The cogito ergo sum of Descartes has been variously regarded by subsequent philosophers, and much discussion has been given to such questions as whether it is an inference or not, and, if it is, what conclusion is drawn from what premise (or premises) and whether the inference is valid or invalid. As we shall see, the difficulties here are largely due to rationalistic confusion as to what inference is; but they can best be resolved by an examination of the line of argument by which the formula itself is arrived at, and an exhibition of the confusions which that argument involves. It can thus be shown that the cogito as it appears in Descartes (i.e., as a “principle” or an “intuition” or a “rational certainty”) is utterly without foundation, and that, as a consequence, certain conceptions which have quite a wide currency among present-day philosophers — particularly the conceptions of “subject” and the “subjective” — must likewise be abandoned.

First of all, however, we may observe that Descartes's own presentation of the case would naturally give rise to divergent interpretations, the point being that he has no consistent view, that he exhibits the instability which is characteristic of rationalism. Thus it might be said that, in spite of its containing ergo, the cogito cannot be taken as an inference, since Descartes specifically refers to it as a proposition which he has discovered to be true, and uses it to illustrate the general conditions under which a proposition can be true and certain. Again, it might seem fairly easy, even if the cogito were taken as an inference, to dismiss that interpretation of it according to which it proceeds from an activity to a substance which has that activity (cogito ergo ego, so to speak). For, as Gilson emphasises in the notes to his edition of the Discourse, the Latin text reads Ego cogito, ergo sum, sive existo, so that the ego is in the premise, if premise it be. Moreover, as Descartes immediately goes on to determine the substance as one “whose whole essence or nature consists in thinking”, this would seem to reduce the “inference” to cogito ergo cogito. Nevertheless, remembering that rationalism is a philosophy of essences or identities, we may consider that the reduction of inference to identity is what Descartes's argument really amounts to, and that the conception of “substance” is only one of the devices by which the emptiness of the position is concealed. A more detailed explanation will, I think, bear out these points.

  ― 102 ―
Descartes, as we have noted, wished to establish the cogito as a true and certain proposition, as contrasted with propositions which may or may not be true but at least can be doubted. He had decided, in the search for truth, “to reject as absolutely false all opinions in regard to which I could suppose the least ground for doubt, in order to ascertain whether after that there remained aught in my belief that was wholly indubitable”. On this basis he found he could reject the evidence of the senses, the reasonings of geometry, and, indeed, all the objects that ever entered his mind. “But immediately upon this I observed that, whilst I thus wished to think that all was false, it was absolutely necessary that I, who thus thought, should be somewhat.” This result he formulates as the cogito (“I think, hence I am”); this is what remains in his belief that is wholly indubitable.

The mechanism of the doubting process, then, is the recognition of something in our belief, some proposition that we entertain, which it is possible to consider false. Thus we can say “I think that grass is green, but I may be wrong” or “I think that the sum of the angles of a triangle is equal to two right angles, but I may be wrong”; these doubts are quite possible, i.e., they contain no internal contradiction. But we cannot say “I think that I think, but I may be wrong”, because even in being wrong we should be thinking, and thus there is contradiction in supposing that we are wrong in this case. It is worth noting that, in following Descartes's own version of the procedure of doubting, we have arrived simply at I think as that which is not subject to doubt. As Descartes indicates, it is “I, who thus thought” that must be something, and this something (as previously pointed out) is “a substance whose whole essence or nature consists only in thinking”; so that what must be something is only my thinking — or, if we care to press the argument further, only thinking. Waiving the latter point, we must insist that the formula that Descartes has been able to reject as involving contradiction is “I think that cogito, but I may be wrong” and not “I think that cogito ergo sum, but I may be wrong”.

The rejection of views on the ground of the contradictions they involve is, of course, a regular part of rationalistic or identity-philosophy, and always depends on ambiguity or confusion of some kind. Descartes's confusions can be more clearly exposed if we emphasise the differences between the concluding doubts which, in the above examples, have been expressed by the same formula, “I may be wrong”. The question, we have noted, is one of possibility and necessity. Thus doubting that grass is green, or considering that our belief to that effect is not certain, may be expressed as “I think that grass is green, but it is possible that grass is not green”. The corresponding attempt to doubt that we think would take the form “I think that I think, but it is possible that I do not think”. And the argument is that, whereas there is no contradiction in the first case and thus it is possible to accept the contradictory of “Grass is green”, there is a contradiction involved in the second case and thus it is not possible to accept the contradictory of “I think”. In other words, to deny

  ― 103 ―
that “I think” is self-refuting, and “I think” is certain or, as it is sometimes put, “follows from the principle of contradiction”. Actually, of course, nothing follows from the principle of contradiction, and no proposition is “self-contradictory”. It may here be remarked that those who contend that a “self-contradictory” proposition must be rejected as false, and thus its contradictory accepted as true and certain, do not observe that, since it is its own contradictory, its contradictory also must be false. But, while it is important to observe that there is confusion in any postulation of “rational” or self-evident truths, our immediate concern is with the confusions underlying the cogito in particular.

Returning, then, to our last formulation of the Cartesian doubt, and putting in brackets the proposition to be examined in each case, we have “I think (that grass is green), but it is possible that grass is not green”; “I think (that I think), but it is possible that I do not think”. The latter formula is rejected as absurd, and it is asserted, on the contrary, that “I think (that I think), and it is not possible that I do not think”, i.e., it is necessary that I think. Now, as regards the rejected formula, the point is that its absurdity depends not on the second, bracketed “I think”, but on the first, and the second, which was the one to be examined and established, has not been examined at all. This is clearly seen if we put something else in the brackets; if we say “I think (that grass is green), but it is possible that I do not think”, the absurdity remains exactly as before. On the other hand, if we retain the second “I think” and substitute something else for the first, if we say, for example, “It is said (that I think), but it is possible that I do not think”, there is now no absurdity. It appears, then, that the bracketed part has nothing to do with the rejection of the formula, and that the contradiction on which Descartes's “demonstration” depends is that which is involved in saying “I think, but it is possible that I do not think” — a contradiction which would appear equally in the assertion, “Grass is green, but it is possible that grass is not green”.

Thus it is not the case that the proposition “I think” has been subjected to the doubting process and has withstood the test and emerged as indubitable. It is the first “I think”, the one that is common to all the formulae and stands outside the brackets, that remains at the end of the process; and it remains merely asserted, as it was to begin with. Descartes, as he has told us, has set out to examine his thoughts, and, in saying that he is doing so, he implies that he has thoughts or does think. We take this as a piece of information that he possesses, something he believes or has found to be the case, just as he might find that grass is green. Formally the two beliefs are equally capable of being contradicted; the proposition “I am not thinking” is no more absurd (even if it be false) than the proposition “Grass is not green”. But it is possible that in actual fact we do not contradict or doubt either of the original propositions, that we regard them both as pieces of correct information; and if Descartes really doubts or disbelieves that grass is green, it is only because of other beliefs that he holds. In any case, whatever may be the beliefs and doubts of any particular person, no distinction at all has been indicated between

  ― 104 ―
propositions dubitable and indubitable in themselves, between mere contingencies and “first principles”.

It may be noted, incidentally, that at best the proposition “I, Descartes, think” would be certain only for Descartes, that anyone else might doubt it freely, so that Descartes would not on his own showing have established any proposition that was certain in itself. That is why he has to go on from his self-certainty to the self-certain (the “perfect being”); he has to set up a universal (or essential) Essence in order to support the conception of himself as a particular essence, viz., a thinking essence. On the other hand, he has to import his essence theory into some empirical field in order to appear to have anything definite to say; a vague, general rationalism, in the form, say, that ultimately there must be a “reason” for everything, would not have the force of the cogito — and, of course, would not have the same mischievous effect on the empirical study of mind.

The method, however, is the same in either case; “perfection” and cogito are alike made to appear to be self-establishing (even if the latter has afterwards to be additionally established by the former), and this can only be done by means of some sort of equivocation, the removal of which shows that nothing at all has been established. In the case we have considered, when the second “I think” is removed as irrelevant, we are left with identity masquerading as inference; we are left with “I think, and it is not possible that I do not think” or “I think, and it is necessary that I think” — in other words, we are left, after all, with cogito ergo cogito, and we then see that the ergo cogito is superfluous, and that the first cogito is not established but merely asserted. More generally, it can be said that the very notion of “certainties” or “necessary truths” is an attempted amalgamation of truth and implication, a uniting of a proposition with a relation between propositions in the supposition that it has that relation to itself — just as the theory of “ideas” attempts to unite being true and being believed in the supposition of something whose truth resides in its being believed. The conception of “that which establishes itself by thinking itself” is, then, only a special case of the general confusion of character and relation. And it is on the same logical footing as that which, as we may put it, establishes itself by establishing itself — the necessary being which must exist (or necessarily is) because it is a necessary being.

This last conception may be further elucidated by the consideration, hitherto deferred, of the use of sum in the Cartesian formula. Why, it may be asked, does not Descartes confine himself to the “I think” which his method proves, if it proves anything? As his own further argument shows, nothing else that can be attributed to his ego can stand the test; it is on that account that he is said to be a thinking essence and nothing more. And that is why such formulae as “I walk, therefore I am” are not accepted as alternatives to the cogito. Logically, as far as sum is concerned, there is no difference between the two cases; “I am walking” implies “I am” no more and no less than “I am thinking” does. In actually distinguishing

  ― 105 ―
them, in saying that my walking does not imply existence whereas my thinking does, Descartes is saying that the latter supposition is bound to be true, while the former is not. And the only support for this contention is to be found in the non-contradiction of “I think that I walk, but it is possible that I do not walk”, i.e., as we have seen, in the non-contradiction of “I think, but it is possible that I do not walk”. This, of course, is no more reason for describing my walking as doubtful than is “I walk, but it is possible that I do not think” for describing my thinking as doubtful. But while the fact that my walking is not my thinking is the sole reason for doubting that I walk, and the identity that my thinking is my thinking is the sole reason for asserting that I think, while, in fact, sum is equally irrelevant to what is rejected and to what is accepted in accordance with the “method”, its inclusion in the latter has the advantage of making it appear something more than an identity, and at the same time conveys its “acceptability”.

Sum, then, in the first instance, appears in the formula to convey the notion of truth, of a truth which is added to the “content”, and which, as added by means of ergo, appears further as a necessary or guaranteed truth, though nothing has been said or can be said to justify such a notion. The addition, moreover, helps to conceal the fact that truth is already conveyed by the copula of any assertion; so that to assert “I am walking” is to assert the truth of “I am walking”, and, though this can be questioned, the question is as to the truth of “I am walking” or the truth of “I am not walking”, and nothing can be added to the belief in one or the other. There are various propositions which we believe and the contradictories of which we disbelieve; but the sum is introduced to make it appear that we can do something more than believing. In the second place it is introduced to give some particularity, some positive character, to the doctrine which is being developed, by emphasising the ego as distinct from the cogitans or the cogitatio. Actually it is impossible on Descartes's theory to distinguish one cogitator or cogitation from another. When he says that he is a substance whose whole essence consists in thinking, he annuls any possible distinction between himself and another thinking substance; yet he is compelled to say so, because his thinking is the one thing that fulfils his conditions of indubitability or “essentiality”. The sum, then, is a means of smuggling in distinctions, of which, strictly speaking, an essence or “identity” theory can make nothing. In the same way, Socrates falls into inconsistencies when he attempts to distinguish one man from another and, at the same time, to make manhood the essence of all of them. But the fact that we are immediately aware of such distinctions tends to make us, in accepting them, overlook their incompatibility with whatever rationalistic suppositions are in question.

Thirdly, the use of sum in the formula is a step towards the establishment of the perfect or necessary. In reducing an empirical fact to an essence, Descartes has amalgamated subject and predicate; the fact of “my thinking” is equated to the fact (existence) of “myself”. But so long as the distinction remains between existence and what exists, so long as the

  ― 106 ―
copula appears even in the distorted form of an attribute, there is a danger of the re-establishment of the distinction between subject and predicate, i.e., of the recognition of facts (“brute” facts, as the phrase goes) which simply have to be learned, instead of being spun out of some essence or “explained” by a “reason”. Consistently with this recognition, of course, anything that can be called an explanation is simply a relation among facts and is itself a “brute” fact. To save the Cartesian doctrine of “transparency”, then, that which is to be seen through has to be further reduced; the step has to be taken of amalgamating subject and existence, finding something whose essence “comprises” existence. And so we have the verbalism of the ontological argument, the very emptiness of which gives it plausibility, while leaving those who accept it free to fill it out with any material that they prefer — to choose which of two opposite conditions they will call a “perfection” and regard as sustained by the perfect. The very fact that there is nothing there makes the essential existence of the essentially existing all the more “transparent” to uncritical minds. And, while there is no logical passage from this to minor transparencies, while Descartes is involved in fresh equivocations in trying to show how there can be any essence but Essence, how, in particular, there can be an ego, his principle device, as has been suggested, is already present in the notion of “perfection”. The sum, then, has served its purpose in the fabrication of this rationalistic edifice, both by its identification of an actual thing with an essence and by its treatment of truth as an attribute.

The foregoing argument, if it be sound, shows that criticism of Descartes must be on grounds of logic, and the persistence of rationalistic confusions of the functions of subject, predicate and copula (as in the distinction of an “is” of existence from the “is” of predication, and so forth) shows that such criticism is not of merely historical interest. But it is on the psychological side that Descartes has been specially influential; and, though criticism of rationalistic psychology must still be logical (i.e., must be criticism of rationalism), it is important to bring out the particular ways in which mental events are confused and obscured by Cartesian assumptions. In the first instance, however, it should be observed that those who have followed Descartes in this matter have not in general repeated his argument, that many of them, indeed, have so disguised the introduction of the cogito into their theories that they appear to be anti-Cartesian. Thus Berkeley, in the second paragraph of his Principles, remarks that “besides all that endless variety of ideas or objects of knowledge, there is likewise something which knows or perceives them, and exercises divers operations, as willing, imagining, remembering about them”; and here the cogito is introduced by the use of the simple word besides. Again, Hume, in spite of his criticism, later in the Treatise, of those philosophers “who imagine we are every moment intimately conscious of what we call our self”, has assumed the cogito from the very beginning in speaking of the objects of our knowledge as “perceptions of the human mind”; i.e., he has assumed, like Berkeley, that they are known

  ― 107 ―
as known and thus as relative to something else which is on a different footing. And Reid, while in the Introduction to his Inquiry he raises explicit objections to the cogito, implicitly accepts it, in the same place, in putting forward a doctrine of inner knowledge (when he says, e.g., that a man's own mind “is the only subject he can look into”), a doctrine which he adheres to throughout. In all these views, “subject” is set over against “object”, self-knowledge against other-knowledge, and thus the Cartesian confusion persists.

The problem naturally arises of relating these two forms of knowledge, and it is just as insoluble as the general problem of relating two essences. We may take for example the way in which Berkeley completes his statement of the position in the paragraph referred to. “This perceiving, active being is what I call mind, spirit, soul, or myself. By which words I do not denote any one of my ideas, but a thing entirely distinct from them, wherein they exist, or, which is the same thing, whereby they are perceived; for the existence of an idea consists in being perceived.” (Italics in text.) The minor problem of how we know the relation between the thinking essence and a thought essence is one that Berkeley cannot solve. For when he says “I see the table”, and since he holds that the table is known by way of “ideas” and the “I” by a “notion” (another form in which he introduces the cogito), then, whether the “see” is to be known by way of “idea” or “notion”, he cannot say that the whole situation “I see the table” can be known in either way — not, at least, without undermining his whole distinction — and so cannot take it to be known at all. Alexander is placed in a similar difficulty in his Cartesian theory of “enjoyment”, since, on his assumptions, “I contemplate the table” can be known neither by enjoyment nor by contemplation — so that he can arrive at a consistent position only by giving up Cartesianism and admitting that there is no knowledge but contemplation.

The more important question, however, is that of the occurrence of the relation itself. For Berkeley, “I” exist in knowing, and the table exists in being known; but the whole situation “I see the table” cannot exist in knowing and cannot exist in being known — it cannot exist as “subject” or as “object”. That is to say, with the retention of the essence doctrine, it cannot exist at all. On the other hand, if it does exist, then both “I” and the table exist in that situation, and therefore do not exist, respectively, “in knowing” and “in being known”. The difficulty is not met by any “supra-relational” theory, any doctrine of the “relativity” of relations or of objects as “aspects” (or expressions of the essence) of a subject. Any such theory depends for its very statement on the assertion of relations. If the “identity behind the diversity” is not to be the empty “essentially essential”, if it is really to be diversely “expressed”, then the various expressions are variously related to one another and to that “of” which they are expressions — in fact, we have a set of interrelated situations, no one of which can be “higher” or more essential than any other, since, as we have seen, its superiority would then have no way, higher or lower, of existing. The case can be met only by a logic of situations, which treats

  ― 108 ―
mental situations, and non-mental situations, and situations embracing the mental and the non-mental, as all of the same order, none having any peculiar “inwardness” or “outwardness”. If we care to put it so, we can call each of them “inward”, as having its own character, and “outward”, as being distinct from other things. Externality is, of course, a symmetrical relation; when people say that certain things are external to them (or are “in the external world”), they imply that they are external to the things. Thus, if we could take existing in the external world simply to mean existing under conditions of externality, we could say that the mental exists in the external world. And in treating A, B, and A knows B, as alike situations, we are confronted with no greater difficulty than in treating thunder, lightning, and thunder follows lightning, as alike situations — remembering always that situations interpenetrate and exist in wider situations.

Any question of mentality, then, is a question of fact; if we know the mental, we know it as a certain sort of thing, distinct from and related to other sorts of things. And that means that we know it not “inwardly” but outwardly, that it is one of the things which, when we recognise that they are known, we call “objects”. If that is not so, if it is a question not of our knowing mentality as a particular sort of thing but of our knowing “mentally”, if, in other words, being mental is not a distinguishing-mark of what is known on one occasion rather than another, then it is equally not a distinguishing-mark of what knows but is just another expression for existing. The term “mental”, in short, would simply have to be dropped; that is the upshot of Idealism. However, we do speak of the mental, and we do say that minds know, and we think, in saying so, that both the relation and the thing related are different from other things and other relations (e.g., that trees grow). And here as elsewhere, in our knowledge of minds as in our knowledge of trees, it is a question of the assertion of propositions any one of which can be significantly denied, i.e., of the raising of issues which are settled, in each case, by what we believe to be true. We may directly observe the truth in question, or, when we do not (and even sometimes when we do, viz., when other people disagree with us), we may draw inferences from what we have observed (or from what is agreed on). This is the method of settling issues whether in the mental or in the non-mental field, whether the subject is admittedly one of controversy or is allegedly one of “certainty”.

Here, incidentally, we may remark on how the procedure of actual inquiry differs from the Cartesian “method”. There is, it must be emphasised, no such thing as absolute doubt; there is no question of adopting an attitude of doubting in general or of finding a proposition to be doubtful in itself. We doubt only in relation to what we believe, i.e., to what we do not doubt. If we are doubtful whether A is B, this is possible only if we have some knowledge of both A and B. Descartes himself cannot express his doubts except in terms of what he does not doubt. As we have seen, he assumes from the beginning, as a matter of information, that he thinks; and every word he uses in working out his argument

  ― 109 ―
implies something that he does not doubt. It is only because there are certain things that he believes, without doubting, that he can arrive at the view that his senses sometimes deceive him; and there is no question of his thinking that “all is false”, since, if a proposition is false, its contradictory is true. Doubt arises, then, only in particular cases, and is settled not by what is indubitable but by what is believed. Propositions are not doubtful or certain; we doubt and are certain — and sometimes when we are certain, we are wrong. Thus we may hold with assurance certain propositions about ourselves or our minds; alternatively, we may be doubtful about them, or we may have our assertions challenged. Such an issue can be settled only by observation and inference from observations. And, in particular, the class of propositions (variously placed and dated) “I am thinking” has among its members some that have been doubted and none that could not be made a matter of controversy. As we have seen, it is only the mechanism of identity and ambiguity that makes this appear to be an exceptional case.

It may be argued, of course, that, even if the matter is a controversial one, even if we can make mistakes about the “subject” as we can about “objects”, that is no reason for confusing the two — that there can be an empirical inquiry into the facts of the “subjective” life. What we have to remember here is that the “subjectivists” are dealing, however confusedly, with certain actualities. In the same way, Descartes started with knowledge of his thinking as a fact, and it was this fact that he erected into an essence or certainty; this explains how some who would reject the ontological argument, are still prepared to accept the cogito, though actually without the general notion of that which is in itself there would be no special notion of that which thinks itself. So, the familiar arguments regarding “subject” and “object” start from the fact that in the situation of knowledge there are the two things, what knows and what is known, and thus that we are falsifying the situation if we confuse the two, just as we should do if, in the relation of eating, we confused the eater with the eaten. But, erroneous as this would be, and while it is possible that an eater should never be eaten, it also happens that some eaters are eaten. Now, when the subjectivists say that we are falsifying a “subject” if we represent it as an object, they are saying that what knows can never be known. And while, if they adhered to that, they would have to say that there can be no theory of what knows, they make confusion worse confounded by holding that the knower is always known — only not known as an object, but known in its true character as knower; in other words, known by the inner or identical knowledge of the cogito.

The virulent rationalism of this doctrine is evident. It is only on the assumption of a knowing essence that it could be supposed that, if X knows, to know X is to know that it knows. Further, it is only if knowing is the “whole nature” of what knows that it can be argued that the very same thing which had an object could not be an object in another relationship. Eater and eaten are distinct, yet what eats may be eaten. Above and below are distinct, yet what is above something is always below

  ― 110 ―
something else. The “domain” and the “converse domain” of a relation can have any relation from co-extension to exclusion; we can determine the matter only empirically in each case, and we do constantly make such statements as that we know X and X knows Y. Indeed, in saying these things, we profess to have knowledge of what knows and what is known by a single act and in a single situation, i.e., as objects occurring within a more extensive object, and the bare assertion that “the knower is not the known” casts not the slightest doubt on these contentions. The weakness of the theory of “subject” appears finally in the treatment of a relation, knowing, as the “nature” of what knows. Certainly, as Leibniz saw, if we are going to have a theory of natures, we shall have to bring its relations within each nature — but he himself could explain away some relations only by bringing in others, and it is, as we have noted, on the question of relations that rationalism most conspicuously breaks down. As regards knowledge, then, there is nothing in the subjectivist argument (and likewise nothing in experience) to confute the views that, if something knows, it need not be known, that, if it is known, it need not be known that it knows, and that, if it is known and even known to know, it is what we call an “object” — or, what is more important, it is a situation or occurrence, differing in quality from other things that know and from things that do not know, just as they differ from one another.

Thus there is no question of setting mind apart from objects, of giving it a special way of being known (any more than of existing), viz., “in itself”; the only questions are whether minds are known and, if so, what is known about them, what situations they are found to exist in, what other things they are distinguished from and connected with. The theory of “subject”, denying that minds are anything in particular, leads straight to the denial of mind altogether in the doctrine of behaviourism. Now behaviourism is to be commended in that it insists on dealing with what is observed and rejecting the “inner light”. Those who hold that there are two sources of psychological knowledge, “introspection” of ourselves and observation of the behaviour of others (and of ourselves also, since this forms the link between the other two), are in as untenable a position as we have seen Berkeley's to be; they can establish no connection between the “introspected” and the observed — or else they are supposing such a combination of inner and outer knowledge as simply annuls the distinction. But, since the behaviourists do not realise that minds are observed just as other things are, they have to treat the mental as simply a name for certain sequences, so that, whereas in the ordinary course of things A is observed to be followed by B, in the “mental” course of things C and not B is found to follow. On this view, “mind” means a magical setting aside of the course of nature. The only scientific attitude is to admit a difference not in the types of sequence, but in the antecedent conditions in the two cases, and to try to discover what that difference is. And this is something that we quite frequently do; we have indeed a wide knowledge of mental qualities (kinds of emotion); the only thing we are not acquainted with is mental “inwardness”.

  ― 111 ―
It is worth emphasising that knowledge of mind is obtained as much by observation of other people's minds as by observation of our own. The knowledge of other minds presents insuperable difficulties for the subjectivist. Unless it is to be said that we are “members of one another” (i.e., unless we are to adopt the “aspect” theory, the theory of one ultimate Subject, which culminates in the denial of all distinctions), it must be admitted that we cannot have inner knowledge of another “subject”, and yet, according to subjectivism, we should be falsifying it in knowing it as an object. Alexander's attempt to get over the difficulty by his theory of “assurance” merely adds to the original problem of relating the enjoyed to the contemplated an equally insoluble problem of relating the “assured” to either. The theory is so far valuable, however, in that “assurance” is taken to be a form of direct knowledge; what is required to complete the argument is the rejection of the conception of “subject” (in the form, in this particular case, of the “enjoyed”) or of “self” — as if a mind itself and a stone itself were not alike just a mind and a stone. Once this is done, it can no more be contended that, as physiologists, we are interested in other minds only in a secondary and subsidiary way, and that our own minds remain as the essential subject of our inquiries and the test of all the rest, than that, as psychologists, we are primarily concerned with own bodies and only secondarily with others. On the contrary, much of our knowledge of ourselves is arrived at through our knowledge of others, and, while we can make mistakes in either connection, we are at least (as Marx has pointed out) not in the habit of taking a man's own estimate of himself as likely, on account of its “intimacy”, to be the best we can obtain — least of all, as being “certain”, though on the subjectivist theory it would have to be.

As already indicated, the question of the knowledge of mind has been especially confused by the taking of a relation, thinking, as the mind's “nature”; and this applies equally to the question of the conditions under which minds exist. Thus the doctrine of psycho-physical parallelism is vitiated from the start by the fact that what is supposed to run parallel to brain processes is “thought”, and this, in the usual sense of what is thought, is commonly not mental at all. There is, of course, the additional objection that parallelism (like correspondence) merely suggests a relation without stating what the relation is — just as the contention that “thought is a function of the brain” is defective in not stating what function it is. Nevertheless, none but subjectivist objections have ever been urged to the view that it is certain brain processes that think and also that it is certain brain processes that are mental (emotional), i.e., that in the parallelist theory the psychical has been put on the wrong side. The main logical point is that, when we say “A is parallel to B”, we are asserting the existence of a situation of the same order as any that can be said to exist in either of the “parallel series” and so are failing to keep the series separate. But this does not prevent but rather assists the putting forward of the psychological position that, in the case of what is thought, the “parallelism” between it and brain processes is

  ― 112 ―
just the cognitive relation, i.e., that they know or think it, and that, in the case of what thinks (assuming this to be mental), the “parallelism” is just predication, i.e., that certain brain processes are mental. As has been said, the only objections to this commonsense view are based on the doctrine of “subject”, which is also responsible for the widespread belief that what is mental must think — a belief which has had a retarding influence on the acceptance of the important contributions of Freud and his followers to psychology.

In this connection we may consider the arguments of Feuerbach (as outlined by Sidney Hooknote) against “absolute materialism” — all the more because it is principally with this form of materialism, and with some justice, according to Hook, in view of his later work, that the name of Feuerbach is associated. “To say that thought is a material activity is as senseless as to say that gravitation has a taste or smell, for according to Feuerbach a proposition has meaning only when the predicate is of the same generic kind (Gattung) as the subject.” Here we have a direct, but easily answered, attack on a situational logic. There is no such thing as gravitation in general, there is only what gravitates; and it is not senseless to say that what gravitates has a taste or smell. Any difficulty there is in calling thought a material activity is due to the fact that thought is a relation, just as there would be difficulty in saying that “on” or “after” was a material activity. But there is no difficulty at all, when A is on B, in saying that A is a material activity, and B is too, and while it might be false to say that A's being on B is a material activity, it would not be senseless. In other words, A, B, and A's being on B are alike situations, which we have to observe to see what characters they respectively have and what they have not. Similarly, there is nothing senseless in saying that what thinks is a material activity or that what is thought is a material activity or even (though it may not be true) that A's thinking B is a material activity. We can settle such questions not on the basis of forms of predication (ways of being, characterisation of a thing in its own “categories”), which, as we have seen, can never be related, but only on the basis of observation — assuming, in this particular case, that we know what quality “materiality” is.

“To call thinking a function of the brain”, says Feuerbach, “is to say nothing about what thinking is.…Such a characterisation does not characterise.…Thought must be something more than, something quite different from, a mere activity of the brain”. (Italics in Hook's text.) We have seen that there is a certain amount of force in the objection to the “function” formula; and, even if we say that the brain thinks, it is true that we are not then saying what thinking is, just as, when we say that A is on B, we are not saying what “being on” is — though we may know it perfectly well. But certainly we cannot know relations without having some knowledge of the things related, and thus it is false to

  ― 113 ―
say that “it is only in terms of thought that the nature of thinking can be understood”. This contention is plausible only because “thought” is commonly used to mean both terms of the relation as well as the relation itself. If we adhere strictly to Feuerbach's description of thought as the product of thinking, we cannot admit that anything is to be understood by understanding something else that it is related to, whether the relation be “producing” or any other.

Again, Feuerbach takes up an equivocal position in the matter of the “more” and the “different”. If thought is not a brain-activity, there is no point in saying that it is “more” than a “mere” brain-activity. If it comes to that, any brain-activity is more than a mere brain-activity, any X is more than a mere X, i.e., it is not the “nature”, X. In other words, any brain process differs from other brain processes, and any mental process, if it is a brain process, will differ from other mental processes, from other brain processes, and from other processes of whatever kind — while at the same time there will be respects in which it does not differ from such other processes. Thus there is no question of “mereness” (except for rationalists); there is only the question whether mental processes actually are brain processes, and Feuerbach has said nothing to show that they are not.

It is Feuerbach, indeed, who is treating mentality as “mere”, or treating it merely as producing “ideas”. It is not true that “the clue to our mentality is what we think”, though it is certainly a clue, granted that we already have some knowledge of mentality and its relations. There can be no objection to the “study of ideas” (i.e., of our demands), a study which will involve us in a broader consideration of the movements in which we are caught up, and which greatly affect our mentality. It has always to be remembered that minds exist under conditions, and that we cannot study them without taking account of some of these conditions. Actually, this is just what Feuerbach seems to neglect; at least, although he holds that without body there can be no mind, the proposal to treat of mind in its own terms leads logically not, as Hook says, to a relative autonomy of thought but to an unconditioned or self-conditioned mental sphere — in fact, it leaves us with nothing more or less than the cogito. Rejecting rationalism, however, recognising the interconnection of situations, we have still to insist on the distinction, as well as the connection, between mind and its surroundings — and also, of course, on the fact that it conditions or affects them just as they affect it. This is a matter which has been inadequately grasped in the Marxist movement; the interest in mentality has been almost entirely subordinated to an interest in the larger social movements into which minds enter. And, while this has occasioned many mistakes in regard to these movements themselves, it is itself facilitated by the cogito, from which no Marxist thinkers have shaken themselves free. For, when mind is treated as a bare identity, it is natural that anything else will be discussed rather than mind — on this assumption, indeed, there is nothing to discuss.

It will be seen, then, that acceptance of the cogito has been an abiding

  ― 114 ―
fetter on the observation of minds, and that its decisive rejection is a condition of any considerable progress in psychological science. It has been observed that there is no logical difference between observation of mental processes in oneself and observation of them in others; the ordinary person does both, and it would appear that the psychologist advances most rapidly when he develops both. Certainly there are difficulties, constant possibilities of error, involved in the observation of minds; we regularly make mistakes both about ourselves and about other people, and even the trained analyst, though he has found out many of the mistakes to avoid, is obviously not infallible. But this brings us at once to the point that there are difficulties in the observation of anything; and the way to meet them is just to follow out the general conditions of inquiry — particularly, the forming of hypotheses and the testing of them by what we can observe. Acceptance of the cogito, however, places insuperable difficulties in the way of psychological science; and, in so far as it has been accepted, i.e., granted that even those who have accepted it have inevitably reverted from time to time to commonsense views and that only so have discoveries been made, it has been a tremendous hindrance to inquiry.

Much could be said about the psychological basis of the doctrine itself — about the fetishism which lies at the root of all rationalism, about the motives which lead men to seek the “safe and certain”, about the very close connection between the notion of “salvation” and that of the ego. Much could be said, again, about its social connections, about the appearance of the cogito in a period of rising individualism, and so forth. But, interesting as these questions may be, they are at any rate subsequent to its logical rebuttal. And, considering it simply from that point of view, we can still describe it as one of the greatest impositions in the history of human thinking.