Classics in the History of Psychology
An internet resource developed by
Christopher D. Green
York University, Toronto, Ontario
(Return to Classics index)

The Ego as Cause
John Dewey (1894)
First published in Philosophical Review, 3, 337-341.

Pretty much all libertarians nowadays insist that their doctrine of freedom of will is quite distinct from the older theory of indifferent choice. They suggest that their opponents are quite out of date in devoting their attention to the latter doctrine, which, under present conditions, is wholly a man of straw; they profess themselves quite as devoted adherents of the doctrine of causation as are the determinists, holding that the sole difference is as to the nature of the cause involved in volition.[1] Now, in one sense, I believe this latter contention to be quite correct; only I should go a step further and say the idea of 'causation' as implying a productive agency or determining force has no standing whatever in science -- that it is a superstition, and accordingly the libertian is the only believer in causation. Much of the opposition to determinism is due, I believe, to the fact that the determinist either is understood to, or actually does, carry over into his use of the term 'cause' this sense of efficient agency, instead of using it in its sole justifiable scientific meaning -- the analysis of a vague and unrelated fact into definite and cohering conditions.[2] For my own part, I wish by 'causation' to mean nothing more nor less than the possibility of analyzing the vague undefined datum of a volition into a group of specific and concrete conditions, that is, factors.

Admitting then, for sake of argument, the libertarian's position that the ego is an efficient cause of volition, I wish to make a confession of ignorance and a request for information. My confession is that I cannot frame to myself any conception of freedom of will (in the libertarian sense) which does not come in the end to the old-fashioned doctrine of a freedom of indifference. My request is that some libertarian who sees the distinction clearly will point it out to me. Let me indicate the special point where I need light. For the sake of argument, it is conceded that the ego is the cause of volition [p. 338] in general that, then, is not the problem. The libertarian, however, puts great stress upon choice between alternatives; as I understand (or if I understand) him, the possibility of such choice is the essence of freedom. Now, in order to avoid pure undeterminism (or the freedom of indifference), it becomes necessary to find a cause for this preference of one alternative over the other. What is the cause of the choice of one rather than the other? The ego simply as ego in general may be (ex hypothesi) the cause of the volition; but exactly the same ego cannot be the cause of two different and even quite opposing effects; there must be some difference in the cause when it operates to bring about one effect from that which would be operative in case the other is effected. I say, 'cannot be' and 'must be'; the reader will please understand this not in a dogmatic sense, but as expressing, my difficulty; I do not see how identically the same cause, with no additional qualification whatever, can be regarded as a sufficient explanation of the choice of a rather than of b, except on the basis of indifferentism. A stroke at billiards may be given so as to make a ball move either to the right or the left; if the ball is so struck that it moves to the left, it is because some further qualification has entered in other than that involved in case it moves to the right. Is the case the same or otherwise with the choice between alternatives? Does identically the same ego, without any further modification or qualification, choose to steal a loaf of bread that would also have chosen to go hungry? If yes, then how does the Neo-libertarian differ from the old-fashioned indifferentist ? If no, how does he differ from a determinist -- from a determinist that is, who sees that the introduction of this further modification is simply a further step in the concrete analysis of the act?

Be it remembered, it is not a cause for volition in general which is wanted; it is a cause for this volition rather than that: for choosing hunger rather than dishonesty. The old-fashioned indifferentist has an answer before which I stand rebuked. He, I imagine, would exclaim: "What, do you think to catch me in this easy way? When I tell you that the essence of freedom is the ability to choose either a or b without any further cause, am I supposed to be so simple as at once to contradict myself by attempting to assign a cause?" I should not know what the indifferentist means, but his meaning (if there be any meaning) would at least be self-consistent. But when I am told both that freedom consists in the ability of an independent ego to choose between alternatives, and that the [p. 339] reference to ego the meets the scientific demand with reference to the principle of causation, I feel as if I were being gratuitously fooled with. My libertarian informant must know as well as myself that the question is concerning motivation as to choice; if there is adequate statement for the choice of a rather than b, surely there is determinism; if there is not, surely there is freedom of indifference. The power of attention is now the favorite philosopher's resort. Putting the question in terms of attention: is there any reason in the conditions of the case, any specific or assignable reason, why attention gives its little boost to this side rather than to that? any assignable condition on the basis of which it gives a jog in this direction rather than in that? As I understand the matter, the whole question lies here: In considering the relation of attention to a given choice, can we (or if foiled in a given instance are we still to try) carry back our analysis to scientific conditions, or must we stop at a given point because we have come upon a force of entirely a different order -an independent ego as entity in itself ? If the action of the latter in swaying mental emphasis this way or that is one of the conditions, can we analyze this condition any further, or is it an ultimate fact? If the former, it seems to me an awkward determinism ; if the latter, a frank indifferentism.[3] [p. 340]

The same point may be briefly repeated from the ethical side. When one man says to another, "You did that, and I shall hold you responsible for it," he means by his "you," not a metaphysical ego, but a definite individual -- John Smith. Every step away from the concrete individual, John Smith, with his special aptitudes, habits, desires, ideas, and ignorances, every step towards an ego in general, means a weakening of the connection between the man and the act, and a release of the man from responsibility for the act. Determinism means that the individual and his act are one. What does libertarianism mean? Will not some libertarian explain to me the agency of the ego in volition in terms of some concrete self, instead of in terms of a metaphysical ego?

One point more. Why does the libertarian change his standpoint so completely when considering the act before and after its performance? When considering the process of volition prior to the overt act, the presence in consciousness of two alternatives, the presence there of two attracting, yet incompatible ends, he treats as a fact in itself outside the freedom of will; it is capable of being accounted for on the ordinary principles of habit, association of ideas and desire. It is, he insists, an occasion for the exercise of freedom, but in itself lies outside of will proper. If he admitted the presence of the two alternatives to be an adequate basis for freedom, there would, of course, be no need whatever to call upon the outside entity, the ego. But if this consciousness of different ends, of competing interests, with the process of reflection upon them to ascertain their respective values, does not prove freedom, why use the memory of such consciousness -the conviction that we might have acted otherwise -- to prove freedom? No determinist (that I know of) denies the facts of conflict of desire, denies that different ends with competing interests attaching to them come to consciousness, or denies the existence of deliberation or a tentative rehearsal of the different acts. He simply urges that choice, when it appears, is the normal psychological conclusion of this same process; that it no more requires the intervention of an outside faculty or entity as efficient cause, than the drawing of a conclusion from theoretical data requires more than recognition of the full meaning of the data.[4] [p. 341]

In any case, we should have one interpretation or the other; not a mixture of two contradictory conceptions. Let us say, if we please, that our consciousness of ability to have acted otherwise does prove freedom, because the presence in consciousness of alternative ends with the reflection which that calls out, is freedom; or, let us say that since this consciousness cannot prove freedom, no subsequent revival of it in memory can prove freedom. In either case, the role of ego as separate efficient agent in causation seems to be excluded.


[1] See, for example, the discussion by Dr. Gulliver in the Jan. (1894) No. of this review.

[2] With reference to this point I may be permitted to refer to an article in the Monist for April, 1893, on "The Superstition of Necessity."

[3] It is somewhat aside from the point in discussion, buy when Professor James says that views like the one quoted from Mr. John Fiske, on p. 577, Vol. II, of his Psychology are caricatures, arising from "not distinguishing between the possibles which really tempt him not at all," and that "free-will, like psychology, deals with the former possibles exclusively," this seems partially only a mitigation of the scientific havoc wrought by the idea of free-will; and partly to be entering on the deterministic path - and a mitigation only so far as the deterministic path is entered upon.  From the anti-libertarian standpoint there is no break in the process; the fact of temptation and the fact of choice are related as the more undefined and the more definate establishing of relations within the self, of "between our Self and our state of mind" (p. 568).  Surely the determinist as well as the libertarian may recognize facts of uncertainty, of hesitation, of tentative action, of first trying on this and then that.  And it is difficult to see why uncertainty will not do everything in giving zest and sting to life, that James thinks can be given only by sheer liberty (Psy., Vol., p. 453).  Our feeling that matters are "really being decided" looks to the future, not to the past; consequences do depend upon whether we act this way or that -- and this fact is one of the determining factors.  When Mr. James puts as the alternative to libertarianism "the rattling off of a chain that was forged innumerable ages ago," he must have in mind, not logical determinism, but theological predeterminism.  And the theological view harks back to an independent entity or ego as Cause -- to the 'free will' doctrine -- not to the determinism of knowledge

[4] It is strange that Professor James, who recognizes so far as knowledge is concerned the entire uselessness of an ego outside and behind, who indeed has given that theory the hardest knocks it has yet received from the psychological side (Vol. I, pp. 360-370), should feel bound to set up its correlate when he comes to deal with will.  If the stream of thought can run itself in one case, the stream of conduct may administer itself in another.  Why should he deny to the transcendentalist ego in knowing a power which he claims for attention in acting?  Historically, I think the independent Ego in knowledge is a survival and transference from the action of an entity of Will in choice.