIS OF MIND\n\nBy Bertrand Russell\n\n1921\n\n\n\n\nMUIRHEAD LIBRARY OF PHILOSOPHY\n\n\nAn admirable statement of the aims of the Library of Philosophy was\nprovided by the first editor, the late Professor J. H. Muirhead, in his\ndescription of the original programme printed in Erdmann's History of\nPhilosophy under the date 1890. This was slightly modified in subsequent\nvolumes to take the form of the following statement:\n\n\"The Muirhead Library of Philosophy was designed as a contribution to\nthe History of Modern Philosophy under the heads: first of Different\nSchools of Thought--Sensationalist, Realist, Idealist, Intuitivist;\nsecondly of different Subjects--Psychology, Ethics, Aesthetics,\nPolitical Philosophy, Theology. While much had been done in England in\ntracing the course of evolution in nature, history, economics, morals\nand religion, little had been done in tracing the development of thought\non these subjects. Yet 'the evolution of opinion is part of the whole\nevolution'.\n\n\"By the co-operation of different writers in carrying out this plan it\nwas hoped that a thoroughness and completeness of treatment, otherwise\nunattainable, might be secured. It was believed also that from writers\nmainly British and American fuller consideration of English Philosophy\nthan it had hitherto received might be looked for. In the earlier series\nof books containing, among others, Bosanquet's \"History of Aesthetic,\"\nPfleiderer's \"Rational Theology since Kant,\" Albee's \"History of English\nUtilitarianism,\" Bonar's \"Philosophy and Political Economy,\" Brett's\n\"History of Psychology,\" Ritchie's \"Natural Rights,\" these objects were\nto a large extent effected.\n\n\"In the meantime original work of a high order was being produced both\nin England and America by such writers as Bradley, Stout, Bertrand\nRussell, Baldwin, Urban, Montague, and others, and a new interest in\nforeign works, German, French and Italian, which had either become\nclassical or were attracting public attention, had developed. The scope\nof the Library thus became extended into something more international,\nand it is entering on the fifth decade of its existence in the hope that\nit may contribute to that mutual understanding between countries which\nis so pressing a need of the present time.\"\n\nThe need which Professor Muirhead stressed is no less pressing to-day,\nand few will deny that philosophy has much to do with enabling us to\nmeet it, although no one, least of all Muirhead himself, would regard\nthat as the sole, or even the main, object of philosophy. As Professor\nMuirhead continues to lend the distinction of his name to the Library\nof Philosophy it seemed not inappropriate to allow him to recall us to\nthese aims in his own words. The emphasis on the history of thought also\nseemed to me very timely; and the number of important works promised\nfor the Library in the very near future augur well for the continued\nfulfilment, in this and other ways, of the expectations of the original\neditor.\n\nH. D. Lewis\n\n\n\n\nPREFACE\n\nThis book has grown out of an attempt to harmonize two different\ntendencies, one in psychology, the other in physics, with both of which\nI find myself in sympathy, although at first sight they might seem\ninconsistent. On the one hand, many psychologists, especially those\nof the behaviourist school, tend to adopt what is essentially a\nmaterialistic position, as a matter of method if not of metaphysics.\nThey make psychology increasingly dependent on physiology and external\nobservation, and tend to think of matter as something much more solid\nand indubitable than mind. Meanwhile the physicists, especially Einstein\nand other exponents of the theory of relativity, have been making\n\"matter\" less and less material. Their world consists of \"events,\" from\nwhich \"matter\" is derived by a logical construction. Whoever reads, for\nexample, Professor Eddington's \"Space, Time and Gravitation\" (Cambridge\nUniversity Press, 1920), will see that an old-fashioned materialism can\nreceive no support from modern physics. I think that what has permanent\nvalue in the outlook of the behaviourists is the feeling that physics is\nthe most fundamental science at present in existence. But this position\ncannot be called materialistic, if, as seems to be the case, physics\ndoes not assume the existence of matter.\n\nThe view that seems to me to reconcile the materialistic tendency of\npsychology with the anti-materialistic tendency of physics is the view\nof William James and the American new realists, according to which the\n\"stuff\" of the world is neither mental nor material, but a \"neutral\nstuff,\" out of which both are constructed. I have endeavoured in this\nwork to develop this view in some detail as regards the phenomena with\nwhich psychology is concerned.\n\nMy thanks are due to Professor John B. Watson and to Dr. T. P. Nunn\nfor reading my MSS. at an early stage and helping me with many valuable\nsuggestions; also to Mr. A. Wohlgemuth for much very useful information\nas regards important literature. I have also to acknowledge the help\nof the editor of this Library of Philosophy, Professor Muirhead, for\nseveral suggestions by which I have profited.\n\nThe work has been given in the form of lectures both in London and\nPeking, and one lecture, that on Desire, has been published in the\nAthenaeum.\n\nThere are a few allusions to China in this book, all of which were\nwritten before I had been in China, and are not intended to be taken by\nthe reader as geographically accurate. I have used \"China\" merely as\na synonym for \"a distant country,\" when I wanted illustrations of\nunfamiliar things.\n\nPeking, January 1921.\n\n\n\nCONTENTS\n\n  I.   Recent Criticisms of \"Consciousness\"\n  II.  Instinct and Habit\n  III. Desire and Feeling\n  IV.  Influence of Past History on Present Occurrences\n            in Living Organisms\n  V.   Psychological and Physical Causal Laws\n  VI.  Introspection\n  VII. The Definition of Perception\n  VIII.Sensations and Images\n  IX.  Memory\n  X.   Words and Meaning\n  XI.  General Ideas and Thought\n  XII. Belief\n  XIII.Truth and Falsehood\n  XIV. Emotions and Will\n  XV.  Characteristics of Mental Phenomena\n\n\n\n\nTHE ANALYSIS OF MIND\n\n\n\n\nLECTURE I. RECENT CRITICISMS OF \"CONSCIOUSNESS\"\n\nThere are certain occurrences which we are in the habit of calling\n\"mental.\" Among these we may take as typical BELIEVING and DESIRING.\nThe exact definition of the word \"mental\" will, I hope, emerge as\nthe lectures proceed; for the present, I shall mean by it whatever\noccurrences would commonly be called mental.\n\nI wish in these lectures to analyse as fully as I can what it is that\nreally takes place when we, e.g. believe or desire. In this first\nlecture I shall be concerned to refute a theory which is widely held,\nand which I formerly held myself: the theory that the essence of\neverything mental is a certain quite peculiar something called\n\"consciousness,\" conceived either as a relation to objects, or as a\npervading quality of psychical phenomena.\n\nThe reasons which I shall give against this theory will be mainly\nderived from previous authors. There are two sorts of reasons, which\nwill divide my lecture into two parts:\n\n(1) Direct reasons, derived from analysis and its difficulties;\n\n(2) Indirect reasons, derived from observation of animals (comparative\npsychology) and of the insane and hysterical (psycho-analysis).\n\nFew things are more firmly established in popular philosophy than the\ndistinction between mind and matter. Those who are not professional\nmetaphysicians are willing to confess that they do not know what mind\nactually is, or how matter is constituted; but they remain convinced\nthat there is an impassable gulf between the two, and that both belong\nto what actually exists in the world. Philosophers, on the other hand,\nhave maintained often that matter is a mere fiction imagined by mind,\nand sometimes that mind is a mere property of a certain kind of matter.\nThose who maintain that mind is the reality and matter an evil dream are\ncalled \"idealists\"--a word which has a different meaning in philosophy\nfrom that which it bears in ordinary life. Those who argue that matter\nis the reality and mind a mere property of protoplasm are called\n\"materialists.\" They have been rare among philosophers, but common,\nat certain periods, among men of science. Idealists, materialists, and\nordinary mortals have been in agreement on one point: that they knew\nsufficiently what they meant by the words \"mind\" and \"matter\" to be able\nto conduct their debate intelligently. Yet it was just in this point, as\nto which they were at one, that they seem to me to have been all alike\nin error.\n\nThe stuff of which the world of our experience is composed is, in my\nbelief, neither mind nor matter, but something more primitive than\neither. Both mind and matter seem to be composite, and the stuff of\nwhich they are compounded lies in a sense between the two, in a sense\nabove them both, like a common ancestor. As regards matter, I have set\nforth my reasons for this view on former occasions,* and I shall not now\nrepeat them. But the question of mind is more difficult, and it is this\nquestion that I propose to discuss in these lectures. A great deal of\nwhat I shall have to say is not original; indeed, much recent work, in\nvarious fields, has tended to show the necessity of such theories as\nthose which I shall be advocating. Accordingly in this first lecture\nI shall try to give a brief description of the systems of ideas within\nwhich our investigation is to be carried on.\n\n     * \"Our Knowledge of the External World\" (Allen & Unwin),\n     Chapters III and IV. Also \"Mysticism and Logic,\" Essays VII\n     and VIII.\n\nIf there is one thing that may be said, in the popular estimation, to\ncharacterize mind, that one thing is \"consciousness.\" We say that we are\n\"conscious\" of what we see and hear, of what we remember, and of our own\nthoughts and feelings. Most of us believe that tables and chairs are\nnot \"conscious.\" We think that when we sit in a chair, we are aware\nof sitting in it, but it is not aware of being sat in. It cannot for\na moment be doubted that we are right in believing that there is SOME\ndifference between us and the chair in this respect: so much may be\ntaken as fact, and as a datum for our inquiry. But as soon as we try to\nsay what exactly the difference is, we become involved in perplexities.\nIs \"consciousness\" ultimate and simple, something to be merely accepted\nand contemplated? Or is it something complex, perhaps consisting in our\nway of behaving in the presence of objects, or, alternatively, in the\nexistence in us of things called \"ideas,\" having a certain relation\nto objects, though different from them, and only symbolically\nrepresentative of them? Such questions are not easy to answer; but until\nthey are answered we cannot profess to know what we mean by saying that\nwe are possessed of \"consciousness.\"\n\nBefore considering modern theories, let us look first at consciousness\nfrom the standpoint of conventional psychology, since this embodies\nviews which naturally occur when we begin to reflect upon the subject.\nFor this purpose, let us as a preliminary consider different ways of\nbeing conscious.\n\nFirst, there is the way of PERCEPTION. We \"perceive\" tables and chairs,\nhorses and dogs, our friends, traffic passing in the street--in short,\nanything which we recognize through the senses. I leave on one side for\nthe present the question whether pure sensation is to be regarded as a\nform of consciousness: what I am speaking of now is perception, where,\naccording to conventional psychology, we go beyond the sensation to the\n\"thing\" which it represents. When you hear a donkey bray, you not only\nhear a noise, but realize that it comes from a donkey. When you see a\ntable, you not only see a coloured surface, but realize that it is hard.\nThe addition of these elements that go beyond crude sensation is said to\nconstitute perception. We shall have more to say about this at a later\nstage. For the moment, I am merely concerned to note that perception\nof objects is one of the most obvious examples of what is called\n\"consciousness.\" We are \"conscious\" of anything that we perceive.\n\nWe may take next the way of MEMORY. If I set to work to recall what\nI did this morning, that is a form of consciousness different from\nperception, since it is concerned with the past. There are various\nproblems as to how we can be conscious now of what no longer exists.\nThese will be dealt with incidentally when we come to the analysis of\nmemory.\n\nFrom memory it is an easy step to what are called \"ideas\"--not in the\nPlatonic sense, but in that of Locke, Berkeley and Hume, in which they\nare opposed to \"impressions.\" You may be conscious of a friend either\nby seeing him or by \"thinking\" of him; and by \"thought\" you can be\nconscious of objects which cannot be seen, such as the human race,\nor physiology. \"Thought\" in the narrower sense is that form of\nconsciousness which consists in \"ideas\" as opposed to impressions or\nmere memories.\n\nWe may end our preliminary catalogue with BELIEF, by which I mean that\nway of being conscious which may be either true or false. We say that a\nman is \"conscious of looking a fool,\" by which we mean that he believes\nhe looks a fool, and is not mistaken in this belief. This is a different\nform of consciousness from any of the earlier ones. It is the form which\ngives \"knowledge\" in the strict sense, and also error. It is, at least\napparently, more complex than our previous forms of consciousness;\nthough we shall find that they are not so separable from it as they\nmight appear to be.\n\nBesides ways of being conscious there are other things that would\nordinarily be called \"mental,\" such as desire and pleasure and pain.\nThese raise problems of their own, which we shall reach in Lecture III.\nBut the hardest problems are those that arise concerning ways of being\n\"conscious.\" These ways, taken together, are called the \"cognitive\"\nelements in mind, and it is these that will occupy us most during the\nfollowing lectures.\n\nThere is one element which SEEMS obviously in common among the different\nways of being conscious, and that is, that they are all directed to\nOBJECTS. We are conscious \"of\" something. The consciousness, it seems,\nis one thing, and that of which we are conscious is another thing.\nUnless we are to acquiesce in the view that we can never be conscious\nof anything outside our own minds, we must say that the object of\nconsciousness need not be mental, though the consciousness must be. (I\nam speaking within the circle of conventional doctrines, not expressing\nmy own beliefs.) This direction towards an object is commonly regarded\nas typical of every form of cognition, and sometimes of mental life\naltogether. We may distinguish two different tendencies in traditional\npsychology. There are those who take mental phenomena naively, just as\nthey would physical phenomena. This school of psychologists tends not to\nemphasize the object. On the other hand, there are those whose primary\ninterest is in the apparent fact that we have KNOWLEDGE, that there is a\nworld surrounding us of which we are aware. These men are interested in\nthe mind because of its relation to the world, because knowledge, if\nit is a fact, is a very mysterious one. Their interest in psychology\nis naturally centred in the relation of consciousness to its object, a\nproblem which, properly, belongs rather to theory of knowledge. We may\ntake as one of the best and most typical representatives of this school\nthe Austrian psychologist Brentano, whose \"Psychology from the Empirical\nStandpoint,\"* though published in 1874, is still influential and was the\nstarting-point of a great deal of interesting work. He says (p. 115):\n\n     * \"Psychologie vom empirischen Standpunkte,\" vol. i, 1874.\n     (The second volume was never published.)\n\n\"Every psychical phenomenon is characterized by what the scholastics of\nthe Middle Ages called the intentional (also the mental) inexistence of\nan object, and what we, although with not quite unambiguous expressions,\nwould call relation to a content, direction towards an object (which is\nnot here to be understood as a reality), or immanent objectivity. Each\ncontains something in itself as an object, though not each in the same\nway. In presentation something is presented, in judgment something is\nacknowledged or rejected, in love something is loved, in hatred hated,\nin desire desired, and so on.\n\n\"This intentional inexistence is exclusively peculiar to psychical\nphenomena. No physical phenomenon shows anything similar. And so we\ncan define psychical phenomena by saying that they are phenomena which\nintentionally contain an object in themselves.\"\n\nThe view here expressed, that relation to an object is an ultimate\nirreducible characteristic of mental phenomena, is one which I shall be\nconcerned to combat. Like Brentano, I am interested in psychology, not\nso much for its own sake, as for the light that it may throw on the\nproblem of knowledge. Until very lately I believed, as he did, that\nmental phenomena have essential reference to objects, except possibly in\nthe case of pleasure and pain. Now I no longer believe this, even in\nthe case of knowledge. I shall try to make my reasons for this rejection\nclear as we proceed. It must be evident at first glance that the\nanalysis of knowledge is rendered more difficult by the rejection; but\nthe apparent simplicity of Brentano's view of knowledge will be found,\nif I am not mistaken, incapable of maintaining itself either against\nan analytic scrutiny or against a host of facts in psycho-analysis and\nanimal psychology. I do not wish to minimize the problems. I will\nmerely observe, in mitigation of our prospective labours, that thinking,\nhowever it is to be analysed, is in itself a delightful occupation,\nand that there is no enemy to thinking so deadly as a false simplicity.\nTravelling, whether in the mental or the physical world, is a joy, and\nit is good to know that, in the mental world at least, there are vast\ncountries still very imperfectly explored.\n\nThe view expressed by Brentano has been held very generally, and\ndeveloped by many writers. Among these we may take as an example his\nAustrian successor Meinong.* According to him there are three elements\ninvolved in the thought of an object. These three he calls the act, the\ncontent and the object. The act is the same in any two cases of the same\nkind of consciousness; for instance, if I think of Smith or think\nof Brown, the act of thinking, in itself, is exactly similar on both\noccasions. But the content of my thought, the particular event that\nis happening in my mind, is different when I think of Smith and when I\nthink of Brown. The content, Meinong argues, must not be confounded with\nthe object, since the content must exist in my mind at the moment when\nI have the thought, whereas the object need not do so. The object may\nbe something past or future; it may be physical, not mental; it may\nbe something abstract, like equality for example; it may be something\nimaginary, like a golden mountain; or it may even be something\nself-contradictory, like a round square. But in all these cases, so\nhe contends, the content exists when the thought exists, and is what\ndistinguishes it, as an occurrence, from other thoughts.\n\n     * See, e.g. his article: \"Ueber Gegenstande hoherer Ordnung\n     und deren Verhaltniss zur inneren Wahrnehmung,\" \"Zeitschrift\n     fur Psychologie and Physiologie der Sinnesorgane,\" vol. xxi,\n     pp. 182-272 (1899), especially pp. 185-8.\n\nTo make this theory concrete, let us suppose that you are thinking of\nSt. Paul's. Then, according to Meinong, we have to distinguish three\nelements which are necessarily combined in constituting the one thought.\nFirst, there is the act of thinking, which would be just the same\nwhatever you were thinking about. Then there is what makes the character\nof the thought as contrasted with other thoughts; this is the content.\nAnd finally there is St. Paul's, which is the object of your thought.\nThere must be a difference between the content of a thought and what it\nis about, since the thought is here and now, whereas what it is about\nmay not be; hence it is clear that the thought is not identical with St.\nPaul's. This seems to show that we must distinguish between content\nand object. But if Meinong is right, there can be no thought without an\nobject: the connection of the two is essential. The object might exist\nwithout the thought, but not the thought without the object: the three\nelements of act, content and object are all required to constitute the\none single occurrence called \"thinking of St. Paul's.\"\n\nThe above analysis of a thought, though I believe it to be mistaken, is\nvery useful as affording a schema in terms of which other theories can\nbe stated. In the remainder of the present lecture I shall state in\noutline the view which I advocate, and show how various other views\nout of which mine has grown result from modifications of the threefold\nanalysis into act, content and object.\n\nThe first criticism I have to make is that the ACT seems unnecessary and\nfictitious. The occurrence of the content of a thought constitutes\nthe occurrence of the thought. Empirically, I cannot discover anything\ncorresponding to the supposed act; and theoretically I cannot see that\nit is indispensable. We say: \"_I_ think so-and-so,\" and this word \"I\"\nsuggests that thinking is the act of a person. Meinong's \"act\" is the\nghost of the subject, or what once was the full-blooded soul. It is\nsupposed that thoughts cannot just come and go, but need a person to\nthink them. Now, of course it is true that thoughts can be collected\ninto bundles, so that one bundle is my thoughts, another is your\nthoughts, and a third is the thoughts of Mr. Jones. But I think\nthe person is not an ingredient in the single thought: he is rather\nconstituted by relations of the thoughts to each other and to the body.\nThis is a large question, which need not, in its entirety, concern us\nat present. All that I am concerned with for the moment is that the\ngrammatical forms \"I think,\" \"you think,\" and \"Mr. Jones thinks,\" are\nmisleading if regarded as indicating an analysis of a single thought.\nIt would be better to say \"it thinks in me,\" like \"it rains here\"; or\nbetter still, \"there is a thought in me.\" This is simply on the\nground that what Meinong calls the act in thinking is not empirically\ndiscoverable, or logically deducible from what we can observe.\n\nThe next point of criticism concerns the relation of content and object.\nThe reference of thoughts to objects is not, I believe, the simple\ndirect essential thing that Brentano and Meinong represent it as being.\nIt seems to me to be derivative, and to consist largely in BELIEFS:\nbeliefs that what constitutes the thought is connected with various\nother elements which together make up the object. You have, say, an\nimage of St. Paul's, or merely the word \"St. Paul's\" in your head. You\nbelieve, however vaguely and dimly, that this is connected with what\nyou would see if you went to St. Paul's, or what you would feel if you\ntouched its walls; it is further connected with what other people see\nand feel, with services and the Dean and Chapter and Sir Christopher\nWren. These things are not mere thoughts of yours, but your thought\nstands in a relation to them of which you are more or less aware. The\nawareness of this relation is a further thought, and constitutes\nyour feeling that the original thought had an \"object.\" But in pure\nimagination you can get very similar thoughts without these accompanying\nbeliefs; and in this case your thoughts do not have objects or seem to\nhave them. Thus in such instances you have content without object. On\nthe other hand, in seeing or hearing it would be less misleading to\nsay that you have object without content, since what you see or hear is\nactually part of the physical world, though not matter in the sense of\nphysics. Thus the whole question of the relation of mental occurrences\nto objects grows very complicated, and cannot be settled by regarding\nreference to objects as of the essence of thoughts. All the above\nremarks are merely preliminary, and will be expanded later.\n\nSpeaking in popular and unphilosophical terms, we may say that the\ncontent of a thought is supposed to be something in your head when you\nthink the thought, while the object is usually something in the outer\nworld. It is held that knowledge of the outer world is constituted by\nthe relation to the object, while the fact that knowledge is different\nfrom what it knows is due to the fact that knowledge comes by way of\ncontents. We can begin to state the difference between realism and\nidealism in terms of this opposition of contents and objects. Speaking\nquite roughly and approximately, we may say that idealism tends to\nsuppress the object, while realism tends to suppress the content.\nIdealism, accordingly, says that nothing can be known except thoughts,\nand all the reality that we know is mental; while realism maintains that\nwe know objects directly, in sensation certainly, and perhaps also in\nmemory and thought. Idealism does not say that nothing can be known\nbeyond the present thought, but it maintains that the context of vague\nbelief, which we spoke of in connection with the thought of St. Paul's,\nonly takes you to other thoughts, never to anything radically different\nfrom thoughts. The difficulty of this view is in regard to sensation,\nwhere it seems as if we came into direct contact with the outer world.\nBut the Berkeleian way of meeting this difficulty is so familiar that I\nneed not enlarge upon it now. I shall return to it in a later lecture,\nand will only observe, for the present, that there seem to me no valid\ngrounds for regarding what we see and hear as not part of the physical\nworld.\n\nRealists, on the other hand, as a rule, suppress the content, and\nmaintain that a thought consists either of act and object alone, or of\nobject alone. I have been in the past a realist, and I remain a realist\nas regards sensation, but not as regards memory or thought. I will try\nto explain what seem to me to be the reasons for and against various\nkinds of realism.\n\nModern idealism professes to be by no means confined to the present\nthought or the present thinker in regard to its knowledge; indeed, it\ncontends that the world is so organic, so dove-tailed, that from any\none portion the whole can be inferred, as the complete skeleton of an\nextinct animal can be inferred from one bone. But the logic by which\nthis supposed organic nature of the world is nominally demonstrated\nappears to realists, as it does to me, to be faulty. They argue that,\nif we cannot know the physical world directly, we cannot really know\nany thing outside our own minds: the rest of the world may be merely our\ndream. This is a dreary view, and they there fore seek ways of escaping\nfrom it. Accordingly they maintain that in knowledge we are in direct\ncontact with objects, which may be, and usually are, outside our own\nminds. No doubt they are prompted to this view, in the first place, by\nbias, namely, by the desire to think that they can know of the existence\nof a world outside themselves. But we have to consider, not what led\nthem to desire the view, but whether their arguments for it are valid.\n\nThere are two different kinds of realism, according as we make a thought\nconsist of act and object, or of object alone. Their difficulties are\ndifferent, but neither seems tenable all through. Take, for the sake of\ndefiniteness, the remembering of a past event. The remembering occurs\nnow, and is therefore necessarily not identical with the past event.\nSo long as we retain the act, this need cause no difficulty. The act\nof remembering occurs now, and has on this view a certain essential\nrelation to the past event which it remembers. There is no LOGICAL\nobjection to this theory, but there is the objection, which we spoke\nof earlier, that the act seems mythical, and is not to be found by\nobservation. If, on the other hand, we try to constitute memory without\nthe act, we are driven to a content, since we must have something that\nhappens NOW, as opposed to the event which happened in the past. Thus,\nwhen we reject the act, which I think we must, we are driven to a theory\nof memory which is more akin to idealism. These arguments, however, do\nnot apply to sensation. It is especially sensation, I think, which is\nconsidered by those realists who retain only the object.* Their views,\nwhich are chiefly held in America, are in large measure derived from\nWilliam James, and before going further it will be well to consider\nthe revolutionary doctrine which he advocated. I believe this doctrine\ncontains important new truth, and what I shall have to say will be in a\nconsiderable measure inspired by it.\n\n     * This is explicitly the case with Mach's \"Analysis of\n     Sensations,\" a book of fundamental importance in the present\n     connection. (Translation of fifth German edition, Open Court\n     Co., 1914. First German edition, 1886.)\n\n\nWilliam James's view was first set forth in an essay called \"Does\n'consciousness' exist?\"* In this essay he explains how what used to be\nthe soul has gradually been refined down to the \"transcendental ego,\"\nwhich, he says, \"attenuates itself to a thoroughly ghostly condition,\nbeing only a name for the fact that the 'content' of experience IS\nKNOWN. It loses personal form and activity--these passing over to the\ncontent--and becomes a bare Bewusstheit or Bewusstsein uberhaupt, of\nwhich in its own right absolutely nothing can be said. I believe (he\ncontinues) that 'consciousness,' when once it has evaporated to this\nestate of pure diaphaneity, is on the point of disappearing altogether.\nIt is the name of a nonentity, and has no right to a place among first\nprinciples. Those who still cling to it are clinging to a mere echo,\nthe faint rumour left behind by the disappearing 'soul' upon the air of\nphilosophy\"(p. 2).\n\n     * \"Journal of Philosophy, Psychology and Scientific\n     Methods,\" vol. i, 1904. Reprinted in \"Essays in Radical\n     Empiricism\" (Longmans, Green & Co., 1912), pp. 1-38, to\n     which references in what follows refer.\n\n\nHe explains that this is no sudden change in his opinions. \"For twenty\nyears past,\" he says, \"I have mistrusted 'consciousness' as an entity;\nfor seven or eight years past I have suggested its non-existence to my\nstudents, and tried to give them its pragmatic equivalent in realities\nof experience. It seems to me that the hour is ripe for it to be openly\nand universally discarded\"(p. 3).\n\nHis next concern is to explain away the air of paradox, for James\nwas never wilfully paradoxical. \"Undeniably,\" he says, \"'thoughts' do\nexist.\" \"I mean only to deny that the word stands for an entity, but to\ninsist most emphatically that it does stand for a function. There is, I\nmean, no aboriginal stuff or quality of being, contrasted with that of\nwhich material objects are made, out of which our thoughts of them are\nmade; but there is a function in experience which thoughts perform,\nand for the performance of which this quality of being is invoked. That\nfunction is KNOWING\"(pp. 3-4).\n\nJames's view is that the raw material out of which the world is built\nup is not of two sorts, one matter and the other mind, but that it is\narranged in different patterns by its inter-relations, and that some\narrangements may be called mental, while others may be called physical.\n\n\"My thesis is,\" he says, \"that if we start with the supposition that\nthere is only one primal stuff or material in the world, a stuff\nof which everything is composed, and if we call that stuff 'pure\nexperience,' then knowing can easily be explained as a particular sort\nof relation towards one another into which portions of pure experience\nmay enter. The relation itself is a part of pure experience; one of its\n'terms' becomes the subject or bearer of the knowledge, the knower, the\nother becomes the object known\"(p. 4).\n\nAfter mentioning the duality of subject and object, which is supposed\nto constitute consciousness, he proceeds in italics: \"EXPERIENCE, I\nBELIEVE, HAS NO SUCH INNER DUPLICITY; AND THE SEPARATION OF IT INTO\nCONSCIOUSNESS AND CONTENT COMES, NOT BY WAY OF SUBTRACTION, BUT BY WAY\nOF ADDITION\"(p. 9).\n\nHe illustrates his meaning by the analogy of paint as it appears in a\npaint-shop and as it appears in a picture: in the one case it is just\n\"saleable matter,\" while in the other it \"performs a spiritual function.\nJust so, I maintain (he continues), does a given undivided portion\nof experience, taken in one context of associates, play the part of a\nknower, of a state of mind, of 'consciousness'; while in a different\ncontext the same undivided bit of experience plays the part of a thing\nknown, of an objective 'content.' In a word, in one group it figures as\na thought, in another group as a thing\"(pp. 9-10).\n\nHe does not believe in the supposed immediate certainty of thought. \"Let\nthe case be what it may in others,\" he says, \"I am as confident as I am\nof anything that, in myself, the stream of thinking (which I recognize\nemphatically as a phenomenon) is only a careless name for what, when\nscrutinized, reveals itself to consist chiefly of the stream of my\nbreathing. The 'I think' which Kant said must be able to accompany all\nmy objects, is the 'I breathe' which actually does accompany them\"(pp.\n36-37).\n\nThe same view of \"consciousness\" is set forth in the succeeding essay,\n\"A World of Pure Experience\" (ib., pp. 39-91). The use of the phrase\n\"pure experience\" in both essays points to a lingering influence of\nidealism. \"Experience,\" like \"consciousness,\" must be a product, not\npart of the primary stuff of the world. It must be possible, if James is\nright in his main contentions, that roughly the same stuff, differently\narranged, would not give rise to anything that could be called\n\"experience.\" This word has been dropped by the American realists, among\nwhom we may mention specially Professor R. B. Perry of Harvard and Mr.\nEdwin B. Holt. The interests of this school are in general philosophy\nand the philosophy of the sciences, rather than in psychology; they have\nderived a strong impulsion from James, but have more interest than he\nhad in logic and mathematics and the abstract part of philosophy. They\nspeak of \"neutral\" entities as the stuff out of which both mind and\nmatter are constructed. Thus Holt says: \"If the terms and propositions\nof logic must be substantialized, they are all strictly of one\nsubstance, for which perhaps the least dangerous name is neutral-stuff.\nThe relation of neutral-stuff to matter and mind we shall have presently\nto consider at considerable length.\" *\n\n     * \"The Concept of Consciousness\" (Geo. Allen & Co., 1914),\n     p. 52.\n\nMy own belief--for which the reasons will appear in subsequent\nlectures--is that James is right in rejecting consciousness as an\nentity, and that the American realists are partly right, though not\nwholly, in considering that both mind and matter are composed of a\nneutral-stuff which, in isolation, is neither mental nor material. I\nshould admit this view as regards sensations: what is heard or seen\nbelongs equally to psychology and to physics. But I should say that\nimages belong only to the mental world, while those occurrences (if any)\nwhich do not form part of any \"experience\" belong only to the physical\nworld. There are, it seems to me, prima facie different kinds of causal\nlaws, one belonging to physics and the other to psychology. The law\nof gravitation, for example, is a physical law, while the law of\nassociation is a psychological law. Sensations are subject to both kinds\nof laws, and are therefore truly \"neutral\" in Holt's sense. But entities\nsubject only to physical laws, or only to psychological laws, are not\nneutral, and may be called respectively purely material and purely\nmental. Even those, however, which are purely mental will not have that\nintrinsic reference to objects which Brentano assigns to them and which\nconstitutes the essence of \"consciousness\" as ordinarily understood. But\nit is now time to pass on to other modern tendencies, also hostile to\n\"consciousness.\"\n\nThere is a psychological school called \"Behaviourists,\" of whom the\nprotagonist is Professor John B. Watson,* formerly of the Johns Hopkins\nUniversity. To them also, on the whole, belongs Professor John Dewey,\nwho, with James and Dr. Schiller, was one of the three founders of\npragmatism. The view of the \"behaviourists\" is that nothing can be known\nexcept by external observation. They deny altogether that there is a\nseparate source of knowledge called \"introspection,\" by which we can\nknow things about ourselves which we could never observe in others.\nThey do not by any means deny that all sorts of things MAY go on in\nour minds: they only say that such things, if they occur, are not\nsusceptible of scientific observation, and do not therefore concern\npsychology as a science. Psychology as a science, they say, is only\nconcerned with BEHAVIOUR, i.e. with what we DO; this alone, they\ncontend, can be accurately observed. Whether we think meanwhile, they\ntell us, cannot be known; in their observation of the behaviour of human\nbeings, they have not so far found any evidence of thought. True, we\ntalk a great deal, and imagine that in so doing we are showing that we\ncan think; but behaviourists say that the talk they have to listen to\ncan be explained without supposing that people think. Where you might\nexpect a chapter on \"thought processes\" you come instead upon a chapter\non \"The Language Habit.\" It is humiliating to find how terribly adequate\nthis hypothesis turns out to be.\n\n     * See especially his \"Behavior: an Introduction to\n     Comparative Psychology,\" New York, 1914.\n\nBehaviourism has not, however, sprung from observing the folly of men.\nIt is the wisdom of animals that has suggested the view. It has always\nbeen a common topic of popular discussion whether animals \"think.\" On\nthis topic people are prepared to take sides without having the vaguest\nidea what they mean by \"thinking.\" Those who desired to investigate such\nquestions were led to observe the behaviour of animals, in the hope that\ntheir behaviour would throw some light on their mental faculties.\nAt first sight, it might seem that this is so. People say that a\ndog \"knows\" its name because it comes when it is called, and that it\n\"remembers\" its master, because it looks sad in his absence, but wags\nits tail and barks when he returns. That the dog behaves in this way is\nmatter of observation, but that it \"knows\" or \"remembers\" anything is an\ninference, and in fact a very doubtful one. The more such inferences are\nexamined, the more precarious they are seen to be. Hence the study of\nanimal behaviour has been gradually led to abandon all attempt at mental\ninterpretation. And it can hardly be doubted that, in many cases of\ncomplicated behaviour very well adapted to its ends, there can be no\nprevision of those ends. The first time a bird builds a nest, we can\nhardly suppose it knows that there will be eggs to be laid in it, or\nthat it will sit on the eggs, or that they will hatch into young birds.\nIt does what it does at each stage because instinct gives it an impulse\nto do just that, not because it foresees and desires the result of its\nactions.*\n\n     * An interesting discussion of the question whether\n     instinctive actions, when first performed, involve any\n     prevision, however vague, will be found in Lloyd Morgan's\n     \"Instinct and Experience\" (Methuen, 1912), chap. ii.\n\nCareful observers of animals, being anxious to avoid precarious\ninferences, have gradually discovered more and more how to give\nan account of the actions of animals without assuming what we call\n\"consciousness.\" It has seemed to the behaviourists that similar methods\ncan be applied to human behaviour, without assuming anything not open\nto external observation. Let us give a crude illustration, too crude for\nthe authors in question, but capable of affording a rough insight into\ntheir meaning. Suppose two children in a school, both of whom are asked\n\"What is six times nine?\" One says fifty-four, the other says fifty-six.\nThe one, we say, \"knows\" what six times nine is, the other does not. But\nall that we can observe is a certain language-habit. The one child has\nacquired the habit of saying \"six times nine is fifty-four\"; the other\nhas not. There is no more need of \"thought\" in this than there is when\na horse turns into his accustomed stable; there are merely more numerous\nand complicated habits. There is obviously an observable fact called\n\"knowing\" such-and-such a thing; examinations are experiments for\ndiscovering such facts. But all that is observed or discovered is a\ncertain set of habits in the use of words. The thoughts (if any) in the\nmind of the examinee are of no interest to the examiner; nor has the\nexaminer any reason to suppose even the most successful examinee capable\nof even the smallest amount of thought.\n\nThus what is called \"knowing,\" in the sense in which we can ascertain\nwhat other people \"know,\" is a phenomenon exemplified in their physical\nbehaviour, including spoken and written words. There is no reason--so\nWatson argues--to suppose that their knowledge IS anything beyond the\nhabits shown in this behaviour: the inference that other people\nhave something nonphysical called \"mind\" or \"thought\" is therefore\nunwarranted.\n\nSo far, there is nothing particularly repugnant to our prejudices in the\nconclusions of the behaviourists. We are all willing to admit that\nother people are thoughtless. But when it comes to ourselves, we feel\nconvinced that we can actually perceive our own thinking. \"Cogito, ergo\nsum\" would be regarded by most people as having a true premiss. This,\nhowever, the behaviourist denies. He maintains that our knowledge of\nourselves is no different in kind from our knowledge of other people.\nWe may see MORE, because our own body is easier to observe than that of\nother people; but we do not see anything radically unlike what we see\nof others. Introspection, as a separate source of knowledge, is entirely\ndenied by psychologists of this school. I shall discuss this question at\nlength in a later lecture; for the present I will only observe that\nit is by no means simple, and that, though I believe the behaviourists\nsomewhat overstate their case, yet there is an important element of\ntruth in their contention, since the things which we can discover by\nintrospection do not seem to differ in any very fundamental way from the\nthings which we discover by external observation.\n\nSo far, we have been principally concerned with knowing. But it might\nwell be maintained that desiring is what is really most characteristic\nof mind. Human beings are constantly engaged in achieving some end\nthey feel pleasure in success and pain in failure. In a purely material\nworld, it may be said, there would be no opposition of pleasant and\nunpleasant, good and bad, what is desired and what is feared. A man's\nacts are governed by purposes. He decides, let us suppose, to go to a\ncertain place, whereupon he proceeds to the station, takes his ticket\nand enters the train. If the usual route is blocked by an accident,\nhe goes by some other route. All that he does is determined--or so it\nseems--by the end he has in view, by what lies in front of him, rather\nthan by what lies behind. With dead matter, this is not the case.\nA stone at the top of a hill may start rolling, but it shows no\npertinacity in trying to get to the bottom. Any ledge or obstacle will\nstop it, and it will exhibit no signs of discontent if this happens. It\nis not attracted by the pleasantness of the valley, as a sheep or cow\nmight be, but propelled by the steepness of the hill at the place\nwhere it is. In all this we have characteristic differences between the\nbehaviour of animals and the behaviour of matter as studied by physics.\n\nDesire, like knowledge, is, of course, in one sense an observable\nphenomenon. An elephant will eat a bun, but not a mutton chop; a duck\nwill go into the water, but a hen will not. But when we think of our\nown desires, most people believe that we can know them by an immediate\nself-knowledge which does not depend upon observation of our actions.\nYet if this were the case, it would be odd that people are so often\nmistaken as to what they desire. It is matter of common observation that\n\"so-and-so does not know his own motives,\" or that \"A is envious of B\nand malicious about him, but quite unconscious of being so.\" Such people\nare called self-deceivers, and are supposed to have had to go through\nsome more or less elaborate process of concealing from themselves what\nwould otherwise have been obvious. I believe that this is an entire\nmistake. I believe that the discovery of our own motives can only be\nmade by the same process by which we discover other people's, namely,\nthe process of observing our actions and inferring the desire which\ncould prompt them. A desire is \"conscious\" when we have told ourselves\nthat we have it. A hungry man may say to himself: \"Oh, I do want my\nlunch.\" Then his desire is \"conscious.\" But it only differs from an\n\"unconscious\" desire by the presence of appropriate words, which is by\nno means a fundamental difference.\n\nThe belief that a motive is normally conscious makes it easier to be\nmistaken as to our own motives than as to other people's. When some\ndesire that we should be ashamed of is attributed to us, we notice that\nwe have never had it consciously, in the sense of saying to ourselves,\n\"I wish that would happen.\" We therefore look for some other\ninterpretation of our actions, and regard our friends as very unjust\nwhen they refuse to be convinced by our repudiation of what we hold to\nbe a calumny. Moral considerations greatly increase the difficulty of\nclear thinking in this matter. It is commonly argued that people are not\nto blame for unconscious motives, but only for conscious ones. In order,\ntherefore, to be wholly virtuous it is only necessary to repeat virtuous\nformulas. We say: \"I desire to be kind to my friends, honourable in\nbusiness, philanthropic towards the poor, public-spirited in politics.\"\nSo long as we refuse to allow ourselves, even in the watches of the\nnight, to avow any contrary desires, we may be bullies at home, shady in\nthe City, skinflints in paying wages and profiteers in dealing with the\npublic; yet, if only conscious motives are to count in moral valuation,\nwe shall remain model characters. This is an agreeable doctrine, and\nit is not surprising that men are un willing to abandon it. But moral\nconsiderations are the worst enemies of the scientific spirit and we\nmust dismiss them from our minds if we wish to arrive at truth.\n\nI believe--as I shall try to prove in a later lecture--that desire,\nlike force in mechanics, is of the nature of a convenient fiction\nfor describing shortly certain laws of behaviour. A hungry animal is\nrestless until it finds food; then it becomes quiescent. The thing which\nwill bring a restless condition to an end is said to be what is desired.\nBut only experience can show what will have this sedative effect, and it\nis easy to make mistakes. We feel dissatisfaction, and think that\nsuch and-such a thing would remove it; but in thinking this, we are\ntheorizing, not observing a patent fact. Our theorizing is often\nmistaken, and when it is mistaken there is a difference between what we\nthink we desire and what in fact will bring satisfaction. This is such\na common phenomenon that any theory of desire which fails to account for\nit must be wrong.\n\nWhat have been called \"unconscious\" desires have been brought very much\nto the fore in recent years by psycho-analysis. Psycho-analysis, as\nevery one knows, is primarily a method of understanding hysteria and\ncertain forms of insanity*; but it has been found that there is much\nin the lives of ordinary men and women which bears a humiliating\nresemblance to the delusions of the insane. The connection of dreams,\nirrational beliefs and foolish actions with unconscious wishes has been\nbrought to light, though with some exaggeration, by Freud and Jung and\ntheir followers. As regards the nature of these unconscious wishes,\nit seems to me--though as a layman I speak with diffidence--that many\npsycho-analysts are unduly narrow; no doubt the wishes they emphasize\nexist, but others, e.g. for honour and power, are equally operative and\nequally liable to concealment. This, however, does not affect the\nvalue of their general theories from the point of view of theoretic\npsychology, and it is from this point of view that their results are\nimportant for the analysis of mind.\n\n     * There is a wide field of \"unconscious\" phenomena which\n     does not depend upon psycho-analytic theories. Such\n     occurrences as automatic writing lead Dr. Morton Prince to\n     say: \"As I view this question of the subconscious, far too\n     much weight is given to the point of awareness or not\n     awareness of our conscious processes. As a matter of fact,\n     we find entirely identical phenomena, that is, identical in\n     every respect but one-that of awareness in which sometimes\n     we are aware of these conscious phenomena and sometimes\n     not\"(p. 87 of \"Subconscious Phenomena,\" by various authors,\n     Rebman). Dr. Morton Price conceives that there may be\n     \"consciousness\" without \"awareness.\" But this is a difficult\n     view, and one which makes some definition of \"consciousness\"\n     imperative. For nay part, I cannot see how to separate\n     consciousness from awareness.\n\nWhat, I think, is clearly established, is that a man's actions and\nbeliefs may be wholly dominated by a desire of which he is quite\nunconscious, and which he indignantly repudiates when it is suggested\nto him. Such a desire is generally, in morbid cases, of a sort which\nthe patient would consider wicked; if he had to admit that he had the\ndesire, he would loathe himself. Yet it is so strong that it must force\nan outlet for itself; hence it becomes necessary to entertain whole\nsystems of false beliefs in order to hide the nature of what is desired.\nThe resulting delusions in very many cases disappear if the hysteric or\nlunatic can be made to face the facts about himself. The consequence\nof this is that the treatment of many forms of insanity has grown more\npsychological and less physiological than it used to be. Instead of\nlooking for a physical defect in the brain, those who treat delusions\nlook for the repressed desire which has found this contorted mode\nof expression. For those who do not wish to plunge into the somewhat\nrepulsive and often rather wild theories of psychoanalytic pioneers, it\nwill be worth while to read a little book by Dr. Bernard Hart on \"The\nPsychology of Insanity.\"* On this question of the mental as opposed to\nthe physiological study of the causes of insanity, Dr. Hart says:\n\n     * Cambridge, 1912; 2nd edition, 1914. The following\n     references are to the second edition.\n\n\"The psychological conception [of insanity] is based on the view that\nmental processes can be directly studied without any reference to the\naccompanying changes which are presumed to take place in the brain, and\nthat insanity may therefore be properly attacked from the standpoint of\npsychology\"(p. 9).\n\nThis illustrates a point which I am anxious to make clear from the\noutset. Any attempt to classify modern views, such as I propose to\nadvocate, from the old standpoint of materialism and idealism, is only\nmisleading. In certain respects, the views which I shall be setting\nforth approximate to materialism; in certain others, they approximate to\nits opposite. On this question of the study of delusions, the practical\neffect of the modern theories, as Dr. Hart points out, is emancipation\nfrom the materialist method. On the other hand, as he also points\nout (pp. 38-9), imbecility and dementia still have to be considered\nphysiologically, as caused by defects in the brain. There is no\ninconsistency in this If, as we maintain, mind and matter are neither of\nthem the actual stuff of reality, but different convenient groupings of\nan underlying material, then, clearly, the question whether, in regard\nto a given phenomenon, we are to seek a physical or a mental cause, is\nmerely one to be decided by trial. Metaphysicians have argued endlessly\nas to the interaction of mind and matter. The followers of Descartes\nheld that mind and matter are so different as to make any action of the\none on the other impossible. When I will to move my arm, they said,\nit is not my will that operates on my arm, but God, who, by His\nomnipotence, moves my arm whenever I want it moved. The modern doctrine\nof psychophysical parallelism is not appreciably different from this\ntheory of the Cartesian school. Psycho-physical parallelism is the\ntheory that mental and physical events each have causes in their own\nsphere, but run on side by side owing to the fact that every state of\nthe brain coexists with a definite state of the mind, and vice versa.\nThis view of the reciprocal causal independence of mind and matter has\nno basis except in metaphysical theory.* For us, there is no necessity\nto make any such assumption, which is very difficult to harmonize with\nobvious facts. I receive a letter inviting me to dinner: the letter is\na physical fact, but my apprehension of its meaning is mental. Here we\nhave an effect of matter on mind. In consequence of my apprehension of\nthe meaning of the letter, I go to the right place at the right time;\nhere we have an effect of mind on matter. I shall try to persuade you,\nin the course of these lectures, that matter is not so material and mind\nnot so mental as is generally supposed. When we are speaking of matter,\nit will seem as if we were inclining to idealism; when we are speaking\nof mind, it will seem as if we were inclining to materialism. Neither\nis the truth. Our world is to be constructed out of what the American\nrealists call \"neutral\" entities, which have neither the hardness and\nindestructibility of matter, nor the reference to objects which is\nsupposed to characterize mind.\n\n     * It would seem, however, that Dr. Hart accepts this theory\n     as 8 methodological precept. See his contribution to\n     \"Subconscious Phenomena\" (quoted above), especially pp. 121-2.\n\nThere is, it is true, one objection which might be felt, not indeed to\nthe action of matter on mind, but to the action of mind on matter. The\nlaws of physics, it may be urged, are apparently adequate to explain\neverything that happens to matter, even when it is matter in a man's\nbrain. This, however, is only a hypothesis, not an established theory.\nThere is no cogent empirical reason for supposing that the laws\ndetermining the motions of living bodies are exactly the same as those\nthat apply to dead matter. Sometimes, of course, they are clearly the\nsame. When a man falls from a precipice or slips on a piece of orange\npeel, his body behaves as if it were devoid of life. These are the\noccasions that make Bergson laugh. But when a man's bodily movements\nare what we call \"voluntary,\" they are, at any rate prima facie, very\ndifferent in their laws from the movements of what is devoid of life.\nI do not wish to say dogmatically that the difference is irreducible;\nI think it highly probable that it is not. I say only that the study of\nthe behaviour of living bodies, in the present state of our knowledge,\nis distinct from physics. The study of gases was originally quite\ndistinct from that of rigid bodies, and would never have advanced to its\npresent state if it had not been independently pursued. Nowadays both\nthe gas and the rigid body are manufactured out of a more primitive and\nuniversal kind of matter. In like manner, as a question of methodology,\nthe laws of living bodies are to be studied, in the first place, without\nany undue haste to subordinate them to the laws of physics. Boyle's law\nand the rest had to be discovered before the kinetic theory of gases\nbecame possible. But in psychology we are hardly yet at the stage\nof Boyle's law. Meanwhile we need not be held up by the bogey of\nthe universal rigid exactness of physics. This is, as yet, a mere\nhypothesis, to be tested empirically without any preconceptions. It may\nbe true, or it may not. So far, that is all we can say.\n\nReturning from this digression to our main topic, namely, the criticism\nof \"consciousness,\" we observe that Freud and his followers, though they\nhave demonstrated beyond dispute the immense importance of \"unconscious\"\ndesires in determining our actions and beliefs, have not attempted the\ntask of telling us what an \"unconscious\" desire actually is, and have\nthus invested their doctrine with an air of mystery and mythology which\nforms a large part of its popular attractiveness. They speak always as\nthough it were more normal for a desire to be conscious, and as though\na positive cause had to be assigned for its being unconscious. Thus\n\"the unconscious\" becomes a sort of underground prisoner, living in a\ndungeon, breaking in at long intervals upon our daylight respectability\nwith dark groans and maledictions and strange atavistic lusts. The\nordinary reader, almost inevitably, thinks of this underground person as\nanother consciousness, prevented by what Freud calls the \"censor\" from\nmaking his voice heard in company, except on rare and dreadful occasions\nwhen he shouts so loud that every one hears him and there is a scandal.\nMost of us like the idea that we could be desperately wicked if only we\nlet ourselves go. For this reason, the Freudian \"unconscious\" has been a\nconsolation to many quiet and well-behaved persons.\n\nI do not think the truth is quite so picturesque as this. I believe an\n\"unconscious\" desire is merely a causal law of our behaviour,* namely,\nthat we remain restlessly active until a certain state of affairs is\nrealized, when we achieve temporary equilibrium If we know beforehand\nwhat this state of affairs is, our desire is conscious; if not,\nunconscious. The unconscious desire is not something actually existing,\nbut merely a tendency to a certain behaviour; it has exactly the same\nstatus as a force in dynamics. The unconscious desire is in no way\nmysterious; it is the natural primitive form of desire, from which the\nother has developed through our habit of observing and theorizing (often\nwrongly). It is not necessary to suppose, as Freud seems to do,\nthat every unconscious wish was once conscious, and was then, in his\nterminology, \"repressed\" because we disapproved of it. On the contrary,\nwe shall suppose that, although Freudian \"repression\" undoubtedly occurs\nand is important, it is not the usual reason for unconsciousness of our\nwishes. The usual reason is merely that wishes are all, to begin with,\nunconscious, and only become known when they are actively noticed.\nUsually, from laziness, people do not notice, but accept the theory\nof human nature which they find current, and attribute to themselves\nwhatever wishes this theory would lead them to expect. We used to be\nfull of virtuous wishes, but since Freud our wishes have become, in\nthe words of the Prophet Jeremiah, \"deceitful above all things and\ndesperately wicked.\" Both these views, in most of those who have held\nthem, are the product of theory rather than observation, for observation\nrequires effort, whereas repeating phrases does not.\n\n     * Cf. Hart, \"The Psychology of Insanity,\" p. 19.\n\nThe interpretation of unconscious wishes which I have been advocating\nhas been set forth briefly by Professor John B. Watson in an article\ncalled \"The Psychology of Wish Fulfilment,\" which appeared in \"The\nScientific Monthly\" in November, 1916. Two quotations will serve to show\nhis point of view:\n\n\"The Freudians (he says) have made more or less of a 'metaphysical\nentity' out of the censor. They suppose that when wishes are repressed\nthey are repressed into the 'unconscious,' and that this mysterious\ncensor stands at the trapdoor lying between the conscious and the\nunconscious. Many of us do not believe in a world of the unconscious\n(a few of us even have grave doubts about the usefulness of the term\nconsciousness), hence we try to explain censorship along ordinary\nbiological lines. We believe that one group of habits can 'down' another\ngroup of habits--or instincts. In this case our ordinary system of\nhabits--those which we call expressive of our 'real selves'--inhibit\nor quench (keep inactive or partially inactive) those habits and\ninstinctive tendencies which belong largely in the past\"(p. 483).\n\nAgain, after speaking of the frustration of some impulses which is\ninvolved in acquiring the habits of a civilized adult, he continues:\n\n\"It is among these frustrated impulses that I would find the biological\nbasis of the unfulfilled wish. Such 'wishes' need never have been\n'conscious,' and NEED NEVER HAVE BEEN SUPPRESSED INTO FREUD'S REALM\nOF THE UNCONSCIOUS. It may be inferred from this that there is no\nparticular reason for applying the term 'wish' to such tendencies\"(p.\n485).\n\nOne of the merits of the general analysis of mind which we shall\nbe concerned with in the following lectures is that it removes the\natmosphere of mystery from the phenomena brought to light by the\npsycho-analysts. Mystery is delightful, but unscientific, since it\ndepends upon ignorance. Man has developed out of the animals, and\nthere is no serious gap between him and the amoeba. Something closely\nanalogous to knowledge and desire, as regards its effects on behaviour,\nexists among animals, even where what we call \"consciousness\" is hard\nto believe in; something equally analogous exists in ourselves in cases\nwhere no trace of \"consciousness\" can be found. It is therefore\nnatural to suppose that, what ever may be the correct definition of\n\"consciousness,\" \"consciousness\" is not the essence of life or mind. In\nthe following lectures, accordingly, this term will disappear until we\nhave dealt with words, when it will re-emerge as mainly a trivial and\nunimportant outcome of linguistic habits.\n\n\n\n\nLECTURE II. INSTINCT AND HABIT\n\nIn attempting to understand the elements out of which mental phenomena\nare compounded, it is of the greatest importance to remember that from\nthe protozoa to man there is nowhere a very wide gap either in structure\nor in behaviour. From this fact it is a highly probable inference that\nthere is also nowhere a very wide mental gap. It is, of course, POSSIBLE\nthat there may be, at certain stages in evolution, elements which are\nentirely new from the standpoint of analysis, though in their nascent\nform they have little influence on behaviour and no very marked\ncorrelatives in structure. But the hypothesis of continuity in mental\ndevelopment is clearly preferable if no psychological facts make it\nimpossible. We shall find, if I am not mistaken, that there are no facts\nwhich refute the hypothesis of mental continuity, and that, on the other\nhand, this hypothesis affords a useful test of suggested theories as to\nthe nature of mind.\n\nThe hypothesis of mental continuity throughout organic evolution may be\nused in two different ways. On the one hand, it may be held that we\nhave more knowledge of our own minds than those of animals, and that we\nshould use this knowledge to infer the existence of something similar\nto our own mental processes in animals and even in plants. On the other\nhand, it may be held that animals and plants present simpler phenomena,\nmore easily analysed than those of human minds; on this ground it may be\nurged that explanations which are adequate in the case of animals ought\nnot to be lightly rejected in the case of man. The practical effects of\nthese two views are diametrically opposite: the first leads us to level\nup animal intelligence with what we believe ourselves to know about our\nown intelligence, while the second leads us to attempt a levelling down\nof our own intelligence to something not too remote from what we can\nobserve in animals. It is therefore important to consider the relative\njustification of the two ways of applying the principle of continuity.\n\nIt is clear that the question turns upon another, namely, which can we\nknow best, the psychology of animals or that of human beings? If we\ncan know most about animals, we shall use this knowledge as a basis for\ninference about human beings; if we can know most about human beings, we\nshall adopt the opposite procedure. And the question whether we can know\nmost about the psychology of human beings or about that of animals turns\nupon yet another, namely: Is introspection or external observation\nthe surer method in psychology? This is a question which I propose to\ndiscuss at length in Lecture VI; I shall therefore content myself now\nwith a statement of the conclusions to be arrived at.\n\nWe know a great many things concerning ourselves which we cannot know\nnearly so directly concerning animals or even other people. We know when\nwe have a toothache, what we are thinking of, what dreams we have when\nwe are asleep, and a host of other occurrences which we only know about\nothers when they tell us of them, or otherwise make them inferable\nby their behaviour. Thus, so far as knowledge of detached facts is\nconcerned, the advantage is on the side of self-knowledge as against\nexternal observation.\n\nBut when we come to the analysis and scientific understanding of the\nfacts, the advantages on the side of self-knowledge become far less\nclear. We know, for example, that we have desires and beliefs, but we\ndo not know what constitutes a desire or a belief. The phenomena are so\nfamiliar that it is difficult to realize how little we really know about\nthem. We see in animals, and to a lesser extent in plants, behaviour\nmore or less similar to that which, in us, is prompted by desires and\nbeliefs, and we find that, as we descend in the scale of evolution,\nbehaviour becomes simpler, more easily reducible to rule, more\nscientifically analysable and predictable. And just because we are not\nmisled by familiarity we find it easier to be cautious in interpreting\nbehaviour when we are dealing with phenomena remote from those of our\nown minds: Moreover, introspection, as psychoanalysis has demonstrated,\nis extraordinarily fallible even in cases where we feel a high degree of\ncertainty. The net result seems to be that, though self-knowledge has\na definite and important contribution to make to psychology, it is\nexceedingly misleading unless it is constantly checked and controlled\nby the test of external observation, and by the theories which such\nobservation suggests when applied to animal behaviour. On the whole,\ntherefore, there is probably more to be learnt about human psychology\nfrom animals than about animal psychology from human beings; but this\nconclusion is one of degree, and must not be pressed beyond a point.\n\nIt is only bodily phenomena that can be directly observed in animals,\nor even, strictly speaking, in other human beings. We can observe such\nthings as their movements, their physiological processes, and the sounds\nthey emit. Such things as desires and beliefs, which seem obvious\nto introspection, are not visible directly to external observation.\nAccordingly, if we begin our study of psychology by external\nobservation, we must not begin by assuming such things as desires and\nbeliefs, but only such things as external observation can reveal, which\nwill be characteristics of the movements and physiological processes of\nanimals. Some animals, for example, always run away from light and hide\nthemselves in dark places. If you pick up a mossy stone which is lightly\nembedded in the earth, you will see a number of small animals scuttling\naway from the unwonted daylight and seeking again the darkness of which\nyou have deprived them. Such animals are sensitive to light, in the\nsense that their movements are affected by it; but it would be rash to\ninfer that they have sensations in any way analogous to our sensations\nof sight. Such inferences, which go beyond the observable facts, are to\nbe avoided with the utmost care.\n\nIt is customary to divide human movements into three classes, voluntary,\nreflex and mechanical. We may illustrate the distinction by a quotation\nfrom William James (\"Psychology,\" i, 12):\n\n\"If I hear the conductor calling 'all aboard' as I enter the depot, my\nheart first stops, then palpitates, and my legs respond to the air-waves\nfalling on my tympanum by quickening their movements. If I stumble as I\nrun, the sensation of falling provokes a movement of the hands towards\nthe direction of the fall, the effect of which is to shield the body\nfrom too sudden a shock. If a cinder enter my eye, its lids close\nforcibly and a copious flow of tears tends to wash it out.\n\n\"These three responses to a sensational stimulus differ, however, in\nmany respects. The closure of the eye and the lachrymation are quite\ninvoluntary, and so is the disturbance of the heart. Such involuntary\nresponses we know as 'reflex' acts. The motion of the arms to break the\nshock of falling may also be called reflex, since it occurs too quickly\nto be deliberately intended. Whether it be instinctive or whether it\nresult from the pedestrian education of childhood may be doubtful; it\nis, at any rate, less automatic than the previous acts, for a man might\nby conscious effort learn to perform it more skilfully, or even to\nsuppress it altogether. Actions of this kind, with which instinct and\nvolition enter upon equal terms, have been called 'semi-reflex.' The\nact of running towards the train, on the other hand, has no instinctive\nelement about it. It is purely the result of education, and is preceded\nby a consciousness of the purpose to be attained and a distinct mandate\nof the will. It is a 'voluntary act.' Thus the animal's reflex and\nvoluntary performances shade into each other gradually, being connected\nby acts which may often occur automatically, but may also be modified by\nconscious intelligence.\n\n\"An outside observer, unable to perceive the accompanying consciousness,\nmight be wholly at a loss to discriminate between the automatic acts and\nthose which volition escorted. But if the criterion of mind's existence\nbe the choice of the proper means for the attainment of a supposed\nend, all the acts alike seem to be inspired by intelligence, for\nAPPROPRIATENESS characterizes them all alike.\"\n\nThere is one movement, among those that James mentions at first, which\nis not subsequently classified, namely, the stumbling. This is the\nkind of movement which may be called \"mechanical\"; it is evidently of a\ndifferent kind from either reflex or voluntary movements, and more akin\nto the movements of dead matter. We may define a movement of an animal's\nbody as \"mechanical\" when it proceeds as if only dead matter were\ninvolved. For example, if you fall over a cliff, you move under the\ninfluence of gravitation, and your centre of gravity describes just as\ncorrect a parabola as if you were already dead. Mechanical movements\nhave not the characteristic of appropriateness, unless by accident, as\nwhen a drunken man falls into a waterbutt and is sobered. But reflex\nand voluntary movements are not ALWAYS appropriate, unless in some very\nrecondite sense. A moth flying into a lamp is not acting sensibly; no\nmore is a man who is in such a hurry to get his ticket that he cannot\nremember the name of his destination. Appropriateness is a complicated\nand merely approximate idea, and for the present we shall do well to\ndismiss it from our thoughts.\n\nAs James states, there is no difference, from the point of view of\nthe outside observer, between voluntary and reflex movements. The\nphysiologist can discover that both depend upon the nervous system,\nand he may find that the movements which we call voluntary depend upon\nhigher centres in the brain than those that are reflex. But he\ncannot discover anything as to the presence or absence of \"will\" or\n\"consciousness,\" for these things can only be seen from within, if\nat all. For the present, we wish to place ourselves resolutely in the\nposition of outside observers; we will therefore ignore the distinction\nbetween voluntary and reflex movements. We will call the two together\n\"vital\" movements. We may then distinguish \"vital\" from mechanical\nmovements by the fact that vital movements depend for their causation\nupon the special properties of the nervous system, while mechanical\nmovements depend only upon the properties which animal bodies share with\nmatter in general.\n\nThere is need for some care if the distinction between mechanical and\nvital movements is to be made precise. It is quite likely that, if we\nknew more about animal bodies, we could deduce all their movements from\nthe laws of chemistry and physics. It is already fairly easy to see how\nchemistry reduces to physics, i.e. how the differences between different\nchemical elements can be accounted for by differences of physical\nstructure, the constituents of the structure being electrons which are\nexactly alike in all kinds of matter. We only know in part how to reduce\nphysiology to chemistry, but we know enough to make it likely that the\nreduction is possible. If we suppose it effected, what would become of\nthe difference between vital and mechanical movements?\n\nSome analogies will make the difference clear. A shock to a mass of\ndynamite produces quite different effects from an equal shock to a mass\nof steel: in the one case there is a vast explosion, while in the other\ncase there is hardly any noticeable disturbance. Similarly, you may\nsometimes find on a mountain-side a large rock poised so delicately that\na touch will set it crashing down into the valley, while the rocks all\nround are so firm that only a considerable force can dislodge them What\nis analogous in these two cases is the existence of a great store of\nenergy in unstable equilibrium ready to burst into violent motion by\nthe addition of a very slight disturbance. Similarly, it requires only\na very slight expenditure of energy to send a post-card with the words\n\"All is discovered; fly!\" but the effect in generating kinetic energy\nis said to be amazing. A human body, like a mass of dynamite, contains\na store of energy in unstable equilibrium, ready to be directed in this\ndirection or that by a disturbance which is physically very small,\nsuch as a spoken word. In all such cases the reduction of behaviour to\nphysical laws can only be effected by entering into great minuteness; so\nlong as we confine ourselves to the observation of comparatively\nlarge masses, the way in which the equilibrium will be upset cannot be\ndetermined. Physicists distinguish between macroscopic and microscopic\nequations: the former determine the visible movements of bodies of\nordinary size, the latter the minute occurrences in the smallest parts.\nIt is only the microscopic equations that are supposed to be the same\nfor all sorts of matter. The macroscopic equations result from a process\nof averaging out, and may be different in different cases. So, in our\ninstance, the laws of macroscopic phenomena are different for mechanical\nand vital movements, though the laws of microscopic phenomena may be the\nsame.\n\nWe may say, speaking somewhat roughly, that a stimulus applied to the\nnervous system, like a spark to dynamite, is able to take advantage of\nthe stored energy in unstable equilibrium, and thus to produce movements\nout of proportion to the proximate cause. Movements produced in this way\nare vital movements, while mechanical movements are those in which the\nstored energy of a living body is not involved. Similarly dynamite may\nbe exploded, thereby displaying its characteristic properties, or may\n(with due precautions) be carted about like any other mineral. The\nexplosion is analogous to vital movements, the carting about to\nmechanical movements.\n\nMechanical movements are of no interest to the psychologist, and it has\nonly been necessary to define them in order to be able to exclude them.\nWhen a psychologist studies behaviour, it is only vital movements\nthat concern him. We shall, therefore, proceed to ignore mechanical\nmovements, and study only the properties of the remainder.\n\nThe next point is to distinguish between movements that are instinctive\nand movements that are acquired by experience. This distinction also is\nto some extent one of degree. Professor Lloyd Morgan gives the following\ndefinition of \"instinctive behaviour\":\n\n\"That which is, on its first occurrence, independent of prior\nexperience; which tends to the well-being of the individual and the\npreservation of the race; which is similarly performed by all members\nof the same more or less restricted group of animals; and which may be\nsubject to subsequent modification under the guidance of experience.\" *\n\n     * \"Instinct and Experience\" (Methuen, 1912) p. 5.\n\nThis definition is framed for the purposes of biology, and is in\nsome respects unsuited to the needs of psychology. Though perhaps\nunavoidable, allusion to \"the same more or less restricted group\nof animals\" makes it impossible to judge what is instinctive in the\nbehaviour of an isolated individual. Moreover, \"the well-being of\nthe individual and the preservation of the race\" is only a usual\ncharacteristic, not a universal one, of the sort of movements that, from\nour point of view, are to be called instinctive; instances of harmful\ninstincts will be given shortly. The essential point of the definition,\nfrom our point of view, is that an instinctive movement is in dependent\nof prior experience.\n\nWe may say that an \"instinctive\" movement is a vital movement performed\nby an animal the first time that it finds itself in a novel situation;\nor, more correctly, one which it would perform if the situation were\nnovel.* The instincts of an animal are different at different periods of\nits growth, and this fact may cause changes of behaviour which are\nnot due to learning. The maturing and seasonal fluctuation of the\nsex-instinct affords a good illustration. When the sex-instinct first\nmatures, the behaviour of an animal in the presence of a mate is\ndifferent from its previous behaviour in similar circumstances, but is\nnot learnt, since it is just the same if the animal has never previously\nbeen in the presence of a mate.\n\n     * Though this can only be decided by comparison with other\n     members of the species, and thus exposes us to the need of\n     comparison which we thought an objection to Professor Lloyd\n     Morgan's definition.\n\nOn the other hand, a movement is \"learnt,\" or embodies a \"habit,\" if it\nis due to previous experience of similar situations, and is not what it\nwould be if the animal had had no such experience.\n\nThere are various complications which blur the sharpness of this\ndistinction in practice. To begin with, many instincts mature gradually,\nand while they are immature an animal may act in a fumbling manner which\nis very difficult to distinguish from learning. James (\"Psychology,\" ii,\n407) maintains that children walk by instinct, and that the awkwardness\nof their first attempts is only due to the fact that the instinct has\nnot yet ripened. He hopes that \"some scientific widower, left alone with\nhis offspring at the critical moment, may ere long test this suggestion\non the living subject.\" However this may be, he quotes evidence to show\nthat \"birds do not LEARN to fly,\" but fly by instinct when they reach\nthe appropriate age (ib., p. 406). In the second place, instinct often\ngives only a rough outline of the sort of thing to do, in which case\nlearning is necessary in order to acquire certainty and precision in\naction. In the third place, even in the clearest cases of acquired\nhabit, such as speaking, some instinct is required to set in motion\nthe process of learning. In the case of speaking, the chief instinct\ninvolved is commonly supposed to be that of imitation, but this may be\nquestioned. (See Thorndike's \"Animal Intelligence,\" p. 253 ff.)\n\nIn spite of these qualifications, the broad distinction between instinct\nand habit is undeniable. To take extreme cases, every animal at birth\ncan take food by instinct, before it has had opportunity to learn; on\nthe other hand, no one can ride a bicycle by instinct, though, after\nlearning, the necessary movements become just as automatic as if they\nwere instinctive.\n\nThe process of learning, which consists in the acquisition of habits,\nhas been much studied in various animals.* For example: you put a hungry\nanimal, say a cat, in a cage which has a door that can be opened by\nlifting a latch; outside the cage you put food. The cat at first dashes\nall round the cage, making frantic efforts to force a way out. At last,\nby accident, the latch is lifted and the cat pounces on the food. Next\nday you repeat the experiment, and you find that the cat gets out much\nmore quickly than the first time, although it still makes some random\nmovements. The third day it gets out still more quickly, and before long\nit goes straight to the latch and lifts it at once. Or you make a model\nof the Hampton Court maze, and put a rat in the middle, assaulted by the\nsmell of food on the outside. The rat starts running down the passages,\nand is constantly stopped by blind alleys, but at last, by persistent\nattempts, it gets out. You repeat this experiment day after day; you\nmeasure the time taken by the rat in reaching the food; you find that\nthe time rapidly diminishes, and that after a while the rat ceases to\nmake any wrong turnings. It is by essentially similar processes that we\nlearn speaking, writing, mathematics, or the government of an empire.\n\n     * The scientific study of this subject may almost be said to\n     begin with Thorndike's \"Animal Intelligence\" (Macmillan,\n     1911).\n\nProfessor Watson (\"Behavior,\" pp. 262-3) has an ingenious theory as to\nthe way in which habit arises out of random movements. I think there is\na reason why his theory cannot be regarded as alone sufficient, but it\nseems not unlikely that it is partly correct. Suppose, for the sake of\nsimplicity, that there are just ten random movements which may be made\nby the animal--say, ten paths down which it may go--and that only one of\nthese leads to food, or whatever else represents success in the case in\nquestion. Then the successful movement always occurs during the animal's\nattempts, whereas each of the others, on the average, occurs in only\nhalf the attempts. Thus the tendency to repeat a previous performance\n(which is easily explicable without the intervention of \"consciousness\")\nleads to a greater emphasis on the successful movement than on any\nother, and in time causes it alone to be performed. The objection to\nthis view, if taken as the sole explanation, is that on improvement\nought to set in till after the SECOND trial, whereas experiment shows\nthat already at the second attempt the animal does better than the\nfirst time. Something further is, therefore, required to account for the\ngenesis of habit from random movements; but I see no reason to suppose\nthat what is further required involves \"consciousness.\"\n\nMr. Thorndike (op. cit., p. 244) formulates two \"provisional laws of\nacquired behaviour or learning,\" as follows:\n\n\"The Law of Effect is that: Of several responses made to the same\nsituation, those which are accompanied or closely followed by\nsatisfaction to the animal will, other things being equal, be more\nfirmly connected with the situation, so that, when it recurs, they will\nbe more likely to recur; those which are accompanied or closely followed\nby discomfort to the animal will, other things being equal, have their\nconnections with that situation weakened, so that, when it recurs,\nthey will be less likely to occur. The greater the satisfaction or\ndiscomfort, the greater the strengthening or weakening of the bond.\n\n\"The Law of Exercise is that: Any response to a situation will, other\nthings being equal, be more strongly connected with the situation\nin proportion to the number of times it has been connected with that\nsituation and to the average vigour and duration of the connections.\"\n\nWith the explanation to be presently given of the meaning of\n\"satisfaction\" and \"discomfort,\" there seems every reason to accept\nthese two laws.\n\nWhat is true of animals, as regards instinct and habit, is equally\ntrue of men. But the higher we rise in the evolutionary scale, broadly\nspeaking, the greater becomes the power of learning, and the fewer are\nthe occasions when pure instinct is exhibited unmodified in adult life.\nThis applies with great force to man, so much so that some have thought\ninstinct less important in the life of man than in that of animals.\nThis, however, would be a mistake. Learning is only possible when\ninstinct supplies the driving-force. The animals in cages, which\ngradually learn to get out, perform random movements at first, which\nare purely instinctive. But for these random movements, they would never\nacquire the experience which afterwards enables them to produce the\nright movement. (This is partly questioned by Hobhouse*--wrongly, I\nthink.) Similarly, children learning to talk make all sorts of sounds,\nuntil one day the right sound comes by accident. It is clear that the\noriginal making of random sounds, without which speech would never be\nlearnt, is instinctive. I think we may say the same of all the habits\nand aptitudes that we acquire in all of them there has been present\nthroughout some instinctive activity, prompting at first rather\ninefficient movements, but supplying the driving force while more and\nmore effective methods are being acquired. A cat which is hungry smells\nfish, and goes to the larder. This is a thoroughly efficient method when\nthere is fish in the larder, and it is often successfully practised by\nchildren. But in later life it is found that merely going to the larder\ndoes not cause fish to be there; after a series of random movements it\nis found that this result is to be caused by going to the City in the\nmorning and coming back in the evening. No one would have guessed a\npriori that this movement of a middle-aged man's body would cause fish\nto come out of the sea into his larder, but experience shows that it\ndoes, and the middle-aged man therefore continues to go to the City,\njust as the cat in the cage continues to lift the latch when it has once\nfound it. Of course, in actual fact, human learning is rendered easier,\nthough psychologically more complex, through language; but at bottom\nlanguage does not alter the essential character of learning, or of the\npart played by instinct in promoting learning. Language, however, is a\nsubject upon which I do not wish to speak until a later lecture.\n\n     * \"Mind in Evolution\" (Macmillan, 1915), pp. 236-237.\n\nThe popular conception of instinct errs by imagining it to be infallible\nand preternaturally wise, as well as incapable of modification. This is\na complete delusion. Instinct, as a rule, is very rough and ready, able\nto achieve its result under ordinary circumstances, but easily misled by\nanything unusual. Chicks follow their mother by instinct, but when they\nare quite young they will follow with equal readiness any moving\nobject remotely resembling their mother, or even a human being (James,\n\"Psychology,\" ii, 396). Bergson, quoting Fabre, has made play with the\nsupposed extraordinary accuracy of the solitary wasp Ammophila, which\nlays its eggs in a caterpillar. On this subject I will quote from\nDrever's \"Instinct in Man,\" p. 92:\n\n\"According to Fabre's observations, which Bergson accepts, the Ammophila\nstings its prey EXACTLY and UNERRINGLY in EACH of the nervous centres.\nThe result is that the caterpillar is paralyzed, but not immediately\nkilled, the advantage of this being that the larva cannot be injured by\nany movement of the caterpillar, upon which the egg is deposited, and is\nprovided with fresh meat when the time comes.\n\n\"Now Dr. and Mrs. Peckham have shown that the sting of the wasp is NOT\nUNERRING, as Fabre alleges, that the number of stings is NOT CONSTANT,\nthat sometimes the caterpillar is NOT PARALYZED, and sometimes it is\nKILLED OUTRIGHT, and that THE DIFFERENT CIRCUMSTANCES DO NOT APPARENTLY\nMAKE ANY DIFFERENCE TO THE LARVA, which is not injured by slight\nmovements of the caterpillar, nor by consuming food decomposed rather\nthan fresh caterpillar.\"\n\nThis illustrates how love of the marvellous may mislead even so careful\nan observer as Fabre and so eminent a philosopher as Bergson.\n\nIn the same chapter of Dr. Drever's book there are some interesting\nexamples of the mistakes made by instinct. I will quote one as a sample:\n\n\"The larva of the Lomechusa beetle eats the young of the ants, in whose\nnest it is reared. Nevertheless, the ants tend the Lomechusa larvae\nwith the same care they bestow on their own young. Not only so, but they\napparently discover that the methods of feeding, which suit their own\nlarvae, would prove fatal to the guests, and accordingly they change\ntheir whole system of nursing\" (loc. cit., p. 106).\n\nSemon (\"Die Mneme,\" pp. 207-9) gives a good illustration of an instinct\ngrowing wiser through experience. He relates how hunters attract stags\nby imitating the sounds of other members of their species, male or\nfemale, but find that the older a stag becomes the more difficult it\nis to deceive him, and the more accurate the imitation has to be. The\nliterature of instinct is vast, and illustrations might be multiplied\nindefinitely. The main points as regards instinct, which need to be\nemphasized as against the popular conceptions of it, are:\n\n(1) That instinct requires no prevision of the biological end which it\nserves;\n\n(2) That instinct is only adapted to achieve this end in the usual\ncircumstances of the animal in question, and has no more precision than\nis necessary for success AS A RULE;\n\n(3) That processes initiated by instinct often come to be performed\nbetter after experience;\n\n(4) That instinct supplies the impulses to experimental movements which\nare required for the process of learning;\n\n(5) That instincts in their nascent stages are easily modifiable, and\ncapable of being attached to various sorts of objects.\n\nAll the above characteristics of instinct can be established by purely\nexternal observation, except the fact that instinct does not require\nprevision. This, though not strictly capable of being PROVED by\nobservation, is irresistibly suggested by the most obvious phenomena.\nWho can believe, for example, that a new-born baby is aware of the\nnecessity of food for preserving life? Or that insects, in laying eggs,\nare concerned for the preservation of their species? The essence of\ninstinct, one might say, is that it provides a mechanism for\nacting without foresight in a manner which is usually advantageous\nbiologically. It is partly for this reason that it is so important to\nunderstand the fundamental position of instinct in prompting both animal\nand human behaviour.\n\n\n\n\nLECTURE III. DESIRE AND FEELING\n\nDesire is a subject upon which, if I am not mistaken, true views can\nonly be arrived at by an almost complete reversal of the ordinary\nunreflecting opinion. It is natural to regard desire as in its essence\nan attitude towards something which is imagined, not actual; this\nsomething is called the END or OBJECT of the desire, and is said to be\nthe PURPOSE of any action resulting from the desire. We think of the\ncontent of the desire as being just like the content of a belief, while\nthe attitude taken up towards the content is different. According to\nthis theory, when we say: \"I hope it will rain,\" or \"I expect it will\nrain,\" we express, in the first case, a desire, and in the second, a\nbelief, with an identical content, namely, the image of rain. It would\nbe easy to say that, just as belief is one kind of feeling in relation\nto this content, so desire is another kind. According to this view, what\ncomes first in desire is something imagined, with a specific feeling\nrelated to it, namely, that specific feeling which we call \"desiring\"\nit. The discomfort associated with unsatisfied desire, and the actions\nwhich aim at satisfying desire, are, in this view, both of them effects\nof the desire. I think it is fair to say that this is a view against\nwhich common sense would not rebel; nevertheless, I believe it to be\nradically mistaken. It cannot be refuted logically, but various facts\ncan be adduced which make it gradually less simple and plausible, until\nat last it turns out to be easier to abandon it wholly and look at the\nmatter in a totally different way.\n\nThe first set of facts to be adduced against the common sense view of\ndesire are those studied by psycho-analysis. In all human beings, but\nmost markedly in those suffering from hysteria and certain forms of\ninsanity, we find what are called \"unconscious\" desires, which are\ncommonly regarded as showing self-deception. Most psycho-analysts\npay little attention to the analysis of desire, being interested in\ndiscovering by observation what it is that people desire, rather than in\ndiscovering what actually constitutes desire. I think the strangeness of\nwhat they report would be greatly diminished if it were expressed in the\nlanguage of a behaviourist theory of desire, rather than in the language\nof every-day beliefs. The general description of the sort of phenomena\nthat bear on our present question is as follows: A person states that\nhis desires are so-and-so, and that it is these desires that inspire his\nactions; but the outside observer perceives that his actions are such\nas to realize quite different ends from those which he avows, and\nthat these different ends are such as he might be expected to desire.\nGenerally they are less virtuous than his professed desires, and are\ntherefore less agreeable to profess than these are. It is accordingly\nsupposed that they really exist as desires for ends, but in a\nsubconscious part of the mind, which the patient refuses to admit into\nconsciousness for fear of having to think ill of himself. There are\nno doubt many cases to which such a supposition is applicable without\nobvious artificiality. But the deeper the Freudians delve into the\nunderground regions of instinct, the further they travel from anything\nresembling conscious desire, and the less possible it becomes to believe\nthat only positive self-deception conceals from us that we really wish\nfor things which are abhorrent to our explicit life.\n\nIn the cases in question we have a conflict between the outside observer\nand the patient's consciousness. The whole tendency of psycho-analysis\nis to trust the outside observer rather than the testimony of\nintrospection. I believe this tendency to be entirely right, but to\ndemand a re-statement of what constitutes desire, exhibiting it as a\ncausal law of our actions, not as something actually existing in our\nminds.\n\nBut let us first get a clearer statement of the essential characteristic\nof the phenomena.\n\nA person, we find, states that he desires a certain end A, and that he\nis acting with a view to achieving it. We observe, however, that his\nactions are such as are likely to achieve a quite different end B, and\nthat B is the sort of end that often seems to be aimed at by animals and\nsavages, though civilized people are supposed to have discarded it. We\nsometimes find also a whole set of false beliefs, of such a kind as to\npersuade the patient that his actions are really a means to A, when in\nfact they are a means to B. For example, we have an impulse to inflict\npain upon those whom we hate; we therefore believe that they are wicked,\nand that punishment will reform them. This belief enables us to act upon\nthe impulse to inflict pain, while believing that we are acting upon\nthe desire to lead sinners to repentance. It is for this reason that the\ncriminal law has been in all ages more severe than it would have been if\nthe impulse to ameliorate the criminal had been what really inspired\nit. It seems simple to explain such a state of affairs as due to\n\"self-deception,\" but this explanation is often mythical. Most people,\nin thinking about punishment, have had no more need to hide their\nvindictive impulses from themselves than they have had to hide\nthe exponential theorem. Our impulses are not patent to a casual\nobservation, but are only to be discovered by a scientific study of our\nactions, in the course of which we must regard ourselves as objectively\nas we should the motions of the planets or the chemical reactions of a\nnew element.\n\nThe study of animals reinforces this conclusion, and is in many ways\nthe best preparation for the analysis of desire. In animals we are\nnot troubled by the disturbing influence of ethical considerations. In\ndealing with human beings, we are perpetually distracted by being told\nthat such-and-such a view is gloomy or cynical or pessimistic: ages of\nhuman conceit have built up such a vast myth as to our wisdom and virtue\nthat any intrusion of the mere scientific desire to know the facts is\ninstantly resented by those who cling to comfortable illusions. But no\none cares whether animals are virtuous or not, and no one is under the\ndelusion that they are rational. Moreover, we do not expect them to be\nso \"conscious,\" and are prepared to admit that their instincts prompt\nuseful actions without any prevision of the ends which they achieve. For\nall these reasons, there is much in the analysis of mind which is more\neasily discovered by the study of animals than by the observation of\nhuman beings.\n\nWe all think that, by watching the behaviour of animals, we can discover\nmore or less what they desire. If this is the case--and I fully agree\nthat it is--desire must be capable of being exhibited in actions, for it\nis only the actions of animals that we can observe. They MAY have minds\nin which all sorts of things take place, but we can know nothing about\ntheir minds except by means of inferences from their actions; and the\nmore such inferences are examined, the more dubious they appear. It\nwould seem, therefore, that actions alone must be the test of the\ndesires of animals. From this it is an easy step to the conclusion that\nan animal's desire is nothing but a characteristic of a certain series\nof actions, namely, those which would be commonly regarded as inspired\nby the desire in question. And when it has been shown that this view\naffords a satisfactory account of animal desires, it is not difficult\nto see that the same explanation is applicable to the desires of human\nbeings.\n\nWe judge easily from the behaviour of an animal of a familiar\nkind whether it is hungry or thirsty, or pleased or displeased, or\ninquisitive or terrified. The verification of our judgment, so far\nas verification is possible, must be derived from the immediately\nsucceeding actions of the animal. Most people would say that they infer\nfirst something about the animal's state of mind--whether it is hungry\nor thirsty and so on--and thence derive their expectations as to its\nsubsequent conduct. But this detour through the animal's supposed mind\nis wholly unnecessary. We can say simply: The animal's behaviour during\nthe last minute has had those characteristics which distinguish what\nis called \"hunger,\" and it is likely that its actions during the next\nminute will be similar in this respect, unless it finds food, or is\ninterrupted by a stronger impulse, such as fear. An animal which is\nhungry is restless, it goes to the places where food is often to be\nfound, it sniffs with its nose or peers with its eyes or otherwise\nincreases the sensitiveness of its sense-organs; as soon as it is near\nenough to food for its sense-organs to be affected, it goes to it with\nall speed and proceeds to eat; after which, if the quantity of food has\nbeen sufficient, its whole demeanour changes it may very likely lie\ndown and go to sleep. These things and others like them are observable\nphenomena distinguishing a hungry animal from one which is not hungry.\nThe characteristic mark by which we recognize a series of actions\nwhich display hunger is not the animal's mental state, which we cannot\nobserve, but something in its bodily behaviour; it is this observable\ntrait in the bodily behaviour that I am proposing to call \"hunger,\"\nnot some possibly mythical and certainly unknowable ingredient of the\nanimal's mind.\n\nGeneralizing what occurs in the case of hunger, we may say that what\nwe call a desire in an animal is always displayed in a cycle of actions\nhaving certain fairly well marked characteristics. There is first a\nstate of activity, consisting, with qualifications to be mentioned\npresently, of movements likely to have a certain result; these\nmovements, unless interrupted, continue until the result is achieved,\nafter which there is usually a period of comparative quiescence. A cycle\nof actions of this sort has marks by which it is broadly distinguished\nfrom the motions of dead matter. The most notable of these marks\nare--(1) the appropriateness of the actions for the realization of a\ncertain result; (2) the continuance of action until that result has been\nachieved. Neither of these can be pressed beyond a point. Either may\nbe (a) to some extent present in dead matter, and (b) to a considerable\nextent absent in animals, while vegetable are intermediate, and display\nonly a much fainter form of the behaviour which leads us to attribute\ndesire to animals. (a) One might say rivers \"desire\" the sea water,\nroughly speaking, remains in restless motion until it reaches either\nthe sea or a place from which it cannot issue without going uphill, and\ntherefore we might say that this is what it wishes while it is flowing.\nWe do not say so, because we can account for the behaviour of water by\nthe laws of physics; and if we knew more about animals, we might equally\ncease to attribute desires to them, since we might find physical and\nchemical reactions sufficient to account for their behaviour. (b) Many\nof the movements of animals do not exhibit the characteristics of the\ncycles which seem to embody desire. There are first of all the movements\nwhich are \"mechanical,\" such as slipping and falling, where ordinary\nphysical forces operate upon the animal's body almost as if it were\ndead matter. An animal which falls over a cliff may make a number of\ndesperate struggles while it is in the air, but its centre of gravity\nwill move exactly as it would if the animal were dead. In this case, if\nthe animal is killed at the end of the fall, we have, at first sight,\njust the characteristics of a cycle of actions embodying desire, namely,\nrestless movement until the ground is reached, and then quiescence.\nNevertheless, we feel no temptation to say that the animal desired what\noccurred, partly because of the obviously mechanical nature of the whole\noccurrence, partly because, when an animal survives a fall, it tends not\nto repeat the experience.\n\nThere may be other reasons also, but of them I do not wish to speak yet.\nBesides mechanical movements, there are interrupted movements, as when\na bird, on its way to eat your best peas, is frightened away by the boy\nwhom you are employing for that purpose. If interruptions are frequent\nand completion of cycles rare, the characteristics by which cycles\nare observed may become so blurred as to be almost unrecognizable. The\nresult of these various considerations is that the differences\nbetween animals and dead matter, when we confine ourselves to external\nunscientific observation of integral behaviour, are a matter of degree\nand not very precise. It is for this reason that it has always been\npossible for fanciful people to maintain that even stocks and stones\nhave some vague kind of soul. The evidence that animals have souls is\nso very shaky that, if it is assumed to be conclusive, one might just as\nwell go a step further and extend the argument by analogy to all matter.\nNevertheless, in spite of vagueness and doubtful cases, the existence\nof cycles in the behaviour of animals is a broad characteristic by which\nthey are prima facie distinguished from ordinary matter; and I think it\nis this characteristic which leads us to attribute desires to animals,\nsince it makes their behaviour resemble what we do when (as we say) we\nare acting from desire.\n\nI shall adopt the following definitions for describing the behaviour of\nanimals:\n\nA \"behaviour-cycle\" is a series of voluntary or reflex movements of an\nanimal, tending to cause a certain result, and continuing until that\nresult is caused, unless they are interrupted by death, accident,\nor some new behaviour-cycle. (Here \"accident\" may be defined as the\nintervention of purely physical laws causing mechanical movements.)\n\nThe \"purpose\" of a behaviour-cycle is the result which brings it to an\nend, normally by a condition of temporary quiescence-provided there is\nno interruption.\n\nAn animal is said to \"desire\" the purpose of a behaviour cycle while the\nbehaviour-cycle is in progress.\n\nI believe these definitions to be adequate also to human purposes and\ndesires, but for the present I am only occupied with animals and with\nwhat can be learnt by external observation. I am very anxious that no\nideas should be attached to the words \"purpose\" and \"desire\" beyond\nthose involved in the above definitions.\n\nWe have not so far considered what is the nature of the initial stimulus\nto a behaviour-cycle. Yet it is here that the usual view of desire seems\non the strongest ground. The hungry animal goes on making movements\nuntil it gets food; it seems natural, therefore, to suppose that the\nidea of food is present throughout the process, and that the thought of\nthe end to be achieved sets the whole process in motion. Such a view,\nhowever, is obviously untenable in many cases, especially where instinct\nis concerned. Take, for example, reproduction and the rearing of the\nyoung. Birds mate, build a nest, lay eggs in it, sit on the eggs, feed\nthe young birds, and care for them until they are fully grown. It\nis totally impossible to suppose that this series of actions, which\nconstitutes one behaviour-cycle, is inspired by any prevision of the\nend, at any rate the first time it is performed.* We must suppose that\nthe stimulus to the performance of each act is an impulsion from behind,\nnot an attraction from the future. The bird does what it does, at each\nstage, because it has an impulse to that particular action, not because\nit perceives that the whole cycle of actions will contribute to the\npreservation of the species. The same considerations apply to other\ninstincts. A hungry animal feels restless, and is led by instinctive\nimpulses to perform the movements which give it nourishment; but the act\nof seeking food is not sufficient evidence from which to conclude that\nthe animal has the thought of food in its \"mind.\"\n\n     * For evidence as to birds' nests, cf. Semon, \"Die Mneme,\"\n     pp. 209, 210.\n\nComing now to human beings, and to what we know about our own actions,\nit seems clear that what, with us, sets a behaviour-cycle in motion is\nsome sensation of the sort which we call disagreeable. Take the case\nof hunger: we have first an uncomfortable feeling inside, producing a\ndisinclination to sit still, a sensitiveness to savoury smells, and an\nattraction towards any food that there may be in our neighbourhood. At\nany moment during this process we may become aware that we are hungry,\nin the sense of saying to ourselves, \"I am hungry\"; but we may have been\nacting with reference to food for some time before this moment. While we\nare talking or reading, we may eat in complete unconsciousness; but we\nperform the actions of eating just as we should if we were conscious,\nand they cease when our hunger is appeased. What we call \"consciousness\"\nseems to be a mere spectator of the process; even when it issues orders,\nthey are usually, like those of a wise parent, just such as would have\nbeen obeyed even if they had not been given. This view may seem at first\nexaggerated, but the more our so-called volitions and their causes are\nexamined, the more it is forced upon us. The part played by words in all\nthis is complicated, and a potent source of confusions; I shall return\nto it later. For the present, I am still concerned with primitive\ndesire, as it exists in man, but in the form in which man shows his\naffinity to his animal ancestors.\n\nConscious desire is made up partly of what is essential to desire,\npartly of beliefs as to what we want. It is important to be clear as to\nthe part which does not consist of beliefs.\n\nThe primitive non-cognitive element in desire seems to be a push, not\na pull, an impulsion away from the actual, rather than an attraction\ntowards the ideal. Certain sensations and other mental occurrences have\na property which we call discomfort; these cause such bodily movements\nas are likely to lead to their cessation. When the discomfort ceases,\nor even when it appreciably diminishes, we have sensations possessing a\nproperty which we call PLEASURE. Pleasurable sensations either stimulate\nno action at all, or at most stimulate such action as is likely to\nprolong them. I shall return shortly to the consideration of what\ndiscomfort and pleasure are in themselves; for the present, it is\ntheir connection with action and desire that concerns us. Abandoning\nmomentarily the standpoint of behaviourism, we may presume that hungry\nanimals experience sensations involving discomfort, and stimulating such\nmovements as seem likely to bring them to the food which is outside the\ncages. When they have reached the food and eaten it, their discomfort\nceases and their sensations become pleasurable. It SEEMS, mistakenly, as\nif the animals had had this situation in mind throughout, when in fact\nthey have been continually pushed by discomfort. And when an animal\nis reflective, like some men, it comes to think that it had the final\nsituation in mind throughout; sometimes it comes to know what situation\nwill bring satisfaction, so that in fact the discomfort does bring the\nthought of what will allay it. Nevertheless the sensation involving\ndiscomfort remains the prime mover.\n\nThis brings us to the question of the nature of discomfort and pleasure.\nSince Kant it has been customary to recognize three great divisions of\nmental phenomena, which are typified by knowledge, desire and feeling,\nwhere \"feeling\" is used to mean pleasure and discomfort. Of course,\n\"knowledge\" is too definite a word: the states of mind concerned are\ngrouped together as \"cognitive,\" and are to embrace not only beliefs,\nbut perceptions, doubts, and the understanding of concepts. \"Desire,\"\nalso, is narrower than what is intended: for example, WILL is to be\nincluded in this category, and in fact every thing that involves any\nkind of striving, or \"conation\" as it is technically called. I do not\nmyself believe that there is any value in this threefold division of the\ncontents of mind. I believe that sensations (including images) supply\nall the \"stuff\" of the mind, and that everything else can be analysed\ninto groups of sensations related in various ways, or characteristics of\nsensations or of groups of sensations. As regards belief, I shall give\ngrounds for this view in later lectures. As regards desires, I have\ngiven some grounds in this lecture. For the present, it is pleasure and\ndiscomfort that concern us. There are broadly three theories that might\nbe held in regard to them. We may regard them as separate existing\nitems in those who experience them, or we may regard them as intrinsic\nqualities of sensations and other mental occurrences, or we may regard\nthem as mere names for the causal characteristics of the occurrences\nwhich are uncomfortable or pleasant. The first of these theories,\nnamely, that which regards discomfort and pleasure as actual contents in\nthose who experience them, has, I think, nothing conclusive to be said\nin its favour.* It is suggested chiefly by an ambiguity in the word\n\"pain,\" which has misled many people, including Berkeley, whom it\nsupplied with one of his arguments for subjective idealism. We may use\n\"pain\" as the opposite of \"pleasure,\" and \"painful\" as the opposite of\n\"pleasant,\" or we may use \"pain\" to mean a certain sort of sensation, on\na level with the sensations of heat and cold and touch. The latter use\nof the word has prevailed in psychological literature, and it is now\nno longer used as the opposite of \"pleasure.\" Dr. H. Head, in a recent\npublication, has stated this distinction as follows:**\n\n     * Various arguments in its favour are advanced by A.\n     Wohlgemuth, \"On the feelings and their neural correlate,\n     with an examination of the nature of pain,\" \"British Journal\n     of Psychology,\" viii, 4. (1917). But as these arguments are\n     largely a reductio ad absurdum of other theories, among\n     which that which I am advocating is not included, I cannot\n     regard them as establishing their contention.\n\n     ** \"Sensation and the Cerebral Cortex,\" \"Brain,\" vol. xli,\n     part ii (September, 1918), p. 90. Cf. also Wohlgemuth, loc.\n     cit. pp. 437, 450.\n\n\"It is necessary at the outset to distinguish clearly between\n'discomfort' and 'pain.' Pain is a distinct sensory quality equivalent\nto heat and cold, and its intensity can be roughly graded according to\nthe force expended in stimulation. Discomfort, on the other hand,\nis that feeling-tone which is directly opposed to pleasure. It may\naccompany sensations not in themselves essentially painful; as for\ninstance that produced by tickling the sole of the foot. The reaction\nproduced by repeated pricking contains both these elements; for it\nevokes that sensory quality known as pain, accompanied by a disagreeable\nfeeling-tone, which we have called discomfort. On the other hand,\nexcessive pressure, except when applied directly over some nerve-trunk,\ntends to excite more discomfort than pain.\"\n\nThe confusion between discomfort and pain has made people regard\ndiscomfort as a more substantial thing than it is, and this in turn has\nreacted upon the view taken of pleasure, since discomfort and pleasure\nare evidently on a level in this respect. As soon as discomfort is\nclearly distinguished from the sensation of pain, it becomes more\nnatural to regard discomfort and pleasure as properties of mental\noccurrences than to regard them as separate mental occurrences on their\nown account. I shall therefore dismiss the view that they are separate\nmental occurrences, and regard them as properties of such experiences as\nwould be called respectively uncomfortable and pleasant.\n\nIt remains to be examined whether they are actual qualities of such\noccurrences, or are merely differences as to causal properties. I do not\nmyself see any way of deciding this question; either view seems equally\ncapable of accounting for the facts. If this is true, it is safer to\navoid the assumption that there are such intrinsic qualities of\nmental occurrences as are in question, and to assume only the causal\ndifferences which are undeniable. Without condemning the intrinsic\ntheory, we can define discomfort and pleasure as consisting in causal\nproperties, and say only what will hold on either of the two theories.\nFollowing this course, we shall say:\n\n\"Discomfort\" is a property of a sensation or other mental occurrence,\nconsisting in the fact that the occurrence in question stimulates\nvoluntary or reflex movements tending to produce some more or less\ndefinite change involving the cessation of the occurrence.\n\n\"Pleasure\" is a property of a sensation or other mental occurrence,\nconsisting in the fact that the occurrence in question either does not\nstimulate any voluntary or reflex movement, or, if it does, stimulates\nonly such as tend to prolong the occurrence in question.*\n\n     * Cf. Thorndike, op. cit., p. 243.\n\n\"Conscious\" desire, which we have now to consider, consists of desire\nin the sense hitherto discussed, together with a true belief as to its\n\"purpose,\" i.e. as to the state of affairs that will bring quiescence\nwith cessation of the discomfort. If our theory of desire is correct,\na belief as to its purpose may very well be erroneous, since only\nexperience can show what causes a discomfort to cease. When the\nexperience needed is common and simple, as in the case of hunger, a\nmistake is not very probable. But in other cases--e.g. erotic desire in\nthose who have had little or no experience of its satisfaction--mistakes\nare to be expected, and do in fact very often occur. The practice of\ninhibiting impulses, which is to a great extent necessary to civilized\nlife, makes mistakes easier, by preventing experience of the actions to\nwhich a desire would otherwise lead, and by often causing the inhibited\nimpulses themselves to be unnoticed or quickly forgotten. The perfectly\nnatural mistakes which thus arise constitute a large proportion of what\nis, mistakenly in part, called self-deception, and attributed by Freud\nto the \"censor.\"\n\nBut there is a further point which needs emphasizing, namely, that a\nbelief that something is desired has often a tendency to cause the very\ndesire that is believed in. It is this fact that makes the effect of\n\"consciousness\" on desire so complicated.\n\nWhen we believe that we desire a certain state of affairs, that often\ntends to cause a real desire for it. This is due partly to the influence\nof words upon our emotions, in rhetoric for example, and partly to the\ngeneral fact that discomfort normally belongs to the belief that we\ndesire such-and-such a thing that we do not possess. Thus what was\noriginally a false opinion as to the object of a desire acquires a\ncertain truth: the false opinion generates a secondary subsidiary\ndesire, which nevertheless becomes real. Let us take an illustration.\nSuppose you have been jilted in a way which wounds your vanity. Your\nnatural impulsive desire will be of the sort expressed in Donne's poem:\n\n     When by thy scorn, O Murderess, I am dead,\n\nin which he explains how he will haunt the poor lady as a ghost, and\nprevent her from enjoying a moment's peace. But two things stand in\nthe way of your expressing yourself so naturally: on the one hand, your\nvanity, which will not acknowledge how hard you are hit; on the other\nhand, your conviction that you are a civilized and humane person,\nwho could not possibly indulge so crude a desire as revenge. You will\ntherefore experience a restlessness which will at first seem quite\naimless, but will finally resolve itself in a conscious desire to change\nyour profession, or go round the world, or conceal your identity and\nlive in Putney, like Arnold Bennett's hero. Although the prime cause of\nthis desire is a false judgment as to your previous unconscious desire,\nyet the new conscious desire has its own derivative genuineness, and may\ninfluence your actions to the extent of sending you round the world.\nThe initial mistake, however, will have effects of two kinds. First,\nin uncontrolled moments, under the influence of sleepiness or drink\nor delirium, you will say things calculated to injure the faithless\ndeceiver. Secondly, you will find travel disappointing, and the East\nless fascinating than you had hoped--unless, some day, you hear that the\nwicked one has in turn been jilted. If this happens, you will believe\nthat you feel sincere sympathy, but you will suddenly be much more\ndelighted than before with the beauties of tropical islands or the\nwonders of Chinese art. A secondary desire, derived from a false\njudgment as to a primary desire, has its own power of influencing\naction, and is therefore a real desire according to our definition.\nBut it has not the same power as a primary desire of bringing thorough\nsatisfaction when it is realized; so long as the primary desire remains\nunsatisfied, restlessness continues in spite of the secondary desire's\nsuccess. Hence arises a belief in the vanity of human wishes: the vain\nwishes are those that are secondary, but mistaken beliefs prevent us\nfrom realizing that they are secondary.\n\nWhat may, with some propriety, be called self-deception arises through\nthe operation of desires for beliefs. We desire many things which it is\nnot in our power to achieve: that we should be universally popular and\nadmired, that our work should be the wonder of the age, and that the\nuniverse should be so ordered as to bring ultimate happiness to all,\nthough not to our enemies until they have repented and been purified\nby suffering. Such desires are too large to be achieved through our own\nefforts. But it is found that a considerable portion of the satisfaction\nwhich these things would bring us if they were realized is to be\nachieved by the much easier operation of believing that they are or\nwill be realized. This desire for beliefs, as opposed to desire for the\nactual facts, is a particular case of secondary desire, and, like all\nsecondary desire its satisfaction does not lead to a complete cessation\nof the initial discomfort. Nevertheless, desire for beliefs, as opposed\nto desire for facts, is exceedingly potent both individually and\nsocially. According to the form of belief desired, it is called vanity,\noptimism, or religion. Those who have sufficient power usually imprison\nor put to death any one who tries to shake their faith in their own\nexcellence or in that of the universe; it is for this reason that\nseditious libel and blasphemy have always been, and still are, criminal\noffences.\n\nIt is very largely through desires for beliefs that the primitive\nnature of desire has become so hidden, and that the part played by\nconsciousness has been so confusing and so exaggerated.\n\nWe may now summarize our analysis of desire and feeling.\n\nA mental occurrence of any kind--sensation, image, belief, or\nemotion--may be a cause of a series of actions, continuing, unless\ninterrupted, until some more or less definite state of affairs is\nrealized. Such a series of actions we call a \"behaviour-cycle.\" The\ndegree of definiteness may vary greatly: hunger requires only food in\ngeneral, whereas the sight of a particular piece of food raises a desire\nwhich requires the eating of that piece of food. The property of causing\nsuch a cycle of occurrences is called \"discomfort\"; the property of the\nmental occurrences in which the cycle ends is called \"pleasure.\" The\nactions constituting the cycle must not be purely mechanical, i.e. they\nmust be bodily movements in whose causation the special properties\nof nervous tissue are involved. The cycle ends in a condition of\nquiescence, or of such action as tends only to preserve the status quo.\nThe state of affairs in which this condition of quiescence is achieved\nis called the \"purpose\" of the cycle, and the initial mental occurrence\ninvolving discomfort is called a \"desire\" for the state of affairs that\nbrings quiescence. A desire is called \"conscious\" when it is accompanied\nby a true belief as to the state of affairs that will bring quiescence;\notherwise it is called \"unconscious.\" All primitive desire is\nunconscious, and in human beings beliefs as to the purposes of desires\nare often mistaken. These mistaken beliefs generate secondary desires,\nwhich cause various interesting complications in the psychology of human\ndesire, without fundamentally altering the character which it shares\nwith animal desire.\n\n\n\n\nLECTURE IV. INFLUENCE OF PAST HISTORY ON PRESENT OCCURRENCES IN LIVING\nORGANISMS\n\nIn this lecture we shall be concerned with a very general characteristic\nwhich broadly, though not absolutely, distinguishes the behaviour\nof living organisms from that of dead matter. The characteristic in\nquestion is this:\n\nThe response of an organism to a given stimulus is very often dependent\nupon the past history of the organism, and not merely upon the stimulus\nand the HITHERTO DISCOVERABLE present state of the organism.\n\nThis characteristic is embodied in the saying \"a burnt child fears the\nfire.\" The burn may have left no visible traces, yet it modifies the\nreaction of the child in the presence of fire. It is customary to assume\nthat, in such cases, the past operates by modifying the structure of the\nbrain, not directly. I have no wish to suggest that this hypothesis is\nfalse; I wish only to point out that it is a hypothesis. At the end of\nthe present lecture I shall examine the grounds in its favour. If we\nconfine ourselves to facts which have been actually observed, we must\nsay that past occurrences, in addition to the present stimulus and\nthe present ascertainable condition of the organism, enter into the\ncausation of the response.\n\nThe characteristic is not wholly confined to living organisms. For\nexample, magnetized steel looks just like steel which has not been\nmagnetized, but its behaviour is in some ways different. In the case\nof dead matter, however, such phenomena are less frequent and important\nthan in the case of living organisms, and it is far less difficult\nto invent satisfactory hypotheses as to the microscopic changes of\nstructure which mediate between the past occurrence and the present\nchanged response. In the case of living organisms, practically\neverything that is distinctive both of their physical and of their\nmental behaviour is bound up with this persistent influence of the past.\nFurther, speaking broadly, the change in response is usually of a kind\nthat is biologically advantageous to the organism.\n\nFollowing a suggestion derived from Semon (\"Die Mneme,\" Leipzig, 1904;\n2nd edition, 1908, English translation, Allen & Unwin, 1921; \"Die\nmnemischen Empfindungen,\" Leipzig, 1909), we will give the name of\n\"mnemic phenomena\" to those responses of an organism which, so far as\nhitherto observed facts are concerned, can only be brought under causal\nlaws by including past occurrences in the history of the organism as\npart of the causes of the present response. I do not mean merely--what\nwould always be the case--that past occurrences are part of a CHAIN of\ncauses leading to the present event. I mean that, in attempting to state\nthe PROXIMATE cause of the present event, some past event or events\nmust be included, unless we take refuge in hypothetical modifications of\nbrain structure. For example: you smell peat-smoke, and you recall some\noccasion when you smelt it before. The cause of your recollection, so\nfar as hitherto observable phenomena are concerned, consists both\nof the peat smoke (present stimulus) and of the former occasion (past\nexperience). The same stimulus will not produce the same recollection\nin another man who did not share your former experience, although the\nformer experience left no OBSERVABLE traces in the structure of the\nbrain. According to the maxim \"same cause, same effect,\" we cannot\ntherefore regard the peat-smoke alone as the cause of your recollection,\nsince it does not have the same effect in other cases. The cause of\nyour recollection must be both the peat-smoke and the past occurrence.\nAccordingly your recollection is an instance of what we are calling\n\"mnemic phenomena.\"\n\nBefore going further, it will be well to give illustrations of different\nclasses of mnemic phenomena.\n\n(a) ACQUIRED HABITS.--In Lecture II we saw how animals can learn by\nexperience how to get out of cages or mazes, or perform other actions\nwhich are useful to them but not provided for by their instincts alone.\nA cat which is put into a cage of which it has had experience behaves\ndifferently from the way in which it behaved at first. We can easily\ninvent hypotheses, which are quite likely to be true, as to connections\nin the brain caused by past experience, and themselves causing the\ndifferent response. But the observable fact is that the stimulus of\nbeing in the cage produces differing results with repetition, and that\nthe ascertainable cause of the cat's behaviour is not merely the cage\nand its own ascertainable organization, but also its past history\nin regard to the cage. From our present point of view, the matter is\nindependent of the question whether the cat's behaviour is due to some\nmental fact called \"knowledge,\" or displays a merely bodily habit. Our\nhabitual knowledge is not always in our minds, but is called up by the\nappropriate stimuli. If we are asked \"What is the capital of France?\"\nwe answer \"Paris,\" because of past experience; the past experience is as\nessential as the present question in the causation of our response. Thus\nall our habitual knowledge consists of acquired habits, and comes under\nthe head of mnemic phenomena.\n\n(b) IMAGES.--I shall have much to say about images in a later lecture;\nfor the present I am merely concerned with them in so far as they are\n\"copies\" of past sensations. When you hear New York spoken of, some\nimage probably comes into your mind, either of the place itself (if you\nhave been there), or of some picture of it (if you have not). The image\nis due to your past experience, as well as to the present stimulus of\nthe words \"New York.\" Similarly, the images you have in dreams are\nall dependent upon your past experience, as well as upon the present\nstimulus to dreaming. It is generally believed that all images, in their\nsimpler parts, are copies of sensations; if so, their mnemic character\nis evident. This is important, not only on its own account, but also\nbecause, as we shall see later, images play an essential part in what is\ncalled \"thinking.\"\n\n(c) ASSOCIATION.--The broad fact of association, on the mental side, is\nthat when we experience something which we have experienced before,\nit tends to call up the context of the former experience. The smell of\npeat-smoke recalling a former scene is an instance which we discussed a\nmoment ago. This is obviously a mnemic phenomenon. There is also a more\npurely physical association, which is indistinguishable from physical\nhabit. This is the kind studied by Mr. Thorndike in animals, where a\ncertain stimulus is associated with a certain act. This is the sort\nwhich is taught to soldiers in drilling, for example. In such a case\nthere need not be anything mental, but merely a habit of the body.\nThere is no essential distinction between association and habit, and the\nobservations which we made concerning habit as a mnemic phenomenon are\nequally applicable to association.\n\n(d) NON-SENSATIONAL ELEMENTS IN PERCEPTION.--When we perceive any object\nof a familiar kind, much of what appears subjectively to be immediately\ngiven is really derived from past experience. When we see an object, say\na penny, we seem to be aware of its \"real\" shape we have the impression\nof something circular, not of something elliptical. In learning to draw,\nit is necessary to acquire the art of representing things according\nto the sensation, not according to the perception. And the visual\nappearance is filled out with feeling of what the object would be like\nto touch, and so on. This filling out and supplying of the \"real\" shape\nand so on consists of the most usual correlates of the sensational core\nin our perception. It may happen that, in the particular case, the real\ncorrelates are unusual; for example, if what we are seeing is a\ncarpet made to look like tiles. If so, the non-sensational part of our\nperception will be illusory, i.e. it will supply qualities which the\nobject in question does not in fact have. But as a rule objects do\nhave the qualities added by perception, which is to be expected,\nsince experience of what is usual is the cause of the addition. If our\nexperience had been different, we should not fill out sensation in\nthe same way, except in so far as the filling out is instinctive,\nnot acquired. It would seem that, in man, all that makes up space\nperception, including the correlation of sight and touch and so on, is\nalmost entirely acquired. In that case there is a large mnemic element\nin all the common perceptions by means of which we handle common\nobjects. And, to take another kind of instance, imagine what our\nastonishment would be if we were to hear a cat bark or a dog mew. This\nemotion would be dependent upon past experience, and would therefore be\na mnemic phenomenon according to the definition.\n\n(e) MEMORY AS KNOWLEDGE.--The kind of memory of which I am now speaking\nis definite knowledge of some past event in one's own experience.\nFrom time to time we remember things that have happened to us, because\nsomething in the present reminds us of them. Exactly the same present\nfact would not call up the same memory if our past experience had been\ndifferent. Thus our remembering is caused by--\n\n(1) The present stimulus,\n\n(2) The past occurrence.\n\nIt is therefore a mnemic phenomenon according to our definition. A\ndefinition of \"mnemic phenomena\" which did not include memory would,\nof course, be a bad one. The point of the definition is not that it\nincludes memory, but that it includes it as one of a class of phenomena\nwhich embrace all that is characteristic in the subject matter of\npsychology.\n\n(f) EXPERIENCE.--The word \"experience\" is often used very vaguely.\nJames, as we saw, uses it to cover the whole primal stuff of the world,\nbut this usage seems objection able, since, in a purely physical world,\nthings would happen without there being any experience. It is only\nmnemic phenomena that embody experience. We may say that an animal\n\"experiences\" an occurrence when this occurrence modifies the animal's\nsubsequent behaviour, i.e. when it is the mnemic portion of the cause of\nfuture occurrences in the animal's life. The burnt child that fears the\nfire has \"experienced\" the fire, whereas a stick that has been thrown on\nand taken off again has not \"experienced\" anything, since it offers\nno more resistance than before to being thrown on. The essence of\n\"experience\" is the modification of behaviour produced by what is\nexperienced. We might, in fact, define one chain of experience, or one\nbiography, as a series of occurrences linked by mnemic causation. I\nthink it is this characteristic, more than any other, that distinguishes\nsciences dealing with living organisms from physics.\n\nThe best writer on mnemic phenomena known to me is Richard Semon, the\nfundamental part of whose theory I shall endeavour to summarize before\ngoing further:\n\nWhen an organism, either animal or plant, is subjected to a stimulus,\nproducing in it some state of excitement, the removal of the stimulus\nallows it to return to a condition of equilibrium. But the new state\nof equilibrium is different from the old, as may be seen by the changed\ncapacity for reaction. The state of equilibrium before the stimulus may\nbe called the \"primary indifference-state\"; that after the cessation\nof the stimulus, the \"secondary indifference-state.\" We define the\n\"engraphic effect\" of a stimulus as the effect in making a difference\nbetween the primary and secondary indifference-states, and this\ndifference itself we define as the \"engram\" due to the stimulus. \"Mnemic\nphenomena\" are defined as those due to engrams; in animals, they are\nspecially associated with the nervous system, but not exclusively, even\nin man.\n\nWhen two stimuli occur together, one of them, occurring afterwards,\nmay call out the reaction for the other also. We call this an \"ekphoric\ninfluence,\" and stimuli having this character are called \"ekphoric\nstimuli.\" In such a case we call the engrams of the two stimuli\n\"associated.\" All simultaneously generated engrams are associated; there\nis also association of successively aroused engrams, though this is\nreducible to simultaneous association. In fact, it is not an isolated\nstimulus that leaves an engram, but the totality of the stimuli at any\nmoment; consequently any portion of this totality tends, if it recurs,\nto arouse the whole reaction which was aroused before. Semon holds that\nengrams can be inherited, and that an animal's innate habits may be due\nto the experience of its ancestors; on this subject he refers to Samuel\nButler.\n\nSemon formulates two \"mnemic principles.\" The first, or \"Law of\nEngraphy,\" is as follows: \"All simultaneous excitements in an organism\nform a connected simultaneous excitement-complex, which as such works\nengraphically, i.e. leaves behind a connected engram-complex, which\nin so far forms a whole\" (\"Die mnemischen Empfindungen,\" p. 146).\nThe second mnemic principle, or \"Law of Ekphory,\" is as follows:\n\"The partial return of the energetic situation which formerly worked\nengraphically operates ekphorically on a simultaneous engram-complex\"\n(ib., p. 173). These two laws together represent in part a hypothesis\n(the engram), and in part an observable fact. The observable fact is\nthat, when a certain complex of stimuli has originally caused a certain\ncomplex of reactions, the recurrence of part of the stimuli tends to\ncause the recurrence of the whole of the reactions.\n\nSemon's applications of his fundamental ideas in various directions are\ninteresting and ingenious. Some of them will concern us later, but for\nthe present it is the fundamental character of mnemic phenomena that is\nin question.\n\nConcerning the nature of an engram, Semon confesses that at present it\nis impossible to say more than that it must consist in some material\nalteration in the body of the organism (\"Die mnemischen Empfindungen,\"\np. 376). It is, in fact, hypothetical, invoked for theoretical uses, and\nnot an outcome of direct observation. No doubt physiology, especially\nthe disturbances of memory through lesions in the brain, affords grounds\nfor this hypothesis; nevertheless it does remain a hypothesis, the\nvalidity of which will be discussed at the end of this lecture.\n\nI am inclined to think that, in the present state of physiology, the\nintroduction of the engram does not serve to simplify the account of\nmnemic phenomena. We can, I think, formulate the known laws of such\nphenomena in terms, wholly, of observable facts, by recognizing\nprovisionally what we may call \"mnemic causation.\" By this I mean that\nkind of causation of which I spoke at the beginning of this lecture,\nthat kind, namely, in which the proximate cause consists not merely of a\npresent event, but of this together with a past event. I do not wish to\nurge that this form of causation is ultimate, but that, in the present\nstate of our knowledge, it affords a simplification, and enables us\nto state laws of behaviour in less hypothetical terms than we should\notherwise have to employ.\n\nThe clearest instance of what I mean is recollection of a past event.\nWhat we observe is that certain present stimuli lead us to recollect\ncertain occurrences, but that at times when we are not recollecting\nthem, there is nothing discoverable in our minds that could be called\nmemory of them. Memories, as mental facts, arise from time to time,\nbut do not, so far as we can see, exist in any shape while they are\n\"latent.\" In fact, when we say that they are \"latent,\" we mean merely\nthat they will exist under certain circumstances. If, then, there is\nto be some standing difference between the person who can remember a\ncertain fact and the person who cannot, that standing difference must\nbe, not in anything mental, but in the brain. It is quite probable that\nthere is such a difference in the brain, but its nature is unknown and\nit remains hypothetical. Everything that has, so far, been made matter\nof observation as regards this question can be put together in the\nstatement: When a certain complex of sensations has occurred to a man,\nthe recurrence of part of the complex tends to arouse the recollection\nof the whole. In like manner, we can collect all mnemic phenomena in\nliving organisms under a single law, which contains what is hitherto\nverifiable in Semon's two laws. This single law is:\n\nIF A COMPLEX STIMULUS A HAS CAUSED A COMPLEX REACTION B IN AN ORGANISM,\nTHE OCCURRENCE OF A PART OF A ON A FUTURE OCCASION TENDS TO CAUSE THE\nWHOLE REACTION B.\n\nThis law would need to be supplemented by some account of the influence\nof frequency, and so on; but it seems to contain the essential\ncharacteristic of mnemic phenomena, without admixture of anything\nhypothetical.\n\nWhenever the effect resulting from a stimulus to an organism differs\naccording to the past history of the organism, without our being able\nactually to detect any relevant difference in its present structure,\nwe will speak of \"mnemic causation,\" provided we can discover laws\nembodying the influence of the past. In ordinary physical causation,\nas it appears to common sense, we have approximate uniformities of\nsequence, such as \"lightning is followed by thunder,\" \"drunkenness\nis followed by headache,\" and so on. None of these sequences are\ntheoretically invariable, since something may intervene to disturb\nthem. In order to obtain invariable physical laws, we have to proceed to\ndifferential equations, showing the direction of change at each moment,\nnot the integral change after a finite interval, however short. But\nfor the purposes of daily life many sequences are to all in tents and\npurposes invariable. With the behaviour of human beings, however, this\nis by no means the case. If you say to an Englishman, \"You have a smut\non your nose,\" he will proceed to remove it, but there will be no such\neffect if you say the same thing to a Frenchman who knows no English.\nThe effect of words upon the hearer is a mnemic phenomena, since it\ndepends upon the past experience which gave him understanding of the\nwords. If there are to be purely psychological causal laws, taking no\naccount of the brain and the rest of the body, they will have to be of\nthe form, not \"X now causes Y now,\" but--\n\n\"A, B, C,... in the past, together with X now, cause Y now.\" For it\ncannot be successfully maintained that our understanding of a word, for\nexample, is an actual existent content of the mind at times when we\nare not thinking of the word. It is merely what may be called a\n\"disposition,\" i.e. it is capable of being aroused whenever we hear the\nword or happen to think of it. A \"disposition\" is not something actual,\nbut merely the mnemic portion of a mnemic causal law.\n\nIn such a law as \"A, B, C,... in the past, together with X now, cause\nY now,\" we will call A, B, C,... the mnemic cause, X the occasion or\nstimulus, and Y the reaction. All cases in which experience influences\nbehaviour are instances of mnemic causation.\n\nBelievers in psycho-physical parallelism hold that psychology can\ntheoretically be freed entirely from all dependence on physiology or\nphysics. That is to say, they believe that every psychical event has\na psychical cause and a physical concomitant. If there is to be\nparallelism, it is easy to prove by mathematical logic that the\ncausation in physical and psychical matters must be of the same sort,\nand it is impossible that mnemic causation should exist in psychology\nbut not in physics. But if psychology is to be independent of\nphysiology, and if physiology can be reduced to physics, it would seem\nthat mnemic causation is essential in psychology. Otherwise we shall be\ncompelled to believe that all our knowledge, all our store of images\nand memories, all our mental habits, are at all times existing in some\nlatent mental form, and are not merely aroused by the stimuli which lead\nto their display. This is a very difficult hypothesis. It seems to me\nthat if, as a matter of method rather than metaphysics, we desire to\nobtain as much independence for psychology as is practically feasible,\nwe shall do better to accept mnemic causation in psychology protem,\nand therefore reject parallelism, since there is no good ground for\nadmitting mnemic causation in physics.\n\nIt is perhaps worth while to observe that mnemic causation is what led\nBergson to deny that there is causation at all in the psychical sphere.\nHe points out, very truly, that the same stimulus, repeated, does not\nhave the same consequences, and he argues that this is contrary to the\nmaxim, \"same cause, same effect.\" It is only necessary, however, to take\naccount of past occurrences and include them with the cause, in order\nto re-establish the maxim, and the possibility of psychological causal\nlaws. The metaphysical conception of a cause lingers in our manner of\nviewing causal laws: we want to be able to FEEL a connection between\ncause and effect, and to be able to imagine the cause as \"operating.\"\nThis makes us unwilling to regard causal laws as MERELY observed\nuniformities of sequence; yet that is all that science has to offer.\nTo ask why such-and-such a kind of sequence occurs is either to ask a\nmeaningless question, or to demand some more general kind of sequence\nwhich includes the one in question. The widest empirical laws of\nsequence known at any time can only be \"explained\" in the sense of being\nsubsumed by later discoveries under wider laws; but these wider laws,\nuntil they in turn are subsumed, will remain brute facts, resting solely\nupon observation, not upon some supposed inherent rationality.\n\nThere is therefore no a priori objection to a causal law in which part\nof the cause has ceased to exist. To argue against such a law on the\nground that what is past cannot operate now, is to introduce the old\nmetaphysical notion of cause, for which science can find no place. The\nonly reason that could be validly alleged against mnemic causation would\nbe that, in fact, all the phenomena can be explained without it. They\nare explained without it by Semon's \"engram,\" or by any theory which\nregards the results of experience as embodied in modifications of\nthe brain and nerves. But they are not explained, unless with extreme\nartificiality, by any theory which regards the latent effects of\nexperience as psychical rather than physical. Those who desire to make\npsychology as far as possible independent of physiology would do well,\nit seems to me, if they adopted mnemic causation. For my part, however,\nI have no such desire, and I shall therefore endeavour to state the\ngrounds which occur to me in favour of some such view as that of the\n\"engram.\"\n\nOne of the first points to be urged is that mnemic phenomena are just\nas much to be found in physiology as in psychology. They are even to\nbe found in plants, as Sir Francis Darwin pointed out (cf. Semon, \"Die\nMneme,\" 2nd edition, p. 28 n.). Habit is a characteristic of the body\nat least as much as of the mind. We should, therefore, be compelled\nto allow the intrusion of mnemic causation, if admitted at all, into\nnon-psychological regions, which ought, one feels, to be subject only to\ncausation of the ordinary physical sort. The fact is that a great deal\nof what, at first sight, distinguishes psychology from physics is found,\non examination, to be common to psychology and physiology; this whole\nquestion of the influence of experience is a case in point. Now it\nis possible, of course, to take the view advocated by Professor J. S.\nHaldane, who contends that physiology is not theoretically reducible to\nphysics and chemistry.* But the weight of opinion among physiologists\nappears to be against him on this point; and we ought certainly to\nrequire very strong evidence before admitting any such breach of\ncontinuity as between living and dead matter. The argument from the\nexistence of mnemic phenomena in physiology must therefore be allowed a\ncertain weight against the hypothesis that mnemic causation is ultimate.\n\n     * See his \"The New Physiology and Other Addresses,\" Griffin,\n     1919, also the symposium, \"Are Physical, Biological and\n     Psychological Categories Irreducible?\" in \"Life and Finite\n     Individuality,\" edited for the Aristotelian Society, with an\n     Introduction. By H. Wildon Carr, Williams & Norgate, 1918.\n\nThe argument from the connection of brain-lesions with loss of memory is\nnot so strong as it looks, though it has also, some weight. What we\nknow is that memory, and mnemic phenomena generally, can be disturbed or\ndestroyed by changes in the brain. This certainly proves that the brain\nplays an essential part in the causation of memory, but does not prove\nthat a certain state of the brain is, by itself, a sufficient condition\nfor the existence of memory. Yet it is this last that has to be proved.\nThe theory of the engram, or any similar theory, has to maintain that,\ngiven a body and brain in a suitable state, a man will have a certain\nmemory, without the need of any further conditions. What is known,\nhowever, is only that he will not have memories if his body and brain\nare not in a suitable state. That is to say, the appropriate state\nof body and brain is proved to be necessary for memory, but not to be\nsufficient. So far, therefore, as our definite knowledge goes, memory\nmay require for its causation a past occurrence as well as a certain\npresent state of the brain.\n\nIn order to prove conclusively that mnemic phenomena arise whenever\ncertain physiological conditions are fulfilled, we ought to be able\nactually to see differences between the brain of a man who speaks\nEnglish and that of a man who speaks French, between the brain of a man\nwho has seen New York and can recall it, and that of a man who has never\nseen that city. It may be that the time will come when this will be\npossible, but at present we are very far removed from it. At present,\nthere is, so far as I am aware, no good evidence that every difference\nbetween the knowledge possessed by A and that possessed by B is\nparalleled by some difference in their brains. We may believe that\nthis is the case, but if we do, our belief is based upon analogies\nand general scientific maxims, not upon any foundation of detailed\nobservation. I am myself inclined, as a working hypothesis, to adopt\nthe belief in question, and to hold that past experience only affects\npresent behaviour through modifications of physiological structure. But\nthe evidence seems not quite conclusive, so that I do not think we ought\nto forget the other hypothesis, or to reject entirely the possibility\nthat mnemic causation may be the ultimate explanation of mnemic\nphenomena. I say this, not because I think it LIKELY that mnemic\ncausation is ultimate, but merely because I think it POSSIBLE, and\nbecause it often turns out important to the progress of science to\nremember hypotheses which have previously seemed improbable.\n\n\n\n\nLECTURE V. PSYCHOLOGICAL AND PHYSICAL CAUSAL LAWS\n\nThe traditional conception of cause and effect is one which modern\nscience shows to be fundamentally erroneous, and requiring to be\nreplaced by a quite different notion, that of LAWS OF CHANGE. In the\ntraditional conception, a particular event A caused a particular event\nB, and by this it was implied that, given any event B, some earlier\nevent A could be discovered which had a relation to it, such that--\n\n(1) Whenever A occurred, it was followed by B;\n\n(2) In this sequence, there was something \"necessary,\" not a mere de\nfacto occurrence of A first and then B.\n\nThe second point is illustrated by the old discussion as to whether\nit can be said that day causes night, on the ground that day is always\nfollowed by night. The orthodox answer was that day could not be called\nthe cause of night, because it would not be followed by night if the\nearth's rotation were to cease, or rather to grow so slow that one\ncomplete rotation would take a year. A cause, it was held, must be such\nthat under no conceivable circumstances could it fail to be followed by\nits effect.\n\nAs a matter of fact, such sequences as were sought by believers in the\ntraditional form of causation have not so far been found in nature.\nEverything in nature is apparently in a state of continuous change,* so\nthat what we call one \"event\" turns out to be really a process. If this\nevent is to cause another event, the two will have to be contiguous in\ntime; for if there is any interval between them, something may happen\nduring that interval to prevent the expected effect. Cause and effect,\ntherefore, will have to be temporally contiguous processes. It is\ndifficult to believe, at any rate where physical laws are concerned,\nthat the earlier part of the process which is the cause can make any\ndifference to the effect, so long as the later part of the process which\nis the cause remains unchanged. Suppose, for example, that a man dies\nof arsenic poisoning, we say that his taking arsenic was the cause\nof death. But clearly the process by which he acquired the arsenic\nis irrelevant: everything that happened before he swallowed it may be\nignored, since it cannot alter the effect except in so far as it alters\nhis condition at the moment of taking the dose. But we may go further:\nswallowing arsenic is not really the proximate cause of death, since a\nman might be shot through the head immediately after taking the dose,\nand then it would not be of arsenic that he would die. The arsenic\nproduces certain physiological changes, which take a finite time before\nthey end in death. The earlier parts of these changes can be ruled out\nin the same way as we can rule out the process by which the arsenic was\nacquired. Proceeding in this way, we can shorten the process which we\nare calling the cause more and more. Similarly we shall have to shorten\nthe effect. It may happen that immediately after the man's death his\nbody is blown to pieces by a bomb. We cannot say what will happen after\nthe man's death, through merely knowing that he has died as the result\nof arsenic poisoning. Thus, if we are to take the cause as one event and\nthe effect as another, both must be shortened indefinitely. The result\nis that we merely have, as the embodiment of our causal law, a certain\ndirection of change at each moment. Hence we are brought to differential\nequations as embodying causal laws. A physical law does not say \"A will\nbe followed by B,\" but tells us what acceleration a particle will have\nunder given circumstances, i.e. it tells us how the particle's motion is\nchanging at each moment, not where the particle will be at some future\nmoment.\n\n     * The theory of quanta suggests that the continuity is only\n     apparent. If so, we shall be able theoretically to reach\n     events which are not processes. But in what is directly\n     observable there is still apparent continuity, which\n     justifies the above remarks for the prevent.\n\nLaws embodied in differential equations may possibly be exact,\nbut cannot be known to be so. All that we can know empirically is\napproximate and liable to exceptions; the exact laws that are assumed in\nphysics are known to be somewhere near the truth, but are not known to\nbe true just as they stand. The laws that we actually know empirically\nhave the form of the traditional causal laws, except that they are not\nto be regarded as universal or necessary. \"Taking arsenic is followed by\ndeath\" is a good empirical generalization; it may have exceptions, but\nthey will be rare. As against the professedly exact laws of physics,\nsuch empirical generalizations have the advantage that they deal with\nobservable phenomena. We cannot observe infinitesimals, whether in time\nor space; we do not even know whether time and space are infinitely\ndivisible. Therefore rough empirical generalizations have a definite\nplace in science, in spite of not being exact of universal. They are the\ndata for more exact laws, and the grounds for believing that they are\nUSUALLY true are stronger than the grounds for believing that the more\nexact laws are ALWAYS true.\n\nScience starts, therefore, from generalizations of the form, \"A is\nusually followed by B.\" This is the nearest approach that can be made\nto a causal law of the traditional sort. It may happen in any particular\ninstance that A is ALWAYS followed by B, but we cannot know this, since\nwe cannot foresee all the perfectly possible circumstances that might\nmake the sequence fail, or know that none of them will actually occur.\nIf, however, we know of a very large number of cases in which A is\nfollowed by B, and few or none in which the sequence fails, we shall in\nPRACTICE be justified in saying \"A causes B,\" provided we do not attach\nto the notion of cause any of the metaphysical superstitions that have\ngathered about the word.\n\nThere is another point, besides lack of universality and necessity,\nwhich it is important to realize as regards causes in the above sense,\nand that is the lack of uniqueness. It is generally assumed that, given\nany event, there is some one phenomenon which is THE cause of the event\nin question. This seems to be a mere mistake. Cause, in the only\nsense in which it can be practically applied, means \"nearly invariable\nantecedent.\" We cannot in practice obtain an antecedent which is QUITE\ninvariable, for this would require us to take account of the whole\nuniverse, since something not taken account of may prevent the expected\neffect. We cannot distinguish, among nearly invariable antecedents, one\nas THE cause, and the others as merely its concomitants: the attempt to\ndo this depends upon a notion of cause which is derived from will, and\nwill (as we shall see later) is not at all the sort of thing that it is\ngenerally supposed to be, nor is there any reason to think that in the\nphysical world there is anything even remotely analogous to what will is\nsupposed to be. If we could find one antecedent, and only one, that was\nQUITE invariable, we could call that one THE cause without introducing\nany notion derived from mistaken ideas about will. But in fact we cannot\nfind any antecedent that we know to be quite invariable, and we can find\nmany that are nearly so. For example, men leave a factory for dinner\nwhen the hooter sounds at twelve o'clock. You may say the hooter is\nTHE cause of their leaving. But innumerable other hooters in other\nfactories, which also always sound at twelve o'clock, have just as\ngood a right to be called the cause. Thus every event has many nearly\ninvariable antecedents, and therefore many antecedents which may be\ncalled its cause.\n\nThe laws of traditional physics, in the form in which they deal with\nmovements of matter or electricity, have an apparent simplicity which\nsomewhat conceals the empirical character of what they assert. A piece\nof matter, as it is known empirically, is not a single existing thing,\nbut a system of existing things. When several people simultaneously see\nthe same table, they all see something different; therefore \"the\" table,\nwhich they are supposed all to see, must be either a hypothesis or\na construction. \"The\" table is to be neutral as between different\nobservers: it does not favour the aspect seen by one man at the expense\nof that seen by another. It was natural, though to my mind mistaken, to\nregard the \"real\" table as the common cause of all the appearances which\nthe table presents (as we say) to different observers. But why should we\nsuppose that there is some one common cause of all these appearances? As\nwe have just seen, the notion of \"cause\" is not so reliable as to allow\nus to infer the existence of something that, by its very nature, can\nnever be observed.\n\nInstead of looking for an impartial source, we can secure neutrality by\nthe equal representation of all parties. Instead of supposing that there\nis some unknown cause, the \"real\" table, behind the different sensations\nof those who are said to be looking at the table, we may take the\nwhole set of these sensations (together possibly with certain other\nparticulars) as actually BEING the table. That is to say, the table\nwhich is neutral as between different observers (actual and possible)\nis the set of all those particulars which would naturally be called\n\"aspects\" of the table from different points of view. (This is a first\napproximation, modified later.)\n\nIt may be said: If there is no single existent which is the source of\nall these \"aspects,\" how are they collected together? The answer is\nsimple: Just as they would be if there were such a single existent. The\nsupposed \"real\" table underlying its appearances is, in any case, not\nitself perceived, but inferred, and the question whether such-and-such\na particular is an \"aspect\" of this table is only to be settled by\nthe connection of the particular in question with the one or more\nparticulars by which the table is defined. That is to say, even if we\nassume a \"real\" table, the particulars which are its aspects have to be\ncollected together by their relations to each other, not to it, since\nit is merely inferred from them. We have only, therefore, to notice how\nthey are collected together, and we can then keep the collection\nwithout assuming any \"real\" table as distinct from the collection. When\ndifferent people see what they call the same table, they see things\nwhich are not exactly the same, owing to difference of point of view,\nbut which are sufficiently alike to be described in the same words, so\nlong as no great accuracy or minuteness is sought. These closely similar\nparticulars are collected together by their similarity primarily\nand, more correctly, by the fact that they are related to each other\napproximately according to the laws of perspective and of reflection and\ndiffraction of light. I suggest, as a first approximation, that these\nparticulars, together with such correlated others as are unperceived,\njointly ARE the table; and that a similar definition applies to all\nphysical objects.*\n\n     *See \"Our Knowledge of the External World\" (Allen & Unwin),\n     chaps. iii and iv.\n\nIn order to eliminate the reference to our perceptions, which introduces\nan irrelevant psychological suggestion, I will take a different\nillustration, namely, stellar photography. A photographic plate exposed\non a clear night reproduces the appearance of the portion of the sky\nconcerned, with more or fewer stars according to the power of the\ntelescope that is being used. Each separate star which is photographed\nproduces its separate effect on the plate, just as it would upon\nourselves if we were looking at the sky. If we assume, as science\nnormally does, the continuity of physical processes, we are forced\nto conclude that, at the place where the plate is, and at all places\nbetween it and a star which it photographs, SOMETHING is happening which\nis specially connected with that star. In the days when the aether was\nless in doubt, we should have said that what was happening was a certain\nkind of transverse vibration in the aether. But it is not necessary\nor desirable to be so explicit: all that we need say is that SOMETHING\nhappens which is specially connected with the star in question. It\nmust be something specially connected with that star, since that star\nproduces its own special effect upon the plate. Whatever it is must be\nthe end of a process which starts from the star and radiates outwards,\npartly on general grounds of continuity, partly to account for the fact\nthat light is transmitted with a certain definite velocity. We thus\narrive at the conclusion that, if a certain star is visible at a certain\nplace, or could be photographed by a sufficiently sensitive plate at\nthat place, something is happening there which is specially connected\nwith that star. Therefore in every place at all times a vast multitude\nof things must be happening, namely, at least one for every physical\nobject which can be seen or photographed from that place. We can\nclassify such happenings on either of two principles:\n\n(1) We can collect together all the happenings in one place, as is done\nby photography so far as light is concerned;\n\n(2) We can collect together all the happenings, in different places,\nwhich are connected in the way that common sense regards as being due to\ntheir emanating from one object.\n\nThus, to return to the stars, we can collect together either--\n\n(1) All the appearances of different stars in a given place, or,\n\n(2) All the appearances of a given star in different places.\n\nBut when I speak of \"appearances,\" I do so only for brevity: I do not\nmean anything that must \"appear\" to somebody, but only that happening,\nwhatever it may be, which is connected, at the place in question, with a\ngiven physical object--according to the old orthodox theory, it would be\na transverse vibration in the aether. Like the different appearances\nof the table to a number of simultaneous observers, the different\nparticulars that belong to one physical object are to be collected\ntogether by continuity and inherent laws of correlation, not by their\nsupposed causal connection with an unknown assumed existent called a\npiece of matter, which would be a mere unnecessary metaphysical thing in\nitself. A piece of matter, according to the definition that I propose,\nis, as a first approximation,* the collection of all those correlated\nparticulars which would normally be regarded as its appearances or\neffects in different places. Some further elaborations are desirable,\nbut we can ignore them for the present. I shall return to them at the\nend of this lecture.\n\n     *The exact definition of a piece of matter as a construction\n     will be given later.\n\nAccording to the view that I am suggesting, a physical object or piece\nof matter is the collection of all those correlated particulars which\nwould be regarded by common sense as its effects or appearances in\ndifferent places. On the other hand, all the happenings in a given place\nrepresent what common sense would regard as the appearances of a number\nof different objects as viewed from that place. All the happenings in\none place may be regarded as the view of the world from that place. I\nshall call the view of the world from a given place a \"perspective.\" A\nphotograph represents a perspective. On the other hand, if photographs\nof the stars were taken in all points throughout space, and in all such\nphotographs a certain star, say Sirius, were picked out whenever it\nappeared, all the different appearances of Sirius, taken together,\nwould represent Sirius. For the understanding of the difference between\npsychology and physics it is vital to understand these two ways of\nclassifying particulars, namely:\n\n(1) According to the place where they occur;\n\n(2) According to the system of correlated particulars in different\nplaces to which they belong, such system being defined as a physical\nobject.\n\nGiven a system of particulars which is a physical object, I shall\ndefine that one of the system which is in a given place (if any) as the\n\"appearance of that object in that place.\"\n\nWhen the appearance of an object in a given place changes, it is found\nthat one or other of two things occurs. The two possibilities may be\nillustrated by an example. You are in a room with a man, whom you see:\nyou may cease to see him either by shutting your eyes or by his going\nout of the room. In the first case, his appearance to other people\nremains unchanged; in the second, his appearance changes from all\nplaces. In the first case, you say that it is not he who has changed,\nbut your eyes; in the second, you say that he has changed. Generalizing,\nwe distinguish--\n\n(1) Cases in which only certain appearances of the object change, while\nothers, and especially appearances from places very near to the object,\ndo not change;\n\n(2) Cases where all, or almost all, the appearances of the object\nundergo a connected change.\n\nIn the first case, the change is attributed to the medium between the\nobject and the place; in the second, it is attributed to the object\nitself.*\n\n     * The application of this distinction to motion raises\n     complications due to relativity, but we may ignore these for\n     our present purposes.\n\nIt is the frequency of the latter kind of change, and the comparatively\nsimple nature of the laws governing the simultaneous alterations\nof appearances in such cases, that have made it possible to treat a\nphysical object as one thing, and to overlook the fact that it is a\nsystem of particulars. When a number of people at a theatre watch an\nactor, the changes in their several perspectives are so similar and so\nclosely correlated that all are popularly regarded as identical with\neach other and with the changes of the actor himself. So long as all\nthe changes in the appearances of a body are thus correlated there is no\npressing prima facie need to break up the system of appearances, or to\nrealize that the body in question is not really one thing but a set of\ncorrelated particulars. It is especially and primarily such changes that\nphysics deals with, i.e. it deals primarily with processes in which\nthe unity of a physical object need not be broken up because all its\nappearances change simultaneously according to the same law--or, if not\nall, at any rate all from places sufficiently near to the object, with\nin creasing accuracy as we approach the object.\n\nThe changes in appearances of an object which are due to changes in the\nintervening medium will not affect, or will affect only very slightly,\nthe appearances from places close to the object. If the appearances\nfrom sufficiently neighbouring places are either wholly un changed,\nor changed to a diminishing extent which has zero for its limit, it\nis usually found that the changes can be accounted for by changes in\nobjects which are between the object in question and the places from\nwhich its appearance has changed appreciably. Thus physics is able\nto reduce the laws of most changes with which it deals to changes in\nphysical objects, and to state most of its fundamental laws in terms of\nmatter. It is only in those cases in which the unity of the system of\nappearances constituting a piece of matter has to be broken up, that the\nstatement of what is happening cannot be made exclusively in terms of\nmatter. The whole of psychology, we shall find, is included among such\ncases; hence their importance for our purposes.\n\nWe can now begin to understand one of the fundamental differences\nbetween physics and psychology. Physics treats as a unit the whole\nsystem of appearances of a piece of matter, whereas psychology is\ninterested in certain of these appearances themselves. Confining\nourselves for the moment to the psychology of perceptions, we observe\nthat perceptions are certain of the appearances of physical objects.\nFrom the point of view that we have been hitherto adopting, we\nmight define them as the appearances of objects at places from which\nsense-organs and the suitable parts of the nervous system form part\nof the intervening medium. Just as a photographic plate receives a\ndifferent impression of a cluster of stars when a telescope is part of\nthe intervening medium, so a brain receives a different impression\nwhen an eye and an optic nerve are part of the intervening medium.\nAn impression due to this sort of intervening medium is called a\nperception, and is interesting to psychology on its own account, not\nmerely as one of the set of correlated particulars which is the physical\nobject of which (as we say) we are having a perception.\n\nWe spoke earlier of two ways of classifying particulars. One way\ncollects together the appearances commonly regarded as a given object\nfrom different places; this is, broadly speaking, the way of physics,\nleading to the construction of physical objects as sets of such\nappearances. The other way collects together the appearances of\ndifferent objects from a given place, the result being what we call a\nperspective. In the particular case where the place concerned is a\nhuman brain, the perspective belonging to the place consists of all the\nperceptions of a certain man at a given time. Thus classification by\nperspectives is relevant to psychology, and is essential in defining\nwhat we mean by one mind.\n\nI do not wish to suggest that the way in which I have been defining\nperceptions is the only possible way, or even the best way. It is the\nway that arose naturally out of our present topic. But when we approach\npsychology from a more introspective standpoint, we have to distinguish\nsensations and perceptions, if possible, from other mental occurrences,\nif any. We have also to consider the psychological effects of\nsensations, as opposed to their physical causes and correlates. These\nproblems are quite distinct from those with which we have been concerned\nin the present lecture, and I shall not deal with them until a later\nstage.\n\nIt is clear that psychology is concerned essentially with actual\nparticulars, not merely with systems of particulars. In this it differs\nfrom physics, which, broadly speaking, is concerned with the cases\nin which all the particulars which make up one physical object can be\ntreated as a single causal unit, or rather the particulars which are\nsufficiently near to the object of which they are appearances can be so\ntreated. The laws which physics seeks can, broadly speaking, be stated\nby treating such systems of particulars as causal units. The laws which\npsychology seeks cannot be so stated, since the particulars themselves\nare what interests the psychologist. This is one of the fundamental\ndifferences between physics and psychology; and to make it clear has\nbeen the main purpose of this lecture.\n\nI will conclude with an attempt to give a more precise definition of\na piece of matter. The appearances of a piece of matter from different\nplaces change partly according to intrinsic laws (the laws of\nperspective, in the case of visual shape), partly according to the\nnature of the intervening medium--fog, blue spectacles, telescopes,\nmicroscopes, sense-organs, etc. As we approach nearer to the object, the\neffect of the intervening medium grows less. In a generalized sense,\nall the intrinsic laws of change of appearance may be called \"laws\nof perspective.\" Given any appearance of an object, we can construct\nhypothetically a certain system of appearances to which the appearance\nin question would belong if the laws of perspective alone were\nconcerned. If we construct this hypothetical system for each appearance\nof the object in turn, the system corresponding to a given appearance\nx will be independent of any distortion due to the medium beyond x, and\nwill only embody such distortion as is due to the medium between x and\nthe object. Thus, as the appearance by which our hypothetical system\nis defined is moved nearer and nearer to the object, the hypothetical\nsystem of appearances defined by its means embodies less and less of the\neffect of the medium. The different sets of appearances resulting from\nmoving x nearer and nearer to the object will approach to a limiting\nset, and this limiting set will be that system of appearances which the\nobject would present if the laws of perspective alone were operative\nand the medium exercised no distorting effect. This limiting set of\nappearances may be defined, for purposes of physics, as the piece of\nmatter concerned.\n\n\n\n\nLECTURE VI. INTROSPECTION\n\nOne of the main purposes of these lectures is to give grounds for\nthe belief that the distinction between mind and matter is not so\nfundamental as is commonly supposed. In the preceding lecture I dealt in\noutline with the physical side of this problem. I attempted to show\nthat what we call a material object is not itself a substance, but is\na system of particulars analogous in their nature to sensations, and in\nfact often including actual sensations among their number. In this\nway the stuff of which physical objects are composed is brought into\nrelation with the stuff of which part, at least, of our mental life is\ncomposed.\n\nThere is, however, a converse task which is equally necessary for our\nthesis, and that is, to show that the stuff of our mental life is devoid\nof many qualities which it is commonly supposed to have, and is not\npossessed of any attributes which make it incapable of forming part of\nthe world of matter. In the present lecture I shall begin the arguments\nfor this view.\n\nCorresponding to the supposed duality of matter and mind, there are, in\northodox psychology, two ways of knowing what exists. One of these, the\nway of sensation and external perception, is supposed to furnish data\nfor our knowledge of matter, the other, called \"introspection,\" is\nsupposed to furnish data for knowledge of our mental processes. To\ncommon sense, this distinction seems clear and easy. When you see a\nfriend coming along the street, you acquire knowledge of an external,\nphysical fact; when you realize that you are glad to meet him, you\nacquire knowledge of a mental fact. Your dreams and memories and\nthoughts, of which you are often conscious, are mental facts, and the\nprocess by which you become aware of them SEEMS to be different from\nsensation. Kant calls it the \"inner sense\"; sometimes it is spoken of\nas \"consciousness of self\"; but its commonest name in modern English\npsychology is \"introspection.\" It is this supposed method of acquiring\nknowledge of our mental processes that I wish to analyse and examine in\nthis lecture.\n\nI will state at the outset the view which I shall aim at establishing.\nI believe that the stuff of our mental life, as opposed to its relations\nand structure, consists wholly of sensations and images. Sensations are\nconnected with matter in the way that I tried to explain in Lecture V,\ni.e. each is a member of a system which is a certain physical object.\nImages, though they USUALLY have certain characteristics, especially\nlack of vividness, that distinguish them from sensations, are not\nINVARIABLY so distinguished, and cannot therefore be defined by these\ncharacteristics. Images, as opposed to sensations, can only be defined\nby their different causation: they are caused by association with a\nsensation, not by a stimulus external to the nervous system--or perhaps\none should say external to the brain, where the higher animals are\nconcerned. The occurrence of a sensation or image does not in itself\nconstitute knowledge but any sensation or image may come to be known\nif the conditions are suitable. When a sensation--like the hearing of a\nclap of thunder--is normally correlated with closely similar sensations\nin our neighbours, we regard it as giving knowledge of the external\nworld, since we regard the whole set of similar sensations as due to\na common external cause. But images and bodily sensations are not so\ncorrelated. Bodily sensations can be brought into a correlation by\nphysiology, and thus take their place ultimately among sources of\nknowledge of the physical world. But images cannot be made to fit in\nwith the simultaneous sensations and images of others. Apart from their\nhypothetical causes in the brain, they have a causal connection\nwith physical objects, through the fact that they are copies of past\nsensations; but the physical objects with which they are thus connected\nare in the past, not in the present. These images remain private in\na sense in which sensations are not. A sensation SEEMS to give us\nknowledge of a present physical object, while an image does not, except\nwhen it amounts to a hallucination, and in this case the seeming is\ndeceptive. Thus the whole context of the two occurrences is different.\nBut in themselves they do not differ profoundly, and there is no reason\nto invoke two different ways of knowing for the one and for the other.\nConsequently introspection as a separate kind of knowledge disappears.\n\nThe criticism of introspection has been in the main the work of American\npsychologists. I will begin by summarizing an article which seems to me\nto afford a good specimen of their arguments, namely, \"The Case against\nIntrospection,\" by Knight Dunlap (\"Psychological Review,\" vol xix, No.\n5, pp. 404-413, September, 1912). After a few historical quotations,\nhe comes to two modern defenders of introspection, Stout and James. He\nquotes from Stout such statements as the following: \"Psychical states as\nsuch become objects only when we attend to them in an introspective way.\nOtherwise they are not themselves objects, but only constituents of the\nprocess by which objects are recognized\" (\"Manual,\" 2nd edition, p. 134.\nThe word \"recognized\" in Dunlap's quotation should be \"cognized.\") \"The\nobject itself can never be identified with the present modification of\nthe individual's consciousness by which it is cognized\" (ib. p. 60).\nThis is to be true even when we are thinking about modifications of\nour own consciousness; such modifications are to be always at least\npartially distinct from the conscious experience in which we think of\nthem.\n\nAt this point I wish to interrupt the account of Knight Dunlap's article\nin order to make some observations on my own account with reference to\nthe above quotations from Stout. In the first place, the conception of\n\"psychical states\" seems to me one which demands analysis of a somewhat\ndestructive character. This analysis I shall give in later lectures as\nregards cognition; I have already given it as regards desire. In the\nsecond place, the conception of \"objects\" depends upon a certain view\nas to cognition which I believe to be wholly mistaken, namely, the view\nwhich I discussed in my first lecture in connection with Brentano.\nIn this view a single cognitive occurrence contains both content and\nobject, the content being essentially mental, while the object is\nphysical except in introspection and abstract thought. I have already\ncriticized this view, and will not dwell upon it now, beyond saying\nthat \"the process by which objects are cognized\" appears to be a very\nslippery phrase. When we \"see a table,\" as common sense would say, the\ntable as a physical object is not the \"object\" (in the psychological\nsense) of our perception. Our perception is made up of sensations,\nimages and beliefs, but the supposed \"object\" is something inferential,\nexternally related, not logically bound up with what is occurring in us.\nThis question of the nature of the object also affects the view we take\nof self-consciousness. Obviously, a \"conscious experience\" is different\nfrom a physical object; therefore it is natural to assume that a thought\nor perception whose object is a conscious experience must be different\nfrom a thought or perception whose object is a physical object. But if\nthe relation to the object is inferential and external, as I maintain,\nthe difference between two thoughts may bear very little relation to\nthe difference between their objects. And to speak of \"the present\nmodification of the individual's consciousness by which an object is\ncognized\" is to suggest that the cognition of objects is a far more\ndirect process, far more intimately bound up with the objects, than I\nbelieve it to be. All these points will be amplified when we come to the\nanalysis of knowledge, but it is necessary briefly to state them now in\norder to suggest the atmosphere in which our analysis of \"introspection\"\nis to be carried on.\n\nAnother point in which Stout's remarks seem to me to suggest what I\nregard as mistakes is his use of \"consciousness.\" There is a view which\nis prevalent among psychologists, to the effect that one can speak of \"a\nconscious experience\" in a curious dual sense, meaning, on the one hand,\nan experience which is conscious of something, and, on the other hand,\nan experience which has some intrinsic nature characteristic of what\nis called \"consciousness.\" That is to say, a \"conscious experience\" is\ncharacterized on the one hand by relation to its object and on the\nother hand by being composed of a certain peculiar stuff, the stuff of\n\"consciousness.\" And in many authors there is yet a third confusion: a\n\"conscious experience,\" in this third sense, is an experience of\nwhich we are conscious. All these, it seems to me, need to be clearly\nseparated. To say that one occurrence is \"conscious\" of another is, to\nmy mind, to assert an external and rather remote relation between them.\nI might illustrate it by the relation of uncle and nephew a man becomes\nan uncle through no effort of his own, merely through an occurrence\nelsewhere. Similarly, when you are said to be \"conscious\" of a table,\nthe question whether this is really the case cannot be decided by\nexamining only your state of mind: it is necessary also to ascertain\nwhether your sensation is having those correlates which past experience\ncauses you to assume, or whether the table happens, in this case, to be\na mirage. And, as I explained in my first lecture, I do not believe that\nthere is any \"stuff\" of consciousness, so that there is no intrinsic\ncharacter by which a \"conscious\" experience could be distinguished from\nany other.\n\nAfter these preliminaries, we can return to Knight Dunlap's article.\nHis criticism of Stout turns on the difficulty of giving any empirical\nmeaning to such notions as the \"mind\" or the \"subject\"; he quotes from\nStout the sentence: \"The most important drawback is that the mind, in\nwatching its own workings, must necessarily have its attention divided\nbetween two objects,\" and he concludes: \"Without question, Stout is\nbringing in here illicitly the concept of a single observer, and his\nintrospection does not provide for the observation of this observer;\nfor the process observed and the observer are distinct\" (p. 407).\nThe objections to any theory which brings in the single observer were\nconsidered in Lecture I, and were acknowledged to be cogent. In so\nfar, therefore, as Stout's theory of introspection rests upon this\nassumption, we are compelled to reject it. But it is perfectly possible\nto believe in introspection without supposing that there is a single\nobserver.\n\nWilliam James's theory of introspection, which Dunlap next examines,\ndoes not assume a single observer. It changed after the publication\nof his \"Psychology,\" in consequence of his abandoning the dualism of\nthought and things. Dunlap summarizes his theory as follows:\n\n\"The essential points in James's scheme of consciousness are SUBJECT,\nOBJECT, and a KNOWING of the object by the subject. The difference\nbetween James's scheme and other schemes involving the same terms is\nthat James considers subject and object to be the same thing, but at\ndifferent times In order to satisfy this requirement James supposes a\nrealm of existence which he at first called 'states of consciousness' or\n'thoughts,' and later, 'pure experience,' the latter term including both\nthe 'thoughts' and the 'knowing.' This scheme, with all its magnificent\nartificiality, James held on to until the end, simply dropping the\nterm consciousness and the dualism between the thought and an external\nreality\"(p. 409).\n\nHe adds: \"All that James's system really amounts to is the\nacknowledgment that a succession of things are known, and that they are\nknown by something. This is all any one can claim, except for the\nfact that the things are known together, and that the knower for the\ndifferent items is one and the same\" (ib.).\n\nIn this statement, to my mind, Dunlap concedes far more than James did\nin his later theory. I see no reason to suppose that \"the knower for\ndifferent items is one and the same,\" and I am convinced that this\nproposition could not possibly be ascertained except by introspection of\nthe sort that Dunlap rejects. The first of these points must wait until\nwe come to the analysis of belief: the second must be considered now.\nDunlap's view is that there is a dualism of subject and object, but that\nthe subject can never become object, and therefore there is no awareness\nof an awareness. He says in discussing the view that introspection\nreveals the occurrence of knowledge: \"There can be no denial of the\nexistence of the thing (knowing) which is alleged to be known or\nobserved in this sort of 'introspection.' The allegation that the\nknowing is observed is that which may be denied. Knowing there certainly\nis; known, the knowing certainly is not\"(p. 410). And again: \"I am\nnever aware of an awareness\" (ib.). And on the next page: \"It may sound\nparadoxical to say that one cannot observe the process (or relation) of\nobservation, and yet may be certain that there is such a process: but\nthere is really no inconsistency in the saying. How do I know that there\nis awareness? By being aware of something. There is no meaning in the\nterm 'awareness' which is not expressed in the statement 'I am aware of\na colour (or what-not).'\"\n\nBut the paradox cannot be so lightly disposed of. The statement \"I am\naware of a colour\" is assumed by Knight Dunlap to be known to be true,\nbut he does not explain how it comes to be known. The argument against\nhim is not conclusive, since he may be able to show some valid way of\ninferring our awareness. But he does not suggest any such way. There is\nnothing odd in the hypothesis of beings which are aware of objects, but\nnot of their own awareness; it is, indeed, highly probable that young\nchildren and the higher animals are such beings. But such beings cannot\nmake the statement \"I am aware of a colour,\" which WE can make. We have,\ntherefore, some knowledge which they lack. It is necessary to Knight\nDunlap's position to maintain that this additional knowledge is purely\ninferential, but he makes no attempt to show how the inference is\npossible. It may, of course, be possible, but I cannot see how. To my\nmind the fact (which he admits) that we know there is awareness, is ALL\nBUT decisive against his theory, and in favour of the view that we can\nbe aware of an awareness.\n\nDunlap asserts (to return to James) that the real ground for James's\noriginal belief in introspection was his belief in two sorts of\nobjects, namely, thoughts and things. He suggests that it was a\nmere inconsistency on James's part to adhere to introspection after\nabandoning the dualism of thoughts and things. I do not wholly agree\nwith this view, but it is difficult to disentangle the difference as to\nintrospection from the difference as to the nature of knowing. Dunlap\nsuggests (p. 411) that what is called introspection really consists of\nawareness of \"images,\" visceral sensations, and so on. This view, in\nessence, seems to me sound. But then I hold that knowing itself consists\nof such constituents suitably related, and that in being aware of them\nwe are sometimes being aware of instances of knowing. For this reason,\nmuch as I agree with his view as to what are the objects of which\nthere is awareness, I cannot wholly agree with his conclusion as to the\nimpossibility of introspection.\n\nThe behaviourists have challenged introspection even more vigorously\nthan Knight Dunlap, and have gone so far as to deny the existence of\nimages. But I think that they have confused various things which\nare very commonly confused, and that it is necessary to make several\ndistinctions before we can arrive at what is true and what false in the\ncriticism of introspection.\n\nI wish to distinguish three distinct questions, any one of which may be\nmeant when we ask whether introspection is a source of knowledge. The\nthree questions are as follows:\n\n(1) Can we observe anything about ourselves which we cannot observe\nabout other people, or is everything we can observe PUBLIC, in the sense\nthat another could also observe it if suitably placed?\n\n(2) Does everything that we can observe obey the laws of physics and\nform part of the physical world, or can we observe certain things that\nlie outside physics?\n\n(3) Can we observe anything which differs in its intrinsic nature from\nthe constituents of the physical world, or is everything that we can\nobserve composed of elements intrinsically similar to the constituents\nof what is called matter?\n\nAny one of these three questions may be used to define introspection. I\nshould favour introspection in the sense of the first question, i.e. I\nthink that some of the things we observe cannot, even theoretically, be\nobserved by any one else. The second question, tentatively and for\nthe present, I should answer in favour of introspection; I think that\nimages, in the actual condition of science, cannot be brought under the\ncausal laws of physics, though perhaps ultimately they may be. The\nthird question I should answer adversely to introspection I think that\nobservation shows us nothing that is not composed of sensations and\nimages, and that images differ from sensations in their causal laws, not\nintrinsically. I shall deal with the three questions successively.\n\n(1) PUBLICITY OR PRIVACY OF WHAT IS OBSERVED. Confining ourselves, for\nthe moment, to sensations, we find that there are different degrees\nof publicity attaching to different sorts of sensations. If you feel a\ntoothache when the other people in the room do not, you are in no way\nsurprised; but if you hear a clap of thunder when they do not, you begin\nto be alarmed as to your mental condition. Sight and hearing are the\nmost public of the senses; smell only a trifle less so; touch, again, a\ntrifle less, since two people can only touch the same spot successively,\nnot simultaneously. Taste has a sort of semi-publicity, since people\nseem to experience similar taste-sensations when they eat similar foods;\nbut the publicity is incomplete, since two people cannot eat actually\nthe same piece of food.\n\nBut when we pass on to bodily sensations--headache, toothache, hunger,\nthirst, the feeling of fatigue, and so on--we get quite away from\npublicity, into a region where other people can tell us what they feel,\nbut we cannot directly observe their feeling. As a natural result of\nthis state of affairs, it has come to be thought that the public senses\ngive us knowledge of the outer world, while the private senses only give\nus knowledge as to our own bodies. As regards privacy, all images, of\nwhatever sort, belong with the sensations which only give knowledge of\nour own bodies, i.e. each is only observable by one observer. This is\nthe reason why images of sight and hearing are more obviously different\nfrom sensations of sight and hearing than images of bodily sensations\nare from bodily sensations; and that is why the argument in favour of\nimages is more conclusive in such cases as sight and hearing than in\nsuch cases as inner speech.\n\nThe whole distinction of privacy and publicity, however, so long as we\nconfine ourselves to sensations, is one of degree, not of kind. No\ntwo people, there is good empirical reason to think, ever have exactly\nsimilar sensations related to the same physical object at the same\nmoment; on the other hand, even the most private sensation has\ncorrelations which would theoretically enable another observer to infer\nit.\n\nThat no sensation is ever completely public, results from differences of\npoint of view. Two people looking at the same table do not get the same\nsensation, because of perspective and the way the light falls. They get\nonly correlated sensations. Two people listening to the same sound do\nnot hear exactly the same thing, because one is nearer to the source of\nthe sound than the other, one has better hearing than the other, and\nso on. Thus publicity in sensations consists, not in having PRECISELY\nsimilar sensations, but in having more or less similar sensations\ncorrelated according to ascertainable laws. The sensations which strike\nus as public are those where the correlated sensations are very similar\nand the correlations are very easy to discover. But even the most\nprivate sensations have correlations with things that others can\nobserve. The dentist does not observe your ache, but he can see the\ncavity which causes it, and could guess that you are suffering even\nif you did not tell him. This fact, however, cannot be used, as Watson\nwould apparently wish, to extrude from science observations which are\nprivate to one observer, since it is by means of many such observations\nthat correlations are established, e.g. between toothaches and cavities.\nPrivacy, therefore does not by itself make a datum unamenable to\nscientific treatment. On this point, the argument against introspection\nmust be rejected.\n\n(2) DOES EVERYTHING OBSERVABLE OBEY THE LAWS OF PHYSICS? We come now to\nthe second ground of objection to introspection, namely, that its data\ndo not obey the laws of physics. This, though less emphasized, is,\nI think, an objection which is really more strongly felt than the\nobjection of privacy. And we obtain a definition of introspection more\nin harmony with usage if we define it as observation of data not subject\nto physical laws than if we define it by means of privacy. No one would\nregard a man as introspective because he was conscious of having a\nstomach ache. Opponents of introspection do not mean to deny the obvious\nfact that we can observe bodily sensations which others cannot observe.\nFor example, Knight Dunlap contends that images are really muscular\ncontractions,* and evidently regards our awareness of muscular\ncontractions as not coming under the head of introspection. I think it\nwill be found that the essential characteristic of introspective data,\nin the sense which now concerns us, has to do with LOCALIZATION: either\nthey are not localized at all, or they are localized, like visual\nimages, in a place already physically occupied by something which would\nbe inconsistent with them if they were regarded as part of the physical\nworld. If you have a visual image of your friend sitting in a chair\nwhich in fact is empty, you cannot locate the image in your body,\nbecause it is visual, nor (as a physical phenomenon) in the chair,\nbecause the chair, as a physical object, is empty. Thus it seems to\nfollow that the physical world does not include all that we are aware\nof, and that images, which are introspective data, have to be regarded,\nfor the present, as not obeying the laws of physics; this is, I think,\none of the chief reasons why an attempt is made to reject them. I\nshall try to show in Lecture VIII that the purely empirical reasons for\naccepting images are overwhelming. But we cannot be nearly so certain\nthat they will not ultimately be brought under the laws of physics. Even\nif this should happen, however, they would still be distinguishable\nfrom sensations by their proximate causal laws, as gases remain\ndistinguishable from solids.\n\n     * \"Psychological Review,\" 1916, \"Thought-Content and\n     Feeling,\" p. 59. See also ib., 1912, \"The Nature of\n     Perceived Relations,\" where he says: \"'Introspection,'\n     divested of its mythological suggestion of the observing of\n     consciousness, is really the observation of bodily\n     sensations (sensibles) and feelings (feelables)\"(p. 427 n.).\n\n(3) CAN WE OBSERVE ANYTHING INTRINSICALLY DIFFERENT FROM SENSATIONS? We\ncome now to our third question concerning introspection. It is commonly\nthought that by looking within we can observe all sorts of things that\nare radically different from the constituents of the physical world,\ne.g. thoughts, beliefs, desires, pleasures, pains and emotions. The\ndifference between mind and matter is increased partly by emphasizing\nthese supposed introspective data, partly by the supposition that matter\nis composed of atoms or electrons or whatever units physics may at the\nmoment prefer. As against this latter supposition, I contend that\nthe ultimate constituents of matter are not atoms or electrons, but\nsensations, and other things similar to sensations as regards extent and\nduration. As against the view that introspection reveals a mental world\nradically different from sensations, I propose to argue that thoughts,\nbeliefs, desires, pleasures, pains and emotions are all built up out\nof sensations and images alone, and that there is reason to think that\nimages do not differ from sensations in their intrinsic character. We\nthus effect a mutual rapprochement of mind and matter, and reduce the\nultimate data of introspection (in our second sense) to images alone. On\nthis third view of the meaning of introspection, therefore, our decision\nis wholly against it.\n\nThere remain two points to be considered concerning introspection. The\nfirst is as to how far it is trustworthy; the second is as to whether,\neven granting that it reveals no radically different STUFF from that\nrevealed by what might be called external perception, it may not reveal\ndifferent RELATIONS, and thus acquire almost as much importance as is\ntraditionally assigned to it.\n\nTo begin with the trustworthiness of introspection. It is common among\ncertain schools to regard the knowledge of our own mental processes as\nincomparably more certain than our knowledge of the \"external\" world;\nthis view is to be found in the British philosophy which descends from\nHume, and is present, somewhat veiled, in Kant and his followers.\nThere seems no reason whatever to accept this view. Our spontaneous,\nunsophisticated beliefs, whether as to ourselves or as to the outer\nworld, are always extremely rash and very liable to error. The\nacquisition of caution is equally necessary and equally difficult in\nboth directions. Not only are we often un aware of entertaining a\nbelief or desire which exists in us; we are often actually mistaken. The\nfallibility of introspection as regards what we desire is made evident\nby psycho-analysis; its fallibility as to what we know is easily\ndemonstrated. An autobiography, when confronted by a careful editor\nwith documentary evidence, is usually found to be full of obviously\ninadvertent errors. Any of us confronted by a forgotten letter written\nsome years ago will be astonished to find how much more foolish our\nopinions were than we had remembered them as being. And as to the\nanalysis of our mental operations--believing, desiring, willing, or what\nnot--introspection unaided gives very little help: it is necessary to\nconstruct hypotheses and test them by their consequences, just as we do\nin physical science. Introspection, therefore, though it is one among\nour sources of knowledge, is not, in isolation, in any degree more\ntrustworthy than \"external\" perception.\n\nI come now to our second question: Does introspection give us materials\nfor the knowledge of relations other than those arrived at by reflecting\nupon external perception? It might be contended that the essence of what\nis \"mental\" consists of relations, such as knowing for example, and that\nour knowledge concerning these essentially mental relations is entirely\nderived from introspection. If \"knowing\" were an unanalysable relation,\nthis view would be incontrovertible, since clearly no such relation\nforms part of the subject matter of physics. But it would seem that\n\"knowing\" is really various relations, all of them complex. Therefore,\nuntil they have been analysed, our present question must remain\nunanswered I shall return to it at the end of the present course of\nlectures.\n\n\n\n\nLECTURE VII. THE DEFINITION OF PERCEPTION\n\nIn Lecture V we found reason to think that the ultimate constituents*\nof the world do not have the characteristics of either mind or matter\nas ordinarily understood: they are not solid persistent objects moving\nthrough space, nor are they fragments of \"consciousness.\" But we found\ntwo ways of grouping particulars, one into \"things\" or \"pieces of\nmatter,\" the other into series of \"perspectives,\" each series being what\nmay be called a \"biography.\" Before we can define either sensations\nor images, it is necessary to consider this twofold classification\nin somewhat greater detail, and to derive from it a definition of\nperception. It should be said that, in so far as the classification\nassumes the whole world of physics (including its unperceived portions),\nit contains hypothetical elements. But we will not linger on the grounds\nfor admitting these, which belong to the philosophy of physics rather\nthan of psychology.\n\n     * When I speak of \"ultimate constituents,\" I do not mean\n     necessarily such as are theoretically incapable of analysis,\n     but only such as, at present, we can see no means of\n     analysing. I speak of such constituents as \"particulars,\" or\n     as \"RELATIVE particulars\" when I wish to emphasize the fact\n     that they may be themselves complex.\n\nThe physical classification of particulars collects together all those\nthat are aspects of one \"thing.\" Given any one particular, it is\nfound often (we do not say always) that there are a number of other\nparticulars differing from this one in gradually increasing degrees.\nThose (or some of those) that differ from it only very slightly will\nbe found to differ approximately according to certain laws which may be\ncalled, in a generalized sense, the laws of \"perspective\"; they include\nthe ordinary laws of perspective as a special case. This approximation\ngrows more and more nearly exact as the difference grows less; in\ntechnical language, the laws of perspective account for the differences\nto the first order of small quantities, and other laws are only\nrequired to account for second-order differences. That is to say, as the\ndifference diminishes, the part of the difference which is not according\nto the laws of perspective diminishes much more rapidly, and bears to\nthe total difference a ratio which tends towards zero as both are made\nsmaller and smaller. By this means we can theoretically collect together\na number of particulars which may be defined as the \"aspects\" or\n\"appearances\" of one thing at one time. If the laws of perspective were\nsufficiently known, the connection between different aspects would be\nexpressed in differential equations.\n\nThis gives us, so far, only those particulars which constitute one thing\nat one time. This set of particulars may be called a \"momentary\nthing.\" To define that series of \"momentary things\" that constitute\nthe successive states of one thing is a problem involving the laws of\ndynamics. These give the laws governing the changes of aspects from\none time to a slightly later time, with the same sort of differential\napproximation to exactness as we obtained for spatially neighbouring\naspects through the laws of perspective. Thus a momentary thing is a set\nof particulars, while a thing (which may be identified with the whole\nhistory of the thing) is a series of such sets of particulars.\nThe particulars in one set are collected together by the laws of\nperspective; the successive sets are collected together by the laws\nof dynamics. This is the view of the world which is appropriate to\ntraditional physics.\n\nThe definition of a \"momentary thing\" involves problems concerning time,\nsince the particulars constituting a momentary thing will not be all\nsimultaneous, but will travel outward from the thing with the velocity\nof light (in case the thing is in vacuo). There are complications\nconnected with relativity, but for our present purpose they are not\nvital, and I shall ignore them.\n\nInstead of first collecting together all the particulars constituting\na momentary thing, and then forming the series of successive sets,\nwe might have first collected together a series of successive aspects\nrelated by the laws of dynamics, and then have formed the set of such\nseries related by the laws of perspective. To illustrate by the case of\nan actor on the stage: our first plan was to collect together all the\naspects which he presents to different spectators at one time, and then\nto form the series of such sets. Our second plan is first to collect\ntogether all the aspects which he presents successively to a given\nspectator, and then to do the same thing for the other spectators, thus\nforming a set of series instead of a series of sets. The first plan\ntells us what he does; the second the impressions he produces. This\nsecond way of classifying particulars is one which obviously has more\nrelevance to psychology than the other. It is partly by this second\nmethod of classification that we obtain definitions of one \"experience\"\nor \"biography\" or \"person.\" This method of classification is also\nessential to the definition of sensations and images, as I shall\nendeavour to prove later on. But we must first amplify the definition of\nperspectives and biographies.\n\nIn our illustration of the actor, we spoke, for the moment, as though\neach spectator's mind were wholly occupied by the one actor. If this\nwere the case, it might be possible to define the biography of one\nspectator as a series of successive aspects of the actor related\naccording to the laws of dynamics. But in fact this is not the case.\nWe are at all times during our waking life receiving a variety of\nimpressions, which are aspects of a variety of things. We have to\nconsider what binds together two simultaneous sensations in one\nperson, or, more generally, any two occurrences which forte part of one\nexperience. We might say, adhering to the standpoint of physics, that\ntwo aspects of different things belong to the same perspective when\nthey are in the same place. But this would not really help us, since a\n\"place\" has not yet been defined. Can we define what is meant by saying\nthat two aspects are \"in the same place,\" without introducing anything\nbeyond the laws of perspective and dynamics?\n\nI do not feel sure whether it is possible to frame such a definition or\nnot; accordingly I shall not assume that it is possible, but shall\nseek other characteristics by which a perspective or biography may be\ndefined.\n\nWhen (for example) we see one man and hear another speaking at the\nsame time, what we see and what we hear have a relation which we\ncan perceive, which makes the two together form, in some sense, one\nexperience. It is when this relation exists that two occurrences become\nassociated. Semon's \"engram\" is formed by all that we experience at one\ntime. He speaks of two parts of this total as having the relation of\n\"Nebeneinander\" (M. 118; M.E. 33 ff.), which is reminiscent of Herbart's\n\"Zusammen.\" I think the relation may be called simply \"simultaneity.\" It\nmight be said that at any moment all sorts of things that are not part\nof my experience are happening in the world, and that therefore the\nrelation we are seeking to define cannot be merely simultaneity.\nThis, however, would be an error--the sort of error that the theory\nof relativity avoids. There is not one universal time, except by an\nelaborate construction; there are only local times, each of which may\nbe taken to be the time within one biography. Accordingly, if I am (say)\nhearing a sound, the only occurrences that are, in any simple sense,\nsimultaneous with my sensation are events in my private world, i.e. in\nmy biography. We may therefore define the \"perspective\" to which\nthe sensation in question belongs as the set of particulars that are\nsimultaneous with this sensation. And similarly we may define the\n\"biography\" to which the sensation belongs as the set of particulars\nthat are earlier or later than, or simultaneous with, the given\nsensation. Moreover, the very same definitions can be applied to\nparticulars which are not sensations. They are actually required for the\ntheory of relativity, if we are to give a philosophical explanation\nof what is meant by \"local time\" in that theory The relations of\nsimultaneity and succession are known to us in our own experience;\nthey may be analysable, but that does not affect their suitability for\ndefining perspectives and biographies. Such time-relations as can be\nconstructed between events in different biographies are of a different\nkind: they are not experienced, and are merely logical, being designed\nto afford convenient ways of stating the correlations between different\nbiographies.\n\nIt is not only by time-relations that the parts of one biography are\ncollected together in the case of living beings. In this case there are\nthe mnemic phenomena which constitute the unity of one \"experience,\" and\ntransform mere occurrences into \"experiences.\" I have already dwelt upon\nthe importance of mnemic phenomena for psychology, and shall not\nenlarge upon them now, beyond observing that they are what transforms a\nbiography (in our technical sense) into a life. It is they that give the\ncontinuity of a \"person\" or a \"mind.\" But there is no reason to suppose\nthat mnemic phenomena are associated with biographies except in the case\nof animals and plants.\n\nOur two-fold classification of particulars gives rise to the dualism of\nbody and biography in regard to everything in the universe, and not only\nin regard to living things. This arises as follows. Every particular of\nthe sort considered by physics is a member of two groups (1) The group\nof particulars constituting the other aspects of the same physical\nobject; (2) The group of particulars that have direct time-relations to\nthe given particular.\n\nEach of these is associated with a place. When I look at a star, my\nsensation is (1) A member of the group of particulars which is the star,\nand which is associated with the place where the star is; (2) A\nmember of the group of particulars which is my biography, and which is\nassociated with the place where I am.*\n\n     *I have explained elsewhere the manner in which space is\n     constructed on this theory, and in which the position of a\n     perspective is brought into relation with the position of a\n     physical object (\"Our Knowledge of the External World,\"\n     Lecture III, pp. 90, 91).\n\nThe result is that every particular of the kind relevant to physics is\nassociated with TWO places; e.g. my sensation of the star is associated\nwith the place where I am and with the place where the star is. This\ndualism has nothing to do with any \"mind\" that I may be supposed to\npossess; it exists in exactly the same sense if I am replaced by a\nphotographic plate. We may call the two places the active and passive\nplaces respectively.* Thus in the case of a perception or photograph\nof a star, the active place is the place where the star is, while the\npassive place is the place where the percipient or photographic plate\nis.\n\n     * I use these as mere names; I do not want to introduce any\n     notion of \"activity.\"\n\nWe can thus, without departing from physics, collect together all the\nparticulars actively at a given place, or all the particulars passively\nat a given place. In our own case, the one group is our body (or\nour brain), while the other is our mind, in so far as it consists of\nperceptions. In the case of the photographic plate, the first group is\nthe plate as dealt with by physics, the second the aspect of the heavens\nwhich it photographs. (For the sake of schematic simplicity, I am\nignoring various complications connected with time, which require some\ntedious but perfectly feasible elaborations.) Thus what may be called\nsubjectivity in the point of view is not a distinctive peculiarity of\nmind: it is present just as much in the photographic plate. And the\nphotographic plate has its biography as well as its \"matter.\" But\nthis biography is an affair of physics, and has none of the peculiar\ncharacteristics by which \"mental\" phenomena are distinguished, with the\nsole exception of subjectivity.\n\nAdhering, for the moment, to the standpoint of physics, we may define a\n\"perception\" of an object as the appearance of the object from a place\nwhere there is a brain (or, in lower animals, some suitable nervous\nstructure), with sense-organs and nerves forming part of the intervening\nmedium. Such appearances of objects are distinguished from appearances\nin other places by certain peculiarities, namely:\n\n(1) They give rise to mnemic phenomena;\n\n(2) They are themselves affected by mnemic phenomena.\n\nThat is to say, they may be remembered and associated or influence our\nhabits, or give rise to images, etc., and they are themselves different\nfrom what they would have been if our past experience had been\ndifferent--for example, the effect of a spoken sentence upon the hearer\ndepends upon whether the hearer knows the language or not, which is\na question of past experience. It is these two characteristics, both\nconnected with mnemic phenomena, that distinguish perceptions from the\nappearances of objects in places where there is no living being.\n\nTheoretically, though often not practically, we can, in our perception\nof an object, separate the part which is due to past experience from the\npart which proceeds without mnemic influences out of the character of\nthe object. We may define as \"sensation\" that part which proceeds in\nthis way, while the remainder, which is a mnemic phenomenon, will have\nto be added to the sensation to make up what is called the \"perception.\"\nAccording to this definition, the sensation is a theoretical core in\nthe actual experience; the actual experience is the perception. It\nis obvious that there are grave difficulties in carrying out these\ndefinitions, but we will not linger over them. We have to pass, as soon\nas we can, from the physical standpoint, which we have been hitherto\nadopting, to the standpoint of psychology, in which we make more use\nof introspection in the first of the three senses discussed in the\npreceding lecture.\n\nBut before making the transition, there are two points which must be\nmade clear. First: Everything outside my own personal biography is\noutside my experience; therefore if anything can be known by me outside\nmy biography, it can only be known in one of two ways:\n\n(1) By inference from things within my biography, or\n\n(2) By some a priori principle independent of experience.\n\nI do not myself believe that anything approaching certainty is to be\nattained by either of these methods, and therefore whatever lies outside\nmy personal biography must be regarded, theoretically, as hypothesis.\nThe theoretical argument for adopting the hypothesis is that it\nsimplifies the statement of the laws according to which events happen\nin our experience. But there is no very good ground for supposing that\na simple law is more likely to be true than a complicated law, though\nthere is good ground for assuming a simple law in scientific practice,\nas a working hypothesis, if it explains the facts as well as another\nwhich is less simple. Belief in the existence of things outside my own\nbiography exists antecedently to evidence, and can only be destroyed, if\nat all, by a long course of philosophic doubt. For purposes of science,\nit is justified practically by the simplification which it introduces\ninto the laws of physics. But from the standpoint of theoretical logic\nit must be regarded as a prejudice, not as a well-grounded theory. With\nthis proviso, I propose to continue yielding to the prejudice.\n\nThe second point concerns the relating of our point of view to that\nwhich regards sensations as caused by stimuli external to the nervous\nsystem (or at least to the brain), and distinguishes images as\n\"centrally excited,\" i.e. due to causes in the brain which cannot be\ntraced back to anything affecting the sense-organs. It is clear that,\nif our analysis of physical objects has been valid, this way of defining\nsensations needs reinterpretation. It is also clear that we must be able\nto find such a new interpretation if our theory is to be admissible.\n\nTo make the matter clear, we will take the simplest possible\nillustration. Consider a certain star, and suppose for the moment\nthat its size is negligible. That is to say, we will regard it as, for\npractical purposes, a luminous point. Let us further suppose that it\nexists only for a very brief time, say a second. Then, according to\nphysics, what happens is that a spherical wave of light travels outward\nfrom the star through space, just as, when you drop a stone into a\nstagnant pond, ripples travel outward from the place where the stone hit\nthe water. The wave of light travels with a certain very nearly constant\nvelocity, roughly 300,000 kilometres per second. This velocity may be\nascertained by sending a flash of light to a mirror, and observing\nhow long it takes before the reflected flash reaches you, just as the\nvelocity of sound may be ascertained by means of an echo.\n\nWhat it is that happens when a wave of light reaches a given place we\ncannot tell, except in the sole case when the place in question is a\nbrain connected with an eye which is turned in the right direction. In\nthis one very special case we know what happens: we have the sensation\ncalled \"seeing the star.\" In all other cases, though we know (more or\nless hypothetically) some of the correlations and abstract properties\nof the appearance of the star, we do not know the appearance itself. Now\nyou may, for the sake of illustration, compare the different appearances\nof the star to the conjugation of a Greek verb, except that the number\nof its parts is really infinite, and not only apparently so to the\ndespairing schoolboy. In vacuo, the parts are regular, and can be\nderived from the (imaginary) root according to the laws of grammar,\ni.e. of perspective. The star being situated in empty space, it may be\ndefined, for purposes of physics, as consisting of all those appearances\nwhich it presents in vacuo, together with those which, according to\nthe laws of perspective, it would present elsewhere if its appearances\nelsewhere were regular. This is merely the adaptation of the definition\nof matter which I gave in an earlier lecture. The appearance of a star\nat a certain place, if it is regular, does not require any cause or\nexplanation beyond the existence of the star. Every regular appearance\nis an actual member of the system which is the star, and its causation\nis entirely internal to that system. We may express this by saying that\na regular appearance is due to the star alone, and is actually part of\nthe star, in the sense in which a man is part of the human race.\n\nBut presently the light of the star reaches our atmosphere. It begins\nto be refracted, and dimmed by mist, and its velocity is slightly\ndiminished. At last it reaches a human eye, where a complicated process\ntakes place, ending in a sensation which gives us our grounds for\nbelieving in all that has gone before. Now, the irregular appearances of\nthe star are not, strictly speaking, members of the system which is the\nstar, according to our definition of matter. The irregular appearances,\nhowever, are not merely irregular: they proceed according to laws which\ncan be stated in terms of the matter through which the light has\npassed on its way. The sources of an irregular appearance are therefore\ntwofold:\n\n(1) The object which is appearing irregularly;\n\n2) The intervening medium.\n\nIt should be observed that, while the conception of a regular appearance\nis perfectly precise, the conception of an irregular appearance is one\ncapable of any degree of vagueness. When the distorting influence of the\nmedium is sufficiently great, the resulting particular can no longer be\nregarded as an appearance of an object, but must be treated on its own\naccount. This happens especially when the particular in question cannot\nbe traced back to one object, but is a blend of two or more. This case\nis normal in perception: we see as one what the microscope or telescope\nreveals to be many different objects. The notion of perception is\ntherefore not a precise one: we perceive things more or less, but always\nwith a very considerable amount of vagueness and confusion.\n\nIn considering irregular appearances, there are certain very natural\nmistakes which must be avoided. In order that a particular may count as\nan irregular appearance of a certain object, it is not necessary that\nit should bear any resemblance to the regular appearances as regard\nits intrinsic qualities. All that is necessary is that it should be\nderivable from the regular appearances by the laws which express\nthe distorting influence of the medium. When it is so derivable,\nthe particular in question may be regarded as caused by the regular\nappearances, and therefore by the object itself, together with the\nmodifications resulting from the medium. In other cases, the particular\nin question may, in the same sense, be regarded as caused by several\nobjects together with the medium; in this case, it may be called a\nconfused appearance of several objects. If it happens to be in a brain,\nit may be called a confused perception of these objects. All actual\nperception is confused to a greater or less extent.\n\nWe can now interpret in terms of our theory the distinction between\nthose mental occurrences which are said to have an external stimulus,\nand those which are said to be \"centrally excited,\" i.e. to have no\nstimulus external to the brain. When a mental occurrence can be regarded\nas an appearance of an object external to the brain, however irregular,\nor even as a confused appearance of several such objects, then we may\nregard it as having for its stimulus the object or objects in question,\nor their appearances at the sense-organ concerned. When, on the other\nhand, a mental occurrence has not sufficient connection with objects\nexternal to the brain to be regarded as an appearance of such objects,\nthen its physical causation (if any) will have to be sought in the\nbrain. In the former case it can be called a perception; in the latter\nit cannot be so called. But the distinction is one of degree, not of\nkind. Until this is realized, no satisfactory theory of perception,\nsensation, or imagination is possible.\n\n\n\n\nLECTURE VIII. SENSATIONS AND IMAGES\n\nThe dualism of mind and matter, if we have been right so far, cannot be\nallowed as metaphysically valid. Nevertheless, we seem to find a certain\ndualism, perhaps not ultimate, within the world as we observe it. The\ndualism is not primarily as to the stuff of the world, but as to causal\nlaws. On this subject we may again quote William James. He points out\nthat when, as we say, we merely \"imagine\" things, there are no such\neffects as would ensue if the things were what we call \"real.\" He takes\nthe case of imagining a fire.\n\n\"I make for myself an experience of blazing fire; I place it near my\nbody; but it does not warm me in the least. I lay a stick upon it and\nthe stick either burns or remains green, as I please. I call up water,\nand pour it on the fire, and absolutely no difference ensues. I account\nfor all such facts by calling this whole train of experiences unreal, a\nmental train. Mental fire is what won't burn real sticks; mental water\nis what won't necessarily (though of course it may) put out even a\nmental fire.... With 'real' objects, on the contrary, consequences\nalways accrue; and thus the real experiences get sifted from the mental\nones, the things from our thoughts of them, fanciful or true, and\nprecipitated together as the stable part of the whole experience--chaos,\nunder the name of the physical world.\"*\n\n     * \"Essays in Radical Empiricism,\" pp. 32-3.\n\nIn this passage James speaks, by mere inadvertence, as though the\nphenomena which he is describing as \"mental\" had NO effects. This is, of\ncourse, not the case: they have their effects, just as much as physical\nphenomena do, but their effects follow different laws. For example,\ndreams, as Freud has shown, are just as much subject to laws as are the\nmotions of the planets. But the laws are different: in a dream you may\nbe transported from one place to another in a moment, or one person\nmay turn into another under your eyes. Such differences compel you to\ndistinguish the world of dreams from the physical world.\n\nIf the two sorts of causal laws could be sharply distinguished, we could\ncall an occurrence \"physical\" when it obeys causal laws appropriate to\nthe physical world, and \"mental\" when it obeys causal laws appropriate\nto the mental world. Since the mental world and the physical world\ninteract, there would be a boundary between the two: there would be\nevents which would have physical causes and mental effects, while there\nwould be others which would have mental causes and physical effects.\nThose that have physical causes and mental effects we should define as\n\"sensations.\" Those that have mental causes and physical effects might\nperhaps be identified with what we call voluntary movements; but they do\nnot concern us at present.\n\nThese definitions would have all the precision that could be desired if\nthe distinction between physical and psychological causation were clear\nand sharp. As a matter of fact, however, this distinction is, as yet, by\nno means sharp. It is possible that, with fuller knowledge, it will be\nfound to be no more ultimate than the distinction between the laws of\ngases and the laws of rigid bodies. It also suffers from the fact that\nan event may be an effect of several causes according to several causal\nlaws we cannot, in general, point to anything unique as THE cause of\nsuch-and-such an event. And finally it is by no means certain that\nthe peculiar causal laws which govern mental events are not really\nphysiological. The law of habit, which is one of the most distinctive,\nmay be fully explicable in terms of the peculiarities of nervous tissue,\nand these peculiarities, in turn, may be explicable by the laws of\nphysics. It seems, therefore, that we are driven to a different kind of\ndefinition. It is for this reason that it was necessary to develop\nthe definition of perception. With this definition, we can define a\nsensation as the non-mnemic elements in a perception.\n\nWhen, following our definition, we try to decide what elements in our\nexperience are of the nature of sensations, we find more difficulty\nthan might have been expected. Prima facie, everything is sensation that\ncomes to us through the senses: the sights we see, the sounds we hear,\nthe smells we smell, and so on; also such things as headache or the\nfeeling of muscular strain. But in actual fact so much interpretation,\nso much of habitual correlation, is mixed with all such experiences,\nthat the core of pure sensation is only to be extracted by careful\ninvestigation. To take a simple illustration: if you go to the theatre\nin your own country, you seem to hear equally well in the stalls or the\ndress circle; in either case you think you miss nothing. But if you go\nin a foreign country where you have a fair knowledge of the language,\nyou will seem to have grown partially deaf, and you will find it\nnecessary to be much nearer the stage than you would need to be in your\nown country. The reason is that, in hearing our own language spoken, we\nquickly and unconsciously fill out what we really hear with inferences\nto what the man must be saying, and we never realize that we have not\nheard the words we have merely inferred. In a foreign language, these\ninferences are more difficult, and we are more dependent upon actual\nsensation. If we found ourselves in a foreign world, where tables looked\nlike cushions and cushions like tables, we should similarly discover how\nmuch of what we think we see is really inference. Every fairly familiar\nsensation is to us a sign of the things that usually go with it, and\nmany of these things will seem to form part of the sensation. I remember\nin the early days of motor-cars being with a friend when a tyre burst\nwith a loud report. He thought it was a pistol, and supported his\nopinion by maintaining that he had seen the flash. But of course there\nhad been no flash. Nowadays no one sees a flash when a tyre bursts.\n\nIn order, therefore, to arrive at what really is sensation in an\noccurrence which, at first sight, seems to contain nothing else, we have\nto pare away all that is due to habit or expectation or interpretation.\nThis is a matter for the psychologist, and by no means an easy matter.\nFor our purposes, it is not important to determine what exactly is the\nsensational core in any case; it is only important to notice that\nthere certainly is a sensational core, since habit, expectation and\ninterpretation are diversely aroused on diverse occasions, and the\ndiversity is clearly due to differences in what is presented to\nthe senses. When you open your newspaper in the morning, the actual\nsensations of seeing the print form a very minute part of what goes\non in you, but they are the starting-point of all the rest, and it\nis through them that the newspaper is a means of information or\nmis-information. Thus, although it may be difficult to determine what\nexactly is sensation in any given experience, it is clear that there is\nsensation, unless, like Leibniz, we deny all action of the outer world\nupon us.\n\nSensations are obviously the source of our knowledge of the world,\nincluding our own body. It might seem natural to regard a sensation as\nitself a cognition, and until lately I did so regard it. When, say, I\nsee a person I know coming towards me in the street, it SEEMS as\nthough the mere seeing were knowledge. It is of course undeniable that\nknowledge comes THROUGH the seeing, but I think it is a mistake to\nregard the mere seeing itself as knowledge. If we are so to regard it,\nwe must distinguish the seeing from what is seen: we must say that, when\nwe see a patch of colour of a certain shape, the patch of colour is one\nthing and our seeing of it is another. This view, however, demands the\nadmission of the subject, or act, in the sense discussed in our first\nlecture. If there is a subject, it can have a relation to the patch of\ncolour, namely, the sort of relation which we might call awareness. In\nthat case the sensation, as a mental event, will consist of awareness of\nthe colour, while the colour itself will remain wholly physical, and\nmay be called the sense-datum, to distinguish it from the sensation.\nThe subject, however, appears to be a logical fiction, like mathematical\npoints and instants. It is introduced, not because observation reveals\nit, but because it is linguistically convenient and apparently demanded\nby grammar. Nominal entities of this sort may or may not exist, but\nthere is no good ground for assuming that they do. The functions that\nthey appear to perform can always be performed by classes or series or\nother logical constructions, consisting of less dubious entities. If we\nare to avoid a perfectly gratuitous assumption, we must dispense with\nthe subject as one of the actual ingredients of the world. But when\nwe do this, the possibility of distinguishing the sensation from\nthe sense-datum vanishes; at least I see no way of preserving the\ndistinction. Accordingly the sensation that we have when we see a patch\nof colour simply is that patch of colour, an actual constituent of the\nphysical world, and part of what physics is concerned with. A patch of\ncolour is certainly not knowledge, and therefore we cannot say that pure\nsensation is cognitive. Through its psychological effects, it is the\ncause of cognitions, partly by being itself a sign of things that\nare correlated with it, as e.g. sensations of sight and touch are\ncorrelated, and partly by giving rise to images and memories after the\nsensation is faded. But in itself the pure sensation is not cognitive.\n\nIn the first lecture we considered the view of Brentano, that \"we may\ndefine psychical phenomena by saying that they are phenomena which\nintentionally contain an object.\" We saw reasons to reject this view in\ngeneral; we are now concerned to show that it must be rejected in the\nparticular case of sensations. The kind of argument which formerly made\nme accept Brentano's view in this case was exceedingly simple. When I\nsee a patch of colour, it seemed to me that the colour is not psychical,\nbut physical, while my seeing is not physical, but psychical. Hence\nI concluded that the colour is something other than my seeing of\nthe colour. This argument, to me historically, was directed against\nidealism: the emphatic part of it was the assertion that the colour is\nphysical, not psychical. I shall not trouble you now with the grounds\nfor holding as against Berkeley that the patch of colour is physical; I\nhave set them forth before, and I see no reason to modify them. But it\ndoes not follow that the patch of colour is not also psychical, unless\nwe assume that the physical and the psychical cannot overlap, which I\nno longer consider a valid assumption. If we admit--as I think we\nshould--that the patch of colour may be both physical and psychical, the\nreason for distinguishing the sense-datum from the sensation disappears,\nand we may say that the patch of colour and our sensation in seeing it\nare identical.\n\nThis is the view of William James, Professor Dewey, and the American\nrealists. Perceptions, says Professor Dewey, are not per se cases of\nknowledge, but simply natural events with no more knowledge status\nthan (say) a shower. \"Let them [the realists] try the experiment of\nconceiving perceptions as pure natural events, not cases of awareness or\napprehension, and they will be surprised to see how little they miss.\"*\nI think he is right in this, except in supposing that the realists will\nbe surprised. Many of them already hold the view he is advocating, and\nothers are very sympathetic to it. At any rate, it is the view which I\nshall adopt in these lectures.\n\n     * Dewey, \"Essays in Experimental Logic,\" pp. 253, 262.\n\nThe stuff of the world, so far as we have experience of it, consists, on\nthe view that I am advocating, of innumerable transient particulars such\nas occur in seeing, hearing, etc., together with images more or less\nresembling these, of which I shall speak shortly. If physics is true,\nthere are, besides the particulars that we experience, others, probably\nequally (or almost equally) transient, which make up that part of the\nmaterial world that does not come into the sort of contact with a\nliving body that is required to turn it into a sensation. But this topic\nbelongs to the philosophy of physics, and need not concern us in our\npresent inquiry.\n\nSensations are what is common to the mental and physical worlds; they\nmay be defined as the intersection of mind and matter. This is by no\nmeans a new view; it is advocated, not only by the American authors I\nhave mentioned, but by Mach in his Analysis of Sensations, which was\npublished in 1886. The essence of sensation, according to the view I am\nadvocating, is its independence of past experience. It is a core in our\nactual experiences, never existing in isolation except possibly in very\nyoung infants. It is not itself knowledge, but it supplies the data for\nour knowledge of the physical world, including our own bodies.\n\nThere are some who believe that our mental life is built up out of\nsensations alone. This may be true; but in any case I think the only\ningredients required in addition to sensations are images. What images\nare, and how they are to be defined, we have now to inquire.\n\nThe distinction between images and sensations might seem at first sight\nby no means difficult. When we shut our eyes and call up pictures of\nfamiliar scenes, we usually have no difficulty, so long as we remain\nawake, in discriminating between what we are imagining and what is\nreally seen. If we imagine some piece of music that we know, we can go\nthrough it in our mind from beginning to end without any discoverable\ntendency to suppose that we are really hearing it. But although such\ncases are so clear that no confusion seems possible, there are many\nothers that are far more difficult, and the definition of images is by\nno means an easy problem.\n\nTo begin with: we do not always know whether what we are experiencing is\na sensation or an image. The things we see in dreams when our eyes are\nshut must count as images, yet while we are dreaming they seem like\nsensations. Hallucinations often begin as persistent images, and only\ngradually acquire that influence over belief that makes the patient\nregard them as sensations. When we are listening for a faint sound--the\nstriking of a distant clock, or a horse's hoofs on the road--we think\nwe hear it many times before we really do, because expectation brings\nus the image, and we mistake it for sensation. The distinction between\nimages and sensations is, therefore, by no means always obvious to\ninspection.*\n\n     * On the distinction between images and sensation, cf.\n     Semon, \"Die mnemischen Empfindungen,\" pp. 19-20.\n\nWe may consider three different ways in which it has been sought to\ndistinguish images from sensations, namely:\n\n(1) By the less degree of vividness in images;\n\n(2) By our absence of belief in their \"physical reality\";\n\n(3) By the fact that their causes and effects are different from those\nof sensations.\n\nI believe the third of these to be the only universally applicable\ncriterion. The other two are applicable in very many cases, but\ncannot be used for purposes of definition because they are liable to\nexceptions. Nevertheless, they both deserve to be carefully considered.\n\n(1) Hume, who gives the names \"impressions\" and \"ideas\" to what may,\nfor present purposes, be identified with our \"sensations\" and \"images,\"\nspeaks of impressions as \"those perceptions which enter with most force\nand violence\" while he defines ideas as \"the faint images of these (i.e.\nof impressions) in thinking and reasoning.\" His immediately following\nobservations, however, show the inadequacy of his criteria of \"force\"\nand \"faintness.\" He says:\n\n\"I believe it will not be very necessary to employ many words in\nexplaining this distinction. Every one of himself will readily perceive\nthe difference betwixt feeling and thinking. The common degrees of these\nare easily distinguished, though it is not impossible but in particular\ninstances they may very nearly approach to each other. Thus in sleep, in\na fever, in madness, or in any very violent emotions of soul, our ideas\nmay approach to our impressions; as, on the other hand, it sometimes\nhappens, that our impressions are so faint and low that we cannot\ndistinguish them from our ideas. But notwithstanding this near\nresemblance in a few instances, they are in general so very different,\nthat no one can make a scruple to rank them under distinct heads, and\nassign to each a peculiar name to mark the difference\" (\"Treatise of\nHuman Nature,\" Part I, Section I).\n\nI think Hume is right in holding that they should be ranked under\ndistinct heads, with a peculiar name for each. But by his own confession\nin the above passage, his criterion for distinguishing them is not\nalways adequate. A definition is not sound if it only applies in cases\nwhere the difference is glaring: the essential purpose of a definition\nis to provide a mark which is applicable even in marginal cases--except,\nof course, when we are dealing with a conception, like, e.g. baldness,\nwhich is one of degree and has no sharp boundaries. But so far we have\nseen no reason to think that the difference between sensations and\nimages is only one of degree.\n\nProfessor Stout, in his \"Manual of Psychology,\" after discussing various\nways of distinguishing sensations and images, arrives at a view which is\na modification of Hume's. He says (I quote from the second edition):\n\n\"Our conclusion is that at bottom the distinction between image\nand percept, as respectively faint and vivid states, is based on a\ndifference of quality. The percept has an aggressiveness which does not\nbelong to the image. It strikes the mind with varying degrees of force\nor liveliness according to the varying intensity of the stimulus. This\ndegree of force or liveliness is part of what we ordinarily mean by\nthe intensity of a sensation. But this constituent of the intensity of\nsensations is absent in mental imagery\"(p. 419).\n\nThis view allows for the fact that sensations may reach any degree of\nfaintness--e.g. in the case of a just visible star or a just audible\nsound--without becoming images, and that therefore mere faintness cannot\nbe the characteristic mark of images. After explaining the sudden shock\nof a flash of lightning or a steam-whistle, Stout says that \"no mere\nimage ever does strike the mind in this manner\"(p. 417). But I believe\nthat this criterion fails in very much the same instances as those in\nwhich Hume's criterion fails in its original form. Macbeth speaks of--\n\n               that suggestion\n           Whose horrid image doth unfix my hair\n     And make my seated heart knock at my ribs\n     Against the use of nature.\n\nThe whistle of a steam-engine could hardly have a stronger effect than\nthis. A very intense emotion will often bring with it--especially\nwhere some future action or some undecided issue is involved--powerful\ncompelling images which may determine the whole course of life, sweeping\naside all contrary solicitations to the will by their capacity for\nexclusively possessing the mind. And in all cases where images,\noriginally recognized as such, gradually pass into hallucinations, there\nmust be just that \"force or liveliness\" which is supposed to be always\nabsent from images. The cases of dreams and fever-delirium are as\nhard to adjust to Professor Stout's modified criterion as to Hume's. I\nconclude therefore that the test of liveliness, however applicable in\nordinary instances, cannot be used to define the differences between\nsensations and images.\n\n(2) We might attempt to distinguish images from sensations by our\nabsence of belief in the \"physical reality\" of images. When we are aware\nthat what we are experiencing is an image, we do not give it the kind of\nbelief that we should give to a sensation: we do not think that it has\nthe same power of producing knowledge of the \"external world.\" Images\nare \"imaginary\"; in SOME sense they are \"unreal.\" But this difference\nis hard to analyse or state correctly. What we call the \"unreality\" of\nimages requires interpretation it cannot mean what would be expressed\nby saying \"there's no such thing.\" Images are just as truly part of the\nactual world as sensations are. All that we really mean by calling an\nimage \"unreal\" is that it does not have the concomitants which it would\nhave if it were a sensation. When we call up a visual image of a chair,\nwe do not attempt to sit in it, because we know that, like Macbeth's\ndagger, it is not \"sensible to feeling as to sight\"--i.e. it does not\nhave the correlations with tactile sensations which it would have if it\nwere a visual sensation and not merely a visual image. But this means\nthat the so-called \"unreality\" of images consists merely in their not\nobeying the laws of physics, and thus brings us back to the causal\ndistinction between images and sensations.\n\nThis view is confirmed by the fact that we only feel images to be\n\"unreal\" when we already know them to be images. Images cannot be\ndefined by the FEELING of unreality, because when we falsely believe an\nimage to be a sensation, as in the case of dreams, it FEELS just as real\nas if it were a sensation. Our feeling of unreality results from our\nhaving already realized that we are dealing with an image, and cannot\ntherefore be the definition of what we mean by an image. As soon as an\nimage begins to deceive us as to its status, it also deceives us as to\nits correlations, which are what we mean by its \"reality.\"\n\n(3) This brings us to the third mode of distinguishing images from\nsensations, namely, by their causes and effects. I believe this to be\nthe only valid ground of distinction. James, in the passage about the\nmental fire which won't burn real sticks, distinguishes images by their\neffects, but I think the more reliable distinction is by their causes.\nProfessor Stout (loc. cit., p. 127) says: \"One characteristic mark of\nwhat we agree in calling sensation is its mode of production. It is\ncaused by what we call a STIMULUS. A stimulus is always some condition\nexternal to the nervous system itself and operating upon it.\" I think\nthat this is the correct view, and that the distinction between images\nand sensations can only be made by taking account of their causation.\nSensations come through sense-organs, while images do not. We cannot\nhave visual sensations in the dark, or with our eyes shut, but we can\nvery well have visual images under these circumstances. Accordingly\nimages have been defined as \"centrally excited sensations,\" i.e.\nsensations which have their physiological cause in the brain only, not\nalso in the sense-organs and the nerves that run from the sense-organs\nto the brain. I think the phrase \"centrally excited sensations\" assumes\nmore than is necessary, since it takes it for granted that an image must\nhave a proximate physiological cause. This is probably true, but it is\nan hypothesis, and for our purposes an unnecessary one. It would seem to\nfit better with what we can immediately observe if we were to say that\nan image is occasioned, through association, by a sensation or another\nimage, in other words that it has a mnemic cause--which does not prevent\nit from also having a physical cause. And I think it will be found that\nthe causation of an image always proceeds according to mnemic laws, i.e.\nthat it is governed by habit and past experience. If you listen to a man\nplaying the pianola without looking at him, you will have images of his\nhands on the keys as if he were playing the piano; if you suddenly look\nat him while you are absorbed in the music, you will experience a shock\nof surprise when you notice that his hands are not touching the notes.\nYour image of his hands is due to the many times that you have heard\nsimilar sounds and at the same time seen the player's hands on the\npiano. When habit and past experience play this part, we are in the\nregion of mnemic as opposed to ordinary physical causation. And I think\nthat, if we could regard as ultimately valid the difference between\nphysical and mnemic causation, we could distinguish images from\nsensations as having mnemic causes, though they may also have physical\ncauses. Sensations, on the other hand, will only have physical causes.\n\nHowever this may be, the practically effective distinction between\nsensations and images is that in the causation of sensations, but not\nof images, the stimulation of nerves carrying an effect into the brain,\nusually from the surface of the body, plays an essential part. And\nthis accounts for the fact that images and sensations cannot always be\ndistinguished by their intrinsic nature.\n\nImages also differ from sensations as regards their effects. Sensations,\nas a rule, have both physical and mental effects. As you watch the train\nyou meant to catch leaving the station, there are both the successive\npositions of the train (physical effects) and the successive waves\nof fury and disappointment (mental effects). Images, on the contrary,\nthough they MAY produce bodily movements, do so according to mnemic\nlaws, not according to the laws of physics. All their effects, of\nwhatever nature, follow mnemic laws. But this difference is less\nsuitable for definition than the difference as to causes.\n\nProfessor Watson, as a logical carrying-out of his behaviourist theory,\ndenies altogether that there are any observable phenomena such as\nimages are supposed to be. He replaces them all by faint sensations, and\nespecially by pronunciation of words sotto voce. When we \"think\" of a\ntable (say), as opposed to seeing it, what happens, according to him, is\nusually that we are making small movements of the throat and tongue\nsuch as would lead to our uttering the word \"table\" if they were more\npronounced. I shall consider his view again in connection with words;\nfor the present I am only concerned to combat his denial of images. This\ndenial is set forth both in his book on \"Behavior\" and in an article\ncalled \"Image and Affection in Behavior\" in the \"Journal of Philosophy,\nPsychology and Scientific Methods,\" vol. x (July, 1913). It seems to me\nthat in this matter he has been betrayed into denying plain facts in\nthe interests of a theory, namely, the supposed impossibility of\nintrospection. I dealt with the theory in Lecture VI; for the present I\nwish to reinforce the view that the facts are undeniable.\n\nImages are of various sorts, according to the nature of the sensations\nwhich they copy. Images of bodily movements, such as we have when we\nimagine moving an arm or, on a smaller scale, pronouncing a word,\nmight possibly be explained away on Professor Watson's lines, as really\nconsisting in small incipient movements such as, if magnified and\nprolonged, would be the movements we are said to be imagining. Whether\nthis is the case or not might even be decided experimentally. If there\nwere a delicate instrument for recording small movements in the mouth\nand throat, we might place such an instrument in a person's mouth and\nthen tell him to recite a poem to himself, as far as possible only in\nimagination. I should not be at all surprised if it were found that\nactual small movements take place while he is \"mentally\" saying over\nthe verses. The point is important, because what is called \"thought\"\nconsists mainly (though I think not wholly) of inner speech. If\nProfessor Watson is right as regards inner speech, this whole region\nis transferred from imagination to sensation. But since the question\nis capable of experimental decision, it would be gratuitous rashness to\noffer an opinion while that decision is lacking.\n\nBut visual and auditory images are much more difficult to deal with in\nthis way, because they lack the connection with physical events in the\nouter world which belongs to visual and auditory sensations. Suppose,\nfor example, that I am sitting in my room, in which there is an empty\narm-chair. I shut my eyes, and call up a visual image of a friend\nsitting in the arm-chair. If I thrust my image into the world of\nphysics, it contradicts all the usual physical laws. My friend reached\nthe chair without coming in at the door in the usual way; subsequent\ninquiry will show that he was somewhere else at the moment. If regarded\nas a sensation, my image has all the marks of the supernatural. My\nimage, therefore, is regarded as an event in me, not as having that\nposition in the orderly happenings of the public world that belongs to\nsensations. By saying that it is an event in me, we leave it possible\nthat it may be PHYSIOLOGICALLY caused: its privacy may be only due to\nits connection with my body. But in any case it is not a public event,\nlike an actual person walking in at the door and sitting down in\nmy chair. And it cannot, like inner speech, be regarded as a SMALL\nsensation, since it occupies just as large an area in my visual field as\nthe actual sensation would do.\n\nProfessor Watson says: \"I should throw out imagery altogether\nand attempt to show that all natural thought goes on in terms of\nsensori-motor processes in the larynx.\" This view seems to me flatly to\ncontradict experience. If you try to persuade any uneducated person that\nshe cannot call up a visual picture of a friend sitting in a chair, but\ncan only use words describing what such an occurrence would be like,\nshe will conclude that you are mad. (This statement is based upon\nexperiment.) Galton, as every one knows, investigated visual imagery,\nand found that education tends to kill it: the Fellows of the Royal\nSociety turned out to have much less of it than their wives. I see no\nreason to doubt his conclusion that the habit of abstract pursuits makes\nlearned men much inferior to the average in power of visualizing, and\nmuch more exclusively occupied with words in their \"thinking.\" And\nProfessor Watson is a very learned man.\n\nI shall henceforth assume that the existence of images is admitted, and\nthat they are to be distinguished from sensations by their causes,\nas well as, in a lesser degree, by their effects. In their intrinsic\nnature, though they often differ from sensations by being more dim\nor vague or faint, yet they do not always or universally differ from\nsensations in any way that can be used for defining them. Their privacy\nneed form no bar to the scientific study of them, any more than the\nprivacy of bodily sensations does. Bodily sensations are admitted by\neven the most severe critics of introspection, although, like images,\nthey can only be observed by one observer. It must be admitted, however,\nthat the laws of the appearance and disappearance of images are little\nknown and difficult to discover, because we are not assisted, as in the\ncase of sensations, by our knowledge of the physical world.\n\nThere remains one very important point concerning images, which will\noccupy us much hereafter, and that is, their resemblance to previous\nsensations. They are said to be \"copies\" of sensations, always as\nregards the simple qualities that enter into them, though not always\nas regards the manner in which these are put together. It is generally\nbelieved that we cannot imagine a shade of colour that we have never\nseen, or a sound that we have never heard. On this subject Hume is the\nclassic. He says, in the definitions already quoted:\n\n\"Those perceptions, which enter with most force and violence, we may\nname IMPRESSIONS; and under this name I comprehend all our sensations,\npassions and emotions, as they make their first appearance in the soul.\nBy IDEAS I mean the faint images of these in thinking and reasoning.\"\n\nHe next explains the difference between simple and complex ideas, and\nexplains that a complex idea may occur without any similar complex\nimpression. But as regards simple ideas, he states that \"every simple\nidea has a simple impression, which resembles it, and every simple\nimpression a correspondent idea.\" He goes on to enunciate the general\nprinciple \"that all our simple ideas in their first appearance are\nderived from simple impressions, which are correspondent to them, and\nwhich they exactly represent\" (\"Treatise of Human Nature,\" Part I,\nSection I).\n\nIt is this fact, that images resemble antecedent sensations, which\nenables us to call them images \"of\" this or that. For the understanding\nof memory, and of knowledge generally, the recognizable resemblance of\nimages and sensations is of fundamental importance.\n\nThere are difficulties in establishing Hume's principles, and doubts\nas to whether it is exactly true. Indeed, he himself signalized an\nexception immediately after stating his maxim. Nevertheless, it is\nimpossible to doubt that in the main simple images are copies of similar\nsimple sensations which have occurred earlier, and that the same is true\nof complex images in all cases of memory as opposed to mere imagination.\nOur power of acting with reference to what is sensibly absent is largely\ndue to this characteristic of images, although, as education advances,\nimages tend to be more and more replaced by words. We shall have much\nto say in the next two lectures on the subject of images as copies of\nsensations. What has been said now is merely by way of reminder that\nthis is their most notable characteristic.\n\nI am by no means confident that the distinction between images and\nsensations is ultimately valid, and I should be glad to be convinced\nthat images can be reduced to sensations of a peculiar kind. I think it\nis clear, however, that, at any rate in the case of auditory and visual\nimages, they do differ from ordinary auditory and visual sensations, and\ntherefore form a recognizable class of occurrences, even if it should\nprove that they can be regarded as a sub-class of sensations. This is\nall that is necessary to validate the use of images to be made in the\nsequel.\n\n\n\n\nLECTURE IX. MEMORY\n\nMemory, which we are to consider to-day, introduces us to knowledge in\none of its forms. The analysis of knowledge will occupy us until the end\nof the thirteenth lecture, and is the most difficult part of our whole\nenterprise.\n\nI do not myself believe that the analysis of knowledge can be effected\nentirely by means of purely external observation, such as behaviourists\nemploy. I shall discuss this question in later lectures. In the present\nlecture I shall attempt the analysis of memory-knowledge, both as an\nintroduction to the problem of knowledge in general, and because memory,\nin some form, is presupposed in almost all other knowledge. Sensation,\nwe decided, is not a form of knowledge. It might, however, have\nbeen expected that we should begin our discussion of knowledge with\nPERCEPTION, i.e. with that integral experience of things in the\nenvironment, out of which sensation is extracted by psychological\nanalysis. What is called perception differs from sensation by the fact\nthat the sensational ingredients bring up habitual associates--images\nand expectations of their usual correlates--all of which are\nsubjectively indistinguishable from the sensation. The FACT of past\nexperience is essential in producing this filling-out of sensation, but\nnot the RECOLLECTION of past experience. The non-sensational elements in\nperception can be wholly explained as the result of habit, produced\nby frequent correlations. Perception, according to our definition in\nLecture VII, is no more a form of knowledge than sensation is, except\nin so far as it involves expectations. The purely psychological problems\nwhich it raises are not very difficult, though they have sometimes been\nrendered artificially obscure by unwillingness to admit the fallibility\nof the non-sensational elements of perception. On the other hand, memory\nraises many difficult and very important problems, which it is necessary\nto consider at the first possible moment.\n\nOne reason for treating memory at this early stage is that it seems to\nbe involved in the fact that images are recognized as \"copies\" of\npast sensible experience. In the preceding lecture I alluded to Hume's\nprinciple \"that all our simple ideas in their first appearance are\nderived from simple impressions, which are correspondent to them, and\nwhich they exactly represent.\" Whether or not this principle is liable\nto exceptions, everyone would agree that is has a broad measure of\ntruth, though the word \"exactly\" might seem an overstatement, and\nit might seem more correct to say that ideas APPROXIMATELY represent\nimpressions. Such modifications of Hume's principle, however, do not\naffect the problem which I wish to present for your consideration,\nnamely: Why do we believe that images are, sometimes or always,\napproximately or exactly, copies of sensations? What sort of evidence is\nthere? And what sort of evidence is logically possible? The difficulty\nof this question arises through the fact that the sensation which an\nimage is supposed to copy is in the past when the image exists, and can\ntherefore only be known by memory, while, on the other hand, memory of\npast sensations seems only possible by means of present images. How,\nthen, are we to find any way of comparing the present image and the past\nsensation? The problem is just as acute if we say that images differ\nfrom their prototypes as if we say that they resemble them; it is the\nvery possibility of comparison that is hard to understand.* We think\nwe can know that they are alike or different, but we cannot bring them\ntogether in one experience and compare them. To deal with this problem,\nwe must have a theory of memory. In this way the whole status of images\nas \"copies\" is bound up with the analysis of memory.\n\n     * How, for example, can we obtain such knowledge as the\n     following: \"If we look at, say, a red nose and perceive it,\n     and after a little while ekphore, its memory-image, we note\n     immediately how unlike, in its likeness, this memory-image\n     is to the original perception\" (A. Wohlgemuth, \"On the\n     Feelings and their Neural Correlate with an Examination of\n     the Nature of Pain,\" \"Journal of Psychology,\" vol. viii,\n     part iv, June, 1917).\n\nIn investigating memory-beliefs, there are certain points which must\nbe borne in mind. In the first place, everything constituting a\nmemory-belief is happening now, not in that past time to which the\nbelief is said to refer. It is not logically necessary to the existence\nof a memory-belief that the event remembered should have occurred,\nor even that the past should have existed at all. There is no logical\nimpossibility in the hypothesis that the world sprang into being five\nminutes ago, exactly as it then was, with a population that \"remembered\"\na wholly unreal past. There is no logically necessary connection between\nevents at different times; therefore nothing that is happening now or\nwill happen in the future can disprove the hypothesis that the world\nbegan five minutes ago. Hence the occurrences which are CALLED knowledge\nof the past are logically independent of the past; they are wholly\nanalysable into present contents, which might, theoretically, be just\nwhat they are even if no past had existed.\n\nI am not suggesting that the non-existence of the past should be\nentertained as a serious hypothesis. Like all sceptical hypotheses, it\nis logically tenable, but uninteresting. All that I am doing is to use\nits logical tenability as a help in the analysis of what occurs when we\nremember.\n\nIn the second place, images without beliefs are insufficient to\nconstitute memory; and habits are still more insufficient. The\nbehaviourist, who attempts to make psychology a record of behaviour, has\nto trust his memory in making the record. \"Habit\" is a concept involving\nthe occurrence of similar events at different times; if the behaviourist\nfeels confident that there is such a phenomenon as habit, that can only\nbe because he trusts his memory, when it assures him that there have\nbeen other times. And the same applies to images. If we are to know as\nit is supposed we do--that images are \"copies,\" accurate or inaccurate,\nof past events, something more than the mere occurrence of images must\ngo to constitute this knowledge. For their mere occurrence, by itself,\nwould not suggest any connection with anything that had happened before.\n\nCan we constitute memory out of images together with suitable beliefs?\nWe may take it that memory-images, when they occur in true memory, are\n(a) known to be copies, (b) sometimes known to be imperfect copies\n(cf. footnote on previous page). How is it possible to know that a\nmemory-image is an imperfect copy, without having a more accurate copy\nby which to replace it? This would SEEM to suggest that we have a way\nof knowing the past which is independent of images, by means of which\nwe can criticize image-memories. But I do not think such an inference is\nwarranted.\n\nWhat results, formally, from our knowledge of the past through images\nof which we recognize the inaccuracy, is that such images must have two\ncharacteristics by which we can arrange them in two series, of which one\ncorresponds to the more or less remote period in the past to which\nthey refer, and the other to our greater or less confidence in their\naccuracy. We will take the second of these points first.\n\nOur confidence or lack of confidence in the accuracy of a memory-image\nmust, in fundamental cases, be based upon a characteristic of the image\nitself, since we cannot evoke the past bodily and compare it with the\npresent image. It might be suggested that vagueness is the required\ncharacteristic, but I do not think this is the case. We sometimes\nhave images that are by no means peculiarly vague, which yet we do not\ntrust--for example, under the influence of fatigue we may see a friend's\nface vividly and clearly, but horribly distorted. In such a case we\ndistrust our image in spite of its being unusually clear. I think\nthe characteristic by which we distinguish the images we trust is the\nfeeling of FAMILIARITY that accompanies them. Some images, like some\nsensations, feel very familiar, while others feel strange. Familiarity\nis a feeling capable of degrees. In an image of a well-known face,\nfor example, some parts may feel more familiar than others; when this\nhappens, we have more belief in the accuracy of the familiar parts than\nin that of the unfamiliar parts. I think it is by this means that we\nbecome critical of images, not by some imageless memory with which\nwe compare them. I shall return to the consideration of familiarity\nshortly.\n\nI come now to the other characteristic which memory-images must have\nin order to account for our knowledge of the past. They must have some\ncharacteristic which makes us regard them as referring to more or less\nremote portions of the past. That is to say if we suppose that A is the\nevent remembered, B the remembering, and t the interval of time between\nA and B, there must be some characteristic of B which is capable of\ndegrees, and which, in accurately dated memories, varies as t varies.\nIt may increase as t increases, or diminish as t increases. The question\nwhich of these occurs is not of any importance for the theoretic\nserviceability of the characteristic in question.\n\nIn actual fact, there are doubtless various factors that concur in\ngiving us the feeling of greater or less remoteness in some remembered\nevent. There may be a specific feeling which could be called the feeling\nof \"pastness,\" especially where immediate memory is concerned. But apart\nfrom this, there are other marks. One of these is context. A recent\nmemory has, usually, more context than a more distant one. When a\nremembered event has a remembered context, this may occur in two ways,\neither (a) by successive images in the same order as their prototypes,\nor (b) by remembering a whole process simultaneously, in the same way in\nwhich a present process may be apprehended, through akoluthic sensations\nwhich, by fading, acquire the mark of just-pastness in an increasing\ndegree as they fade, and are thus placed in a series while all sensibly\npresent. It will be context in this second sense, more specially, that\nwill give us a sense of the nearness or remoteness of a remembered\nevent.\n\nThere is, of course, a difference between knowing the temporal relation\nof a remembered event to the present, and knowing the time-order of two\nremembered events. Very often our knowledge of the temporal relation\nof a remembered event to the present is inferred from its temporal\nrelations to other remembered events. It would seem that only rather\nrecent events can be placed at all accurately by means of feelings\ngiving their temporal relation to the present, but it is clear that such\nfeelings must play an essential part in the process of dating remembered\nevents.\n\nWe may say, then, that images are regarded by us as more or less\naccurate copies of past occurrences because they come to us with two\nsorts of feelings: (1) Those that may be called feelings of familiarity;\n(2) those that may be collected together as feelings giving a sense of\npastness. The first lead us to trust our memories, the second to assign\nplaces to them in the time-order.\n\nWe have now to analyse the memory-belief, as opposed to the\ncharacteristics of images which lead us to base memory-beliefs upon\nthem.\n\nIf we had retained the \"subject\" or \"act\" in knowledge, the whole\nproblem of memory would have been comparatively simple. We could then\nhave said that remembering is a direct relation between the present act\nor subject and the past occurrence remembered: the act of remembering\nis present, though its object is past. But the rejection of the subject\nrenders some more complicated theory necessary. Remembering has to be\na present occurrence in some way resembling, or related to, what is\nremembered. And it is difficult to find any ground, except a pragmatic\none, for supposing that memory is not sheer delusion, if, as seems to be\nthe case, there is not, apart from memory, any way of ascertaining that\nthere really was a past occurrence having the required relation to our\npresent remembering. What, if we followed Meinong's terminology, we\nshould call the \"object\" in memory, i.e. the past event which we are\nsaid to be remembering, is unpleasantly remote from the \"content,\" i.e.\nthe present mental occurrence in remembering. There is an awkward gulf\nbetween the two, which raises difficulties for the theory of knowledge.\nBut we must not falsify observation to avoid theoretical difficulties.\nFor the present, therefore, let us forget these problems, and try to\ndiscover what actually occurs in memory.\n\nSome points may be taken as fixed, and such as any theory of memory must\narrive at. In this case, as in most others, what may be taken as certain\nin advance is rather vague. The study of any topic is like the continued\nobservation of an object which is approaching us along a road: what is\ncertain to begin with is the quite vague knowledge that there is SOME\nobject on the road. If you attempt to be less vague, and to assert that\nthe object is an elephant, or a man, or a mad dog, you run a risk of\nerror; but the purpose of continued observation is to enable you to\narrive at such more precise knowledge. In like manner, in the study of\nmemory, the certainties with which you begin are very vague, and the\nmore precise propositions at which you try to arrive are less certain\nthan the hazy data from which you set out. Nevertheless, in spite of the\nrisk of error, precision is the goal at which we must aim.\n\nThe first of our vague but indubitable data is that there is knowledge\nof the past. We do not yet know with any precision what we mean by\n\"knowledge,\" and we must admit that in any given instance our memory may\nbe at fault. Nevertheless, whatever a sceptic might urge in theory,\nwe cannot practically doubt that we got up this morning, that we did\nvarious things yesterday, that a great war has been taking place, and so\non. How far our knowledge of the past is due to memory, and how far to\nother sources, is of course a matter to be investigated, but there can\nbe no doubt that memory forms an indispensable part of our knowledge of\nthe past.\n\nThe second datum is that we certainly have more capacity for knowing the\npast than for knowing the future. We know some things about the future,\nfor example what eclipses there will be; but this knowledge is a matter\nof elaborate calculation and inference, whereas some of our knowledge of\nthe past comes to us without effort, in the same sort of immediate way\nin which we acquire knowledge of occurrences in our present environment.\nWe might provisionally, though perhaps not quite correctly, define\n\"memory\" as that way of knowing about the past which has no analogue in\nour knowledge of the future; such a definition would at least serve to\nmark the problem with which we are concerned, though some expectations\nmay deserve to rank with memory as regards immediacy.\n\nA third point, perhaps not quite so certain as our previous two, is that\nthe truth of memory cannot be wholly practical, as pragmatists wish\nall truth to be. It seems clear that some of the things I remember are\ntrivial and without any visible importance for the future, but that my\nmemory is true (or false) in virtue of a past event, not in virtue of\nany future consequences of my belief. The definition of truth as the\ncorrespondence between beliefs and facts seems peculiarly evident in the\ncase of memory, as against not only the pragmatist definition but also\nthe idealist definition by means of coherence. These considerations,\nhowever, are taking us away from psychology, to which we must now\nreturn.\n\nIt is important not to confuse the two forms of memory which Bergson\ndistinguishes in the second chapter of his \"Matter and Memory,\"\nnamely the sort that consists of habit, and the sort that consists of\nindependent recollection. He gives the instance of learning a lesson\nby heart: when I know it by heart I am said to \"remember\" it, but this\nmerely means that I have acquired certain habits; on the other hand,\nmy recollection of (say) the second time I read the lesson while I was\nlearning it is the recollection of a unique event, which occurred only\nonce. The recollection of a unique event cannot, so Bergson contends,\nbe wholly constituted by habit, and is in fact something radically\ndifferent from the memory which is habit. The recollection alone is true\nmemory. This distinction is vital to the understanding of memory. But\nit is not so easy to carry out in practice as it is to draw in theory.\nHabit is a very intrusive feature of our mental life, and is often\npresent where at first sight it seems not to be. There is, for example,\na habit of remembering a unique event. When we have once described the\nevent, the words we have used easily become habitual. We may even have\nused words to describe it to ourselves while it was happening; in that\ncase, the habit of these words may fulfil the function of Bergson's true\nmemory, while in reality it is nothing but habit-memory. A gramophone,\nby the help of suitable records, might relate to us the incidents of its\npast; and people are not so different from gramophones as they like to\nbelieve.\n\nIn spite, however, of a difficulty in distinguishing the two forms of\nmemory in practice, there can be no doubt that both forms exist. I can\nset to work now to remember things I never remembered before, such\nas what I had to eat for breakfast this morning, and it can hardly be\nwholly habit that enables me to do this. It is this sort of occurrence\nthat constitutes the essence of memory Until we have analysed what\nhappens in such a case as this, we have not succeeded in understanding\nmemory.\n\nThe sort of memory with which we are here concerned is the sort which is\na form of knowledge. Whether knowledge itself is reducible to habit is\na question to which I shall return in a later lecture; for the present\nI am only anxious to point out that, whatever the true analysis of\nknowledge may be, knowledge of past occurrences is not proved by\nbehaviour which is due to past experience. The fact that a man can\nrecite a poem does not show that he remembers any previous occasion on\nwhich he has recited or read it. Similarly, the performances of animals\nin getting out of cages or mazes to which they are accustomed do not\nprove that they remember having been in the same situation before.\nArguments in favour of (for example) memory in plants are only arguments\nin favour of habit-memory, not of knowledge-memory. Samuel Butler's\narguments in favour of the view that an animal remembers something of\nthe lives of its ancestors* are, when examined, only arguments in favour\nof habit-memory. Semon's two books, mentioned in an earlier lecture, do\nnot touch knowledge-memory at all closely. They give laws according to\nwhich images of past occurrences come into our minds, but do not discuss\nour belief that these images refer to past occurrences, which is what\nconstitutes knowledge-memory. It is this that is of interest to theory\nof knowledge. I shall speak of it as \"true\" memory, to distinguish it\nfrom mere habit acquired through past experience. Before considering\ntrue memory, it will be well to consider two things which are on the way\ntowards memory, namely the feeling of familiarity and recognition.\n\n     * See his \"Life and Habit and Unconscious Memory.\"\n\nWe often feel that something in our sensible environment is familiar,\nwithout having any definite recollection of previous occasions on which\nwe have seen it. We have this feeling normally in places where we have\noften been before--at home, or in well-known streets. Most people and\nanimals find it essential to their happiness to spend a good deal of\ntheir time in familiar surroundings, which are especially comforting\nwhen any danger threatens. The feeling of familiarity has all sorts\nof degrees, down to the stage where we dimly feel that we have seen a\nperson before. It is by no means always reliable; almost everybody\nhas at some time experienced the well-known illusion that all that is\nhappening now happened before at some time. There are occasions when\nfamiliarity does not attach itself to any definite object, when there is\nmerely a vague feeling that SOMETHING is familiar. This is illustrated\nby Turgenev's \"Smoke,\" where the hero is long puzzled by a haunting\nsense that something in his present is recalling something in his past,\nand at last traces it to the smell of heliotrope. Whenever the sense of\nfamiliarity occurs without a definite object, it leads us to search the\nenvironment until we are satisfied that we have found the appropriate\nobject, which leads us to the judgment: \"THIS is familiar.\" I think\nwe may regard familiarity as a definite feeling, capable of existing\nwithout an object, but normally standing in a specific relation to some\nfeature of the environment, the relation being that which we express in\nwords by saying that the feature in question is familiar. The judgment\nthat what is familiar has been experienced before is a product of\nreflection, and is no part of the feeling of familiarity, such as a\nhorse may be supposed to have when he returns to his stable. Thus\nno knowledge as to the past is to be derived from the feeling of\nfamiliarity alone.\n\nA further stage is RECOGNITION. This may be taken in two senses,\nthe first when a thing not merely feels familiar, but we know it is\nsuch-and-such. We recognize our friend Jones, we know cats and dogs\nwhen we see them, and so on. Here we have a definite influence of past\nexperience, but not necessarily any actual knowledge of the past. When\nwe see a cat, we know it is a cat because of previous cats we have\nseen, but we do not, as a rule, recollect at the moment any particular\noccasion when we have seen a cat. Recognition in this sense does not\nnecessarily involve more than a habit of association: the kind of object\nwe are seeing at the moment is associated with the word \"cat,\" or with\nan auditory image of purring, or whatever other characteristic we may\nhappen to recognize in the cat of the moment. We are, of course, in\nfact able to judge, when we recognize an object, that we have seen it\nbefore, but this judgment is something over and above recognition in\nthis first sense, and may very probably be impossible to animals that\nnevertheless have the experience of recognition in this first sense of\nthe word.\n\nThere is, however, another sense of the word, in which we mean by\nrecognition, not knowing the name of a thing or some other property of\nit, but knowing that we have seen it before In this sense recognition\ndoes involve knowledge about the Fast. This knowledge is memory in\none sense, though in another it is not. It does not involve a definite\nmemory of a definite past event, but only the knowledge that something\nhappening now is similar to something that happened before. It differs\nfrom the sense of familiarity by being cognitive; it is a belief or\njudgment, which the sense of familiarity is not. I do not wish to\nundertake the analysis of belief at present, since it will be the\nsubject of the twelfth lecture; for the present I merely wish to\nemphasize the fact that recognition, in our second sense, consists in\na belief, which we may express approximately in the words: \"This has\nexisted before.\"\n\nThere are, however, several points in which such an account of\nrecognition is inadequate. To begin with, it might seem at first sight\nmore correct to define recognition as \"I have seen this before\" than\nas \"this has existed before.\" We recognize a thing (it may be urged) as\nhaving been in our experience before, whatever that may mean; we do not\nrecognize it as merely having been in the world before. I am not sure\nthat there is anything substantial in this point. The definition of \"my\nexperience\" is difficult; broadly speaking, it is everything that is\nconnected with what I am experiencing now by certain links, of which\nthe various forms of memory are among the most important. Thus, if I\nrecognize a thing, the occasion of its previous existence in virtue\nof which I recognize it forms part of \"my experience\" by DEFINITION:\nrecognition will be one of the marks by which my experience is singled\nout from the rest of the world. Of course, the words \"this has existed\nbefore\" are a very inadequate translation of what actually happens when\nwe form a judgment of recognition, but that is unavoidable: words are\nframed to express a level of thought which is by no means primitive,\nand are quite incapable of expressing such an elementary occurrence as\nrecognition. I shall return to what is virtually the same question in\nconnection with true memory, which raises exactly similar problems.\n\nA second point is that, when we recognize something, it was not in fact\nthe very same thing, but only something similar, that we experienced on\na former occasion. Suppose the object in question is a friend's face. A\nperson's face is always changing, and is not exactly the same on any two\noccasions. Common sense treats it as one face with varying expressions;\nbut the varying expressions actually exist, each at its proper time,\nwhile the one face is merely a logical construction. We regard two\nobjects as the same, for common-sense purposes, when the reaction they\ncall for is practically the same. Two visual appearances, to both\nof which it is appropriate to say: \"Hullo, Jones!\" are treated as\nappearances of one identical object, namely Jones. The name \"Jones\" is\napplicable to both, and it is only reflection that shows us that many\ndiverse particulars are collected together to form the meaning of the\nname \"Jones.\" What we see on any one occasion is not the whole series of\nparticulars that make up Jones, but only one of them (or a few in quick\nsuccession). On another occasion we see another member of the series,\nbut it is sufficiently similar to count as the same from the standpoint\nof common sense. Accordingly, when we judge \"I have seen THIS\nbefore,\" we judge falsely if \"this\" is taken as applying to the actual\nconstituent of the world that we are seeing at the moment. The\nword \"this\" must be interpreted vaguely so as to include anything\nsufficiently like what we are seeing at the moment. Here, again, we\nshall find a similar point as regards true memory; and in connection\nwith true memory we will consider the point again. It is sometimes\nsuggested, by those who favour behaviourist views, that recognition\nconsists in behaving in the same way when a stimulus is repeated as we\nbehaved on the first occasion when it occurred. This seems to be the\nexact opposite of the truth. The essence of recognition is in the\nDIFFERENCE between a repeated stimulus and a new one. On the first\noccasion there is no recognition; on the second occasion there is. In\nfact, recognition is another instance of the peculiarity of causal laws\nin psychology, namely, that the causal unit is not a single event, but\ntwo or more events Habit is the great instance of this, but recognition\nis another. A stimulus occurring once has a certain effect; occurring\ntwice, it has the further effect of recognition. Thus the phenomenon\nof recognition has as its cause the two occasions when the stimulus has\noccurred; either alone is insufficient. This complexity of causes\nin psychology might be connected with Bergson's arguments against\nrepetition in the mental world. It does not prove that there are no\ncausal laws in psychology, as Bergson suggests; but it does prove that\nthe causal laws of psychology are Prima facie very different from those\nof physics. On the possibility of explaining away the difference as due\nto the peculiarities of nervous tissue I have spoken before, but this\npossibility must not be forgotten if we are tempted to draw unwarranted\nmetaphysical deductions.\n\nTrue memory, which we must now endeavour to understand, consists of\nknowledge of past events, but not of all such knowledge. Some knowledge\nof past events, for example what we learn through reading history, is\non a par with the knowledge we can acquire concerning the future: it\nis obtained by inference, not (so to speak) spontaneously. There is\na similar distinction in our knowledge of the present: some of it is\nobtained through the senses, some in more indirect ways. I know that\nthere are at this moment a number of people in the streets of New York,\nbut I do not know this in the immediate way in which I know of the\npeople whom I see by looking out of my window. It is not easy to state\nprecisely wherein the difference between these two sorts of knowledge\nconsists, but it is easy to feel the difference. For the moment, I shall\nnot stop to analyse it, but shall content myself with saying that, in\nthis respect, memory resembles the knowledge derived from the senses.\nIt is immediate, not inferred, not abstract; it differs from perception\nmainly by being referred to the past.\n\nIn regard to memory, as throughout the analysis of knowledge, there are\ntwo very distinct problems, namely (1) as to the nature of the present\noccurrence in knowing; (2) as to the relation of this occurrence to what\nis known. When we remember, the knowing is now, while what is known is\nin the past. Our two questions are, in the case of memory:\n\n(1) What is the present occurrence when we remember?\n\n(2) What is the relation of this present occurrence to the past event\nwhich is remembered?\n\nOf these two questions, only the first concerns the psychologist; the\nsecond belongs to theory of knowledge. At the same time, if we accept\nthe vague datum with which we began, to the effect that, in some sense,\nthere is knowledge of the past, we shall have to find, if we can, such\nan account of the present occurrence in remembering as will make it not\nimpossible for remembering to give us knowledge of the past. For the\npresent, however, we shall do well to forget the problems concerning\ntheory of knowledge, and concentrate upon the purely psychological\nproblem of memory.\n\nBetween memory-image and sensation there is an intermediate experience\nconcerning the immediate past. For example, a sound that we have just\nheard is present to us in a way which differs both from the sensation\nwhile we are hearing the sound and from the memory-image of something\nheard days or weeks ago. James states that it is this way of\napprehending the immediate past that is \"the ORIGINAL of our experience\nof pastness, from whence we get the meaning of the term\"(\"Psychology,\"\ni, p. 604). Everyone knows the experience of noticing (say) that\nthe clock HAS BEEN striking, when we did not notice it while it was\nstriking. And when we hear a remark spoken, we are conscious of the\nearlier words while the later ones are being uttered, and this retention\nfeels different from recollection of something definitely past. A\nsensation fades gradually, passing by continuous gradations to the\nstatus of an image. This retention of the immediate past in a condition\nintermediate between sensation and image may be called \"immediate\nmemory.\" Everything belonging to it is included with sensation in what\nis called the \"specious present.\" The specious present includes elements\nat all stages on the journey from sensation to image. It is this fact\nthat enables us to apprehend such things as movements, or the order of\nthe words in a spoken sentence. Succession can occur within the specious\npresent, of which we can distinguish some parts as earlier and others as\nlater. It is to be supposed that the earliest parts are those that have\nfaded most from their original force, while the latest parts are those\nthat retain their full sensational character. At the beginning of a\nstimulus we have a sensation; then a gradual transition; and at the\nend an image. Sensations while they are fading are called \"akoluthic\"\nsensations.* When the process of fading is completed (which happens very\nquickly), we arrive at the image, which is capable of being revived on\nsubsequent occasions with very little change. True memory, as opposed to\n\"immediate memory,\" applies only to events sufficiently distant to\nhave come to an end of the period of fading. Such events, if they are\nrepresented by anything present, can only be represented by images, not\nby those intermediate stages, between sensations and images, which occur\nduring the period of fading.\n\n     * See Semon, \"Die mnemischen Empfindungen,\" chap. vi.\n\nImmediate memory is important both because it provides experience of\nsuccession, and because it bridges the gulf between sensations and\nthe images which are their copies. But it is now time to resume the\nconsideration of true memory.\n\nSuppose you ask me what I ate for breakfast this morning. Suppose,\nfurther, that I have not thought about my breakfast in the meantime, and\nthat I did not, while I was eating it, put into words what it consisted\nof. In this case my recollection will be true memory, not habit-memory.\nThe process of remembering will consist of calling up images of my\nbreakfast, which will come to me with a feeling of belief such as\ndistinguishes memory-images from mere imagination-images. Or sometimes\nwords may come without the intermediary of images; but in this case\nequally the feeling of belief is essential.\n\nLet us omit from our consideration, for the present, the memories\nin which words replace images. These are always, I think, really\nhabit-memories, the memories that use images being the typical true\nmemories.\n\nMemory-images and imagination-images do not differ in their intrinsic\nqualities, so far as we can discover. They differ by the fact that\nthe images that constitute memories, unlike those that constitute\nimagination, are accompanied by a feeling of belief which may be\nexpressed in the words \"this happened.\" The mere occurrence of images,\nwithout this feeling of belief, constitutes imagination; it is the\nelement of belief that is the distinctive thing in memory.*\n\n     * For belief of a specific kind, cf. Dorothy Wrinch \"On the\n     Nature of Memory,\" \"Mind,\" January, 1920.\n\nThere are, if I am not mistaken, at least three different kinds of\nbelief-feeling, which we may call respectively memory, expectation and\nbare assent. In what I call bare assent, there is no time-element in\nthe feeling of belief, though there may be in the content of what is\nbelieved. If I believe that Caesar landed in Britain in B.C. 55, the\ntime-determination lies, not in the feeling of belief, but in what is\nbelieved. I do not remember the occurrence, but have the same feeling\ntowards it as towards the announcement of an eclipse next year. But when\nI have seen a flash of lightning and am waiting for the thunder, I\nhave a belief-feeling analogous to memory, except that it refers to the\nfuture: I have an image of thunder, combined with a feeling which may be\nexpressed in the words: \"this will happen.\" So, in memory, the pastness\nlies, not in the content of what is believed, but in the nature of\nthe belief-feeling. I might have just the same images and expect their\nrealization; I might entertain them without any belief, as in reading a\nnovel; or I might entertain them together with a time-determination, and\ngive bare assent, as in reading history. I shall return to this subject\nin a later lecture, when we come to the analysis of belief. For the\npresent, I wish to make it clear that a certain special kind of belief\nis the distinctive characteristic of memory.\n\n\nThe problem as to whether memory can be explained as habit or\nassociation requires to be considered afresh in connection with the\ncauses of our remembering something. Let us take again the case of my\nbeing asked what I had for breakfast this morning. In this case the\nquestion leads to my setting to work to recollect. It is a little\nstrange that the question should instruct me as to what it is that I am\nto recall. This has to do with understanding words, which will be the\ntopic of the next lecture; but something must be said about it now. Our\nunderstanding of the words \"breakfast this morning\" is a habit, in spite\nof the fact that on each fresh day they point to a different occasion.\n\"This morning\" does not, whenever it is used, mean the same thing, as\n\"John\" or \"St. Paul's\" does; it means a different period of time on\neach different day. It follows that the habit which constitutes\nour understanding of the words \"this morning\" is not the habit of\nassociating the words with a fixed object, but the habit of associating\nthem with something having a fixed time-relation to our present. This\nmorning has, to-day, the same time-relation to my present that yesterday\nmorning had yesterday. In order to understand the phrase \"this morning\"\nit is necessary that we should have a way of feeling time-intervals,\nand that this feeling should give what is constant in the meaning of the\nwords \"this morning.\" This appreciation of time-intervals is, however,\nobviously a product of memory, not a presupposition of it. It will be\nbetter, therefore, if we wish to analyse the causation of memory by\nsomething not presupposing memory, to take some other instance than that\nof a question about \"this morning.\"\n\nLet us take the case of coming into a familiar room where something has\nbeen changed--say a new picture hung on the wall. We may at first\nhave only a sense that SOMETHING is unfamiliar, but presently we shall\nremember, and say \"that picture was not on the wall before.\" In order to\nmake the case definite, we will suppose that we were only in the room\non one former occasion. In this case it seems fairly clear what happens.\nThe other objects in the room are associated, through the former\noccasion, with a blank space of wall where now there is a picture. They\ncall up an image of a blank wall, which clashes with perception of the\npicture. The image is associated with the belief-feeling which we found\nto be distinctive of memory, since it can neither be abolished nor\nharmonized with perception. If the room had remained unchanged, we\nmight have had only the feeling of familiarity without the definite\nremembering; it is the change that drives us from the present to memory\nof the past.\n\nWe may generalize this instance so as to cover the causes of many\nmemories. Some present feature of the environment is associated, through\npast experiences, with something now absent; this absent something comes\nbefore us as an image, and is contrasted with present sensation. In\ncases of this sort, habit (or association) explains why the present\nfeature of the environment brings up the memory-image, but it does\nnot explain the memory-belief. Perhaps a more complete analysis could\nexplain the memory-belief also on lines of association and habit, but\nthe causes of beliefs are obscure, and we cannot investigate them\nyet. For the present we must content ourselves with the fact that the\nmemory-image can be explained by habit. As regards the memory-belief,\nwe must, at least provisionally, accept Bergson's view that it cannot be\nbrought under the head of habit, at any rate when it first occurs, i.e.\nwhen we remember something we never remembered before.\n\nWe must now consider somewhat more closely the content of a\nmemory-belief. The memory-belief confers upon the memory-image something\nwhich we may call \"meaning;\" it makes us feel that the image points to\nan object which existed in the past. In order to deal with this topic\nwe must consider the verbal expression of the memory-belief. We might\nbe tempted to put the memory-belief into the words: \"Something like\nthis image occurred.\" But such words would be very far from an accurate\ntranslation of the simplest kind of memory-belief. \"Something like this\nimage\" is a very complicated conception. In the simplest kind of memory\nwe are not aware of the difference between an image and the sensation\nwhich it copies, which may be called its \"prototype.\" When the image\nis before us, we judge rather \"this occurred.\" The image is not\ndistinguished from the object which existed in the past: the word \"this\"\ncovers both, and enables us to have a memory-belief which does not\nintroduce the complicated notion \"something like this.\"\n\nIt might be objected that, if we judge \"this occurred\" when in fact\n\"this\" is a present image, we judge falsely, and the memory-belief,\nso interpreted, becomes deceptive. This, however, would be a mistake,\nproduced by attempting to give to words a precision which they do not\npossess when used by unsophisticated people. It is true that the image\nis not absolutely identical with its prototype, and if the word \"this\"\nmeant the image to the exclusion of everything else, the judgment \"this\noccurred\" would be false. But identity is a precise conception, and no\nword, in ordinary speech, stands for anything precise. Ordinary speech\ndoes not distinguish between identity and close similarity. A word\nalways applies, not only to one particular, but to a group of associated\nparticulars, which are not recognized as multiple in common thought or\nspeech. Thus primitive memory, when it judges that \"this occurred,\" is\nvague, but not false.\n\nVague identity, which is really close similarity, has been a source\nof many of the confusions by which philosophy has lived. Of a vague\nsubject, such as a \"this,\" which is both an image and its prototype,\ncontradictory predicates are true simultaneously: this existed and does\nnot exist, since it is a thing remembered, but also this exists and did\nnot exist, since it is a present image. Hence Bergson's\ninterpenetration of the present by the past, Hegelian continuity and\nidentity-in-diversity, and a host of other notions which are thought to\nbe profound because they are obscure and confused. The contradictions\nresulting from confounding image and prototype in memory force us to\nprecision. But when we become precise, our remembering becomes different\nfrom that of ordinary life, and if we forget this we shall go wrong in\nthe analysis of ordinary memory.\n\nVagueness and accuracy are important notions, which it is very necessary\nto understand. Both are a matter of degree. All thinking is vague\nto some extent, and complete accuracy is a theoretical ideal not\npractically attainable. To understand what is meant by accuracy, it will\nbe well to consider first instruments of measurement, such as a balance\nor a thermometer. These are said to be accurate when they give different\nresults for very slightly different stimuli.* A clinical thermometer\nis accurate when it enables us to detect very slight differences in the\ntemperature of the blood. We may say generally that an instrument\nis accurate in proportion as it reacts differently to very slightly\ndifferent stimuli. When a small difference of stimulus produces a great\ndifference of reaction, the instrument is accurate; in the contrary case\nit is not.\n\n     * This is a necessary but not a sufficient condition. The\n     subject of accuracy and vagueness will be considered again\n     in Lecture XIII.\n\nExactly the same thing applies in defining accuracy of thought\nor perception. A musician will respond differently to very minute\ndifferences in playing which would be quite imperceptible to the\nordinary mortal. A negro can see the difference between one negro and\nanother one is his friend, another his enemy. But to us such different\nresponses are impossible: we can merely apply the word \"negro\"\nindiscriminately. Accuracy of response in regard to any particular kind\nof stimulus is improved by practice. Understanding a language is a\ncase in point. Few Frenchmen can hear any difference between the sounds\n\"hall\" and \"hole,\" which produce quite different impressions upon us.\nThe two statements \"the hall is full of water\" and \"the hole is full\nof water\" call for different responses, and a hearing which cannot\ndistinguish between them is inaccurate or vague in this respect.\n\nPrecision and vagueness in thought, as in perception, depend upon the\ndegree of difference between responses to more or less similar stimuli.\nIn the case of thought, the response does not follow immediately upon\nthe sensational stimulus, but that makes no difference as regards our\npresent question. Thus to revert to memory: A memory is \"vague\" when\nit is appropriate to many different occurrences: for instance, \"I met a\nman\" is vague, since any man would verify it. A memory is \"precise\" when\nthe occurrences that would verify it are narrowly circumscribed: for\ninstance, \"I met Jones\" is precise as compared to \"I met a man.\" A\nmemory is \"accurate\" when it is both precise and true, i.e. in the above\ninstance, if it was Jones I met. It is precise even if it is false,\nprovided some very definite occurrence would have been required to make\nit true.\n\nIt follows from what has been said that a vague thought has more\nlikelihood of being true than a precise one. To try and hit an object\nwith a vague thought is like trying to hit the bull's eye with a lump of\nputty: when the putty reaches the target, it flattens out all over it,\nand probably covers the bull's eye along with the rest. To try and hit\nan object with a precise thought is like trying to hit the bull's\neye with a bullet. The advantage of the precise thought is that it\ndistinguishes between the bull's eye and the rest of the target. For\nexample, if the whole target is represented by the fungus family and the\nbull's eye by mushrooms, a vague thought which can only hit the target\nas a whole is not much use from a culinary point of view. And when I\nmerely remember that I met a man, my memory may be very inadequate to my\npractical requirements, since it may make a great difference whether I\nmet Brown or Jones. The memory \"I met Jones\" is relatively precise. It\nis accurate if I met Jones, inaccurate if I met Brown, but precise in\neither case as against the mere recollection that I met a man.\n\nThe distinction between accuracy and precision is however, not\nfundamental. We may omit precision from out thoughts and confine\nourselves to the distinction between accuracy and vagueness. We may then\nset up the following definitions:\n\nAn instrument is \"reliable\" with respect to a given set of stimuli when\nto stimuli which are not relevantly different it gives always responses\nwhich are not relevantly different.\n\nAn instrument is a \"measure\" of a set of stimuli which are serially\nordered when its responses, in all cases where they are relevantly\ndifferent, are arranged in a series in the same order.\n\nThe \"degree of accuracy\" of an instrument which is a reliable measurer\nis the ratio of the difference of response to the difference of stimulus\nin cases where the difference of stimulus is small.* That is to say, if\na small difference of stimulus produces a great difference of response,\nthe instrument is very accurate; in the contrary case, very inaccurate.\n\n     * Strictly speaking, the limit of this, i.e. the derivative\n     of the response with respect to the stimulus.\n\nA mental response is called \"vague\" in proportion to its lack of\naccuracy, or rather precision.\n\nThese definitions will be found useful, not only in the case of memory,\nbut in almost all questions concerned with knowledge.\n\nIt should be observed that vague beliefs, so far from being necessarily\nfalse, have a better chance of truth than precise ones, though their\ntruth is less valuable than that of precise beliefs, since they do not\ndistinguish between occurrences which may differ in important ways.\n\nThe whole of the above discussion of vagueness and accuracy was\noccasioned by the attempt to interpret the word \"this\" when we judge in\nverbal memory that \"this occurred.\" The word \"this,\" in such a judgment,\nis a vague word, equally applicable to the present memory-image and to\nthe past occurrence which is its prototype. A vague word is not to be\nidentified with a general word, though in practice the distinction\nmay often be blurred. A word is general when it is understood to be\napplicable to a number of different objects in virtue of some common\nproperty. A word is vague when it is in fact applicable to a number of\ndifferent objects because, in virtue of some common property, they\nhave not appeared, to the person using the word, to be distinct. I\nemphatically do not mean that he has judged them to be identical, but\nmerely that he has made the same response to them all and has not judged\nthem to be different. We may compare a vague word to a jelly and\na general word to a heap of shot. Vague words precede judgments\nof identity and difference; both general and particular words are\nsubsequent to such judgments. The word \"this\" in the primitive\nmemory-belief is a vague word, not a general word; it covers both the\nimage and its prototype because the two are not distinguished.*\n\n     * On the vague and the general cf. Ribot: \"Evolution of\n     General Ideas,\" Open Court Co., 1899, p. 32: \"The sole\n     permissible formula is this: Intelligence progresses from\n     the indefinite to the definite. If 'indefinite' is taken as\n     synonymous with general, it may be said that the particular\n     does not appear at the outset, but neither does the general\n     in any exact sense: the vague would be more appropriate. In\n     other words, no sooner has the intellect progressed beyond\n     the moment of perception and of its immediate reproduction\n     in memory, than the generic image makes its appearance, i.e.\n     a state intermediate between the particular and the general,\n     participating in the nature of the one and of the other--a\n     confused simplification.\"\n\nBut we have not yet finished our analysis of the memory-belief. The\ntense in the belief that \"this occurred\" is provided by the nature of\nthe belief-feeling involved in memory; the word \"this,\" as we have seen,\nhas a vagueness which we have tried to describe. But we must still ask\nwhat we mean by \"occurred.\" The image is, in one sense, occurring now;\nand therefore we must find some other sense in which the past event\noccurred but the image does not occur.\n\nThere are two distinct questions to be asked: (1) What causes us to say\nthat a thing occurs? (2) What are we feeling when we say this? As to the\nfirst question, in the crude use of the word, which is what concerns us,\nmemory-images would not be said to occur; they would not be noticed\nin themselves, but merely used as signs of the past event. Images are\n\"merely imaginary\"; they have not, in crude thought, the sort of reality\nthat belongs to outside bodies. Roughly speaking, \"real\" things would\nbe those that can cause sensations, those that have correlations of the\nsort that constitute physical objects. A thing is said to be \"real\"\nor to \"occur\" when it fits into a context of such correlations. The\nprototype of our memory-image did fit into a physical context, while\nour memory-image does not. This causes us to feel that the prototype was\n\"real,\" while the image is \"imaginary.\"\n\nBut the answer to our second question, namely as to what we are feeling\nwhen we say a thing \"occurs\" or is \"real,\" must be somewhat different.\nWe do not, unless we are unusually reflective, think about the presence\nor absence of correlations: we merely have different feelings which,\nintellectualized, may be represented as expectations of the presence\nor absence of correlations. A thing which \"feels real\" inspires us with\nhopes or fears, expectations or curiosities, which are wholly absent\nwhen a thing \"feels imaginary.\" The feeling of reality is a feeling akin\nto respect: it belongs PRIMARILY to whatever can do things to us without\nour voluntary co-operation. This feeling of reality, related to\nthe memory-image, and referred to the past by the specific kind of\nbelief-feeling that is characteristic of memory, seems to be what\nconstitutes the act of remembering in its pure form.\n\nWe may now summarize our analysis of pure memory.\n\nMemory demands (a) an image, (b) a belief in past existence. The belief\nmay be expressed in the words \"this existed.\"\n\nThe belief, like every other, may be analysed into (1) the believing,\n(2) what is believed. The believing is a specific feeling or sensation\nor complex of sensations, different from expectation or bare assent in\na way that makes the belief refer to the past; the reference to the\npast lies in the belief-feeling, not in the content believed. There is\na relation between the belief-feeling and the content, making the\nbelief-feeling refer to the content, and expressed by saying that the\ncontent is what is believed.\n\nThe content believed may or may not be expressed in words. Let us\ntake first the case when it is not. In that case, if we are merely\nremembering that something of which we now have an image occurred,\nthe content consists of (a) the image, (b) the feeling, analogous\nto respect, which we translate by saying that something is \"real\" as\nopposed to \"imaginary,\" (c) a relation between the image and the feeling\nof reality, of the sort expressed when we say that the feeling refers\nto the image. This content does not contain in itself any\ntime-determination.\n\nThe time-determination lies in the nature of the belief feeling, which\nis that called \"remembering\" or (better) \"recollecting.\" It is only\nsubsequent reflection upon this reference to the past that makes us\nrealize the distinction between the image and the event recollected.\nWhen we have made this distinction, we can say that the image \"means\"\nthe past event.\n\nThe content expressed in words is best represented by the words \"the\nexistence of this,\" since these words do not involve tense, which\nbelongs to the belief-feeling, not to the content. Here \"this\" is\na vague term, covering the memory-image and anything very like it,\nincluding its prototype. \"Existence\" expresses the feeling of a\n\"reality\" aroused primarily by whatever can have effects upon us without\nour voluntary co-operation. The word \"of\" in the phrase \"the existence\nof this\" represents the relation which subsists between the feeling of\nreality and the \"this.\"\n\nThis analysis of memory is probably extremely faulty, but I do not know\nhow to improve it.\n\nNOTE.-When I speak of a FEELING of belief, I use the word \"feeling\" in\na popular sense, to cover a sensation or an image or a complex of\nsensations or images or both; I use this word because I do not wish to\ncommit myself to any special analysis of the belief-feeling.\n\n\n\n\nLECTURE X. WORDS AND MEANING\n\nThe problem with which we shall be concerned in this lecture is the\nproblem of determining what is the relation called \"meaning.\" The word\n\"Napoleon,\" we say, \"means\" a certain person. In saying this, we are\nasserting a relation between the word \"Napoleon\" and the person so\ndesignated. It is this relation that we must now investigate.\n\nLet us first consider what sort of object a word is when considered\nsimply as a physical thing, apart from its meaning. To begin with, there\nare many instances of a word, namely all the different occasions when it\nis employed. Thus a word is not something unique and particular, but a\nset of occurrences. If we confine ourselves to spoken words, a word has\ntwo aspects, according as we regard it from the point of view of the\nspeaker or from that of the hearer. From the point of view of the\nspeaker, a single instance of the use of a word consists of a certain\nset of movements in the throat and mouth, combined with breath. From\nthe point of view of the hearer, a single instance of the use of a\nword consists of a certain series of sounds, each being approximately\nrepresented by a single letter in writing, though in practice a letter\nmay represent several sounds, or several letters may represent one\nsound. The connection between the spoken word and the word as it reaches\nthe hearer is causal. Let us confine ourselves to the spoken word, which\nis the more important for the analysis of what is called \"thought.\"\nThen we may say that a single instance of the spoken word consists of\na series of movements, and the word consists of a whole set of such\nseries, each member of the set being very similar to each other member.\nThat is to say, any two instances of the word \"Napoleon\" are very\nsimilar, and each instance consists of a series of movements in the\nmouth.\n\nA single word, accordingly, is by no means simple it is a class of\nsimilar series of movements (confining ourselves still to the spoken\nword). The degree of similarity required cannot be precisely defined:\na man may pronounce the word \"Napoleon\" so badly that it can hardly be\ndetermined whether he has really pronounced it or not. The instances\nof a word shade off into other movements by imperceptible degrees. And\nexactly analogous observations apply to words heard or written or read.\nBut in what has been said so far we have not even broached the\nquestion of the DEFINITION of a word, since \"meaning\" is clearly what\ndistinguishes a word from other sets of similar movements, and \"meaning\"\nremains to be defined.\n\nIt is natural to think of the meaning of a word as something\nconventional. This, however, is only true with great limitations. A new\nword can be added to an existing language by a mere convention, as\nis done, for instance, with new scientific terms. But the basis of\na language is not conventional, either from the point of view of the\nindividual or from that of the community. A child learning to speak is\nlearning habits and associations which are just as much determined by\nthe environment as the habit of expecting dogs to bark and cocks to\ncrow. The community that speaks a language has learnt it, and modified\nit by processes almost all of which are not deliberate, but the results\nof causes operating according to more or less ascertainable laws. If\nwe trace any Indo-European language back far enough, we arrive\nhypothetically (at any rate according to some authorities) at the stage\nwhen language consisted only of the roots out of which subsequent words\nhave grown. How these roots acquired their meanings is not known, but a\nconventional origin is clearly just as mythical as the social contract\nby which Hobbes and Rousseau supposed civil government to have been\nestablished. We can hardly suppose a parliament of hitherto speechless\nelders meeting together and agreeing to call a cow a cow and a wolf a\nwolf. The association of words with their meanings must have grown up\nby some natural process, though at present the nature of the process is\nunknown.\n\nSpoken and written words are, of course, not the only way of conveying\nmeaning. A large part of one of Wundt's two vast volumes on language in\nhis \"Volkerpsychologie\" is concerned with gesture-language. Ants appear\nto be able to communicate a certain amount of information by means of\ntheir antennae. Probably writing itself, which we now regard as merely\na way of representing speech, was originally an independent language,\nas it has remained to this day in China. Writing seems to have consisted\noriginally of pictures, which gradually became conventionalized, coming\nin time to represent syllables, and finally letters on the telephone\nprinciple of \"T for Tommy.\" But it would seem that writing nowhere\nbegan as an attempt to represent speech it began as a direct pictorial\nrepresentation of what was to be expressed. The essence of language\nlies, not in the use of this or that special means of communication,\nbut in the employment of fixed associations (however these may have\noriginated) in order that something now sensible--a spoken word, a\npicture, a gesture, or what not--may call up the \"idea\" of something\nelse. Whenever this is done, what is now sensible may be called a \"sign\"\nor \"symbol,\" and that of which it is intended to call up the \"idea\" may\nbe called its \"meaning.\" This is a rough outline of what constitutes\n\"meaning.\" But we must fill in the outline in various ways. And,\nsince we are concerned with what is called \"thought,\" we must pay more\nattention than we otherwise should do to the private as opposed to the\nsocial use of language. Language profoundly affects our thoughts, and\nit is this aspect of language that is of most importance to us in our\npresent inquiry. We are almost more concerned with the internal speech\nthat is never uttered than we are with the things said out loud to other\npeople.\n\nWhen we ask what constitutes meaning, we are not asking what is the\nmeaning of this or that particular word. The word \"Napoleon\" means a\ncertain individual; but we are asking, not who is the individual meant,\nbut what is the relation of the word to the individual which makes the\none mean the other. But just as it is useful to realize the nature of a\nword as part of the physical world, so it is useful to realize the sort\nof thing that a word may mean. When we are clear both as to what a word\nis in its physical aspect, and as to what sort of thing it can mean, we\nare in a better position to discover the relation of the two which is\nmeaning.\n\nThe things that words mean differ more than words do. There are\ndifferent sorts of words, distinguished by the grammarians; and there\nare logical distinctions, which are connected to some extent, though not\nso closely as was formerly supposed, with the grammatical distinctions\nof parts of speech. It is easy, however, to be misled by grammar,\nparticularly if all the languages we know belong to one family. In some\nlanguages, according to some authorities, the distinction of parts of\nspeech does not exist; in many languages it is widely different from\nthat to which we are accustomed in the Indo-European languages. These\nfacts have to be borne in mind if we are to avoid giving metaphysical\nimportance to mere accidents of our own speech.\n\nIn considering what words mean, it is natural to start with proper\nnames, and we will again take \"Napoleon\" as our instance. We commonly\nimagine, when we use a proper name, that we mean one definite entity,\nthe particular individual who was called \"Napoleon.\" But what we know\nas a person is not simple. There MAY be a single simple ego which was\nNapoleon, and remained strictly identical from his birth to his death.\nThere is no way of proving that this cannot be the case, but there is\nalso not the slightest reason to suppose that it is the case. Napoleon\nas he was empirically known consisted of a series of gradually changing\nappearances: first a squalling baby, then a boy, then a slim and\nbeautiful youth, then a fat and slothful person very magnificently\ndressed This series of appearances, and various occurrences having\ncertain kinds of causal connections with them, constitute Napoleon as\nempirically known, and therefore are Napoleon in so far as he forms\npart of the experienced world. Napoleon is a complicated series of\noccurrences, bound together by causal laws, not, like instances of a\nword, by similarities. For although a person changes gradually, and\npresents similar appearances on two nearly contemporaneous occasions,\nit is not these similarities that constitute the person, as appears from\nthe \"Comedy of Errors\" for example.\n\nThus in the case of a proper name, while the word is a set of similar\nseries of movements, what it means is a series of occurrences bound\ntogether by causal laws of that special kind that makes the occurrences\ntaken together constitute what we call one person, or one animal or\nthing, in case the name applies to an animal or thing instead of to\na person. Neither the word nor what it names is one of the ultimate\nindivisible constituents of the world. In language there is no direct\nway of designating one of the ultimate brief existents that go to make\nup the collections we call things or persons. If we want to speak of\nsuch existents--which hardly happens except in philosophy--we have to do\nit by means of some elaborate phrase, such as \"the visual sensation\nwhich occupied the centre of my field of vision at noon on January 1,\n1919.\" Such ultimate simples I call \"particulars.\" Particulars MIGHT\nhave proper names, and no doubt would have if language had been invented\nby scientifically trained observers for purposes of philosophy and\nlogic. But as language was invented for practical ends, particulars have\nremained one and all without a name.\n\nWe are not, in practice, much concerned with the actual particulars\nthat come into our experience in sensation; we are concerned rather\nwith whole systems to which the particulars belong and of which they\nare signs. What we see makes us say \"Hullo, there's Jones,\" and the fact\nthat what we see is a sign of Jones (which is the case because it is one\nof the particulars that make up Jones) is more interesting to us than\nthe actual particular itself. Hence we give the name \"Jones\" to the\nwhole set of particulars, but do not trouble to give separate names to\nthe separate particulars that make up the set.\n\nPassing on from proper names, we come next to general names, such as\n\"man,\" \"cat,\" \"triangle.\" A word such as \"man\" means a whole class\nof such collections of particulars as have proper names. The several\nmembers of the class are assembled together in virtue of some similarity\nor common property. All men resemble each other in certain important\nrespects; hence we want a word which shall be equally applicable to all\nof them. We only give proper names to the individuals of a species when\nthey differ inter se in practically important respects. In other cases\nwe do not do this. A poker, for instance, is just a poker; we do not\ncall one \"John\" and another \"Peter.\"\n\nThere is a large class of words, such as \"eating,\" \"walking,\"\n\"speaking,\" which mean a set of similar occurrences. Two instances of\nwalking have the same name because they resemble each other, whereas\ntwo instances of Jones have the same name because they are causally\nconnected. In practice, however, it is difficult to make any precise\ndistinction between a word such as \"walking\" and a general name such as\n\"man.\" One instance of walking cannot be concentrated into an instant:\nit is a process in time, in which there is a causal connection between\nthe earlier and later parts, as between the earlier and later parts\nof Jones. Thus an instance of walking differs from an instance of man\nsolely by the fact that it has a shorter life. There is a notion that an\ninstance of walking, as compared with Jones, is unsubstantial, but this\nseems to be a mistake. We think that Jones walks, and that there could\nnot be any walking unless there were somebody like Jones to perform\nthe walking. But it is equally true that there could be no Jones unless\nthere were something like walking for him to do. The notion that actions\nare performed by an agent is liable to the same kind of criticism as\nthe notion that thinking needs a subject or ego, which we rejected in\nLecture I. To say that it is Jones who is walking is merely to say that\nthe walking in question is part of the whole series of occurrences which\nis Jones. There is no LOGICAL impossibility in walking occurring as an\nisolated phenomenon, not forming part of any such series as we call a\n\"person.\"\n\nWe may therefore class with \"eating,\" \"walking,\" \"speaking\" words\nsuch as \"rain,\" \"sunrise,\" \"lightning,\" which do not denote what would\ncommonly be called actions. These words illustrate, incidentally, how\nlittle we can trust to the grammatical distinction of parts of speech,\nsince the substantive \"rain\" and the verb \"to rain\" denote precisely the\nsame class of meteorological occurrences. The distinction between the\nclass of objects denoted by such a word and the class of objects denoted\nby a general name such as \"man,\" \"vegetable,\" or \"planet,\" is that the\nsort of object which is an instance of (say) \"lightning\" is much simpler\nthan (say) an individual man. (I am speaking of lightning as a sensible\nphenomenon, not as it is described in physics.) The distinction is one\nof degree, not of kind. But there is, from the point of view of ordinary\nthought, a great difference between a process which, like a flash of\nlightning, can be wholly comprised within one specious present and a\nprocess which, like the life of a man, has to be pieced together by\nobservation and memory and the apprehension of causal connections.\nWe may say broadly, therefore, that a word of the kind we have been\ndiscussing denotes a set of similar occurrences, each (as a rule) much\nmore brief and less complex than a person or thing. Words themselves, as\nwe have seen, are sets of similar occurrences of this kind. Thus there\nis more logical affinity between a word and what it means in the case of\nwords of our present sort than in any other case.\n\nThere is no very great difference between such words as we have just\nbeen considering and words denoting qualities, such as \"white\" or\n\"round.\" The chief difference is that words of this latter sort do not\ndenote processes, however brief, but static features of the world. Snow\nfalls, and is white; the falling is a process, the whiteness is not.\nWhether there is a universal, called \"whiteness,\" or whether white\nthings are to be defined as those having a certain kind of similarity to\na standard thing, say freshly fallen snow, is a question which need\nnot concern us, and which I believe to be strictly insoluble. For our\npurposes, we may take the word \"white\" as denoting a certain set of\nsimilar particulars or collections of particulars, the similarity being\nin respect of a static quality, not of a process.\n\nFrom the logical point of view, a very important class of words\nare those that express relations, such as \"in,\" \"above,\" \"before,\"\n\"greater,\" and so on. The meaning of one of these words differs very\nfundamentally from the meaning of one of any of our previous classes,\nbeing more abstract and logically simpler than any of them. If our\nbusiness were logic, we should have to spend much time on these words.\nBut as it is psychology that concerns us, we will merely note their\nspecial character and pass on, since the logical classification of words\nis not our main business.\n\nWe will consider next the question what is implied by saying that a\nperson \"understands\" a word, in the sense in which one understands\na word in one's own language, but not in a language of which one is\nignorant. We may say that a person understands a word when (a) suitable\ncircumstances make him use it, (b) the hearing of it causes suitable\nbehaviour in him. We may call these two active and passive understanding\nrespectively. Dogs often have passive understanding of some words, but\nnot active understanding, since they cannot use words.\n\nIt is not necessary, in order that a man should \"understand\" a word,\nthat he should \"know what it means,\" in the sense of being able to say\n\"this word means so-and-so.\" Understanding words does not consist in\nknowing their dictionary definitions, or in being able to specify the\nobjects to which they are appropriate. Such understanding as this may\nbelong to lexicographers and students, but not to ordinary mortals\nin ordinary life. Understanding language is more like understanding\ncricket*: it is a matter of habits, acquired in oneself and rightly\npresumed in others. To say that a word has a meaning is not to say that\nthose who use the word correctly have ever thought out what the meaning\nis: the use of the word comes first, and the meaning is to be distilled\nout of it by observation and analysis. Moreover, the meaning of a word\nis not absolutely definite: there is always a greater or less degree of\nvagueness. The meaning is an area, like a target: it may have a bull's\neye, but the outlying parts of the target are still more or less within\nthe meaning, in a gradually diminishing degree as we travel further from\nthe bull's eye. As language grows more precise, there is less and less\nof the target outside the bull's eye, and the bull's eye itself grows\nsmaller and smaller; but the bull's eye never shrinks to a point, and\nthere is always a doubtful region, however small, surrounding it.**\n\n     * This point of view, extended to the analysis of \"thought\"\n     is urged with great force by J. B. Watson, both in his\n     \"Behavior,\" and in \"Psychology from the Standpoint of a\n     Behaviorist\" (Lippincott. 1919), chap. ix.\n\n     ** On the understanding of words, a very admirable little\n     book is Ribot's \"Evolution of General Ideas,\" Open Court\n     Co., 1899. Ribot says (p. 131): \"We learn to understand a\n     concept as we learn to walk, dance, fence or play a musical\n     instrument: it is a habit, i.e. an organized memory. General\n     terms cover an organized, latent knowledge which is the\n     hidden capital without which we should be in a state of\n     bankruptcy, manipulating false money or paper of no value.\n     General ideas are habits in the intellectual order.\"\n\nA word is used \"correctly\" when the average hearer will be affected\nby it in the way intended. This is a psychological, not a literary,\ndefinition of \"correctness.\" The literary definition would substitute,\nfor the average hearer, a person of high education living a long time\nago; the purpose of this definition is to make it difficult to speak or\nwrite correctly.\n\nThe relation of a word to its meaning is of the nature of a causal law\ngoverning our use of the word and our actions when we hear it used.\nThere is no more reason why a person who uses a word correctly should\nbe able to tell what it means than there is why a planet which is moving\ncorrectly should know Kepler's laws.\n\nTo illustrate what is meant by \"understanding\" words and sentences, let\nus take instances of various situations.\n\nSuppose you are walking in London with an absent-minded friend, and\nwhile crossing a street you say, \"Look out, there's a motor coming.\"\nHe will glance round and jump aside without the need of any \"mental\"\nintermediary. There need be no \"ideas,\" but only a stiffening of the\nmuscles, followed quickly by action. He \"understands\" the words, because\nhe does the right thing. Such \"understanding\" may be taken to belong to\nthe nerves and brain, being habits which they have acquired while the\nlanguage was being learnt. Thus understanding in this sense may be\nreduced to mere physiological causal laws.\n\nIf you say the same thing to a Frenchman with a slight knowledge of\nEnglish he will go through some inner speech which may be represented by\n\"Que dit-il? Ah, oui, une automobile!\" After this, the rest follows as\nwith the Englishman. Watson would contend that the inner speech must be\nincipiently pronounced; we should argue that it MIGHT be merely imaged.\nBut this point is not important in the present connection.\n\nIf you say the same thing to a child who does not yet know the word\n\"motor,\" but does know the other words you are using, you produce a\nfeeling of anxiety and doubt you will have to point and say, \"There,\nthat's a motor.\" After that the child will roughly understand the word\n\"motor,\" though he may include trains and steam-rollers If this is the\nfirst time the child has heard the word \"motor,\" he may for a long time\ncontinue to recall this scene when he hears the word.\n\nSo far we have found four ways of understanding words:\n\n(1) On suitable occasions you use the word properly.\n\n(2) When you hear it you act appropriately.\n\n(3) You associate the word with another word (say in a different\nlanguage) which has the appropriate effect on behaviour.\n\n(4) When the word is being first learnt, you may associate it with an\nobject, which is what it \"means,\" or a representative of various objects\nthat it \"means.\"\n\nIn the fourth case, the word acquires, through association, some of the\nsame causal efficacy as the object. The word \"motor\" can make you\nleap aside, just as the motor can, but it cannot break your bones. The\neffects which a word can share with its object are those which proceed\naccording to laws other than the general laws of physics, i.e. those\nwhich, according to our terminology, involve vital movements as opposed\nto merely mechanical movements. The effects of a word that we understand\nare always mnemic phenomena in the sense explained in Lecture IV, in\nso far as they are identical with, or similar to, the effects which the\nobject itself might have.\n\nSo far, all the uses of words that we have considered can be accounted\nfor on the lines of behaviourism.\n\nBut so far we have only considered what may be called the\n\"demonstrative\" use of language, to point out some feature in the\npresent environment. This is only one of the ways in which language\nmay be used. There are also its narrative and imaginative uses, as\nin history and novels. Let us take as an instance the telling of some\nremembered event.\n\nWe spoke a moment ago of a child who hears the word \"motor\" for\nthe first time when crossing a street along which a motor-car is\napproaching. On a later occasion, we will suppose, the child remembers\nthe incident and relates it to someone else. In this case, both the\nactive and passive understanding of words is different from what it is\nwhen words are used demonstratively. The child is not seeing a motor,\nbut only remembering one; the hearer does not look round in expectation\nof seeing a motor coming, but \"understands\" that a motor came at some\nearlier time. The whole of this occurrence is much more difficult to\naccount for on behaviourist lines. It is clear that, in so far as the\nchild is genuinely remembering, he has a picture of the past occurrence,\nand his words are chosen so as to describe the picture; and in so far\nas the hearer is genuinely apprehending what is said, the hearer is\nacquiring a picture more or less like that of the child. It is true that\nthis process may be telescoped through the operation of the word-habit.\nThe child may not genuinely remember the incident, but only have the\nhabit of the appropriate words, as in the case of a poem which we know\nby heart, though we cannot remember learning it. And the hearer also\nmay only pay attention to the words, and not call up any corresponding\npicture. But it is, nevertheless, the possibility of a memory-image in\nthe child and an imagination-image in the hearer that makes the essence\nof the narrative \"meaning\" of the words. In so far as this is absent,\nthe words are mere counters, capable of meaning, but not at the moment\npossessing it.\n\nYet this might perhaps be regarded as something of an overstatement. The\nwords alone, without the use of images, may cause appropriate emotions\nand appropriate behaviour. The words have been used in an environment\nwhich produced certain emotions; by a telescoped process, the words\nalone are now capable of producing similar emotions. On these lines it\nmight be sought to show that images are unnecessary. I do not believe,\nhowever, that we could account on these lines for the entirely different\nresponse produced by a narrative and by a description of present facts.\nImages, as contrasted with sensations, are the response expected during\na narrative; it is understood that present action is not called\nfor. Thus it seems that we must maintain our distinction words used\ndemonstratively describe and are intended to lead to sensations, while\nthe same words used in narrative describe and are only intended to lead\nto images.\n\nWe have thus, in addition to our four previous ways in which words can\nmean, two new ways, namely the way of memory and the way of imagination.\nThat is to say:\n\n(5) Words may be used to describe or recall a memory-image: to describe\nit when it already exists, or to recall it when the words exist as a\nhabit and are known to be descriptive of some past experience.\n\n(6) Words may be used to describe or create an imagination-image: to\ndescribe it, for example, in the case of a poet or novelist, or to\ncreate it in the ordinary case for giving information-though, in the\nlatter case, it is intended that the imagination-image, when created,\nshall be accompanied by belief that something of the sort occurred.\n\nThese two ways of using words, including their occurrence in inner\nspeech, may be spoken of together as the use of words in \"thinking.\"\nIf we are right, the use of words in thinking depends, at least in its\norigin, upon images, and cannot be fully dealt with on behaviourist\nlines. And this is really the most essential function of words, namely\nthat, originally through their connection with images, they bring us\ninto touch with what is remote in time or space. When they operate\nwithout the medium of images, this seems to be a telescoped process.\nThus the problem of the meaning of words is brought into connection with\nthe problem of the meaning of images.\n\nTo understand the function that words perform in what is called\n\"thinking,\" we must understand both the causes and the effects of their\noccurrence. The causes of the occurrence of words require somewhat\ndifferent treatment according as the object designated by the word is\nsensibly present or absent. When the object is present, it may itself\nbe taken as the cause of the word, through association. But when it is\nabsent there is more difficulty in obtaining a behaviourist theory of\nthe occurrence of the word. The language-habit consists not merely\nin the use of words demonstratively, but also in their use to express\nnarrative or desire. Professor Watson, in his account of the acquisition\nof the language-habit, pays very little attention to the use of words in\nnarrative and desire. He says (\"Behavior,\" pp. 329-330):\n\n\"The stimulus (object) to which the child often responds, a box, e.g. by\nmovements such as opening and closing and putting objects into it, may\nserve to illustrate our argument. The nurse, observing that the child\nreacts with his hands, feet, etc., to the box, begins to say 'box' when\nthe child is handed the box, 'open box' when the child opens it, 'close\nbox' when he closes it, and 'put doll in box' when that act is executed.\nThis is repeated over and over again. In the process of time it comes\nabout that without any other stimulus than that of the box which\noriginally called out the bodily habits, he begins to say 'box' when he\nsees it, 'open box' when he opens it, etc. The visible box now becomes\na stimulus capable of releasing either the bodily habits or the\nword-habit, i.e. development has brought about two things: (1) a series\nof functional connections among arcs which run from visual receptor to\nmuscles of throat, and (2) a series of already earlier connected arcs\nwhich run from the same receptor to the bodily muscles.... The object\nmeets the child's vision. He runs to it and tries to reach it and says\n'box.'... Finally the word is uttered without the movement of going\ntowards the box being executed.... Habits are formed of going to the\nbox when the arms are full of toys. The child has been taught to deposit\nthem there. When his arms are laden with toys and no box is there, the\nword-habit arises and he calls 'box'; it is handed to him, and he opens\nit and deposits the toys therein. This roughly marks what we would call\nthe genesis of a true language-habit.\"(pp. 329-330).*\n\n     * Just the same account of language is given in Professor\n     Watson's more recent book (reference above).\n\nWe need not linger over what is said in the above passage as to the use\nof the word \"box\" in the presence of the box. But as to its use in the\nabsence of the box, there is only one brief sentence, namely: \"When his\narms are laden with toys and no box is there, the word-habit arises and\nhe calls 'box.'\" This is inadequate as it stands, since the habit has\nbeen to use the word when the box is present, and we have to explain its\nextension to cases in which the box is absent.\n\nHaving admitted images, we may say that the word \"box,\" in the absence\nof the box, is caused by an image of the box. This may or may not\nbe true--in fact, it is true in some cases but not in others. Even,\nhowever, if it were true in all cases, it would only slightly shift our\nproblem: we should now have to ask what causes an image of the box to\narise. We might be inclined to say that desire for the box is the cause.\nBut when this view is investigated, it is found that it compels us to\nsuppose that the box can be desired without the child's having either an\nimage of the box or the word \"box.\" This will require a theory of desire\nwhich may be, and I think is, in the main true, but which removes desire\nfrom among things that actually occur, and makes it merely a convenient\nfiction, like force in mechanics.* With such a view, desire is no longer\na true cause, but merely a short way of describing certain processes.\n\n     * See Lecture III, above.\n\nIn order to explain the occurrence of either the word or the image\nin the absence of the box, we have to assume that there is something,\neither in the environment or in our own sensations, which has frequently\noccurred at about the same time as the word \"box.\" One of the laws which\ndistinguish psychology (or nerve-physiology?) from physics is the\nlaw that, when two things have frequently existed in close temporal\ncontiguity, either comes in time to cause the other.* This is the basis\nboth of habit and of association. Thus, in our case, the arms full of\ntoys have frequently been followed quickly by the box, and the box in\nturn by the word \"box.\" The box itself is subject to physical laws, and\ndoes not tend to be caused by the arms full of toys, however often it\nmay in the past have followed them--always provided that, in the case in\nquestion, its physical position is such that voluntary movements cannot\nlead to it. But the word \"box\" and the image of the box are subject to\nthe law of habit; hence it is possible for either to be caused by the\narms full of toys. And we may lay it down generally that, whenever we\nuse a word, either aloud or in inner speech, there is some sensation\nor image (either of which may be itself a word) which has frequently\noccurred at about the same time as the word, and now, through habit,\ncauses the word. It follows that the law of habit is adequate to account\nfor the use of words in the absence of their objects; moreover, it would\nbe adequate even without introducing images. Although, therefore, images\nseem undeniable, we cannot derive an additional argument in their favour\nfrom the use of words, which could, theoretically, be explained without\nintroducing images.\n\n     *For a more exact statement of this law, with the\n     limitations suggested by experiment, see A. Wohlgemuth, \"On\n     Memory and the Direction of Associations,\" \"British Journal\n     of Psychology,\" vol. v, part iv (March, 1913).\n\nWhen we understand a word, there is a reciprocal association between\nit and the images of what it \"means.\" Images may cause us to use words\nwhich mean them, and these words, heard or read, may in turn cause the\nappropriate images. Thus speech is a means of producing in our hearers\nthe images which are in us. Also, by a telescoped process, words come in\ntime to produce directly the effects which would have been produced\nby the images with which they were associated. The general law of\ntelescoped processes is that, if A causes B and B causes C, it will\nhappen in time that A will cause C directly, without the intermediary\nof B. This is a characteristic of psychological and neural causation.\nIn virtue of this law, the effects of images upon our actions come to\nbe produced by words, even when the words do not call up appropriate\nimages. The more familiar we are with words, the more our \"thinking\"\ngoes on in words instead of images. We may, for example, be able to\ndescribe a person's appearance correctly without having at any time had\nany image of him, provided, when we saw him, we thought of words which\nfitted him; the words alone may remain with us as a habit, and enable\nus to speak as if we could recall a visual image of the man. In this and\nother ways the understanding of a word often comes to be quite free from\nimagery; but in first learning the use of language it would seem that\nimagery always plays a very important part.\n\nImages as well as words may be said to have \"meaning\"; indeed, the\nmeaning of images seems more primitive than the meaning of words. What\nwe call (say) an image of St. Paul's may be said to \"mean\" St. Paul's.\nBut it is not at all easy to say exactly what constitutes the meaning of\nan image. A memory-image of a particular occurrence, when accompanied\nby a memory-belief, may be said to mean the occurrence of which it is an\nimage. But most actual images do not have this degree of definiteness.\nIf we call up an image of a dog, we are very likely to have a vague\nimage, which is not representative of some one special dog, but of dogs\nin general. When we call up an image of a friend's face, we are not\nlikely to reproduce the expression he had on some one particular\noccasion, but rather a compromise expression derived from many\noccasions. And there is hardly any limit to the vagueness of which\nimages are capable. In such cases, the meaning of the image, if defined\nby relation to the prototype, is vague: there is not one definite\nprototype, but a number, none of which is copied exactly.*\n\n     * Cf. Semon, Mnemische Empfindungen, chap. xvi, especially\n     pp. 301-308.\n\nThere is, however, another way of approaching the meaning of images,\nnamely through their causal efficacy. What is called an image \"of\"\nsome definite object, say St. Paul's, has some of the effects which the\nobject would have. This applies especially to the effects that depend\nupon association. The emotional effects, also, are often similar:\nimages may stimulate desire almost as strongly as do the objects they\nrepresent. And conversely desire may cause images*: a hungry man will\nhave images of food, and so on. In all these ways the causal laws\nconcerning images are connected with the causal laws concerning the\nobjects which the images \"mean.\" An image may thus come to fulfil the\nfunction of a general idea. The vague image of a dog, which we spoke of\na moment ago, will have effects which are only connected with dogs in\ngeneral, not the more special effects which would be produced by some\ndogs but not by others. Berkeley and Hume, in their attack on general\nideas, do not allow for the vagueness of images: they assume that every\nimage has the definiteness that a physical object would have This is not\nthe case, and a vague image may well have a meaning which is general.\n\n     * This phrase is in need of interpretation, as appears from\n     the analysis of desire. But the reader can easily supply the\n     interpretation for himself.\n\nIn order to define the \"meaning\" of an image, we have to take account\nboth of its resemblance to one or more prototypes, and of its causal\nefficacy. If there were such a thing as a pure imagination-image,\nwithout any prototype whatever, it would be destitute of meaning. But\naccording to Hume's principle, the simple elements in an image,\nat least, are derived from prototypes-except possibly in very rare\nexceptional cases. Often, in such instances as our image of a friend's\nface or of a nondescript dog, an image is not derived from one\nprototype, but from many; when this happens, the image is vague, and\nblurs the features in which the various prototypes differ. To arrive\nat the meaning of the image in such a case, we observe that there are\ncertain respects, notably associations, in which the effects of images\nresemble those of their prototypes. If we find, in a given case, that\nour vague image, say, of a nondescript dog, has those associative\neffects which all dogs would have, but not those belonging to any\nspecial dog or kind of dog, we may say that our image means \"dog\" in\ngeneral. If it has all the associations appropriate to spaniels but\nno others, we shall say it means \"spaniel\"; while if it has all the\nassociations appropriate to one particular dog, it will mean that dog,\nhowever vague it may be as a picture. The meaning of an image, according\nto this analysis, is constituted by a combination of likeness and\nassociations. It is not a sharp or definite conception, and in many\ncases it will be impossible to decide with any certainty what an image\nmeans. I think this lies in the nature of things, and not in defective\nanalysis.\n\nWe may give somewhat more precision to the above account of the meaning\nof images, and extend it to meaning in general. We find sometimes that,\nIN MNEMIC CAUSATION, an image or word, as stimulus, has the same effect\n(or very nearly the same effect) as would belong to some object, say,\na certain dog. In that case we say that the image or word means that\nobject. In other cases the mnemic effects are not all those of one\nobject, but only those shared by objects of a certain kind, e.g. by all\ndogs. In this case the meaning of the image or word is general: it means\nthe whole kind. Generality and particularity are a matter of degree. If\ntwo particulars differ sufficiently little, their mnemic effects will be\nthe same; therefore no image or word can mean the one as opposed to the\nother; this sets a bound to the particularity of meaning. On the other\nhand, the mnemic effects of a number of sufficiently dissimilar objects\nwill have nothing discoverable in common; hence a word which aims at\ncomplete generality, such as \"entity\" for example, will have to be\ndevoid of mnemic effects, and therefore of meaning. In practice, this is\nnot the case: such words have VERBAL associations, the learning of which\nconstitutes the study of metaphysics.\n\nThe meaning of a word, unlike that of an image, is wholly constituted\nby mnemic causal laws, and not in any degree by likeness (except in\nexceptional cases). The word \"dog\" bears no resemblance to a dog, but\nits effects, like those of an image of a dog, resemble the effects of an\nactual dog in certain respects. It is much easier to say definitely\nwhat a word means than what an image means, since words, however they\noriginated, have been framed in later times for the purpose of having\nmeaning, and men have been engaged for ages in giving increased\nprecision to the meanings of words. But although it is easier to\nsay what a word means than what an image means, the relation which\nconstitutes meaning is much the same in both cases. A word, like an\nimage, has the same associations as its meaning has. In addition to\nother associations, it is associated with images of its meaning, so that\nthe word tends to call up the image and the image tends to call up the\nword., But this association is not essential to the intelligent use of\nwords. If a word has the right associations with other objects, we shall\nbe able to use it correctly, and understand its use by others, even if\nit evokes no image. The theoretical understanding of words involves only\nthe power of associating them correctly with other words; the practical\nunderstanding involves associations with other bodily movements.\n\nThe use of words is, of course, primarily social, for the purpose of\nsuggesting to others ideas which we entertain or at least wish them to\nentertain. But the aspect of words that specially concerns us is their\npower of promoting our own thought. Almost all higher intellectual\nactivity is a matter of words, to the nearly total exclusion of\neverything else. The advantages of words for purposes of thought are so\ngreat that I should never end if I were to enumerate them. But a few of\nthem deserve to be mentioned.\n\nIn the first place, there is no difficulty in producing a word, whereas\nan image cannot always be brought into existence at will, and when it\ncomes it often contains much irrelevant detail. In the second place,\nmuch of our thinking is concerned with abstract matters which do not\nreadily lend themselves to imagery, and are apt to be falsely conceived\nif we insist upon finding images that may be supposed to represent them.\nThe word is always concrete and sensible, however abstract its\nmeaning may be, and thus by the help of words we are able to dwell on\nabstractions in a way which would otherwise be impossible. In the third\nplace, two instances of the same word are so similar that neither has\nassociations not capable of being shared by the other. Two instances of\nthe word \"dog\" are much more alike than (say) a pug and a great dane;\nhence the word \"dog\" makes it much easier to think about dogs in\ngeneral. When a number of objects have a common property which is\nimportant but not obvious, the invention of a name for the common\nproperty helps us to remember it and to think of the whole set of\nobjects that possess it. But it is unnecessary to prolong the catalogue\nof the uses of language in thought.\n\nAt the same time, it is possible to conduct rudimentary thought by\nmeans of images, and it is important, sometimes, to check purely verbal\nthought by reference to what it means. In philosophy especially the\ntyranny of traditional words is dangerous, and we have to be on our\nguard against assuming that grammar is the key to metaphysics, or that\nthe structure of a sentence corresponds at all accurately with the\nstructure of the fact that it asserts. Sayce maintained that all\nEuropean philosophy since Aristotle has been dominated by the fact that\nthe philosophers spoke Indo-European languages, and therefore supposed\nthe world, like the sentences they were used to, necessarily divisible\ninto subjects and predicates. When we come to the consideration of truth\nand falsehood, we shall see how necessary it is to avoid assuming too\nclose a parallelism between facts and the sentences which assert them.\nAgainst such errors, the only safeguard is to be able, once in a way, to\ndiscard words for a moment and contemplate facts more directly through\nimages. Most serious advances in philosophic thought result from some\nsuch comparatively direct contemplation of facts. But the outcome has\nto be expressed in words if it is to be communicable. Those who have\na relatively direct vision of facts are often incapable of translating\ntheir vision into words, while those who possess the words have\nusually lost the vision. It is partly for this reason that the highest\nphilosophical capacity is so rare: it requires a combination of vision\nwith abstract words which is hard to achieve, and too quickly lost in\nthe few who have for a moment achieved it.\n\n\n\n\nLECTURE XI. GENERAL IDEAS AND THOUGHT\n\nIt is said to be one of the merits of the human mind that it is capable\nof framing abstract ideas, and of conducting nonsensational thought.\nIn this it is supposed to differ from the mind of animals. From Plato\nonward the \"idea\" has played a great part in the systems of idealizing\nphilosophers. The \"idea\" has been, in their hands, always something\nnoble and abstract, the apprehension and use of which by man confers\nupon him a quite special dignity.\n\nThe thing we have to consider to-day is this: seeing that there\ncertainly are words of which the meaning is abstract, and seeing that we\ncan use these words intelligently, what must be assumed or inferred, or\nwhat can be discovered by observation, in the way of mental content to\naccount for the intelligent use of abstract words?\n\nTaken as a problem in logic, the answer is, of course, that absolutely\nnothing in the way of abstract mental content is inferable from the\nmere fact that we can use intelligently words of which the meaning\nis abstract. It is clear that a sufficiently ingenious person could\nmanufacture a machine moved by olfactory stimuli which, whenever a dog\nappeared in its neighbourhood, would say, \"There is a dog,\" and when\na cat appeared would throw stones at it. The act of saying \"There is a\ndog,\" and the act of throwing stones, would in such a case be equally\nmechanical. Correct speech does not of itself afford any better evidence\nof mental content than the performance of any other set of biologically\nuseful movements, such as those of flight or combat. All that is\ninferable from language is that two instances of a universal, even when\nthey differ very greatly, may cause the utterance of two instances\nof the same word which only differ very slightly. As we saw in the\npreceding lecture, the word \"dog\" is useful, partly, because two\ninstances of this word are much more similar than (say) a pug and a\ngreat dane. The use of words is thus a method of substituting for two\nparticulars which differ widely, in spite of being instances of the same\nuniversal, two other particulars which differ very little, and which\nare also instances of a universal, namely the name of the previous\nuniversal. Thus, so far as logic is concerned, we are entirely free to\nadopt any theory as to general ideas which empirical observation may\nrecommend.\n\nBerkeley and Hume made a vigorous onslaught on \"abstract ideas.\" They\nmeant by an idea approximately what we should call an image. Locke\nhaving maintained that he could form an idea of triangle in general,\nwithout deciding what sort of triangle it was to be, Berkeley contended\nthat this was impossible. He says:\n\n\"Whether others, have this wonderful faculty of abstracting their ideas,\nthey best can tell: for myself, I dare be confident I have it not. I\nfind, indeed, I have indeed a faculty of imagining, or representing to\nmyself, the ideas of those particular things I have perceived, and of\nvariously compounding and dividing them. I can imagine a man with two\nheads, or the upper parts of a man joined to the body of a horse. I\ncan consider the hand, the eye, the nose, each by itself abstracted or\nseparated from the rest of the body. But, then, whatever hand or eye\nI imagine, it must have some particular shape and colour. Likewise the\nidea of a man that I frame to myself must be either of a white, or a\nblack, or a tawny, a straight, or a crooked, a tall, or a low, or\na middle-sized man. I cannot by any effort of thought conceive the\nabstract idea above described. And it is equally impossible for me to\nform the abstract idea of motion distinct from the body moving, and\nwhich is neither swift nor slow, curvilinear nor rectilinear; and the\nlike may be said of all other abstract general ideas whatsoever. To be\nplain, I own myself able to abstract in one sense, as when I consider\nsome particular parts of qualities separated from others, with which,\nthough they are united in some object, yet it is possible they may\nreally exist without them. But I deny that I can abstract from one\nanother, or conceive separately, those qualities which it is impossible\nshould exist so separated; or that I can frame a general notion, by\nabstracting from particulars in the manner aforesaid--which last are\nthe two proper acceptations of ABSTRACTION. And there is ground to think\nmost men will acknowledge themselves to be in my case. The generality of\nmen which are simple and illiterate never pretend to ABSTRACT NOTIONS.\nIt is said they are difficult and not to be attained without pains and\nstudy; we may therefore reasonably conclude that, if such there be, they\nare confined only to the learned.\n\n\"I proceed to examine what can be alleged in defence of the doctrine of\nabstraction, and try if I can discover what it is that inclines the men\nof speculation to embrace an opinion so remote from common sense as that\nseems to be. There has been a late excellent and deservedly esteemed\nphilosopher who, no doubt, has given it very much countenance, by\nseeming to think the having abstract general ideas is what puts the\nwidest difference in point of understanding betwixt man and beast.\n'The having of general ideas,' saith he, 'is that which puts a perfect\ndistinction betwixt man and brutes, and is an excellency which the\nfaculties of brutes do by no means attain unto. For, it is evident\nwe observe no footsteps in them of making use of general signs for\nuniversal ideas; from which we have reason to imagine that they have not\nthe faculty of abstracting, or making general ideas, since they have\nno use of words or any other general signs.' And a little after:\n'Therefore, I think, we may suppose that it is in this that the species\nof brutes are discriminated from men, and it is that proper difference\nwherein they are wholly separated, and which at last widens to so wide a\ndistance. For, if they have any ideas at all, and are not bare machines\n(as some would have them), we cannot deny them to have some reason. It\nseems as evident to me that they do, some of them, in certain instances\nreason as that they have sense; but it is only in particular ideas, just\nas they receive them from their senses. They are the best of them tied\nup within those narrow bounds, and have not (as I think) the faculty\nto enlarge them by any kind of abstraction.* (\"Essay on Human\nUnderstanding,\" Bk. II, chap. xi, paragraphs 10 and 11.) I readily agree\nwith this learned author, that the faculties of brutes can by no means\nattain to abstraction. But, then, if this be made the distinguishing\nproperty of that sort of animals, I fear a great many of those that\npass for men must be reckoned into their number. The reason that is here\nassigned why we have no grounds to think brutes have abstract general\nideas is, that we observe in them no use of words or any other general\nsigns; which is built on this supposition-that the making use of words\nimplies the having general ideas. From which it follows that men who use\nlanguage are able to abstract or generalize their ideas. That this is\nthe sense and arguing of the author will further appear by his answering\nthe question he in another place puts: 'Since all things that exist are\nonly particulars, how come we by general terms?' His answer is: 'Words\nbecome general by being made the signs of general ideas.' (\"Essay on\nHuman Understanding,\" Bk. III, chap. III, paragraph 6.) But it seems\nthat a word becomes general by being made the sign, not of an abstract\ngeneral idea, but of several particular ideas, any one of which it\nindifferently suggests to the mind. For example, when it is said 'the\nchange of motion is proportional to the impressed force,' or that\n'whatever has extension is divisible,' these propositions are to be\nunderstood of motion and extension in general; and nevertheless it will\nnot follow that they suggest to my thoughts an idea of motion without\na body moved, or any determinate direction and velocity, or that I must\nconceive an abstract general idea of extension, which is neither line,\nsurface, nor solid, neither great nor small, black, white, nor red,\nnor of any other determinate colour. It is only implied that\nwhatever particular motion I consider, whether it be swift or slow,\nperpendicular, horizontal, or oblique, or in whatever object, the axiom\nconcerning it holds equally true. As does the other of every particular\nextension, it matters not whether line, surface, or solid, whether of\nthis or that magnitude or figure.\n\n\"By observing how ideas become general, we may the better judge how\nwords are made so. And here it is to be noted that I do not deny\nabsolutely there are general ideas, but only that there are any ABSTRACT\ngeneral ideas; for, in the passages we have quoted wherein there is\nmention of general ideas, it is always supposed that they are formed by\nabstraction, after the manner set forth in sections 8 and 9. Now, if\nwe will annex a meaning to our words, and speak only of what we can\nconceive, I believe we shall acknowledge that an idea which, considered\nin itself, is particular, becomes general by being made to represent\nor stand for all other particular ideas of the same sort. To make this\nplain by an example, suppose a geometrician is demonstrating the method\nof cutting a line in two equal parts. He draws, for instance, a black\nline of an inch in length: this, which in itself is a particular line,\nis nevertheless with regard to its signification general, since, as it\nis there used, it represents all particular lines whatsoever; so that\nwhat is demonstrated of it is demonstrated of all lines, or, in other\nwords, of a line in general. And, as THAT PARTICULAR LINE becomes\ngeneral by being made a sign, so the NAME 'line,' which taken absolutely\nis particular, by being a sign is made general. And as the former owes\nits generality not to its being the sign of an abstract or general line,\nbut of all particular right lines that may possibly exist, so the latter\nmust be thought to derive its generality from the same cause, namely,\nthe various particular lines which it indifferently denotes.\" *\n\n     * Introduction to \"A Treatise concerning the Principles of\n     Human Knowledge,\" paragraphs 10, 11, and 12.\n\nBerkeley's view in the above passage, which is essentially the same as\nHume's, does not wholly agree with modern psychology, although it comes\nnearer to agreement than does the view of those who believe that there\nare in the mind single contents which can be called abstract ideas. The\nway in which Berkeley's view is inadequate is chiefly in the fact that\nimages are as a rule not of one definite prototype, but of a number of\nrelated similar prototypes. On this subject Semon has written well.\nIn \"Die Mneme,\" pp. 217 ff., discussing the effect of repeated similar\nstimuli in producing and modifying our images, he says: \"We choose a\ncase of mnemic excitement whose existence we can perceive for ourselves\nby introspection, and seek to ekphore the bodily picture of our nearest\nrelation in his absence, and have thus a pure mnemic excitement before\nus. At first it may seem to us that a determinate quite concrete picture\nbecomes manifest in us, but just when we are concerned with a person\nwith whom we are in constant contact, we shall find that the ekphored\npicture has something so to speak generalized. It is something like\nthose American photographs which seek to display what is general about a\ntype by combining a great number of photographs of different heads over\neach other on one plate. In our opinion, the generalizations happen by\nthe homophonic working of different pictures of the same face which we\nhave come across in the most different conditions and situations, once\npale, once reddened, once cheerful, once earnest, once in this\nlight, and once in that. As soon as we do not let the whole series\nof repetitions resound in us uniformly, but give our attention to one\nparticular moment out of the many... this particular mnemic stimulus at\nonce overbalances its simultaneously roused predecessors and successors,\nand we perceive the face in question with concrete definiteness in that\nparticular situation.\" A little later he says: \"The result is--at least\nin man, but probably also in the higher animals--the development of a\nsort of PHYSIOLOGICAL abstraction. Mnemic homophony gives us, without\nthe addition of other processes of thought, a picture of our friend\nX which is in a certain sense abstract, not the concrete in any one\nsituation, but X cut loose from any particular point of time. If the\ncircle of ekphored engrams is drawn even more widely, abstract pictures\nof a higher order appear: for instance, a white man or a negro. In my\nopinion, the first form of abstract concepts in general is based upon\nsuch abstract pictures. The physiological abstraction which takes\nplace in the above described manner is a predecessor of purely logical\nabstraction. It is by no means a monopoly of the human race, but shows\nitself in various ways also among the more highly organized animals.\"\nThe same subject is treated in more detail in Chapter xvi of \"Die\nmnemischen Empfindungen,\" but what is said there adds nothing vital to\nwhat is contained in the above quotations.\n\nIt is necessary, however, to distinguish between the vague and the\ngeneral. So long as we are content with Semon's composite image, we MAY\nget no farther than the vague. The question whether this image takes us\nto the general or not depends, I think, upon the question whether, in\naddition to the generalized image, we have also particular images\nof some of the instances out of which it is compounded. Suppose, for\nexample, that on a number of occasions you had seen one negro, and\nthat you did not know whether this one was the same or different on\nthe different occasions. Suppose that in the end you had an abstract\nmemory-image of the different appearances presented by the negro on\ndifferent occasions, but no memory-image of any one of the single\nappearances. In that case your image would be vague. If, on the other\nhand, you have, in addition to the generalized image, particular images\nof the several appearances, sufficiently clear to be recognized as\ndifferent, and as instances of the generalized picture, you will then\nnot feel the generalized picture to be adequate to any one particular\nappearance, and you will be able to make it function as a general\nidea rather than a vague idea. If this view is correct, no new general\ncontent needs to be added to the generalized image. What needs to be\nadded is particular images compared and contrasted with the generalized\nimage. So far as I can judge by introspection, this does occur in\npractice. Take for example Semon's instance of a friend's face. Unless\nwe make some special effort of recollection, the face is likely to come\nbefore us with an average expression, very blurred and vague, but we can\nat will recall how our friend looked on some special occasion when he\nwas pleased or angry or unhappy, and this enables us to realize the\ngeneralized character of the vague image.\n\nThere is, however, another way of distinguishing between the vague, the\nparticular and the general, and this is not by their content, but by\nthe reaction which they produce. A word, for example, may be said to be\nvague when it is applicable to a number of different individuals, but to\neach as individuals; the name Smith, for example, is vague: it is always\nmeant to apply to one man, but there are many men to each of whom it\napplies.* The word \"man,\" on the other hand, is general. We say, \"This\nis Smith,\" but we do not say \"This is man,\" but \"This is a man.\" Thus\nwe may say that a word embodies a vague idea when its effects are\nappropriate to an individual, but are the same for various similar\nindividuals, while a word embodies a general idea when its effects are\ndifferent from those appropriate to individuals. In what this difference\nconsists it is, however, not easy to say. I am inclined to think that it\nconsists merely in the knowledge that no one individual is represented,\nso that what distinguishes a general idea from a vague idea is merely\nthe presence of a certain accompanying belief. If this view is correct,\na general idea differs from a vague one in a way analogous to that in\nwhich a memory-image differs from an imagination-image. There also\nwe found that the difference consists merely of the fact that a\nmemory-image is accompanied by a belief, in this case as to the past.\n\n     * \"Smith\" would only be a quite satisfactory representation\n     of vague words if we failed to discriminate between\n     different people called Smith.\n\nIt should also be said that our images even of quite particular\noccurrences have always a greater or a less degree of vagueness. That is\nto say, the occurrence might have varied within certain limits without\ncausing our image to vary recognizably. To arrive at the general it\nis necessary that we should be able to contrast it with a number of\nrelatively precise images or words for particular occurrences; so long\nas all our images and words are vague, we cannot arrive at the contrast\nby which the general is defined. This is the justification for the\nview which I quoted on p. 184 from Ribot (op. cit., p. 32), viz. that\nintelligence progresses from the indefinite to the definite, and that\nthe vague appears earlier than either the particular or the general.\n\nI think the view which I have been advocating, to the effect that a\ngeneral idea is distinguished from a vague one by the presence of a\njudgment, is also that intended by Ribot when he says (op. cit., p. 92):\n\"The generic image is never, the concept is always, a judgment. We know\nthat for logicians (formerly at any rate) the concept is the simple\nand primitive element; next comes the judgment, uniting two or several\nconcepts; then ratiocination, combining two or several judgments. For\nthe psychologists, on the contrary, affirmation is the fundamental\nact; the concept is the result of judgment (explicit or implicit), of\nsimilarities with exclusion of differences.\"\n\nA great deal of work professing to be experimental has been done in\nrecent years on the psychology of thought. A good summary of such\nwork up to the year agog is contained in Titchener's \"Lectures on the\nExperimental Psychology of the Thought Processes\" (1909). Three articles\nin the \"Archiv fur die gesammte Psychologie\" by Watt,* Messer** and\nBuhler*** contain a great deal of the material amassed by the methods\nwhich Titchener calls experimental.\n\n     * Henry J. Watt, \"Experimentelle Beitrage zu einer Theorie\n     des Denkens,\" vol. iv (1905) pp. 289-436.\n\n     ** August Messer, \"Experimentell-psychologische Untersuchu\n     gen uber das Denken,\" vol. iii (1906), pp. 1-224.\n\n     *** Karl Buhler, \"Uber Gedanken,\" vol. ix (1907), pp. 297-365.\n\nFor my part I am unable to attach as much importance to this work as\nmany psychologists do. The method employed appears to me hardly to\nfulfil the conditions of scientific experiment. Broadly speaking, what\nis done is, that a set of questions are asked of various people, their\nanswers are recorded, and likewise their own accounts, based upon\nintrospection, of the processes of thought which led them to give\nthose answers. Much too much reliance seems to me to be placed upon the\ncorrectness of their introspection. On introspection as a method I have\nspoken earlier (Lecture VI). I am not prepared, like Professor Watson,\nto reject it wholly, but I do consider that it is exceedingly fallible\nand quite peculiarly liable to falsification in accordance with\npreconceived theory. It is like depending upon the report of a\nshortsighted person as to whom he sees coming along the road at a moment\nwhen he is firmly convinced that Jones is sure to come. If everybody\nwere shortsighted and obsessed with beliefs as to what was going to be\nvisible, we might have to make the best of such testimony, but we should\nneed to correct its errors by taking care to collect the simultaneous\nevidence of people with the most divergent expectations. There is no\nevidence that this was done in the experiments in question, nor indeed\nthat the influence of theory in falsifying the introspection was at all\nadequately recognized. I feel convinced that if Professor Watson had\nbeen one of the subjects of the questionnaires, he would have given\nanswers totally different from those recorded in the articles in\nquestion. Titchener quotes an opinion of Wundt on these investigations,\nwhich appears to me thoroughly justified. \"These experiments,\" he says,\n\"are not experiments at all in the sense of a scientific methodology;\nthey are counterfeit experiments, that seem methodical simply because\nthey are ordinarily performed in a psychological laboratory, and involve\nthe co-operation of two persons, who purport to be experimenter and\nobserver. In reality, they are as unmethodical as possible; they possess\nnone of the special features by which we distinguish the introspections\nof experimental psychology from the casual introspections of everyday\nlife.\"* Titchener, of course, dissents from this opinion, but I cannot\nsee that his reasons for dissent are adequate. My doubts are only\nincreased by the fact that Buhler at any rate used trained psychologists\nas his subjects. A trained psychologist is, of course, supposed to have\nacquired the habit of observation, but he is at least equally likely to\nhave acquired a habit of seeing what his theories require. We may take\nBuhler's \"Uber Gedanken\" to illustrate the kind of results arrived at\nby such methods. Buhler says (p. 303): \"We ask ourselves the general\nquestion: 'WHAT DO WE EXPERIENCE WHEN WE THINK?' Then we do not at all\nattempt a preliminary determination of the concept 'thought,' but choose\nfor analysis only such processes as everyone would describe as\nprocesses of thought.\" The most important thing in thinking, he says, is\n\"awareness that...\" (Bewusstheit dass), which he calls a thought. It\nis, he says, thoughts in this sense that are essential to thinking.\nThinking, he maintains, does not need language or sensuous\npresentations. \"I assert rather that in principle every object can be\nthought (meant) distinctly, without any help from sensuous presentation\n(Anschauungshilfen). Every individual shade of blue colour on the\npicture that hangs in my room I can think with complete distinctness\nunsensuously (unanschaulich), provided it is possible that the object\nshould be given to me in another manner than by the help of sensations.\nHow that is possible we shall see later.\" What he calls a thought\n(Gedanke) cannot be reduced, according to him, to other psychic\noccurrences. He maintains that thoughts consist for the most part of\nknown rules (p. 342). It is clearly essential to the interest of this\ntheory that the thought or rule alluded to by Buhler should not need to\nbe expressed in words, for if it is expressed in words it is immediately\ncapable of being dealt with on the lines with which the behaviourists\nhave familiarized us. It is clear also that the supposed absence of\nwords rests solely upon the introspective testimony of the persons\nexperimented upon. I cannot think that there is sufficient certainty\nof their reliability in this negative observation to make us accept a\ndifficult and revolutionary view of thought, merely because they have\nfailed to observe the presence of words or their equivalent in their\nthinking. I think it far more likely, especially in view of the fact\nthat the persons concerned were highly educated, that we are concerned\nwith telescoped processes, in which habit has caused a great many\nintermediate terms to be elided or to be passed over so quickly as to\nescape observation.\n\n     * Titchener, op. cit., p. 79.\n\nI am inclined to think that similar remarks apply to the general idea of\n\"imageless thinking,\" concerning which there has been much controversy.\nThe advocates of imageless thinking are not contending merely that there\ncan be thinking which is purely verbal; they are contending that there\ncan be thinking which proceeds neither in words nor in images. My own\nfeeling is that they have rashly assumed the presence of thinking in\ncases where habit has rendered thinking unnecessary. When Thorndike\nexperimented with animals in cages, he found that the associations\nestablished were between a sensory stimulus and a bodily movement (not\nthe idea of it), without the need of supposing any non-physiological\nintermediary (op. cit., p. 100 ff.). The same thing, it seems to me,\napplies to ourselves. A certain sensory situation produces in us a\ncertain bodily movement. Sometimes this movement consists in uttering\nwords. Prejudice leads us to suppose that between the sensory\nstimulus and the utterance of the words a process of thought must have\nintervened, but there seems no good reason for such a supposition. Any\nhabitual action, such as eating or dressing, may be performed on the\nappropriate occasion, without any need of thought, and the same seems\nto be true of a painfully large proportion of our talk. What applies to\nuttered speech applies of course equally to the internal speech which is\nnot uttered. I remain, therefore, entirely unconvinced that there is\nany such phenomenon as thinking which consists neither of images nor of\nwords, or that \"ideas\" have to be added to sensations and images as part\nof the material out of which mental phenomena are built.\n\nThe question of the nature of our consciousness of the universal is\nmuch affected by our view as to the general nature of the relation of\nconsciousness to its object. If we adopt the view of Brentano, according\nto which all mental content has essential reference to an object, it\nis then natural to suppose that there is some peculiar kind of mental\ncontent of which the object is a universal, as oppose to a particular.\nAccording to this view, a particular cat can be PERceived or imagined,\nwhile the universal \"cat\" is CONceived. But this whole manner of viewing\nour dealings with universals has to be abandoned when the relation of\na mental occurrence to its \"object\" is regarded as merely indirect and\ncausal, which is the view that we have adopted. The mental content is,\nof course, always particular, and the question as to what it \"means\"\n(in case it means anything) is one which cannot be settled by merely\nexamining the intrinsic character of the mental content, but only by\nknowing its causal connections in the case of the person concerned. To\nsay that a certain thought \"means\" a universal as opposed to either a\nvague or a particular, is to say something exceedingly complex. A horse\nwill behave in a certain manner whenever he smells a bear, even if\nthe smell is derived from a bearskin. That is to say, any environment\ncontaining an instance of the universal \"smell of a bear\" produces\nclosely similar behaviour in the horse, but we do not say that the horse\nis conscious of this universal. There is equally little reason to\nregard a man as conscious of the same universal, because under the same\ncircumstances he can react by saying, \"I smell a bear.\" This reaction,\nlike that of the horse, is merely closely similar on different occasions\nwhere the environment affords instances of the same universal. Words\nof which the logical meaning is universal can therefore be employed\ncorrectly, without anything that could be called consciousness of\nuniversals. Such consciousness in the only sense in which it can be\nsaid to exist is a matter of reflective judgment consisting in the\nobservation of similarities and differences. A universal never appears\nbefore the mind as a single object in the sort of way in which something\nperceived appears. I THINK a logical argument could be produced to show\nthat universals are part of the structure of the world, but they are\nan inferred part, not a part of our data. What exists in us consists of\nvarious factors, some open to external observation, others only visible\nto introspection. The factors open to external observation are primarily\nhabits, having the peculiarity that very similar reactions are produced\nby stimuli which are in many respects very different from each other. Of\nthis the reaction of the horse to the smell of the bear is an instance,\nand so is the reaction of the man who says \"bear\" under the same\ncircumstances. The verbal reaction is, of course, the most important\nfrom the point of view of what may be called knowledge of universals. A\nman who can always use the word \"dog\" when he sees a dog may be said,\nin a certain sense, to know the meaning of the word \"dog,\" and IN THAT\nSENSE to have knowledge of the universal \"dog.\" But there is, of course,\na further stage reached by the logician in which he not merely reacts\nwith the word \"dog,\" but sets to work to discover what it is in the\nenvironment that causes in him this almost identical reaction on\ndifferent occasions. This further stage consists in knowledge of\nsimilarities and differences: similarities which are necessary to the\napplicability of the word \"dog,\" and differences which are compatible\nwith it. Our knowledge of these similarities and differences is never\nexhaustive, and therefore our knowledge of the meaning of a universal is\nnever complete.\n\nIn addition to external observable habits (including the habit of\nwords), there is also the generic image produced by the superposition,\nor, in Semon's phrase, homophony, of a number of similar perceptions.\nThis image is vague so long as the multiplicity of its prototypes is not\nrecognized, but becomes universal when it exists alongside of the more\nspecific images of its instances, and is knowingly contrasted with them.\nIn this case we find again, as we found when we were discussing words\nin general in the preceding lecture, that images are not logically\nnecessary in order to account for observable behaviour, i.e. in this\ncase intelligent speech. Intelligent speech could exist as a motor\nhabit, without any accompaniment of images, and this conclusion applies\nto words of which the meaning is universal, just as much as to words of\nwhich the meaning is relatively particular. If this conclusion is valid,\nit follows that behaviourist psychology, which eschews introspective\ndata, is capable of being an independent science, and of accounting\nfor all that part of the behaviour of other people which is commonly\nregarded as evidence that they think. It must be admitted that this\nconclusion considerably weakens the reliance which can be placed upon\nintrospective data. They must be accepted simply on account of the\nfact that we seem to perceive them, not on account of their supposed\nnecessity for explaining the data of external observation.\n\nThis, at any rate, is the conclusion to which we are forced, so long\nas, with the behaviourists, we accept common-sense views of the physical\nworld. But if, as I have urged, the physical world itself, as known,\nis infected through and through with subjectivity, if, as the theory\nof relativity suggests, the physical universe contains the diversity of\npoints of view which we have been accustomed to regard as distinctively\npsychological, then we are brought back by this different road to the\nnecessity for trusting observations which are in an important sense\nprivate. And it is the privacy of introspective data which causes much\nof the behaviourists' objection to them.\n\nThis is an example of the difficulty of constructing an adequate\nphilosophy of any one science without taking account of other sciences.\nThe behaviourist philosophy of psychology, though in many respects\nadmirable from the point of view of method, appears to me to fail in\nthe last analysis because it is based upon an inadequate philosophy of\nphysics. In spite, therefore, of the fact that the evidence for images,\nwhether generic or particular, is merely introspective, I cannot\nadmit that images should be rejected, or that we should minimize their\nfunction in our knowledge of what is remote in time or space.\n\n\n\n\nLECTURE XII. BELIEF\n\nBelief, which is our subject to-day, is the central problem in the\nanalysis of mind. Believing seems the most \"mental\" thing we do,\nthe thing most remote from what is done by mere matter. The whole\nintellectual life consists of beliefs, and of the passage from one\nbelief to another by what is called \"reasoning.\" Beliefs give knowledge\nand error; they are the vehicles of truth and falsehood. Psychology,\ntheory of knowledge and metaphysics revolve about belief, and on the\nview we take of belief our philosophical outlook largely depends.\n\nBefore embarking upon the detailed analysis of belief, we shall do well\nto note certain requisites which any theory must fulfil.\n\n(1) Just as words are characterized by meaning, so beliefs are\ncharacterized by truth or falsehood. And just as meaning consists in\nrelation to the object meant, so truth and falsehood consist in\nrelation to something that lies outside the belief. You may believe that\nsuch-and-such a horse will win the Derby. The time comes, and your horse\nwins or does not win; according to the outcome, your belief was true or\nfalse. You may believe that six times nine is fifty-six; in this case\nalso there is a fact which makes your belief false. You may believe that\nAmerica was discovered in 1492, or that it was discovered in 1066. In\nthe one case your belief is true, in the other false; in either case\nits truth or falsehood depends upon the actions of Columbus, not upon\nanything present or under your control. What makes a belief true or\nfalse I call a \"fact.\" The particular fact that makes a given belief\ntrue or false I call its \"objective,\"* and the relation of the belief to\nits objective I call the \"reference\" or the \"objective reference\" of the\nbelief. Thus, if I believe that Columbus crossed the Atlantic in 1492,\nthe \"objective\" of my belief is Columbus's actual voyage, and the\n\"reference\" of my belief is the relation between my belief and the\nvoyage--that relation, namely, in virtue of which the voyage makes my\nbelief true (or, in another case, false). \"Reference\" of beliefs differs\nfrom \"meaning\" of words in various ways, but especially in the fact that\nit is of two kinds, \"true\" reference and \"false\" reference. The truth\nor falsehood of a belief does not depend upon anything intrinsic to\nthe belief, but upon the nature of its relation to its objective. The\nintrinsic nature of belief can be treated without reference to what\nmakes it true or false. In the remainder of the present lecture I shall\nignore truth and falsehood, which will be the subject of Lecture XIII.\nIt is the intrinsic nature of belief that will concern us to-day.\n\n     * This terminology is suggested by Meinong, but is not\n     exactly the same as his.\n\n(2) We must distinguish between believing and what is believed. I may\nbelieve that Columbus crossed the Atlantic, that all Cretans are liars,\nthat two and two are four, or that nine times six is fifty-six; in\nall these cases the believing is just the same, and only the contents\nbelieved are different. I may remember my breakfast this morning, my\nlecture last week, or my first sight of New York. In all these cases the\nfeeling of memory-belief is just the same, and only what is remembered\ndiffers. Exactly similar remarks apply to expectations. Bare assent,\nmemory and expectation are forms of belief; all three are different from\nwhat is believed, and each has a constant character which is independent\nof what is believed.\n\nIn Lecture I we criticized the analysis of a presentation into act,\ncontent and object. But our analysis of belief contains three very\nsimilar elements, namely the believing, what is believed and the\nobjective. The objections to the act (in the case of presentations)\nare not valid against the believing in the case of beliefs, because the\nbelieving is an actual experienced feeling, not something postulated,\nlike the act. But it is necessary first to complete our preliminary\nrequisites, and then to examine the content of a belief. After that, we\nshall be in a position to return to the question as to what constitutes\nbelieving.\n\n(3) What is believed, and the believing, must both consist of present\noccurrences in the believer, no matter what may be the objective of\nthe belief. Suppose I believe, for example, \"that Caesar crossed the\nRubicon.\" The objective of my belief is an event which happened long\nago, which I never saw and do not remember. This event itself is not in\nmy mind when I believe that it happened. It is not correct to say that\nI am believing the actual event; what I am believing is something now\nin my mind, something related to the event (in a way which we shall\ninvestigate in Lecture XIII), but obviously not to be confounded with\nthe event, since the event is not occurring now but the believing is.\nWhat a man is believing at a given moment is wholly determinate if we\nknow the contents of his mind at that moment; but Caesar's crossing of\nthe Rubicon was an historical physical event, which is distinct from the\npresent contents of every present mind. What is believed, however true\nit may be, is not the actual fact that makes the belief true, but a\npresent event related to the fact. This present event, which is what is\nbelieved, I shall call the \"content\" of the belief. We have already had\noccasion to notice the distinction between content and objective in the\ncase of memory-beliefs, where the content is \"this occurred\" and the\nobjective is the past event.\n\n(4) Between content and objective there is sometimes a very wide gulf,\nfor example in the case of \"Caesar crossed the Rubicon.\" This gulf may,\nwhen it is first perceived, give us a feeling that we cannot really\n\"know\" anything about the outer world. All we can \"know,\" it may be\nsaid, is what is now in our thoughts. If Caesar and the Rubicon cannot\nbe bodily in our thoughts, it might seem as though we must remain cut\noff from knowledge of them. I shall not now deal at length with this\nfeeling, since it is necessary first to define \"knowing,\" which cannot\nbe done yet. But I will say, as a preliminary answer, that the feeling\nassumes an ideal of knowing which I believe to be quite mistaken. It\nassumes, if it is thought out, something like the mystic unity of knower\nand known. These two are often said to be combined into a unity by the\nfact of cognition; hence when this unity is plainly absent, it may\nseem as if there were no genuine cognition. For my part, I think such\ntheories and feelings wholly mistaken: I believe knowing to be a very\nexternal and complicated relation, incapable of exact definition,\ndependent upon causal laws, and involving no more unity than there is\nbetween a signpost and the town to which it points. I shall return to\nthis question on a later occasion; for the moment these provisional\nremarks must suffice.\n\n(5) The objective reference of a belief is connected with the fact that\nall or some of the constituents of its content have meaning. If I say\n\"Caesar conquered Gaul,\" a person who knows the meaning of the three\nwords composing my statement knows as much as can be known about the\nnature of the objective which would make my statement true. It is clear\nthat the objective reference of a belief is, in general, in some way\nderivative from the meanings of the words or images that occur in its\ncontent. There are, however, certain complications which must be borne\nin mind. In the first place, it might be contended that a memory-image\nacquires meaning only through the memory-belief, which would seem, at\nleast in the case of memory, to make belief more primitive than the\nmeaning of images. In the second place, it is a very singular thing that\nmeaning, which is single, should generate objective reference, which is\ndual, namely true and false. This is one of the facts which any theory\nof belief must explain if it is to be satisfactory.\n\nIt is now time to leave these preliminary requisites, and attempt the\nanalysis of the contents of beliefs.\n\nThe first thing to notice about what is believed, i.e. about the content\nof a belief, is that it is always complex: We believe that a certain\nthing has a certain property, or a certain relation to something else,\nor that it occurred or will occur (in the sense discussed at the end of\nLecture IX); or we may believe that all the members of a certain class\nhave a certain property, or that a certain property sometimes occurs\namong the members of a class; or we may believe that if one thing\nhappens, another will happen (for example, \"if it rains I shall bring my\numbrella\"), or we may believe that something does not happen, or did not\nor will not happen (for example, \"it won't rain\"); or that one of two\nthings must happen (for example, \"either you withdraw your accusation,\nor I shall bring a libel action\"). The catalogue of the sorts of things\nwe may believe is infinite, but all of them are complex.\n\nLanguage sometimes conceals the complexity of a belief. We say that a\nperson believes in God, and it might seem as if God formed the whole\ncontent of the belief. But what is really believed is that God exists,\nwhich is very far from being simple. Similarly, when a person has a\nmemory-image with a memory-belief, the belief is \"this occurred,\" in\nthe sense explained in Lecture IX; and \"this occurred\" is not simple.\nIn like manner all cases where the content of a belief seems simple at\nfirst sight will be found, on examination, to confirm the view that the\ncontent is always complex.\n\nThe content of a belief involves not merely a plurality of constituents,\nbut definite relations between them; it is not determinate when its\nconstituents alone are given. For example, \"Plato preceded Aristotle\"\nand \"Aristotle preceded Plato\" are both contents which may be believed,\nbut, although they consist of exactly the same constituents, they are\ndifferent, and even incompatible.\n\nThe content of a belief may consist of words only, or of images only, or\nof a mixture of the two, or of either or both together with one or more\nsensations. It must contain at least one constituent which is a word\nor an image, and it may or may not contain one or more sensations as\nconstituents. Some examples will make these various possibilities clear.\n\nWe may take first recognition, in either of the forms \"this is of\nsuch-and-such a kind\" or \"this has occurred before.\" In either case,\npresent sensation is a constituent. For example, you hear a noise, and\nyou say to yourself \"tram.\" Here the noise and the word \"tram\" are both\nconstituents of your belief; there is also a relation between them,\nexpressed by \"is\" in the proposition \"that is a tram.\" As soon as your\nact of recognition is completed by the occurrence of the word \"tram,\"\nyour actions are affected: you hurry if you want the tram, or cease to\nhurry if you want a bus. In this case the content of your belief is a\nsensation (the noise) and a word (\"tram\") related in a way which may be\ncalled predication.\n\nThe same noise may bring into your mind the visual image of a tram,\ninstead of the word \"tram.\" In this case your belief consists of a\nsensation and an image suitable related. Beliefs of this class are what\nare called \"judgments of perception.\" As we saw in Lecture VIII, the\nimages associated with a sensation often come with such spontaneity\nand force that the unsophisticated do not distinguish them from the\nsensation; it is only the psychologist or the skilled observer who is\naware of the large mnemic element that is added to sensation to make\nperception. It may be objected that what is added consists merely of\nimages without belief. This is no doubt sometimes the case, but\nis certainly sometimes not the case. That belief always occurs in\nperception as opposed to sensation it is not necessary for us to\nmaintain; it is enough for our purposes to note that it sometimes\noccurs, and that when it does, the content of our belief consists of a\nsensation and an image suitably related.\n\nIn a PURE memory-belief only images occur. But a mixture of words\nand images is very common in memory. You have an image of the past\noccurrence, and you say to yourself: \"Yes, that's how it was.\" Here the\nimage and the words together make up the content of the belief. And\nwhen the remembering of an incident has become a habit, it may be purely\nverbal, and the memory-belief may consist of words alone.\n\nThe more complicated forms of belief tend to consist only of words.\nOften images of various kinds accompany them, but they are apt to\nbe irrelevant, and to form no part of what is actually believed. For\nexample, in thinking of the Solar System, you are likely to have vague\nimages of pictures you have seen of the earth surrounded by clouds,\nSaturn and his rings, the sun during an eclipse, and so on; but none of\nthese form part of your belief that the planets revolve round the sun\nin elliptical orbits. The only images that form an actual part of such\nbeliefs are, as a rule, images of words. And images of words, for the\nreasons considered in Lecture VIII, cannot be distinguished with any\ncertainty from sensations, when, as is often, if not usually, the case,\nthey are kinaesthetic images of pronouncing the words.\n\nIt is impossible for a belief to consist of sensations alone, except\nwhen, as in the case of words, the sensations have associations which\nmake them signs possessed of meaning. The reason is that objective\nreference is of the essence of belief, and objective reference is\nderived from meaning. When I speak of a belief consisting partly of\nsensations and partly of words, I do not mean to deny that the words,\nwhen they are not mere images, are sensational, but that they occur as\nsigns, not (so to speak) in their own right. To revert to the noise of\nthe tram, when you hear it and say \"tram,\" the noise and the word are\nboth sensations (if you actually pronounce the word), but the noise is\npart of the fact which makes your belief true, whereas the word is not\npart of this fact. It is the MEANING of the word \"tram,\" not the actual\nword, that forms part of the fact which is the objective of your\nbelief. Thus the word occurs in the belief as a symbol, in virtue of\nits meaning, whereas the noise enters into both the belief and its\nobjective. It is this that distinguishes the occurrence of words as\nsymbols from the occurrence of sensations in their own right: the\nobjective contains the sensations that occur in their own right, but\ncontains only the meanings of the words that occur as symbols.\n\nFor the sake of simplicity, we may ignore the cases in which sensations\nin their own right form part of the content of a belief, and confine\nourselves to images and words. We may also omit the cases in which\nboth images and words occur in the content of a belief. Thus we become\nconfined to two cases: (a) when the content consists wholly of images,\n(b) when it consists wholly of words. The case of mixed images and words\nhas no special importance, and its omission will do no harm.\n\nLet us take in illustration a case of memory. Suppose you are thinking\nof some familiar room. You may call up an image of it, and in your image\nthe window may be to the left of the door. Without any intrusion of\nwords, you may believe in the correctness of your image. You then have a\nbelief, consisting wholly of images, which becomes, when put into words,\n\"the window is to the left of the door.\" You may yourself use these\nwords and proceed to believe them. You thus pass from an image-content\nto the corresponding word-content. The content is different in the two\ncases, but its objective reference is the same. This shows the relation\nof image-beliefs to word-beliefs in a very simple case. In more\nelaborate cases the relation becomes much less simple.\n\nIt may be said that even in this very simple case the objective\nreference of the word-content is not quite the same as that of the\nimage-content, that images have a wealth of concrete features which are\nlost when words are substituted, that the window in the image is not a\nmere window in the abstract, but a window of a certain shape and size,\nnot merely to the left of the door, but a certain distance to the left,\nand so on. In reply, it may be admitted at once that there is, as a\nrule, a certain amount of truth in the objection. But two points may be\nurged to minimize its force. First, images do not, as a rule, have that\nwealth of concrete detail that would make it IMPOSSIBLE to express\nthem fully in words. They are vague and fragmentary: a finite number\nof words, though perhaps a large number, would exhaust at least their\nSIGNIFICANT features. For--and this is our second point--images enter\ninto the content of a belief through the fact that they are capable of\nmeaning, and their meaning does not, as a rule, have as much complexity\nas they have: some of their characteristics are usually devoid of\nmeaning. Thus it may well be possible to extract in words all that\nhas meaning in an image-content; in that case the word-content and the\nimage-content will have exactly the same objective reference.\n\nThe content of a belief, when expressed in words, is the same thing (or\nvery nearly the same thing) as what in logic is called a \"proposition.\"\nA proposition is a series of words (or sometimes a single word)\nexpressing the kind of thing that can be asserted or denied. \"That all\nmen are mortal,\" \"that Columbus discovered America,\" \"that Charles I\ndied in his bed,\" \"that all philosophers are wise,\" are propositions.\nNot any series of words is a proposition, but only such series of words\nas have \"meaning,\" or, in our phraseology, \"objective reference.\" Given\nthe meanings of separate words, and the rules of syntax, the meaning of\na proposition is determinate. This is the reason why we can understand\na sentence we never heard before. You probably never heard before the\nproposition \"that the inhabitants of the Andaman Islands habitually\neat stewed hippopotamus for dinner,\" but there is no difficulty in\nunderstanding the proposition. The question of the relation between\nthe meaning of a sentence and the meanings of the separate words is\ndifficult, and I shall not pursue it now; I brought it up solely as\nbeing illustrative of the nature of propositions.\n\nWe may extend the term \"proposition\" so as to cover the image-contents\nof beliefs consisting of images. Thus, in the case of remembering a\nroom in which the window is to the left of the door, when we believe the\nimage-content the proposition will consist of the image of the window\non the left together with the image of the door on the right. We will\ndistinguish propositions of this kind as \"image-propositions\"\nand propositions in words as \"word-propositions.\" We may identify\npropositions in general with the contents of actual and possible\nbeliefs, and we may say that it is propositions that are true or false.\nIn logic we are concerned with propositions rather than beliefs, since\nlogic is not interested in what people do in fact believe, but only\nin the conditions which determine the truth or falsehood of possible\nbeliefs. Whenever possible, except when actual beliefs are in question,\nit is generally a simplification to deal with propositions.\n\nIt would seem that image-propositions are more primitive than\nword-propositions, and may well ante-date language. There is no reason\nwhy memory-images, accompanied by that very simple belief-feeling which\nwe decided to be the essence of memory, should not have occurred before\nlanguage arose; indeed, it would be rash to assert positively that\nmemory of this sort does not occur among the higher animals. Our more\nelementary beliefs, notably those that are added to sensation to make\nperception, often remain at the level of images. For example, most of\nthe visual objects in our neighbourhood rouse tactile images: we have a\ndifferent feeling in looking at a sofa from what we have in looking at\na block of marble, and the difference consists chiefly in different\nstimulation of our tactile imagination. It may be said that the tactile\nimages are merely present, without any accompanying belief; but I think\nthis view, though sometimes correct, derives its plausibility as a\ngeneral proposition from our thinking of explicit conscious belief only.\nMost of our beliefs, like most of our wishes, are \"unconscious,\" in the\nsense that we have never told ourselves that we have them. Such beliefs\ndisplay themselves when the expectations that they arouse fail in any\nway. For example, if someone puts tea (without milk) into a glass, and\nyou drink it under the impression that it is going to be beer; or if you\nwalk on what appears to be a tiled floor, and it turns out to be a soft\ncarpet made to look like tiles. The shock of surprise on an occasion of\nthis kind makes us aware of the expectations that habitually enter into\nour perceptions; and such expectations must be classed as beliefs, in\nspite of the fact that we do not normally take note of them or put them\ninto words. I remember once watching a cock pigeon running over and over\nagain to the edge of a looking-glass to try to wreak vengeance on the\nparticularly obnoxious bird whom he expected to find there, judging by\nwhat he saw in the glass. He must have experienced each time the sort of\nsurprise on finding nothing, which is calculated to lead in time to\nthe adoption of Berkeley's theory that objects of sense are only in the\nmind. His expectation, though not expressed in words, deserved, I think,\nto be called a belief.\n\nI come now to the question what constitutes believing, as opposed to the\ncontent believed.\n\nTo begin with, there are various different attitudes that may be taken\ntowards the same content. Let us suppose, for the sake of argument, that\nyou have a visual image of your breakfast-table. You may expect it while\nyou are dressing in the morning; remember it as you go to your work;\nfeel doubt as to its correctness when questioned as to your powers of\nvisualizing; merely entertain the image, without connecting it with\nanything external, when you are going to sleep; desire it if you are\nhungry, or feel aversion for it if you are ill. Suppose, for the sake of\ndefiniteness, that the content is \"an egg for breakfast.\" Then you have\nthe following attitudes \"I expect there will be an egg for breakfast\";\n\"I remember there was an egg for breakfast\"; \"Was there an egg for\nbreakfast?\" \"An egg for breakfast: well, what of it?\" \"I hope there\nwill be an egg for breakfast\"; \"I am afraid there will be an egg for\nbreakfast and it is sure to be bad.\" I do not suggest that this is a\nlist of all possible attitudes on the subject; I say only that they\nare different attitudes, all concerned with the one content \"an egg for\nbreakfast.\"\n\nThese attitudes are not all equally ultimate. Those that involve desire\nand aversion have occupied us in Lecture III. For the present, we are\nonly concerned with such as are cognitive. In speaking of memory, we\ndistinguished three kinds of belief directed towards the same\ncontent, namely memory, expectation and bare assent without any\ntime-determination in the belief-feeling. But before developing this\nview, we must examine two other theories which might be held concerning\nbelief, and which, in some ways, would be more in harmony with a\nbehaviourist outlook than the theory I wish to advocate.\n\n(1) The first theory to be examined is the view that the differentia of\nbelief consists in its causal efficacy I do not wish to make any author\nresponsible for this theory: I wish merely to develop it hypothetically\nso that we may judge of its tenability.\n\nWe defined the meaning of an image or word by causal efficacy, namely by\nassociations: an image or word acquires meaning, we said, through having\nthe same associations as what it means.\n\nWe propose hypothetically to define \"belief\" by a different kind\nof causal efficacy, namely efficacy in causing voluntary movements.\n(Voluntary movements are defined as those vital movements which are\ndistinguished from reflex movements as involving the higher nervous\ncentres. I do not like to distinguish them by means of such notions as\n\"consciousness\" or \"will,\" because I do not think these notions, in any\ndefinable sense, are always applicable. Moreover, the purpose of the\ntheory we are examining is to be, as far as possible, physiological and\nbehaviourist, and this purpose is not achieved if we introduce such a\nconception as \"consciousness\" or \"will.\" Nevertheless, it is necessary\nfor our purpose to find some way of distinguishing between voluntary and\nreflex movements, since the results would be too paradoxical, if we were\nto say that reflex movements also involve beliefs.) According to this\ndefinition, a content is said to be \"believed\" when it causes us to\nmove. The images aroused are the same if you say to me, \"Suppose there\nwere an escaped tiger coming along the street,\" and if you say to me,\n\"There is an escaped tiger coming along the street.\" But my actions will\nbe very different in the two cases: in the first, I shall remain calm;\nin the second, it is possible that I may not. It is suggested, by the\ntheory we are considering, that this difference of effects constitutes\nwhat is meant by saying that in the second case I believe the\nproposition suggested, while in the first case I do not. According\nto this view, images or words are \"believed\" when they cause bodily\nmovements.\n\nI do not think this theory is adequate, but I think it is suggestive\nof truth, and not so easily refutable as it might appear to be at first\nsight.\n\nIt might be objected to the theory that many things which we certainly\nbelieve do not call for any bodily movements. I believe that Great\nBritain is an island, that whales are mammals, that Charles I was\nexecuted, and so on; and at first sight it seems obvious that such\nbeliefs, as a rule, do not call for any action on my part. But when we\ninvestigate the matter more closely, it becomes more doubtful. To begin\nwith, we must distinguish belief as a mere DISPOSITION from actual\nactive belief. We speak as if we always believed that Charles I was\nexecuted, but that only means that we are always ready to believe it\nwhen the subject comes up. The phenomenon we are concerned to analyse\nis the active belief, not the permanent disposition. Now, what are\nthe occasions when, we actively believe that Charles I was executed?\nPrimarily: examinations, when we perform the bodily movement of writing\nit down; conversation, when we assert it to display our historical\nerudition; and political discourses, when we are engaged in showing what\nSoviet government leads to. In all these cases bodily movements (writing\nor speaking) result from our belief.\n\nBut there remains the belief which merely occurs in \"thinking.\" One may\nset to work to recall some piece of history one has been reading, and\nwhat one recalls is believed, although it probably does not cause any\nbodily movement whatever. It is true that what we believe always MAY\ninfluence action. Suppose I am invited to become King of Georgia: I find\nthe prospect attractive, and go to Cook's to buy a third-class ticket to\nmy new realm. At the last moment I remember Charles I and all the other\nmonarchs who have come to a bad end; I change my mind, and walk out\nwithout completing the transaction. But such incidents are rare, and\ncannot constitute the whole of my belief that Charles I was executed.\nThe conclusion seems to be that, although a belief always MAY influence\naction if it becomes relevant to a practical issue, it often exists\nactively (not as a mere disposition) without producing any voluntary\nmovement whatever. If this is true, we cannot define belief by the\neffect on voluntary movements.\n\nThere is another, more theoretical, ground for rejecting the view we\nare examining. It is clear that a proposition can be either believed or\nmerely considered, and that the content is the same in both cases. We\ncan expect an egg for breakfast, or merely entertain the supposition\nthat there may be an egg for breakfast. A moment ago I considered the\npossibility of being invited to become King of Georgia, but I do not\nbelieve that this will happen. Now, it seems clear that, since believing\nand considering have different effects if one produces bodily movements\nwhile the other does not, there must be some intrinsic difference\nbetween believing and considering*; for if they were precisely similar,\ntheir effects also would be precisely similar. We have seen that the\ndifference between believing a given proposition and merely considering\nit does not lie in the content; therefore there must be, in one case\nor in both, something additional to the content which distinguishes the\noccurrence of a belief from the occurrence of a mere consideration\nof the same content. So far as the theoretical argument goes, this\nadditional element may exist only in belief, or only in consideration,\nor there may be one sort of additional element in the case of belief,\nand another in the case of consideration. This brings us to the second\nview which we have to examine.\n\n     * Cf. Brentano, \"Psychologie vom empirischen Standpunkte,\"\n     p. 268 (criticizing Bain, \"The Emotions and the Will\").\n\n(1) The theory which we have now to consider regards belief as belonging\nto every idea which is entertained, except in so far as some positive\ncounteracting force interferes. In this view belief is not a positive\nphenomenon, though doubt and disbelief are so. What we call belief,\naccording to this hypothesis, involves only the appropriate content,\nwhich will have the effects characteristic of belief unless something\nelse operating simultaneously inhibits them. James (Psychology, vol.\nii, p. 288) quotes with approval, though inaccurately, a passage from\nSpinoza embodying this view:\n\n\"Let us conceive a boy imagining to himself a horse, and taking note of\nnothing else. As this imagination involves the existence of the horse,\nAND THE BOY HAS NO PERCEPTION WHICH ANNULS ITS EXISTENCE [James's\nitalics], he will necessarily contemplate the horse as present, nor will\nhe be able to doubt of its existence, however little certain of it he\nmay be. I deny that a man in so far as he imagines [percipit] affirms\nnothing. For what is it to imagine a winged horse but to affirm that the\nhorse [that horse, namely] has wings? For if the mind had nothing before\nit but the winged horse, it would contemplate the same as present, would\nhave no cause to doubt of its existence, nor any power of dissenting\nfrom its existence, unless the imagination of the winged horse were\njoined to an idea which contradicted [tollit] its existence\" (\"Ethics,\"\nvol. ii, p. 49, Scholium).\n\nTo this doctrine James entirely assents, adding in italics:\n\n\"ANY OBJECT WHICH REMAINS UNCONTRADICTED IS IPSO FACTO BELIEVED AND\nPOSITED AS ABSOLUTE REALITY.\"\n\nIf this view is correct, it follows (though James does not draw\nthe inference) that there is no need of any specific feeling called\n\"belief,\" and that the mere existence of images yields all that is\nrequired. The state of mind in which we merely consider a proposition,\nwithout believing or disbelieving it, will then appear as a\nsophisticated product, the result of some rival force adding to the\nimage-proposition a positive feeling which may be called suspense or\nnon-belief--a feeling which may be compared to that of a man about to\nrun a race waiting for the signal. Such a man, though not moving, is in\na very different condition from that of a man quietly at rest And so the\nman who is considering a proposition without believing it will be in\na state of tension, restraining the natural tendency to act upon the\nproposition which he would display if nothing interfered. In this view\nbelief primarily consists merely in the existence of the appropriate\nimages without any counteracting forces.\n\nThere is a great deal to be said in favour of this view, and I have some\nhesitation in regarding it as inadequate. It fits admirably with the\nphenomena of dreams and hallucinatory images, and it is recommended by\nthe way in which it accords with mental development. Doubt, suspense\nof judgment and disbelief all seem later and more complex than a wholly\nunreflecting assent. Belief as a positive phenomenon, if it exists,\nmay be regarded, in this view, as a product of doubt, a decision after\ndebate, an acceptance, not merely of THIS, but of THIS-RATHER-THAN-THAT.\nIt is not difficult to suppose that a dog has images (possible\nolfactory) of his absent master, or of the rabbit that he dreams of\nhunting. But it is very difficult to suppose that he can entertain mere\nimagination-images to which no assent is given.\n\nI think it must be conceded that a mere image, without the addition of\nany positive feeling that could be called \"belief,\" is apt to have a\ncertain dynamic power, and in this sense an uncombated image has the\nforce of a belief. But although this may be true, it accounts only for\nsome of the simplest phenomena in the region of belief. It will not, for\nexample, explain memory. Nor can it explain beliefs which do not issue\nin any proximate action, such as those of mathematics. I conclude,\ntherefore, that there must be belief-feelings of the same order as those\nof doubt or disbelief, although phenomena closely analogous to those of\nbelief can be produced by mere uncontradicted images.\n\n(3) I come now to the view of belief which I wish to advocate. It seems\nto me that there are at least three kinds of belief, namely memory,\nexpectation and bare assent. Each of these I regard as constituted by\na certain feeling or complex of sensations, attached to the content\nbelieved. We may illustrate by an example. Suppose I am believing,\nby means of images, not words, that it will rain. We have here two\ninterrelated elements, namely the content and the expectation. The\ncontent consists of images of (say) the visual appearance of rain, the\nfeeling of wetness, the patter of drops, interrelated, roughly, as the\nsensations would be if it were raining. Thus the content is a complex\nfact composed of images. Exactly the same content may enter into the\nmemory \"it was raining\" or the assent \"rain occurs.\" The difference of\nthese cases from each other and from expectation does not lie in the\ncontent. The difference lies in the nature of the belief-feeling.\nI, personally, do not profess to be able to analyse the sensations\nconstituting respectively memory, expectation and assent; but I am\nnot prepared to say that they cannot be analysed. There may be other\nbelief-feelings, for example in disjunction and implication; also a\ndisbelief-feeling.\n\nIt is not enough that the content and the belief-feeling should coexist:\nit is necessary that there should be a specific relation between them,\nof the sort expressed by saying that the content is what is believed.\nIf this were not obvious, it could be made plain by an argument. If\nthe mere co-existence of the content and the belief-feeling sufficed,\nwhenever we were having (say) a memory-feeling we should be remembering\nany proposition which came into our minds at the same time. But this is\nnot the case, since we may simultaneously remember one proposition and\nmerely consider another.\n\nWe may sum up our analysis, in the case of bare assent to a proposition\nnot expressed in words, as follows: (a) We have a proposition,\nconsisting of interrelated images, and possibly partly of sensations;\n(b) we have the feeling of assent, which is presumably a complex\nsensation demanding analysis; (c) we have a relation, actually\nsubsisting, between the assent and the proposition, such as is expressed\nby saying that the proposition in question is what is assented to. For\nother forms of belief-feeling or of content, we have only to make the\nnecessary substitutions in this analysis.\n\nIf we are right in our analysis of belief, the use of words in\nexpressing beliefs is apt to be misleading. There is no way of\ndistinguishing, in words, between a memory and an assent to a\nproposition about the past: \"I ate my breakfast\" and \"Caesar conquered\nGaul\" have the same verbal form, though (assuming that I remember my\nbreakfast) they express occurrences which are psychologically very\ndifferent. In the one case, what happens is that I remember the content\n\"eating my breakfast\"; in the other case, I assent to the content\n\"Caesar's conquest of Gaul occurred.\" In the latter case, but not in the\nformer, the pastness is part of the content believed. Exactly similar\nremarks apply to the difference between expectation, such as we have\nwhen waiting for the thunder after a flash of lightning, and assent to a\nproposition about the future, such as we have in all the usual cases of\ninferential knowledge as to what will occur. I think this difficulty in\nthe verbal expression of the temporal aspects of beliefs is one among\nthe causes which have hampered philosophy in the consideration of time.\n\nThe view of belief which I have been advocating contains little that is\nnovel except the distinction of kinds of belief-feeling--such as memory\nand expectation. Thus James says: \"Everyone knows the difference between\nimagining a thing and believing in its existence, between supposing a\nproposition and acquiescing in its truth...IN ITS INNER NATURE, BELIEF,\nOR THE SENSE OF REALITY, IS A SORT OF FEELING MORE ALLIED TO THE\nEMOTIONS THAN TO ANYTHING ELSE\" (\"Psychology,\" vol. ii, p. 283. James's\nitalics). He proceeds to point out that drunkenness, and, still more,\nnitrous-oxide intoxication, will heighten the sense of belief: in the\nlatter case, he says, a man's very soul may sweat with conviction, and\nhe be all the time utterly unable to say what he is convinced of. It\nwould seem that, in such cases, the feeling of belief exists unattached,\nwithout its usual relation to a content believed, just as the feeling\nof familiarity may sometimes occur without being related to any definite\nfamiliar object. The feeling of belief, when it occurs in this separated\nheightened form, generally leads us to look for a content to which to\nattach it. Much of what passes for revelation or mystic insight probably\ncomes in this way: the belief-feeling, in abnormal strength, attaches\nitself, more or less accidentally, to some content which we happen to\nthink of at the appropriate moment. But this is only a speculation, upon\nwhich I do not wish to lay too much stress.\n\n\n\n\nLECTURE XIII. TRUTH AND FALSEHOOD\n\nThe definition of truth and falsehood, which is our topic to-day, lies\nstrictly outside our general subject, namely the analysis of mind. From\nthe psychological standpoint, there may be different kinds of belief,\nand different degrees of certainty, but there cannot be any purely\npsychological means of distinguishing between true and false beliefs.\nA belief is rendered true or false by relation to a fact, which may lie\noutside the experience of the person entertaining the belief. Truth and\nfalsehood, except in the case of beliefs about our own minds, depend\nupon the relations of mental occurrences to outside things, and thus\ntake us beyond the analysis of mental occurrences as they are in\nthemselves. Nevertheless, we can hardly avoid the consideration of truth\nand falsehood. We wish to believe that our beliefs, sometimes at least,\nyield KNOWLEDGE, and a belief does not yield knowledge unless it is\ntrue. The question whether our minds are instruments of knowledge, and,\nif so, in what sense, is so vital that any suggested analysis of mind\nmust be examined in relation to this question. To ignore this question\nwould be like describing a chronometer without regard to its accuracy\nas a time-keeper, or a thermometer without mentioning the fact that it\nmeasures temperature.\n\nMany difficult questions arise in connection with knowledge. It is\ndifficult to define knowledge, difficult to decide whether we have any\nknowledge, and difficult, even if it is conceded that we sometimes have\nknowledge to discover whether we can ever know that we have knowledge\nin this or that particular case. I shall divide the discussion into four\nparts:\n\nI. We may regard knowledge, from a behaviourist standpoint, as exhibited\nin a certain kind of response to the environment. This response must\nhave some characteristics which it shares with those of scientific\ninstruments, but must also have others that are peculiar to knowledge.\nWe shall find that this point of view is important, but not exhaustive\nof the nature of knowledge.\n\nII. We may hold that the beliefs that constitute knowledge are\ndistinguished from such as are erroneous or uncertain by properties\nwhich are intrinsic either to single beliefs or to systems of beliefs,\nbeing in either case discoverable without reference to outside fact.\nViews of this kind have been widely held among philosophers, but we\nshall find no reason to accept them.\n\nIII. We believe that some beliefs are true, and some false. This raises\nthe problem of VERIFIABILITY: are there any circumstances which can\njustifiably give us an unusual degree of certainty that such and such a\nbelief is true? It is obvious that there are circumstances which in fact\ncause a certainty of this sort, and we wish to learn what we can from\nexamining these circumstances.\n\nIV. Finally, there is the formal problem of defining truth and\nfalsehood, and deriving the objective reference of a proposition from\nthe meanings of its component words.\n\nWe will consider these four problems in succession.\n\nI. We may regard a human being as an instrument, which makes various\nresponses to various stimuli. If we observe these responses from\noutside, we shall regard them as showing knowledge when they display\ntwo characteristics, ACCURACY and APPROPRIATENESS. These two are quite\ndistinct, and even sometimes incompatible. If I am being pursued by\na tiger, accuracy is furthered by turning round to look at him, but\nappropriateness by running away without making any search for\nfurther knowledge of the beast. I shall return to the question of\nappropriateness later; for the present it is accuracy that I wish to\nconsider.\n\nWhen we are viewing a man from the outside, it is not his beliefs,\nbut his bodily movements, that we can observe. His knowledge must be\ninferred from his bodily movements, and especially from what he says\nand writes. For the present we may ignore beliefs, and regard a man's\nknowledge as actually consisting in what he says and does. That is to\nsay, we will construct, as far as possible, a purely behaviouristic\naccount of truth and falsehood.\n\nIf you ask a boy \"What is twice two?\" and the boy says \"four,\" you take\nthat as prima facie evidence that the boy knows what twice two is. But\nif you go on to ask what is twice three, twice four, twice five, and so\non, and the boy always answers \"four,\" you come to the conclusion that\nhe knows nothing about it. Exactly similar remarks apply to scientific\ninstruments. I know a certain weather-cock which has the pessimistic\nhabit of always pointing to the north-east. If you were to see it first\non a cold March day, you would think it an excellent weather-cock; but\nwith the first warm day of spring your confidence would be shaken. The\nboy and the weather-cock have the same defect: they do not vary their\nresponse when the stimulus is varied. A good instrument, or a person\nwith much knowledge, will give different responses to stimuli which\ndiffer in relevant ways. This is the first point in defining accuracy of\nresponse.\n\nWe will now assume another boy, who also, when you first question him,\nasserts that twice two is four. But with this boy, instead of asking him\ndifferent questions, you make a practice of asking him the same question\nevery day at breakfast. You find that he says five, or six, or seven, or\nany other number at random, and you conclude that he also does not know\nwhat twice two is, though by good luck he answered right the first time.\nThis boy is like a weather-cock which, instead of being stuck fast, is\nalways going round and round, changing without any change of wind. This\nboy and weather-cock have the opposite defect to that of the previous\npair: they give different responses to stimuli which do not differ in\nany relevant way.\n\nIn connection with vagueness in memory, we already had occasion to\nconsider the definition of accuracy. Omitting some of the niceties of\nour previous discussion, we may say that an instrument is ACCURATE when\nit avoids the defects of the two boys and weather-cocks, that is to say,\nwhen--\n\n(a) It gives different responses to stimuli which differ in relevant\nways;\n\n(b) It gives the same response to stimuli which do not differ in\nrelevant ways.\n\nWhat are relevant ways depends upon the nature and purpose of the\ninstrument. In the case of a weather-cock, the direction of the wind is\nrelevant, but not its strength; in the case of the boy, the meaning of\nthe words of your question is relevant, but not the loudness of your\nvoice, or whether you are his father or his schoolmaster If, however,\nyou were a boy of his own age, that would be relevant, and the\nappropriate response would be different.\n\nIt is clear that knowledge is displayed by accuracy of response to\ncertain kinds of stimuli, e.g. examinations. Can we say, conversely,\nthat it consists wholly of such accuracy of response? I do not think\nwe can; but we can go a certain distance in this direction. For this\npurpose we must define more carefully the kind of accuracy and the kind\nof response that may be expected where there is knowledge.\n\nFrom our present point of view, it is difficult to exclude perception\nfrom knowledge; at any rate, knowledge is displayed by actions based\nupon perception. A bird flying among trees avoids bumping into their\nbranches; its avoidance is a response to visual sensations. This\nresponse has the characteristic of accuracy, in the main, and leads\nus to say that the bird \"knows,\" by sight, what objects are in its\nneighbourhood. For a behaviourist, this must certainly count as\nknowledge, however it may be viewed by analytic psychology. In this\ncase, what is known, roughly, is the stimulus; but in more advanced\nknowledge the stimulus and what is known become different. For example,\nyou look in your calendar and find that Easter will be early next year.\nHere the stimulus is the calendar, whereas the response concerns the\nfuture. Even this can be paralleled among instruments: the behaviour of\nthe barometer has a present stimulus but foretells the future, so that\nthe barometer might be said, in a sense, to know the future. However\nthat may be, the point I am emphasizing as regards knowledge is that\nwhat is known may be quite different from the stimulus, and no part of\nthe cause of the knowledge-response. It is only in sense-knowledge that\nthe stimulus and what is known are, with qualifications, identifiable.\nIn knowledge of the future, it is obvious that they are totally\ndistinct, since otherwise the response would precede the stimulus. In\nabstract knowledge also they are distinct, since abstract facts have no\ndate. In knowledge of the past there are complications, which we must\nbriefly examine.\n\nEvery form of memory will be, from our present point of view, in one\nsense a delayed response. But this phrase does not quite clearly\nexpress what is meant. If you light a fuse and connect it with a heap of\ndynamite, the explosion of the dynamite may be spoken of, in a sense,\nas a delayed response to your lighting of the fuse. But that only means\nthat it is a somewhat late portion of a continuous process of which the\nearlier parts have less emotional interest. This is not the case\nwith habit. A display of habit has two sorts of causes: (a) the past\noccurrences which generated the habit, (b) the present occurrence which\nbrings it into play. When you drop a weight on your toe, and say what\nyou do say, the habit has been caused by imitation of your undesirable\nassociates, whereas it is brought into play by the dropping of the\nweight. The great bulk of our knowledge is a habit in this sense:\nwhenever I am asked when I was born, I reply correctly by mere habit. It\nwould hardly be correct to say that getting born was the stimulus, and\nthat my reply is a delayed response But in cases of memory this way\nof speaking would have an element of truth. In an habitual memory, the\nevent remembered was clearly an essential part of the stimulus to the\nformation of the habit. The present stimulus which brings the habit into\nplay produces a different response from that which it would produce if\nthe habit did not exist. Therefore the habit enters into the causation\nof the response, and so do, at one remove, the causes of the habit. It\nfollows that an event remembered is an essential part of the causes of\nour remembering.\n\nIn spite, however, of the fact that what is known is SOMETIMES an\nindispensable part of the cause of the knowledge, this circumstance is,\nI think, irrelevant to the general question with which we are concerned,\nnamely What sort of response to what sort of stimulus can be regarded\nas displaying knowledge? There is one characteristic which the response\nmust have, namely, it must consist of voluntary movements. The need\nof this characteristic is connected with the characteristic of\nAPPROPRIATENESS, which I do not wish to consider as yet. For the present\nI wish only to obtain a clearer idea of the sort of ACCURACY that a\nknowledge-response must have. It is clear from many instances that\naccuracy, in other cases, may be purely mechanical. The most complete\nform of accuracy consists in giving correct answers to questions, an\nachievement in which calculating machines far surpass human beings. In\nasking a question of a calculating machine, you must use its language:\nyou must not address it in English, any more than you would address\nan Englishman in Chinese. But if you address it in the language it\nunderstands, it will tell you what is 34521 times 19987, without a\nmoment's hesitation or a hint of inaccuracy. We do not say the machine\nKNOWS the answer, because it has no purpose of its own in giving the\nanswer: it does not wish to impress you with its cleverness, or feel\nproud of being such a good machine. But as far as mere accuracy goes,\nthe machine leaves nothing to be desired.\n\nAccuracy of response is a perfectly clear notion in the case of answers\nto questions, but in other cases it is much more obscure. We may say\ngenerally that an object whether animate or inanimate, is \"sensitive\" to\na certain feature of the environment if it behaves differently according\nto the presence or absence of that feature. Thus iron is sensitive to\nanything magnetic. But sensitiveness does not constitute knowledge, and\nknowledge of a fact which is not sensible is not sensitiveness to\nthat fact, as we have seen in distinguishing the fact known from the\nstimulus. As soon as we pass beyond the simple case of question and\nanswer, the definition of knowledge by means of behaviour demands the\nconsideration of purpose. A carrier pigeon flies home, and so we say\nit \"knows\" the way. But if it merely flew to some place at random, we\nshould not say that it \"knew\" the way to that place, any more than a\nstone rolling down hill knows the way to the valley.\n\nOn the features which distinguish knowledge from accuracy of response in\ngeneral, not much can be said from a behaviourist point of view without\nreferring to purpose. But the necessity of SOMETHING besides accuracy of\nresponse may be brought out by the following consideration: Suppose\ntwo persons, of whom one believed whatever the other disbelieved,\nand disbelieved whatever the other believed. So far as accuracy and\nsensitiveness of response alone are concerned, there would be nothing to\nchoose between these two persons. A thermometer which went down for warm\nweather and up for cold might be just as accurate as the usual kind; and\na person who always believes falsely is just as sensitive an instrument\nas a person who always believes truly. The observable and practical\ndifference between them would be that the one who always believed\nfalsely would quickly come to a bad end. This illustrates once more that\naccuracy of response to stimulus does not alone show knowledge, but must\nbe reinforced by appropriateness, i.e. suitability for realizing one's\npurpose. This applies even in the apparently simple case of answering\nquestions: if the purpose of the answers is to deceive, their falsehood,\nnot their truth, will be evidence of knowledge. The proportion of\nthe combination of appropriateness with accuracy in the definition\nof knowledge is difficult; it seems that both enter in, but that\nappropriateness is only required as regards the general type of\nresponse, not as regards each individual instance.\n\nII. I have so far assumed as unquestionable the view that the truth or\nfalsehood of a belief consists in a relation to a certain fact,\nnamely the objective of the belief. This view has, however, been often\nquestioned. Philosophers have sought some intrinsic criterion by which\ntrue and false beliefs could be distinguished.* I am afraid their chief\nreason for this search has been the wish to feel more certainty than\nseems otherwise possible as to what is true and what is false. If\nwe could discover the truth of a belief by examining its intrinsic\ncharacteristics, or those of some collection of beliefs of which it\nforms part, the pursuit of truth, it is thought, would be a less arduous\nbusiness than it otherwise appears to be. But the attempts which\nhave been made in this direction are not encouraging. I will take two\ncriteria which have been suggested, namely, (1) self-evidence, (2)\nmutual coherence. If we can show that these are inadequate, we may\nfeel fairly certain that no intrinsic criterion hitherto suggested will\nsuffice to distinguish true from false beliefs.\n\n     * The view that such a criterion exists is generally held by\n     those whose views are in any degree derived from Hegel. It\n     may be illustrated by the following passage from Lossky,\n     \"The Intuitive Basis of Knowledge\" (Macmillan, 1919), p.\n     268: \"Strictly speaking, a false judgment is not a judgment\n     at all. The predicate does not follow from the subject S\n     alone, but from the subject plus a certain addition C, WHICH\n     IN NO SENSE BELONGS TO THE CONTENT OF THE JUDGMENT. What\n     takes place may be a process of association of ideas, of\n     imagining, or the like, but is not a process of judging. An\n     experienced psychologist will be able by careful observation\n     to detect that in this process there is wanting just the\n     specific element of the objective dependence of the\n     predicate upon the subject which is characteristic of a\n     judgment. It must be admitted, however, that an exceptional\n     power of observation is needed in order to distinguish, by\n     means of introspection, mere combination of ideas from\n     judgments.\"\n\n\n(1) Self-evidence.--Some of our beliefs seem to be peculiarly\nindubitable. One might instance the belief that two and two are four,\nthat two things cannot be in the same place at the same time, nor one\nthing in two places, or that a particular buttercup that we are seeing\nis yellow. The suggestion we are to examine is that such: beliefs have\nsome recognizable quality which secures their truth, and the truth of\nwhatever is deduced from them according to self-evident principles of\ninference. This theory is set forth, for example, by Meinong in his\nbook, \"Ueber die Erfahrungsgrundlagen unseres Wissens.\"\n\nIf this theory is to be logically tenable, self-evidence must not\nconsist merely in the fact that we believe a proposition. We believe\nthat our beliefs are sometimes erroneous, and we wish to be able to\nselect a certain class of beliefs which are never erroneous. If we\nare to do this, it must be by some mark which belongs only to certain\nbeliefs, not to all; and among those to which it belongs there must be\nnone that are mutually inconsistent. If, for example, two propositions\np and q were self-evident, and it were also self-evident that p and q\ncould not both be true, that would condemn self-evidence as a guarantee\nof truth. Again, self-evidence must not be the same thing as the absence\nof doubt or the presence of complete certainty. If we are completely\ncertain of a proposition, we do not seek a ground to support our belief.\nIf self-evidence is alleged as a ground of belief, that implies that\ndoubt has crept in, and that our self-evident proposition has not\nwholly resisted the assaults of scepticism. To say that any given person\nbelieves some things so firmly that he cannot be made to doubt them is\nno doubt true. Such beliefs he will be willing to use as premisses in\nreasoning, and to him personally they will seem to have as much evidence\nas any belief can need. But among the propositions which one man finds\nindubitable there will be some that another man finds it quite possible\nto doubt. It used to seem self-evident that there could not be men at\nthe Antipodes, because they would fall off, or at best grow giddy from\nstanding on their heads. But New Zealanders find the falsehood of this\nproposition self-evident. Therefore, if self-evidence is a guarantee of\ntruth, our ancestors must have been mistaken in thinking their beliefs\nabout the Antipodes self-evident. Meinong meets this difficulty by\nsaying that some beliefs are falsely thought to be self-evident, but in\nthe case of others it is self-evident that they are self-evident, and\nthese are wholly reliable. Even this, however, does not remove the\npractical risk of error, since we may mistakenly believe it self-evident\nthat a certain belief is self-evident. To remove all risk of error, we\nshall need an endless series of more and more complicated self-evident\nbeliefs, which cannot possibly be realized in practice. It would seem,\ntherefore, that self-evidence is useless as a practical criterion for\ninsuring truth.\n\nThe same result follows from examining instances. If we take the four\ninstances mentioned at the beginning of this discussion, we shall\nfind that three of them are logical, while the fourth is a judgment of\nperception. The proposition that two and two are four follows by purely\nlogical deduction from definitions: that means that its truth results,\nnot from the properties of objects, but from the meanings of symbols.\nNow symbols, in mathematics, mean what we choose; thus the feeling of\nself-evidence, in this case, seems explicable by the fact that the whole\nmatter is within our control. I do not wish to assert that this is\nthe whole truth about mathematical propositions, for the question is\ncomplicated, and I do not know what the whole truth is. But I do wish to\nsuggest that the feeling of self-evidence in mathematical propositions\nhas to do with the fact that they are concerned with the meanings of\nsymbols, not with properties of the world such as external observation\nmight reveal.\n\nSimilar considerations apply to the impossibility of a thing being in\ntwo places at once, or of two things being in one place at the same\ntime. These impossibilities result logically, if I am not mistaken, from\nthe definitions of one thing and one place. That is to say, they are not\nlaws of physics, but only part of the intellectual apparatus which we\nhave manufactured for manipulating physics. Their self-evidence, if this\nis so, lies merely in the fact that they represent our decision as to\nthe use of words, not a property of physical objects.\n\nJudgments of perception, such as \"this buttercup is yellow,\" are in\na quite different position from judgments of logic, and their\nself-evidence must have a different explanation. In order to arrive at\nthe nucleus of such a judgment, we will eliminate, as far as possible,\nthe use of words which take us beyond the present fact, such as\n\"buttercup\" and \"yellow.\" The simplest kind of judgment underlying the\nperception that a buttercup is yellow would seem to be the perception of\nsimilarity in two colours seen simultaneously. Suppose we are seeing\ntwo buttercups, and we perceive that their colours are similar. This\nsimilarity is a physical fact, not a matter of symbols or words; and it\ncertainly seems to be indubitable in a way that many judgments are not.\n\nThe first thing to observe, in regard to such judgments, is that as they\nstand they are vague. The word \"similar\" is a vague word, since there\nare degrees of similarity, and no one can say where similarity ends\nand dissimilarity begins. It is unlikely that our two buttercups have\nEXACTLY the same colour, and if we judged that they had we should have\npassed altogether outside the region of self-evidence. To make our\nproposition more precise, let us suppose that we are also seeing a\nred rose at the same time. Then we may judge that the colours of the\nbuttercups are more similar to each other than to the colour of the\nrose. This judgment seems more complicated, but has certainly gained\nin precision. Even now, however, it falls short of complete precision,\nsince similarity is not prima facie measurable, and it would require\nmuch discussion to decide what we mean by greater or less similarity. To\nthis process of the pursuit of precision there is strictly no limit.\n\nThe next thing to observe (although I do not personally doubt that most\nof our judgments of perception are true) is that it is very difficult to\ndefine any class of such judgments which can be known, by its intrinsic\nquality, to be always exempt from error. Most of our judgments of\nperception involve correlations, as when we judge that a certain noise\nis that of a passing cart. Such judgments are all obviously liable to\nerror, since there is no correlation of which we have a right to be\ncertain that it is invariable. Other judgments of perception are derived\nfrom recognition, as when we say \"this is a buttercup,\" or even merely\n\"this is yellow.\" All such judgments entail some risk of error,\nthough sometimes perhaps a very small one; some flowers that look like\nbuttercups are marigolds, and colours that some would call yellow others\nmight call orange. Our subjective certainty is usually a result of\nhabit, and may lead us astray in circumstances which are unusual in ways\nof which we are unaware.\n\nFor such reasons, no form of self-evidence seems to afford an absolute\ncriterion of truth. Nevertheless, it is perhaps true that judgments\nhaving a high degree of subjective certainty are more apt to be true\nthan other judgments. But if this be the case, it is a result to be\ndemonstrated, not a premiss from which to start in defining truth and\nfalsehood. As an initial guarantee, therefore, neither self-evidence nor\nsubjective certainty can be accepted as adequate.\n\n(2) Coherence.--Coherence as the definition of truth is advocated by\nidealists, particularly by those who in the main follow Hegel. It is set\nforth ably in Mr. Joachim's book, \"The Nature of Truth\" (Oxford, 1906).\nAccording to this view, any set of propositions other than the whole\nof truth can be condemned on purely logical grounds, as internally\ninconsistent; a single proposition, if it is what we should ordinarily\ncall false, contradicts itself irremediably, while if it is what we\nshould ordinarily call true, it has implications which compel us to\nadmit other propositions, which in turn lead to others, and so on, until\nwe find ourselves committed to the whole of truth. One might illustrate\nby a very simple example: if I say \"so-and-so is a married man,\" that\nis not a self-subsistent proposition. We cannot logically conceive of a\nuniverse in which this proposition constituted the whole of truth. There\nmust be also someone who is a married woman, and who is married to\nthe particular man in question. The view we are considering regards\neverything that can be said about any one object as relative in the same\nsort of way as \"so-and-so is a married man.\" But everything, according\nto this view, is relative, not to one or two other things, but to all\nother things, so that from one bit of truth the whole can be inferred.\n\nThe fundamental objection to this view is logical, and consists in a\ncriticism of its doctrine as to relations. I shall omit this line of\nargument, which I have developed elsewhere.* For the moment I will\ncontent myself with saying that the powers of logic seem to me very\nmuch less than this theory supposes. If it were taken seriously, its\nadvocates ought to profess that any one truth is logically inferable\nfrom any other, and that, for example, the fact that Caesar conquered\nGaul, if adequately considered, would enable us to discover what the\nweather will be to-morrow. No such claim is put forward in practice, and\nthe necessity of empirical observation is not denied; but according to\nthe theory it ought to be.\n\n     * In the article on \"The Monistic Theory of Truth\" in\n     \"Philosophical Essays\" (Longmans, 1910), reprinted from the\n     \"Proceedings of the Aristotelian Society,\" 1906-7.\n\nAnother objection is that no endeavour is made to show that we cannot\nform a consistent whole composed partly or wholly of false propositions,\nas in a novel. Leibniz's conception of many possible worlds seems to\naccord much better with modern logic and with the practical empiricism\nwhich is now universal. The attempt to deduce the world by pure thought\nis attractive, and in former times was largely supposed capable of\nsuccess. But nowadays most men admit that beliefs must be tested by\nobservation, and not merely by the fact that they harmonize with other\nbeliefs. A consistent fairy-tale is a different thing from truth,\nhowever elaborate it may be. But to pursue this topic would lead us\ninto difficult technicalities; I shall therefore assume, without further\nargument, that coherence is not sufficient as a definition of truth.\n\nIII. Many difficult problems arise as regards the verifiability of\nbeliefs. We believe various things, and while we believe them we think\nwe know them. But it sometimes turns out that we were mistaken, or at\nany rate we come to think we were. We must be mistaken either in our\nprevious opinion or in our subsequent recantation; therefore our beliefs\nare not all correct, and there are cases of belief which are not cases\nof knowledge. The question of verifiability is in essence this: can we\ndiscover any set of beliefs which are never mistaken or any test which,\nwhen applicable, will always enable us to discriminate between true\nand false beliefs? Put thus broadly and abstractly, the answer must be\nnegative. There is no way hitherto discovered of wholly eliminating the\nrisk of error, and no infallible criterion. If we believe we have found\na criterion, this belief itself may be mistaken; we should be begging\nthe question if we tried to test the criterion by applying the criterion\nto itself.\n\nBut although the notion of an absolute criterion is chimerical, there\nmay be relative criteria, which increase the probability of truth.\nCommon sense and science hold that there are. Let us see what they have\nto say.\n\nOne of the plainest cases of verification, perhaps ultimately the only\ncase, consists in the happening of something expected. You go to the\nstation believing that there will be a train at a certain time; you\nfind the train, you get into it, and it starts at the expected time This\nconstitutes verification, and is a perfectly definite experience. It is,\nin a sense, the converse of memory instead of having first sensations\nand then images accompanied by belief, we have first images accompanied\nby belief and then sensations. Apart from differences as to the\ntime-order and the accompanying feelings, the relation between image and\nsensation is closely similar in the two cases of memory and expectation;\nit is a relation of similarity, with difference as to causal\nefficacy--broadly, the image has the psychological but not the physical\neffects that the sensation would have. When an image accompanied by\nan expectation-belief is thus succeeded by a sensation which is the\n\"meaning\" of the image, we say that the expectation-belief has been\nverified. The experience of verification in this sense is exceedingly\nfamiliar; it happens every time that accustomed activities have results\nthat are not surprising, in eating and walking and talking and all our\ndaily pursuits.\n\nBut although the experience in question is common, it is not wholly easy\nto give a theoretical account of it. How do we know that the sensation\nresembles the previous image? Does the image persist in presence of the\nsensation, so that we can compare the two? And even if SOME image does\npersist, how do we know that it is the previous image unchanged? It does\nnot seem as if this line of inquiry offered much hope of a successful\nissue. It is better, I think, to take a more external and causal view of\nthe relation of expectation to expected occurrence. If the occurrence,\nwhen it comes, gives us the feeling of expectedness, and if the\nexpectation, beforehand, enabled us to act in a way which proves\nappropriate to the occurrence, that must be held to constitute the\nmaximum of verification. We have first an expectation, then a sensation\nwith the feeling of expectedness related to memory of the expectation.\nThis whole experience, when it occurs, may be defined as verification,\nand as constituting the truth of the expectation. Appropriate action,\nduring the period of expectation, may be regarded as additional\nverification, but is not essential. The whole process may be illustrated\nby looking up a familiar quotation, finding it in the expected words,\nand in the expected part of the book. In this case we can strengthen\nthe verification by writing down beforehand the words which we expect to\nfind.\n\nI think all verification is ultimately of the above sort. We verify a\nscientific hypothesis indirectly, by deducing consequences as to the\nfuture, which subsequent experience confirms. If somebody were to doubt\nwhether Caesar had crossed the Rubicon, verification could only be\nobtained from the future. We could proceed to display manuscripts to our\nhistorical sceptic, in which it was said that Caesar had behaved in this\nway. We could advance arguments, verifiable by future experience, to\nprove the antiquity of the manuscript from its texture, colour, etc. We\ncould find inscriptions agreeing with the historian on other points,\nand tending to show his general accuracy. The causal laws which our\narguments would assume could be verified by the future occurrence of\nevents inferred by means of them. The existence and persistence of\ncausal laws, it is true, must be regarded as a fortunate accident, and\nhow long it will continue we cannot tell. Meanwhile verification remains\noften practically possible. And since it is sometimes possible, we\ncan gradually discover what kinds of beliefs tend to be verified by\nexperience, and what kinds tend to be falsified; to the former kinds\nwe give an increased degree of assent, to the latter kinds a diminished\ndegree. The process is not absolute or infallible, but it has been\nfound capable of sifting beliefs and building up science. It affords\nno theoretical refutation of the sceptic, whose position must remain\nlogically unassailable; but if complete scepticism is rejected, it gives\nthe practical method by which the system of our beliefs grows gradually\ntowards the unattainable ideal of impeccable knowledge.\n\nIV. I come now to the purely formal definition of the truth or falsehood\nof a belief. For this definition it is necessary first of all to\nconsider the derivation of the objective reference of a proposition from\nthe meanings of its component words or images.\n\nJust as a word has meaning, so a proposition has an objective reference.\nThe objective reference of a proposition is a function (in the\nmathematical sense) of the meanings of its component words. But the\nobjective reference differs from the meaning of a word through the\nduality of truth and falsehood. You may believe the proposition \"to-day\nis Tuesday\" both when, in fact, to-day is Tuesday, and when to-day is\nnot Tuesday. If to-day is not Tuesday, this fact is the objective of\nyour belief that to-day is Tuesday. But obviously the relation of your\nbelief to the fact is different in this case from what it is in the case\nwhen to-day is Tuesday. We may say, metaphorically, that when to-day is\nTuesday, your belief that it is Tuesday points TOWARDS the fact, whereas\nwhen to-day is not Tuesday your belief points AWAY FROM the fact. Thus\nthe objective reference of a belief is not determined by the fact alone,\nbut by the direction of the belief towards or away from the fact.* If,\non a Tuesday, one man believes that it is Tuesday while another believes\nthat it is not Tuesday, their beliefs have the same objective, namely\nthe fact that it is Tuesday but the true belief points towards the fact\nwhile the false one points away from it. Thus, in order to define the\nreference of a proposition we have to take account not only of the\nobjective, but also of the direction of pointing, towards the objective\nin the case of a true proposition and away from it in the case of a\nfalse one.\n\n     * I owe this way of looking at the matter to my friend\n     Ludwig Wittgenstein.\n\nThis mode of stating the nature of the objective reference of a\nproposition is necessitated by the circumstance that there are true and\nfalse propositions, but not true and false facts. If to-day is Tuesday,\nthere is not a false objective \"to-day is not Tuesday,\" which could be\nthe objective of the false belief \"to-day is not Tuesday.\" This is the\nreason why two beliefs which are each other's contradictories have the\nsame objective. There is, however, a practical inconvenience, namely\nthat we cannot determine the objective reference of a proposition,\naccording to this definition, unless we know whether the proposition\nis true or false. To avoid this inconvenience, it is better to adopt\na slightly different phraseology, and say: The \"meaning\" of the\nproposition \"to-day is Tuesday\" consists in pointing to the fact \"to-day\nis Tuesday\" if that is a fact, or away from the fact \"to-day is not\nTuesday\" if that is a fact. The \"meaning\" of the proposition \"to-day is\nnot Tuesday\" will be exactly the opposite. By this hypothetical form\nwe are able to speak of the meaning of a proposition without knowing\nwhether it is true or false. According to this definition, we know the\nmeaning of a proposition when we know what would make it true and what\nwould make it false, even if we do not know whether it is in fact true\nor false.\n\nThe meaning of a proposition is derivative from the meanings of its\nconstituent words. Propositions occur in pairs, distinguished (in\nsimple cases) by the absence or presence of the word \"not.\" Two such\npropositions have the same objective, but opposite meanings: when one is\ntrue, the other is false, and when one is false, the other is true.\n\nThe purely formal definition of truth and falsehood offers little\ndifficulty. What is required is a formal expression of the fact that a\nproposition is true when it points towards its objective, and false when\nit points away from it, In very simple cases we can give a very simple\naccount of this: we can say that true propositions actually resemble\ntheir objectives in a way in which false propositions do not. But for\nthis purpose it is necessary to revert to image-propositions instead of\nword-propositions. Let us take again the illustration of a memory-image\nof a familiar room, and let us suppose that in the image the window is\nto the left of the door. If in fact the window is to the left of the\ndoor, there is a correspondence between the image and the objective;\nthere is the same relation between the window and the door as between\nthe images of them. The image-memory consists of the image of the window\nto the left of the image of the door. When this is true, the very same\nrelation relates the terms of the objective (namely the window and\nthe door) as relates the images which mean them. In this case the\ncorrespondence which constitutes truth is very simple.\n\nIn the case we have just been considering the objective consists of\ntwo parts with a certain relation (that of left-to-right), and the\nproposition consists of images of these parts with the very same\nrelation. The same proposition, if it were false, would have a less\nsimple formal relation to its objective. If the image-proposition\nconsists of an image of the window to the left of an image of the door,\nwhile in fact the window is not to the left of the door, the proposition\ndoes not result from the objective by the mere substitution of images\nfor their prototypes. Thus in this unusually simple case we can say that\na true proposition \"corresponds\" to its objective in a formal sense in\nwhich a false proposition does not. Perhaps it may be possible to modify\nthis notion of formal correspondence in such a way as to be more widely\napplicable, but if so, the modifications required will be by no means\nslight. The reasons for this must now be considered.\n\nTo begin with, the simple type of correspondence we have been exhibiting\ncan hardly occur when words are substituted for images, because, in\nword-propositions, relations are usually expressed by words, which are\nnot themselves relations. Take such a proposition as \"Socrates\nprecedes Plato.\" Here the word \"precedes\" is just as solid as the words\n\"Socrates\" and \"Plato\"; it MEANS a relation, but is not a relation. Thus\nthe objective which makes our proposition true consists of TWO terms\nwith a relation between them, whereas our proposition consists of THREE\nterms with a relation of order between them. Of course, it would be\nperfectly possible, theoretically, to indicate a few chosen relations,\nnot by words, but by relations between the other words. \"Socrates-Plato\"\nmight be used to mean \"Socrates precedes Plato\"; \"Plato-Socrates\" might\nbe used to mean \"Plato was born before Socrates and died after him\"; and\nso on. But the possibilities of such a method would be very limited. For\naught I know, there may be languages that use it, but they are not among\nthe languages with which I am acquainted. And in any case, in view of\nthe multiplicity of relations that we wish to express, no language could\nadvance far without words for relations. But as soon as we have words\nfor relations, word-propositions have necessarily more terms than the\nfacts to which they refer, and cannot therefore correspond so simply\nwith their objectives as some image-propositions can.\n\nThe consideration of negative propositions and negative facts introduces\nfurther complications. An image-proposition is necessarily positive:\nwe can image the window to the left of the door, or to the right of the\ndoor, but we can form no image of the bare negative \"the window not to\nthe left of the door.\" We can DISBELIEVE the image-proposition expressed\nby \"the window to the left of the door,\" and our disbelief will be true\nif the window is not to the left of the door. But we can form no image\nof the fact that the window is not to the left of the door. Attempts\nhave often been made to deny such negative facts, but, for reasons which\nI have given elsewhere,* I believe these attempts to be mistaken, and I\nshall assume that there are negative facts.\n\n     * \"Monist,\" January, 1919, p. 42 ff.\n\nWord-propositions, like image-propositions, are always positive facts.\nThe fact that Socrates precedes Plato is symbolized in English by the\nfact that the word \"precedes\" occurs between the words \"Socrates\" and\n\"Plato.\" But we cannot symbolize the fact that Plato does not precede\nSocrates by not putting the word \"precedes\" between \"Plato\" and\n\"Socrates.\" A negative fact is not sensible, and language, being\nintended for communication, has to be sensible. Therefore we symbolize\nthe fact that Plato does not precede Socrates by putting the words \"does\nnot precede\" between \"Plato\" and \"Socrates.\" We thus obtain a series of\nwords which is just as positive a fact as the series \"Socrates precedes\nPlato.\" The propositions asserting negative facts are themselves\npositive facts; they are merely different positive facts from those\nasserting positive facts.\n\nWe have thus, as regards the opposition of positive and negative, three\ndifferent sorts of duality, according as we are dealing with facts,\nimage-propositions, or word-propositions. We have, namely:\n\n(1) Positive and negative facts;\n\n(2) Image-propositions, which may be believed or disbelieved, but do\nnot allow any duality of content corresponding to positive and negative\nfacts;\n\n(3) Word-propositions, which are always positive facts, but are of two\nkinds: one verified by a positive objective, the other by a negative\nobjective.\n\nOwing to these complications, the simplest type of correspondence is\nimpossible when either negative facts or negative propositions are\ninvolved.\n\nEven when we confine ourselves to relations between two terms which are\nboth imaged, it may be impossible to form an image-proposition in which\nthe relation of the terms is represented by the same relation of the\nimages. Suppose we say \"Caesar was 2,000 years before Foch,\" we express\na certain temporal relation between Caesar and Foch; but we cannot allow\n2,000 years to elapse between our image of Caesar and our image of Foch.\nThis is perhaps not a fair example, since \"2,000 years before\" is not a\ndirect relation. But take a case where the relation is direct, say, \"the\nsun is brighter than the moon.\" We can form visual images of sunshine\nand moonshine, and it may happen that our image of the sunshine is\nthe brighter of the two, but this is by no means either necessary or\nsufficient. The act of comparison, implied in our judgment, is something\nmore than the mere coexistence of two images, one of which is in fact\nbrighter than the other. It would take us too far from our main topic if\nwe were to go into the question what actually occurs when we make this\njudgment. Enough has been said to show that the correspondence between\nthe belief and its objective is more complicated in this case than in\nthat of the window to the left of the door, and this was all that had to\nbe proved.\n\nIn spite of these complications, the general nature of the formal\ncorrespondence which makes truth is clear from our instances. In the\ncase of the simpler kind of propositions, namely those that I call\n\"atomic\" propositions, where there is only one word expressing a\nrelation, the objective which would verify our proposition, assuming\nthat the word \"not\" is absent, is obtained by replacing each word\nby what it means, the word meaning a relation being replaced by this\nrelation among the meanings of the other words. For example, if the\nproposition is \"Socrates precedes Plato,\" the objective which verifies\nit results from replacing the word \"Socrates\" by Socrates, the word\n\"Plato\" by Plato, and the word \"precedes\" by the relation of preceding\nbetween Socrates and Plato. If the result of this process is a fact,\nthe proposition is true; if not, it is false. When our proposition is\n\"Socrates does not precede Plato,\" the conditions of truth and falsehood\nare exactly reversed. More complicated propositions can be dealt with on\nthe same lines. In fact, the purely formal question, which has occupied\nus in this last section, offers no very formidable difficulties.\n\nI do not believe that the above formal theory is untrue, but I do\nbelieve that it is inadequate. It does not, for example, throw any\nlight upon our preference for true beliefs rather than false ones. This\npreference is only explicable by taking account of the causal efficacy\nof beliefs, and of the greater appropriateness of the responses\nresulting from true beliefs. But appropriateness depends upon purpose,\nand purpose thus becomes a vital part of theory of knowledge.\n\n\n\n\nLECTURE XIV. EMOTIONS AND WILL\n\nOn the two subjects of the present lecture I have nothing original to\nsay, and I am treating them only in order to complete the discussion of\nmy main thesis, namely that all psychic phenomena are built up out of\nsensations and images alone.\n\nEmotions are traditionally regarded by psychologists as a separate\nclass of mental occurrences: I am, of course, not concerned to deny\nthe obvious fact that they have characteristics which make a special\ninvestigation of them necessary. What I am concerned with is the\nanalysis of emotions. It is clear that an emotion is essentially\ncomplex, and we have to inquire whether it ever contains any\nnon-physiological material not reducible to sensations and images and\ntheir relations.\n\nAlthough what specially concerns us is the analysis of emotions, we\nshall find that the more important topic is the physiological causation\nof emotions. This is a subject upon which much valuable and exceedingly\ninteresting work has been done, whereas the bare analysis of emotions\nhas proved somewhat barren. In view of the fact that we have defined\nperceptions, sensations, and images by their physiological causation, it\nis evident that our problem of the analysis of the emotions is bound up\nwith the problem of their physiological causation.\n\nModern views on the causation of emotions begin with what is called\nthe James-Lange theory. James states this view in the following terms\n(\"Psychology,\" vol. ii, p. 449):\n\n\"Our natural way of thinking about these coarser emotions, grief, fear,\nrage, love, is that the mental perception of some fact excites the\nmental affection called the emotion, and that this latter state of mind\ngives rise to the bodily expression. My theory, on the contrary, is that\nTHE BODILY CHANGES FOLLOW DIRECTLY THE PERCEPTION OF THE EXCITING FACT,\nAND THAT OUR FEELING OF THE SAME CHANGES AS THEY OCCUR _IS_ THE EMOTION\n(James's italics). Common sense says: we lose our fortune, are sorry\nand weep; we meet a bear, are frightened and run; we are insulted by\na rival, are angry and strike. The hypothesis here to be defended says\nthat this order of sequence is incorrect, that the one mental state is\nnot immediately induced by the other, that the bodily manifestations\nmust first be interposed between, and that the more rational statement\nis that we feel sorry because we cry, angry because we strike, afraid\nbecause we tremble, and not that we cry, strike, or tremble, because\nwe are sorry, angry, or fearful, as the case may be. Without the bodily\nstates following on the perception, the latter would be purely cognitive\nin form, pale, colourless, destitute of emotional warmth.\"\n\nRound this hypothesis a very voluminous literature has grown up. The\nhistory of its victory over earlier criticism, and its difficulties with\nthe modern experimental work of Sherrington and Cannon, is well told\nby James R. Angell in an article called \"A Reconsideration of James's\nTheory of Emotion in the Light of Recent Criticisms.\"* In this article\nAngell defends James's theory and to me--though I speak with diffidence\non a question as to which I have little competence--it appears that his\ndefence is on the whole successful.\n\n     * \"Psychological Review,\" 1916.\n\nSherrington, by experiments on dogs, showed that many of the usual marks\nof emotion were present in their behaviour even when, by severing the\nspinal cord in the lower cervical region, the viscera were cut off from\nall communication with the brain, except that existing through certain\ncranial nerves. He mentions the various signs which \"contributed to\nindicate the existence of an emotion as lively as the animal had ever\nshown us before the spinal operation had been made.\"* He infers that\nthe physiological condition of the viscera cannot be the cause of the\nemotion displayed under such circumstances, and concludes: \"We are\nforced back toward the likelihood that the visceral expression of\nemotion is SECONDARY to the cerebral action occurring with the psychical\nstate.... We may with James accept visceral and organic sensations\nand the memories and associations of them as contributory to primitive\nemotion, but we must regard them as re-enforcing rather than as\ninitiating the psychosis.\"*\n\n     * Quoted by Angell, loc. cit.\n\n\nAngell suggests that the display of emotion in such cases may be due\nto past experience, generating habits which would require only the\nstimulation of cerebral reflex arcs. Rage and some forms of fear,\nhowever, may, he thinks, gain expression without the brain. Rage and\nfear have been especially studied by Cannon, whose work is of the\ngreatest importance. His results are given in his book, \"Bodily Changes\nin Pain, Hunger, Fear and Rage\" (D. Appleton and Co., 1916).\n\nThe most interesting part of Cannon's book consists in the investigation\nof the effects produced by secretion of adrenin. Adrenin is a substance\nsecreted into the blood by the adrenal glands. These are among the\nductless glands, the functions of which, both in physiology and in\nconnection with the emotions, have only come to be known during recent\nyears. Cannon found that pain, fear and rage occurred in circumstances\nwhich affected the supply of adrenin, and that an artificial injection\nof adrenin could, for example, produce all the symptoms of fear. He\nstudied the effects of adrenin on various parts of the body; he found\nthat it causes the pupils to dilate, hairs to stand erect, blood vessels\nto be constricted, and so on. These effects were still produced if\nthe parts in question were removed from the body and kept alive\nartificially.*\n\n     * Cannon's work is not unconnected with that of Mosso, who\n     maintains, as the result of much experimental work, that\n     \"the seat of the emotions lies in the sympathetic nervous\n     system.\" An account of the work of both these men will be\n     found in Goddard's \"Psychology of the Normal and Sub-normal\"\n     (Kegan Paul, 1919), chap. vii and Appendix.\n\nCannon's chief argument against James is, if I understand him rightly,\nthat similar affections of the viscera may accompany dissimilar\nemotions, especially fear and rage. Various different emotions make\nus cry, and therefore it cannot be true to say, as James does, that we\n\"feel sorry because we cry,\" since sometimes we cry when we feel glad.\nThis argument, however, is by no means conclusive against James, because\nit cannot be shown that there are no visceral differences for different\nemotions, and indeed it is unlikely that this is the case.\n\nAs Angell says (loc. cit.): \"Fear and joy may both cause cardiac\npalpitation, but in one case we find high tonus of the skeletal muscles,\nin the other case relaxation and the general sense of weakness.\"\n\nAngell's conclusion, after discussing the experiments of Sherrington\nand Cannon, is: \"I would therefore submit that, so far as concerns\nthe critical suggestions by these two psychologists, James's essential\ncontentions are not materially affected.\" If it were necessary for me to\ntake sides on this question, I should agree with this conclusion; but I\nthink my thesis as to the analysis of emotion can be maintained without\ncoming to a probably premature conclusion upon the doubtful parts of\nthe physiological problem.\n\nAccording to our definitions, if James is right, an emotion may be\nregarded as involving a confused perception of the viscera concerned\nin its causation, while if Cannon and Sherrington are right, an emotion\ninvolves a confused perception of its external stimulus. This follows\nfrom what was said in Lecture VII. We there defined a perception as an\nappearance, however irregular, of one or more objects external to the\nbrain. And in order to be an appearance of one or more objects, it is\nonly necessary that the occurrence in question should be connected\nwith them by a continuous chain, and should vary when they are varied\nsufficiently. Thus the question whether a mental occurrence can be\ncalled a perception turns upon the question whether anything can be\ninferred from it as to its causes outside the brain: if such inference\nis possible, the occurrence in question will come within our definition\nof a perception. And in that case, according to the definition in\nLecture VIII, its non-mnemic elements will be sensations. Accordingly,\nwhether emotions are caused by changes in the viscera or by sensible\nobjects, they contain elements which are sensations according to our\ndefinition.\n\nAn emotion in its entirety is, of course, something much more complex\nthan a perception. An emotion is essentially a process, and it will be\nonly what one may call a cross-section of the emotion that will be a\nperception, of a bodily condition according to James, or (in certain\ncases) of an external object according to his opponents. An emotion in\nits entirety contains dynamic elements, such as motor impulses, desires,\npleasures and pains. Desires and pleasures and pains, according to the\ntheory adopted in Lecture III, are characteristics of processes, not\nseparate ingredients. An emotion--rage, for example--will be a certain\nkind of process, consisting of perceptions and (in general) bodily\nmovements. The desires and pleasures and pains involved are properties\nof this process, not separate items in the stuff of which the emotion\nis composed. The dynamic elements in an emotion, if we are right in our\nanalysis, contain, from our point of view, no ingredients beyond those\ncontained in the processes considered in Lecture III. The ingredients\nof an emotion are only sensations and images and bodily movements\nsucceeding each other according to a certain pattern. With this\nconclusion we may leave the emotions and pass to the consideration of\nthe will.\n\nThe first thing to be defined when we are dealing with Will is a\nVOLUNTARY MOVEMENT. We have already defined vital movements, and we have\nmaintained that, from a behaviourist standpoint, it is impossible to\ndistinguish which among such movements are reflex and which voluntary.\nNevertheless, there certainly is a distinction. When we decide in the\nmorning that it is time to get up, our consequent movement is voluntary.\nThe beating of the heart, on the other hand, is involuntary: we can\nneither cause it nor prevent it by any decision of our own, except\nindirectly, as e.g. by drugs. Breathing is intermediate between the two:\nwe normally breathe without the help of the will, but we can alter or\nstop our breathing if we choose.\n\nJames (\"Psychology,\" chap. xxvi) maintains that the only distinctive\ncharacteristic of a voluntary act is that it involves an idea of the\nmovement to be performed, made up of memory-images of the kinaesthetic\nsensations which we had when the same movement occurred on some former\noccasion. He points out that, on this view, no movement can be made\nvoluntarily unless it has previously occurred involuntarily.*\n\n     * \"Psychology,\" Vol. ii, pp. 492-3.\n\nI see no reason to doubt the correctness of this view. We shall say,\nthen, that movements which are accompanied by kinaesthetic sensations\ntend to be caused by the images of those sensations, and when so caused\nare called VOLUNTARY.\n\nVolition, in the emphatic sense, involves something more than\nvoluntary movement. The sort of case I am thinking of is decision after\ndeliberation. Voluntary movements are a part of this, but not the whole.\nThere is, in addition to them, a judgment: \"This is what I shall\ndo\"; there is also a sensation of tension during doubt, followed by a\ndifferent sensation at the moment of deciding. I see no reason whatever\nto suppose that there is any specifically new ingredient; sensations and\nimages, with their relations and causal laws, yield all that seems to\nbe wanted for the analysis of the will, together with the fact that\nkinaesthetic images tend to cause the movements with which they are\nconnected. Conflict of desires is of course essential in the causation\nof the emphatic kind of will: there will be for a time kinaesthetic\nimages of incompatible movements, followed by the exclusive image of\nthe movement which is said to be willed. Thus will seems to add no new\nirreducible ingredient to the analysis of the mind.\n\n\n\n\nLECTURE XV. CHARACTERISTICS OF MENTAL PHENOMENA\n\nAt the end of our journey it is time to return to the question from\nwhich we set out, namely: What is it that characterizes mind as opposed\nto matter? Or, to state the same question in other terms: How is\npsychology to be distinguished from physics? The answer provisionally\nsuggested at the outset of our inquiry was that psychology and physics\nare distinguished by the nature of their causal laws, not by their\nsubject matter. At the same time we held that there is a certain subject\nmatter, namely images, to which only psychological causal laws are\napplicable; this subject matter, therefore, we assigned exclusively to\npsychology. But we found no way of defining images except through\ntheir causation; in their intrinsic character they appeared to have no\nuniversal mark by which they could be distinguished from sensations.\n\nIn this last lecture I propose to pass in review various suggested\nmethods of distinguishing mind from matter. I shall then briefly sketch\nthe nature of that fundamental science which I believe to be the true\nmetaphysic, in which mind and matter alike are seen to be constructed\nout of a neutral stuff, whose causal laws have no such duality as that\nof psychology, but form the basis upon which both physics and psychology\nare built.\n\nIn search for the definition of \"mental phenomena,\" let us begin with\n\"consciousness,\" which is often thought to be the essence of mind.\nIn the first lecture I gave various arguments against the view that\nconsciousness is fundamental, but I did not attempt to say what\nconsciousness is. We must find a definition of it, if we are to feel\nsecure in deciding that it is not fundamental. It is for the sake of the\nproof that it is not fundamental that we must now endeavour to decide\nwhat it is.\n\n\"Consciousness,\" by those who regard it as fundamental, is taken to be a\ncharacter diffused throughout our mental life, distinct from sensations\nand images, memories, beliefs and desires, but present in all of\nthem.* Dr. Henry Head, in an article which I quoted in Lecture III,\ndistinguishing sensations from purely physiological occurrences, says:\n\"Sensation, in the strict sense of the term, demands the existence of\nconsciousness.\" This statement, at first sight, is one to which we feel\ninclined to assent, but I believe we are mistaken if we do so. Sensation\nis the sort of thing of which we MAY be conscious, but not a thing\nof which we MUST be conscious. We have been led, in the course of our\ninquiry, to admit unconscious beliefs and unconscious desires. There is,\nso far as I can see, no class of mental or other occurrences of which we\nare always conscious whenever they happen.\n\n     * Cf. Lecture VI.\n\nThe first thing to notice is that consciousness must be of something. In\nview of this, I should define \"consciousness\" in terms of that relation\nof an image of a word to an object which we defined, in Lecture XI, as\n\"meaning.\" When a sensation is followed by an image which is a \"copy\" of\nit, I think it may be said that the existence of the image constitutes\nconsciousness of the sensation, provided it is accompanied by that sort\nof belief which, when we reflect upon it, makes us feel that the image\nis a \"sign\" of something other than itself. This is the sort of belief\nwhich, in the case of memory, we expressed in the words \"this occurred\";\nor which, in the case of a judgment of perception, makes us believe\nin qualities correlated with present sensations, as e.g., tactile and\nvisual qualities are correlated. The addition of some element of belief\nseems required, since mere imagination does not involve consciousness of\nanything, and there can be no consciousness which is not of something.\nIf images alone constituted consciousness of their prototypes,\nsuch imagination-images as in fact have prototypes would involve\nconsciousness of them; since this is not the case, an element of belief\nmust be added to the images in defining consciousness. The belief must\nbe of that sort that constitutes objective reference, past or\npresent. An image, together with a belief of this sort concerning it,\nconstitutes, according to our definition, consciousness of the prototype\nof the image.\n\nBut when we pass from consciousness of sensations to consciousness of\nobjects of perception, certain further points arise which demand an\naddition to our definition. A judgment of perception, we may say,\nconsists of a core of sensation, together with associated images, with\nbelief in the present existence of an object to which sensation and\nimages are referred in a way which is difficult to analyse. Perhaps we\nmight say that the belief is not fundamentally in any PRESENT existence,\nbut is of the nature of an expectation: for example, when we see an\nobject, we expect certain sensations to result if we proceed to touch\nit. Perception, then, will consist of a present sensation together with\nexpectations of future sensations. (This, of course, is a reflective\nanalysis, not an account of the way perception appears to unchecked\nintrospection.) But all such expectations are liable to be erroneous,\nsince they are based upon correlations which are usual but not\ninvariable. Any such correlation may mislead us in a particular case,\nfor example, if we try to touch a reflection in a looking-glass under\nthe impression that it is \"real.\" Since memory is fallible, a similar\ndifficulty arises as regards consciousness of past objects. It would\nseem odd to say that we can be \"conscious\" of a thing which does not or\ndid not exist. The only way to avoid this awkwardness is to add to our\ndefinition the proviso that the beliefs involved in consciousness must\nbe TRUE.\n\nIn the second place, the question arises as to whether we can be\nconscious of images. If we apply our definition to this case, it seems\nto demand images of images. In order, for example, to be conscious of\nan image of a cat, we shall require, according to the letter of the\ndefinition, an image which is a copy of our image of the cat, and has\nthis image for its prototype. Now, it hardly seems probable, as a matter\nof observation, that there are images of images, as opposed to images\nof sensations. We may meet this difficulty in two ways, either by boldly\ndenying consciousness of images, or by finding a sense in which, by\nmeans of a different accompanying belief, an image, instead of meaning\nits prototype, can mean another image of the same prototype.\n\nThe first alternative, which denies consciousness of images, has already\nbeen discussed when we were dealing with Introspection in Lecture VI. We\nthen decided that there must be, in some sense, consciousness of images.\nWe are therefore left with the second suggested way of dealing with\nknowledge of images. According to this second hypothesis, there may be\ntwo images of the same prototype, such that one of them means the other,\ninstead of meaning the prototype. It will be remembered that we defined\nmeaning by association a word or image means an object, we said, when it\nhas the same associations as the object. But this definition must not be\ninterpreted too absolutely: a word or image will not have ALL the\nsame associations as the object which it means. The word \"cat\" may\nbe associated with the word \"mat,\" but it would not happen except by\naccident that a cat would be associated with a mat. And in like manner\nan image may have certain associations which its prototype will not\nhave, e.g. an association with the word \"image.\" When these associations\nare active, an image means an image, instead of meaning its prototype.\nIf I have had images of a given prototype many times, I can mean one of\nthese, as opposed to the rest, by recollecting the time and place or any\nother distinctive association of that one occasion. This happens, for\nexample, when a place recalls to us some thought we previously had in\nthat place, so that we remember a thought as opposed to the occurrence\nto which it referred. Thus we may say that we think of an image A when\nwe have a similar image B associated with recollections of circumstances\nconnected with A, but not with its prototype or with other images of the\nsame prototype. In this way we become aware of images without the\nneed of any new store of mental contents, merely by the help of new\nassociations. This theory, so far as I can see, solves the problems of\nintrospective knowledge, without requiring heroic measures such as those\nproposed by Knight Dunlap, whose views we discussed in Lecture VI.\n\nAccording to what we have been saying, sensation itself is not an\ninstance of consciousness, though the immediate memory by which it is\napt to be succeeded is so. A sensation which is remembered becomes an\nobject of consciousness as soon as it begins to be remembered, which\nwill normally be almost immediately after its occurrence (if at all);\nbut while it exists it is not an object of consciousness. If, however,\nit is part of a perception, say of some familiar person, we may say that\nthe person perceived is an object of consciousness. For in this case\nthe sensation is a SIGN of the perceived object in much the same way\nin which a memory-image is a sign of a remembered object. The essential\npractical function of \"consciousness\" and \"thought\" is that they enable\nus to act with reference to what is distant in time or space, even\nthough it is not at present stimulating our senses. This reference\nto absent objects is possible through association and habit. Actual\nsensations, in themselves, are not cases of consciousness, because they\ndo not bring in this reference to what is absent. But their connection\nwith consciousness is very close, both through immediate memory, and\nthrough the correlations which turn sensations into perceptions.\n\nEnough has, I hope, been said to show that consciousness is far too\ncomplex and accidental to be taken as the fundamental characteristic\nof mind. We have seen that belief and images both enter into it. Belief\nitself, as we saw in an earlier lecture, is complex. Therefore, if any\ndefinition of mind is suggested by our analysis of consciousness, images\nare what would naturally suggest themselves. But since we found\nthat images can only be defined causally, we cannot deal with this\nsuggestion, except in connection with the difference between physical\nand psychological causal laws.\n\nI come next to those characteristics of mental phenomena which arise out\nof mnemic causation. The possibility of action with reference to what\nis not sensibly present is one of the things that might be held to\ncharacterize mind. Let us take first a very elementary example. Suppose\nyou are in a familiar room at night, and suddenly the light goes out.\nYou will be able to find your way to the door without much difficulty\nby means of the picture of the room which you have in your mind. In this\ncase visual images serve, somewhat imperfectly it is true, the purpose\nwhich visual sensations would otherwise serve. The stimulus to the\nproduction of visual images is the desire to get out of the room, which,\naccording to what we found in Lecture III, consists essentially of\npresent sensations and motor impulses caused by them. Again, words heard\nor read enable you to act with reference to the matters about which they\ngive information; here, again, a present sensible stimulus, in virtue of\nhabits formed in the past, enables you to act in a manner appropriate\nto an object which is not sensibly present. The whole essence of the\npractical efficiency of \"thought\" consists in sensitiveness to signs:\nthe sensible presence of A, which is a sign of the present or future\nexistence of B, enables us to act in a manner appropriate to B. Of\nthis, words are the supreme example, since their effects as signs are\nprodigious, while their intrinsic interest as sensible occurrences on\ntheir own account is usually very slight. The operation of signs may or\nmay not be accompanied by consciousness. If a sensible stimulus A calls\nup an image of B, and we then act with reference to B, we have what may\nbe called consciousness of B. But habit may enable us to act in a manner\nappropriate to B as soon as A appears, without ever having an image of\nB. In that case, although A operates as a sign, it operates without the\nhelp of consciousness. Broadly speaking, a very familiar sign tends to\noperate directly in this manner, and the intervention of consciousness\nmarks an imperfectly established habit.\n\nThe power of acquiring experience, which characterizes men and animals,\nis an example of the general law that, in mnemic causation, the causal\nunit is not one event at one time, but two or more events at two or more\ntimes.& A burnt child fears the fire, that is to say, the neighbourhood\nof fire has a different effect upon a child which has had the sensations\nof burning than upon one which has not. More correctly, the observed\neffect, when a child which has been burnt is put near a fire, has for\nits cause, not merely the neighbourhood of the fire, but this together\nwith the previous burning. The general formula, when an animal has\nacquired experience through some event A, is that, when B occurs at some\nfuture time, the animal to which A has happened acts differently from\nan animal which A has not happened. Thus A and B together, not either\nseparately, must be regarded as the cause of the animal's behaviour,\nunless we take account of the effect which A has had in altering the\nanimal's nervous tissue, which is a matter not patent to external\nobservation except under very special circumstances. With this\npossibility, we are brought back to causal laws, and to the suggestion\nthat many things which seem essentially mental are really neural.\nPerhaps it is the nerves that acquire experience rather than the mind.\nIf so, the possibility of acquiring experience cannot be used to define\nmind.*\n\n     * Cf. Lecture IV.\n\nVery similar considerations apply to memory, if taken as the essence of\nmind. A recollection is aroused by something which is happening now,\nbut is different from the effect which the present occurrence would\nhave produced if the recollected event had not occurred. This may be\naccounted for by the physical effect of the past event on the brain,\nmaking it a different instrument from that which would have resulted\nfrom a different experience. The causal peculiarities of memory may,\ntherefore, have a physiological explanation. With every special class of\nmental phenomena this possibility meets us afresh. If psychology is\nto be a separate science at all, we must seek a wider ground for its\nseparateness than any that we have been considering hitherto.\n\nWe have found that \"consciousness\" is too narrow to characterize mental\nphenomena, and that mnemic causation is too wide. I come now to a\ncharacteristic which, though difficult to define, comes much nearer to\nwhat we require, namely subjectivity.\n\nSubjectivity, as a characteristic of mental phenomena, was considered in\nLecture VII, in connection with the definition of perception. We there\ndecided that those particulars which constitute the physical world can\nbe collected into sets in two ways, one of which makes a bundle of all\nthose particulars that are appearances of a given thing from different\nplaces, while the other makes a bundle of all those particulars which\nare appearances of different things from a given place. A bundle of\nthis latter sort, at a given time, is called a \"perspective\"; taken\nthroughout a period of time, it is called a \"biography.\" Subjectivity is\nthe characteristic of perspectives and biographies, the characteristic\nof giving the view of the world from a certain place. We saw in Lecture\nVII that this characteristic involves none of the other characteristics\nthat are commonly associated with mental phenomena, such as\nconsciousness, experience and memory. We found in fact that it is\nexhibited by a photographic plate, and, strictly speaking, by any\nparticular taken in conjunction with those which have the same \"passive\"\nplace in the sense defined in Lecture VII. The particulars forming one\nperspective are connected together primarily by simultaneity;\nthose forming one biography, primarily by the existence of direct\ntime-relations between them. To these are to be added relations\nderivable from the laws of perspective. In all this we are clearly not\nin the region of psychology, as commonly understood; yet we are also\nhardly in the region of physics. And the definition of perspectives\nand biographies, though it does not yet yield anything that would\nbe commonly called \"mental,\" is presupposed in mental phenomena, for\nexample in mnemic causation: the causal unit in mnemic causation, which\ngives rise to Semon's engram, is the whole of one perspective--not of\nany perspective, but of a perspective in a place where there is nervous\ntissue, or at any rate living tissue of some sort. Perception also,\nas we saw, can only be defined in terms of perspectives. Thus the\nconception of subjectivity, i.e. of the \"passive\" place of a particular,\nthough not alone sufficient to define mind, is clearly an essential\nelement in the definition.\n\nI have maintained throughout these lectures that the data of psychology\ndo not differ in, their intrinsic character from the data of physics.\nI have maintained that sensations are data for psychology and\nphysics equally, while images, which may be in some sense exclusively\npsychological data, can only be distinguished from sensations by their\ncorrelations, not by what they are in themselves. It is now necessary,\nhowever, to examine the notion of a \"datum,\" and to obtain, if possible,\na definition of this notion.\n\nThe notion of \"data\" is familiar throughout science, and is usually\ntreated by men of science as though it were perfectly clear.\nPsychologists, on the other hand, find great difficulty in the\nconception. \"Data\" are naturally defined in terms of theory of\nknowledge: they are those propositions of which the truth is known\nwithout demonstration, so that they may be used as premisses in proving\nother propositions. Further, when a proposition which is a datum asserts\nthe existence of something, we say that the something is a datum, as\nwell as the proposition asserting its existence. Thus those objects\nof whose existence we become certain through perception are said to be\ndata.\n\nThere is some difficulty in connecting this epistemological definition\nof \"data\" with our psychological analysis of knowledge; but until such\na connection has been effected, we have no right to use the conception\n\"data.\"\n\nIt is clear, in the first place, that there can be no datum apart from a\nbelief. A sensation which merely comes and goes is not a datum; it only\nbecomes a datum when it is remembered. Similarly, in perception, we do\nnot have a datum unless we have a JUDGMENT of perception. In the sense\nin which objects (as opposed to propositions) are data, it would seem\nnatural to say that those objects of which we are conscious are data.\nBut consciousness, as we have seen, is a complex notion, involving\nbeliefs, as well as mnemic phenomena such as are required for perception\nand memory. It follows that no datum is theoretically indubitable, since\nno belief is infallible; it follows also that every datum has a greater\nor less degree of vagueness, since there is always some vagueness in\nmemory and the meaning of images.\n\nData are not those things of which our consciousness is earliest in\ntime. At every period of life, after we have become capable of thought,\nsome of our beliefs are obtained by inference, while others are not.\nA belief may pass from either of these classes into the other, and may\ntherefore become, or cease to be, a belief giving a datum. When, in what\nfollows, I speak of data, I do not mean the things of which we feel sure\nbefore scientific study begins, but the things which, when a science\nis well advanced, appear as affording grounds for other parts of\nthe science, without themselves being believed on any ground except\nobservation. I assume, that is to say, a trained observer, with an\nanalytic attention, knowing the sort of thing to look for, and the sort\nof thing that will be important. What he observes is, at the stage of\nscience which he has reached, a datum for his science. It is just as\nsophisticated and elaborate as the theories which he bases upon it,\nsince only trained habits and much practice enable a man to make\nthe kind of observation that will be scientifically illuminating.\nNevertheless, when once it has been observed, belief in it is not based\non inference and reasoning, but merely upon its having been seen. In\nthis way its logical status differs from that of the theories which are\nproved by its means.\n\nIn any science other than psychology the datum is primarily a\nperception, in which only the sensational core is ultimately and\ntheoretically a datum, though some such accretions as turn the sensation\ninto a perception are practically unavoidable. But if we postulate an\nideal observer, he will be able to isolate the sensation, and treat this\nalone as datum. There is, therefore, an important sense in which we\nmay say that, if we analyse as much as we ought, our data, outside\npsychology, consist of sensations, which include within themselves\ncertain spatial and temporal relations.\n\nApplying this remark to physiology, we see that the nerves and brain\nas physical objects are not truly data; they are to be replaced, in\nthe ideal structure of science, by the sensations through which the\nphysiologist is said to perceive them. The passage from these sensations\nto nerves and brain as physical objects belongs really to the initial\nstage in the theory of physics, and ought to be placed in the reasoned\npart, not in the part supposed to be observed. To say we see the\nnerves is like saying we hear the nightingale; both are convenient but\ninaccurate expressions. We hear a sound which we believe to be causally\nconnected with the nightingale, and we see a sight which we believe\nto be causally connected with a nerve. But in each case it is only\nthe sensation that ought, in strictness, to be called a datum. Now,\nsensations are certainly among the data of psychology. Therefore all the\ndata of the physical sciences are also psychological data. It remains\nto inquire whether all the data of psychology are also data of physical\nscience, and especially of physiology.\n\nIf we have been right in our analysis of mind, the ultimate data of\npsychology are only sensations and images and their relations. Beliefs,\ndesires, volitions, and so on, appeared to us to be complex phenomena\nconsisting of sensations and images variously interrelated. Thus (apart\nfrom certain relations) the occurrences which seem most distinctively\nmental, and furthest removed from physics, are, like physical objects,\nconstructed or inferred, not part of the original stock of data in the\nperfected science. From both ends, therefore, the difference between\nphysical and psychological data is diminished. Is there ultimately\nno difference, or do images remain as irreducibly and exclusively\npsychological? In view of the causal definition of the difference\nbetween images and sensations, this brings us to a new question, namely:\nAre the causal laws of psychology different from those of any other\nscience, or are they really physiological?\n\nCertain ambiguities must be removed before this question can be\nadequately discussed.\n\nFirst, there is the distinction between rough approximate laws and\nsuch as appear to be precise and general. I shall return to the former\npresently; it is the latter that I wish to discuss now.\n\nMatter, as defined at the end of Lecture V, is a logical fiction,\ninvented because it gives a convenient way of stating causal laws.\nExcept in cases of perfect regularity in appearances (of which we can\nhave no experience), the actual appearances of a piece of matter are not\nmembers of that ideal system of regular appearances which is defined\nas being the matter in question. But the matter is, after all, inferred\nfrom its appearances, which are used to VERIFY physical laws. Thus, in\nso far as physics is an empirical and verifiable science, it must assume\nor prove that the inference from appearances to matter is, in general,\nlegitimate, and it must be able to tell us, more or less, what\nappearances to expect. It is through this question of verifiability and\nempirical applicability to experience that we are led to a theory of\nmatter such as I advocate. From the consideration of this question it\nresults that physics, in so far as it is an empirical science, not a\nlogical phantasy, is concerned with particulars of just the same sort\nas those which psychology considers under the name of sensations. The\ncausal laws of physics, so interpreted, differ from those of psychology\nonly by the fact that they connect a particular with other appearances\nin the same piece of matter, rather than with other appearances in the\nsame perspective. That is to say, they group together particulars having\nthe same \"active\" place, while psychology groups together those having\nthe same \"passive\" place. Some particulars, such as images, have no\n\"active\" place, and therefore belong exclusively to psychology.\n\nWe can now understand the distinction between physics and psychology.\nThe nerves and brain are matter: our visual sensations when we look\nat them may be, and I think are, members of the system constituting\nirregular appearances of this matter, but are not the whole of the\nsystem. Psychology is concerned, inter alia, with our sensations when we\nsee a piece of matter, as opposed to the matter which we see. Assuming,\nas we must, that our sensations have physical causes, their causal laws\nare nevertheless radically different from the laws of physics, since\nthe consideration of a single sensation requires the breaking up of\nthe group of which it is a member. When a sensation is used to verify\nphysics, it is used merely as a sign of a certain material phenomenon,\ni.e. of a group of particulars of which it is a member. But when it is\nstudied by psychology, it is taken away from that group and put\ninto quite a different context, where it causes images or voluntary\nmovements. It is primarily this different grouping that is\ncharacteristic of psychology as opposed to all the physical sciences,\nincluding physiology; a secondary difference is that images, which\nbelong to psychology, are not easily to be included among the aspects\nwhich constitute a physical thing or piece of matter.\n\nThere remains, however, an important question, namely: Are mental events\ncausally dependent upon physical events in a sense in which the converse\ndependence does not hold? Before we can discuss the answer to this\nquestion, we must first be clear as to what our question means.\n\nWhen, given A, it is possible to infer B, but given B, it is not\npossible to infer A, we say that B is dependent upon A in a sense in\nwhich A is not dependent upon B. Stated in logical terms, this amounts\nto saying that, when we know a many-one relation of A to B, B is\ndependent upon A in respect of this relation. If the relation is a\ncausal law, we say that B is causally dependent upon A. The illustration\nthat chiefly concerns us is the system of appearances of a physical\nobject. We can, broadly speaking, infer distant appearances from near\nones, but not vice versa. All men look alike when they are a mile away,\nhence when we see a man a mile off we cannot tell what he will look like\nwhen he is only a yard away. But when we see him a yard away, we can\ntell what he will look like a mile away. Thus the nearer view gives us\nmore valuable information, and the distant view is causally dependent\nupon it in a sense in which it is not causally dependent upon the\ndistant view.\n\nIt is this greater causal potency of the near appearance that leads\nphysics to state its causal laws in terms of that system of regular\nappearances to which the nearest appearances increasingly approximate,\nand that makes it value information derived from the microscope or\ntelescope. It is clear that our sensations, considered as irregular\nappearances of physical objects, share the causal dependence belonging\nto comparatively distant appearances; therefore in our sensational life\nwe are in causal dependence upon physical laws.\n\nThis, however, is not the most important or interesting part of our\nquestion. It is the causation of images that is the vital problem. We\nhave seen that they are subject to mnenic causation, and that mnenic\ncausation may be reducible to ordinary physical causation in nervous\ntissue. This is the question upon which our attitude must turn towards\nwhat may be called materialism. One sense of materialism is the view\nthat all mental phenomena are causally dependent upon physical phenomena\nin the above-defined sense of causal dependence. Whether this is the\ncase or not, I do not profess to know. The question seems to me the\nsame as the question whether mnemic causation is ultimate, which we\nconsidered without deciding in Lecture IV. But I think the bulk of the\nevidence points to the materialistic answer as the more probable.\n\nIn considering the causal laws of psychology, the distinction between\nrough generalizations and exact laws is important. There are many rough\ngeneralizations in psychology, not only of the sort by which we\ngovern our ordinary behaviour to each other, but also of a more nearly\nscientific kind. Habit and association belong among such laws. I will\ngive an illustration of the kind of law that can be obtained. Suppose a\nperson has frequently experienced A and B in close temporal contiguity,\nan association will be established, so that A, or an image of A, tends\nto cause an image of B. The question arises: will the association work\nin either direction, or only from the one which has occurred earlier\nto the one which has occurred later? In an article by Mr. Wohlgemuth,\ncalled \"The Direction of Associations\" (\"British Journal of Psychology,\"\nvol. v, part iv, March, 1913), it is claimed to be proved by experiment\nthat, in so far as motor memory (i.e. memory of movements) is concerned,\nassociation works only from earlier to later, while in visual and\nauditory memory this is not the case, but the later of two neighbouring\nexperiences may recall the earlier as well as the earlier the later.\nIt is suggested that motor memory is physiological, while visual and\nauditory memory are more truly psychological. But that is not the point\nwhich concerns us in the illustration. The point which concerns us\nis that a law of association, established by purely psychological\nobservation, is a purely psychological law, and may serve as a sample\nof what is possible in the way of discovering such laws. It is, however,\nstill no more than a rough generalization, a statistical average. It\ncannot tell us what will result from a given cause on a given occasion.\nIt is a law of tendency, not a precise and invariable law such as those\nof physics aim at being.\n\nIf we wish to pass from the law of habit, stated as a tendency or\naverage, to something more precise and invariable, we seem driven to the\nnervous system. We can more or less guess how an occurrence produces a\nchange in the brain, and how its repetition gradually produces something\nanalogous to the channel of a river, along which currents flow more\neasily than in neighbouring paths. We can perceive that in this way, if\nwe had more knowledge, the tendency to habit through repetition might\nbe replaced by a precise account of the effect of each occurrence\nin bringing about a modification of the sort from which habit would\nultimately result. It is such considerations that make students of\npsychophysiology materialistic in their methods, whatever they may be in\ntheir metaphysics. There are, of course, exceptions, such as Professor\nJ. S. Haldane,* who maintains that it is theoretically impossible to\nobtain physiological explanations of psychical phenomena, or physical\nexplanations of physiological phenomena. But I think the bulk of expert\nopinion, in practice, is on the other side.\n\n     *See his book, \"The New Physiology and Other Addresses\"\n     (Charles Griffin & Co., 1919).\n\nThe question whether it is possible to obtain precise causal laws in\nwhich the causes are psychological, not material, is one of detailed\ninvestigation. I have done what I could to make clear the nature of the\nquestion, but I do not believe that it is possible as yet to answer it\nwith any confidence. It seems to be by no means an insoluble question,\nand we may hope that science will be able to produce sufficient grounds\nfor regarding one answer as much more probable than the other. But for\nthe moment I do not see how we can come to a decision.\n\nI think, however, on grounds of the theory of matter explained in\nLectures V and VII, that an ultimate scientific account of what goes on\nin the world, if it were ascertainable, would resemble psychology rather\nthan physics in what we found to be the decisive difference between\nthem. I think, that is to say, that such an account would not be content\nto speak, even formally, as though matter, which is a logical fiction,\nwere the ultimate reality. I think that, if our scientific knowledge\nwere adequate to the task, which it neither is nor is likely to become,\nit would exhibit the laws of correlation of the particulars constituting\na momentary condition of a material unit, and would state the causal\nlaws* of the world in terms of these particulars, not in terms of\nmatter. Causal laws so stated would, I believe, be applicable to\npsychology and physics equally; the science in which they were stated\nwould succeed in achieving what metaphysics has vainly attempted, namely\na unified account of what really happens, wholly true even if not the\nwhole of truth, and free from all convenient fictions or unwarrantable\nassumptions of metaphysical entities. A causal law applicable to\nparticulars would count as a law of physics if it could be stated in\nterms of those fictitious systems of regular appearances which are\nmatter; if this were not the case, it would count as a law of psychology\nif one of the particulars were a sensation or an image, i.e. were\nsubject to mnemic causation. I believe that the realization of the\ncomplexity of a material unit, and its analysis into constituents\nanalogous to sensations, is of the utmost importance to philosophy, and\nvital for any understanding of the relations between mind and matter,\nbetween our perceptions and the world which they perceive. It is in this\ndirection, I am convinced, that we must look for the solution of many\nancient perplexities.\n\n     * In a perfected science, causal laws will take the form of\n     differential equations--or of finite-difference equations,\n     if the theory of quanta should prove correct.\n\nIt is probable that the whole science of mental occurrences, especially\nwhere its initial definitions are concerned, could be simplified by the\ndevelopment of the fundamental unifying science in which the causal laws\nof particulars are sought, rather than the causal laws of those systems\nof particulars that constitute the material units of physics. This\nfundamental science would cause physics to become derivative, in the\nsort of way in which theories of the constitution of the atom make\nchemistry derivative from physics; it would also cause psychology to\nappear less singular and isolated among sciences. If we are right in\nthis, it is a wrong philosophy of matter which has caused many of the\ndifficulties in the philosophy of mind--difficulties which a right\nphilosophy of matter would cause to disappear.\n\nThe conclusions at which we have arrived may be summed up as follows:\n\nI. Physics and psychology are not distinguished by their material. Mind\nand matter alike are logical constructions; the particulars out of which\nthey are constructed, or from which they are inferred, have various\nrelations, some of which are studied by physics, others by psychology.\nBroadly speaking, physics group particulars by their active places,\npsychology by their passive places.\n\nII. The two most essential characteristics of the causal laws which\nwould naturally be called psychological are SUBJECTIVITY and MNEMIC\nCAUSATION; these are not unconnected, since the causal unit in mnemic\ncausation is the group of particulars having a given passive place at\na given time, and it is by this manner of grouping that subjectivity is\ndefined.\n\nIII. Habit, memory and thought are all developments of mnemic causation.\nIt is probable, though not certain, that mnemic causation is derivative\nfrom ordinary physical causation in nervous (and other) tissue.\n\nIV. Consciousness is a complex and far from universal characteristic of\nmental phenomena.\n\nV. Mind is a matter of degree, chiefly exemplified in number and\ncomplexity of habits.\n\nVI. All our data, both in physics and psychology, are subject to\npsychological causal laws; but physical causal laws, at least in\ntraditional physics, can only be stated in terms of matter, which is\nboth inferred and constructed, never a datum. In this respect psychology\nis nearer to what actually exists.","title":"The Analysis of Mind","tradition":"philosophy-ethics","translated_text":null,"word_count":89115}