We have found that the conditions of discourse and inquiry demand the rejection of “pure” science and the assertion that all sciences deal

  ― 12 ―
with facts, in relation to which we assert or deny, prove or suppose. We have found, in other words, that the theory of different ways of being is untenable. But with it falls the theory of different ways of knowing, the distinction between sense and reason. The very slightly empiricist character of the work of those philosophers who are called “the English empiricists” is accounted for by their still making this very unempirical distinction. In maintaining that all our knowledge is derived from sense (a position which, on account of their rationalist preconceptions, they by no means maintained consistently) they took a view of sense which was dependent on its having been regarded as an inferior way of knowing. It was supposed to provide isolated data, materials which reason had to shape into, or subordinate to, the coherent system of knowledge which we call science. And Hume, while admitting that no such coherence could be imposed upon isolated data, still maintained that the data of sense were isolated, and accordingly could not show how science is possible. The rejoinder of idealists like Green that Hume's position leaves out of account the function of the mind as a relating agency, that it takes as real what has not yet been made real by the work of the mind, is no reply. Hume's argument is precisely that neither mind nor any other agency could possibly perform such work on “distinct existences”. And this is the point of departure of the “radical empiricism” of James. Mind is not required to relate things, because things are given as related just as much as they are given as distinguished. Connections and distinctions, in fact, are given together; and those who argue that the work of the mind is required to connect distinct things, might equally well maintain that work had previously been required to distinguish them. Here James is drawing attention to the important fact (important, as well as for other reasons, in view of the persistent misunderstanding of the meaning of empiricism) that there is nothing in the least empirical in the conception of a “distinct existence”. It is on the contrary the rationalist conception of “essence” masquerading as a fact of experience.

If, then, there is to be any question of what is given or presented (though it would be better to speak of what is observed), connections must be included. This is in line with the view already set forth that what can be contemplated or enunciated is always in the form of a proposition; in other words, that we always deal with complex states of affairs and never with “simple entities”. Any theory which refers to the work of the mind, or to rational factors, as contributing, along with sensible or given factors, to making things intelligible, is self-refuting or “unspeakable”. If whatever is intelligible has both connections and distinctions, then in order to speak intelligibly of what is contributed by the mind we shall have to assume that it has both connections and distinctions, and in order to speak intelligibly of what is given by things we shall have to assume that it has both connections and distinctions, so that no “work of the mind” is required to make it intelligible. And in the same way, in speaking intelligibly of “knowledge”, we are speaking of a certain state of affairs, the mental process which knows, as connected

  ― 13 ―
with and distinguished from another state of affairs, the process or situation, mental or non-mental, which is known.

We cannot, then, make any such distinction as between “things as we know them” and “things themselves”. Unless the former are things themselves, we are not entitled to speak of things (and hence to speak) at all. On the other hand, we are entitled to reject, by reference to things themselves, viz., the things we know, any suggestion of an agency whose operation cannot be detected; which we cannot observe acting on some observed situation and bringing about observable changes therein. As “rational factors”, ex hypothesi, cannot be seen at work (since they must have worked before anything can be seen; since they are “conditions of the possibility of experience”), not only can we not assert that there are such ideal entities, but we cannot show what they would do, if there were. An agency whose presence cannot be detected is an agency which it is of no advantage to postulate, as Berkeley showed in regard to Locke's “matter”. We cannot have a “merely inferential” knowledge of it. We must be able to say: “This is the sort of thing which under certain circumstances will act in such and such a way, and under other circumstances will act in a different way”. But if we have never observed it so acting, if we have never been able to distinguish it from its effects on the situation, then the whole content of our knowledge, all that we are in a position to speak about, consists of the circumstances, no longer to be described as effects of “it”, at least. The appeal to inference, or to the distinction between “knowledge by acquaintance” and “knowledge by description”, is futile. We can say, for example, that any man we happen to meet had parents; we can have an indirect knowledge of their existence, though we have never seen them. But this would not have been possible if we had not at some time known individuals who stood in that relation to some one, and had not thereafter come to believe that all beings of a certain sort have parents. We cannot, then, by inference from what we observe, conclude that there is a mind whose function it is to observe these things, i.e., which is purely instrumental, a pure agency. Unless we have observed minds, we cannot speak of them. Having observed them, and having observed that they are related by “knowledge” to other things, we can also consider how they fall into error. But this criticism of the mind's operation in regard to things cannot take the form of “criticism of the instrument”. We cannot, without self-refutation, undertake to criticise the mind's entire knowledge; for it is by our knowledge that we criticise. Criticism, then, can only proceed by our asserting what we find to be the case; we can criticise propositions only by means of propositions, similarly asserted. The distinction of ways of knowing, at least in the form of a distinction among faculties, is therefore untenable. We can, of course, distinguish such attitudes as asserting and supposing. But in every case we are dealing with something which is, or may be, found to be the case, and there is no question of seeking for and fostering some superior instrument.

In terms of this theory it must be said that in psychology, and

  ― 14 ―
likewise in ethics, our knowledge is observational and propositional. The question is of psychological and ethical facts, and not of an ultimate agency or ultimate standards. These sciences, like all others, are nothing if not empirical and experimental. This does not, of course, mean that minds must be studied in a laboratory; they show some of their characteristics better in other social situations. Love, for example, is a very important psychical phenomenon; in fact it may be said that no one can know much about minds who has not taken it into account. But none but the most hardened “experimentalist” will claim that a laboratory is the best place for getting to know its characters and conditions. The main point is that, in order to know minds, we have to observe them and think about them. There is no real distinction between thinking and experiment. In each we require some hypothesis, and in each case we test it by reference to what we believe, or find, to be the case, i.e., by whether or not its consequences are in accordance with facts which we know. In holding that in order to know minds we have to look at them, empiricism is not opposed to “introspection”, the study of our own minds, though it opposes the supposition that in this peculiar case the process which knows and the process which is known are identical; i.e., it insists on the fact that the study of our own minds takes place by means of observation. But, an empiricist will say, there is no more reason for confining ourselves to “introspection” than for considering only our own bodies in studying physiology.

What has chiefly to be emphasised, however, is that the observation of minds, the knowledge of them in propositions, requires the rejection of the “unitary” view of mind. the conception of it as having only one character and being self-contained in that character. This is a rationalist, “unspeakable” view. If we are to have any dealings with minds, we must be able to consider how they act in different situations, i.e., to consider them as having complex characters and activities, as being divisible and determinate. Psychological science will only be possible if we have a variety of psychological truths, between which, and in each of which, connection and distinction are discernible. And the same applies to ethics. These sciences are historical, they are studies of occurrences and activities, they are concerned with situations in space and time. I have thus, without going into detail, indicated the place in the empiricist scheme of the other anti-rationalist theories I mentioned. The general conclusion is that all the objects of science, including minds and goods, are things occurring in space and time (the only reason for regarding minds as not in space being the rationalistic contention that they are indivisible), and that we can study them by virtue of the fact that we come into spatial and temporal relations with them. And therefore all ideals, ultimates, symbols, agencies and the like are to be rejected, and no such distinction as that of facts and principles, or facts and values, can be maintained. There are only facts, i.e., occurrences in space and time.