5 Realism and Some of its Critics (1930)note

“I am convinced that some of the principles on which Realism bases itself and some of the arguments by which it buttresses itself are so palpably unsound as to cause wonder why admittedly competent minds have accepted them.” So says Professor A. C. Fox in leading up to his discussion and rejection of “the cardinal principle of Realism”, particularly as formulated and supported in my paper, “The Knower and the Known” (Proc. Arist. Soc., Vol. XXVII). Professor Fox's treatment of this “cardinal principle” is, as I shall attempt to show, far from being thorough, but it is gratifying to find an approach being made to the discussion of the central issues.

It is indeed the case that there is no mind so competent that it never falls into fallacy and inconsistency, and the realist and the idealist, humanly prone though they are to wonder how any sensible person can hold the view they oppose, can advance matters only by discussing specific issues. As I pointed out in discussing the philosophy of Alexander (“The Non-Existence of Consciousness”; A.J.P.P., Vol. VII, No. lnote), we cannot define Realism by what any particular realist says, for it is perfectly possible for him to make some quite unrealistic errors — as Alexander, I argued, has done, although “in his doctrine of Space-Time he has laid the foundation of a thorough-going realism as a logic of events”. Similarly I do not hold Idealism responsible for all that appears in the articles referred to, but shall endeavour in discussing these articles to keep the main issues clear.

Nevertheless, the fact that discussion is advanced by consideration only of the issue itself, and not of the minds of persons who hold views about it, is evidence of the truth of the realistic position. For Idealism, which makes the settling of every issue depend on the settling of every other, no issue can ever be settled — and thus Idealism itself cannot be upheld. All actual argument implies the independent issue or individually true proposition, and this is the same sort of independence as the realist finds in the terms of the relation, “knowledge”. On this issue there can

  ― 42 ―
be no compromise; the realist and the idealist simply cannot recognise each other's competence. Speaking as a realist, I find myself bound to assert that Idealism, so far from being competent philosophy, is not philosophy at all. But this does not prevent me from recognising that an adherent of Idealism may be acquainted with many philosophical truths, while an adherent of Realism, such as myself, may fall into many philosophical errors.

I. Realism versus Monism

Realism appears first, then, as a pluralistic doctrine or theory of independence; and this brings it into conflict with the monistic doctrine properly called Idealism, which denies independence to everything but the “Absolute” or one true Being. It is this question of independence which is the “cardinal” issue, and on which, though it may be convenient to discuss it with special reference to the question of knowledge, the settling of the dispute about knowledge really turns. This is what I tried to make clear in the paper I read to the Aristotelian Society, but it does not appear at all clearly in Professor Fox's “examination” of my argument. The following summary of what I said may therefore help readers to come to a conclusion.

Logically we are bound to recognise real differences (as opposed to the idealist doctrine of the “merely relative” nature of differences), since otherwise we could not distinguish between affirmative and negative propositions, or indeed make statements at all. If “when we say that A is not B, we are somehow also saying that A is B and B is A…discourse [will] be impossible”. But to recognise real differences, or, what comes to the same, different real things, is not to say that these things are unrelated. On the contrary, any relation has two terms, or holds between different things; and if these things are not “really” different, then there are not really two terms and there is really no relation. Hence there is no argument from relatedness to monism, quite the reverse.

Thus the recognition of the “subject-object” relation, or relation between knower and known, implies that each of these is an independent thing, or thing with an existence and characters of its own, and that it cannot be properly described in terms of the other thing or of the relation between them. This point I expressed (following Marvin; “The New Realism”, p. 473) by saying that the thing which is known, or the “object”, is not constituted by the knower or by being known, nor is the thing which knows, or “subject”, constituted by knowing or by the known. In other words, we cannot define the nature or character or constitution or “what is it” of a thing by saying what relations it has or what it is related to. Hence I concluded that we must reject the notions of “that whose nature it is to know”, or consciousness, and “that whose nature it is to be known”, or idea. The rejection of the theory of constitutive relations leads therefore to a criticism of those “realists” who inconsistently permit themselves to speak about consciousness or ideas, as well as of idealists.

It is as a consequence, then, of a consideration of relations that I put

  ― 43 ―
forward the statement, quoted by Professor Fox, that “nothing is constituted by knowing and nothing by being known”. To say that X has the relation R, or has the relation R to A (where R might, for instance, be knowing or being known), is not to say what sort of thing X is, and, if the above were all the information we had about X, we should not know X at all. The fact that we can in many cases come to a conclusion about X's character, when we are told that it has a certain relation, is due, I argued, to our having the additional information that only things of that character have that relation; but, as I said, we could not have this information unless we could distinguish the character from the relation. For example, we know that only men can be husbands. And it is owing to our having this additional information that we find ourselves, when we are told that X is a husband, knowing some of his own qualities. So far from attempting to meet these contentions Professor Fox does not even mention the question of relations in his criticism of my views.

On the question of being “constituted by knowing”, I contended that to say that A knows or that A knows X is not to give a description of A, and that, unless such an independent description is possible, there is no relation. Professor Fox thinks this is “obviously” false. “That Einstein knows (or believes he knows) certain things is precisely what makes him Einstein; and that I do not nor ever can know these things in the same manner is what hinders me from becoming the same sort of person.” It should be obvious from my exposition that I do not deny, but assert, that only beings of a certain sort have certain relations to other things, e.g., know certain mathematical theorems. But to say that they have these relations is not in the least to tell us what beings they are. We should never succeed in identifying Einstein if all that we knew of him was that he had certain beliefs; we could never find out what beings are mathematically gifted unless we could observe their qualities, as well as their relations to mathematical facts.

In fact, unless things had qualities of their own, there would be nothing to have relations to other things. What I have in effect maintained is that even those who support other views, do unwittingly concede, in the language they employ, the distinction between relations which hold between two things and qualities which belong to a thing itself. Professor Fox passes over this distinction when he says that “to possess human knowledge is (partly) identical with being a man”. All that the phrase “human knowledge” conveys is that there are certain things that only men know, but it does not tell us at all what men are. They do not possess the things they know as qualities; yet if the relation were part of what a man is, the things related to him would also have to be part of what he is. This is the doctrine that Berkeley tried, and failed, to maintain. It is contradicted by the admission of the distinction in the first place, e.g., between Einstein and the “certain things” which he believes. He is not his relation to these things.

What the idealist has, in fact, to show is that there is no real distinction, and the answer is that in that case there is no real relation. But we

  ― 44 ―
may also criticise the arguments used to show that the distinction is “unreal” or “not ultimate”. It is urged, as I pointed out in the paper referred to, that “there is no subject without an object, and no object without a subject. But this merely means that any ‘subject-object’ relation has two terms; it could not for a moment show that knower and known are not two different things, or that anything is to be regarded as in itself either a subject or an object.” It was here that I introduced Hume's “husband and wife” analogy, the point being that, although there is no husband without a wife and no wife without a husband, husband and wife are two different persons, just as cause and effect are two different things.

This, as I stated, is the importance of the analogy for my argument. But I introduced in a subsidiary way the point that was important for Hume. Although “every effect has a cause” and though it would be agreed that effects are events, this does not prove, Hume says, that every event has a cause, any more than it follows, from the facts that every husband has a wife and that husbands are men, that every man has a wife. Similarly, I argued, though every subject has an object, and even if it be granted that subjects or knowers are always minds, it is not proved that all minds know or that anything that can properly be called mental must have an object. Professor Fox translates this argument into the form that “we may believe a being to be a man in order that he may be a husband, but may not believe him to be a husband in order to be a man”, and thus loses the reference to implication, and makes the question only whether men (as he says, instead of minds) do in fact all know something or other. This I should not think of denying, but what I did, in accordance with the Freudian theory, deny in a later part of the paper is that every mental process knows. Professor Fox, however, does not refer at all to what I said was the important point, namely, that knower and known are different things; so that even if minds were always knowing, this constant relation would not constitute them, any more than causes are constituted by having effects or husbands by having wives.

On the question of being “constituted by being known”, Professor Fox proceeds more cautiously. He does not say, “That the moon is known by certain minds is precisely what makes it the moon”, though it is clear that he would desire finally to come to this conclusion. For a beginning he argues that the moon's being known by me (or my knowing the moon) contributes to making it the moon that it is, because, in knowing it, I affect and alter it. “I assume”, says Professor Fox, “that affecting and altering are modes of constituting, and if we can establish these minor modes, the way will be opened (should we desire to pursue it) to a demonstration that the literal constituting of things by some mind is a pre-condition of their knowledge by any mind.” The assumption, in fact, begs the question. Of course, there is just as strong reason for taking “altering” to be a constitutive relation as for holding that view about “knowing”; but no more. “X is altered by me” can no more be a statement of X's own qualities than “X is known by me”. That things should literally be some mind, in

  ― 45 ―
order that a mind may know them, and that a thing should be me (or I should constitute it) in so far as I affect it, are alike consequences that would follow from the view that to be related to a thing is somehow to be it; but on this view, as I said, discourse would be impossible.

In fact, the treatment of relations as forms of identity was definitely exploded by Hume, whose argument on causality is applicable to any other case. If there were in A a “power to produce B”, then B would be in A, and no production would take place. Similarly, as Berkeley saw, if there were in me a knowledge of the moon, then the moon would be in me, and, as Berkeley did not see, no relation of knowledge would hold. There is, as I argued, an identity in the case, namely, the particular state of affairs, “A causes B”, or “I know the moon”, or “Jack is married to Jill”. But this is not the slightest ground for identifying A with B, me with the moon, or Jack with Jill; nor, again, for regarding my identity or that of the moon as in any way inferior to that of my knowledge of the moon, or constituted by it. And the theory of partial identity is only an attempt to hold to the doctrine of identification while avoiding some of its consequences.

The case of Berkeley is instructive here. He argues that it is of the nature of whatever I know, to be known, but he concedes that such an “object” need not be known by me but may have its constitution kept up by other minds. But if I can know that which, however it is constituted, exists independently of my knowledge of it, the basis of the contention that there must be some other mind in whose knowledge it consists is removed. Either I can know things which, in spite of this relation, I do nothing to constitute (and Berkeley is compelled to admit this in the case of other minds and laws of nature), or I can know nothing but what I am — indeed, there is no difference at all between knower and known, and so there is no knowledge. More broadly, if the knowledge, say, of the moon by various minds were precisely what made it the moon, there would be nothing for these minds to know; the position would be that the minds know that the minds know that…To stop the regress, we have to say that they know the moon, and that it is not constituted by their knowing it. If I say, “The moon is made of green cheese”, that is a significant, though false, statement; but if I say, “The moon is made of someone's knowing it”, the unanswerable question arises, “What is it?”

This is aside from the question whether the moon is made by someone, or whether I can make a bit of it. As I have already pointed out, causality is on the same footing as other relations (the cause is not the effect), and the fact that I can cause or affect things does not imply that I “constitute” them. In the paper referred to, I said that “what exists because of me nonetheless exists, apart from or independently of me. The houses which would not have existed, had not men planned and built them (i.e., but for their minds and bodies), are physical and are not private to these men; they stand for other men to see them, and may remain when no one perceives them at all.”

Professor Fox's argument on alteration is therefore beside the point.

  ― 46 ―
If, when we know a thing, we alter it, that means that we cause it to have a character which it did not have before; i.e., its having that character now is an independent fact, and so is its not having that character before. Again, whether we say that “the moon terminates at the confines of its bodily mass” or that “the moon does not terminate at the confines of its bodily mass”, the fact remains that one of these propositions is true; and the moon's so terminating or not so terminating (whichever it does) is an independent fact, in the assertion of which nothing whatever is said about us or our knowing or our altering. It may be noted in passing that if we could not know the moon if it terminated at the confines of its bodily mass, it would follow at once that we could not know that bodily mass, which certainly has these confines. Yet apparently we talk about it; just as we talk about confines and differences in general, in spite of idealists.

Idealism, then, stands or falls with the doctrine of constitutive relations, and I have tried to show that it falls because that doctrine cannot even be consistently stated, because it is contrary to the fact of independence which we have all in some measure to recognise in our discussions. And this is not, as the idealist imagines, to take our modes of recognition as the determinant of fact; on the contrary, to do so would be to recognise them and we can recognise them only as facts, i.e., as having that independence of which we would illogically deprive other things. The doctrine of “principles of understanding”, or of “the world as intelligible”, defeats itself. If it is held, in a Kantian manner, that existence in Space and Time and subjection to categories are our ways of regarding things, the answer is that either things are not under these conditions and so our principles are principles of error — and in that case, moreover, we are wrong even about there being such principles, and in fact know nothing at all — or things are under those conditions, and, while we are right in thinking so, their being so is an independent fact. And this fact, which we do not constitute, we do not even cause; because that would imply that there was a time when things were not under these conditions, which, ex hypothesi, we cannot even imagine to have been the case. That is to say, we believe that everything that is acted upon is subject to these conditions; we recognise that that is so.

This is the answer to Professor Morris Miller who, in his exposition of “The Work of Henry Laurie”, takes exception to those who argue that what we know exists independently. “What we know, Laurie would reply, we know in relation; and what relates and recognises relations can only be the mind.” No realist will deny that what recognises relations, or anything else, can only be a recogniser, and, if it is only minds that are recognisers, that a mind must do so in the case of relations. But what is it that the mind recognises? That certain things are related in a certain way. Now, if they are not, the recogniser is simply mistaken. Otherwise, they are themselves, or independently, related in that way. And, similarly, if what we were recognising was a relation between us and things, we should recognise it, as well as the things, as having independent existence.

As I have argued, this independence would not be affected even if

  ― 47 ―
we ourselves had brought about all the relations we know; we should have brought about nothing, unless the relations proceeded to exist independently. But we all do distinguish between relations we cause (as when we drop a book on the floor) and relations we do not cause (such as the order of the words in the book), and, as I have shown, this distinction gives us no ground for supposing that the latter are caused by other minds. In fact, the idealist says nothing to show that the mind is responsible for the relations it knows, except that it recognises them; and thus Professor Morris Miller's “relates and recognises relations” is merely an attempt to have things both ways — to identify, while appearing to distinguish, the conditions of relation and the conditions of the recognition of relation.

The same confusion appears in Professor Fox's reference to vision. “In general, the moon would remain invisible were there no visual mechanism in the universe. But the fully constituted moon is a visible moon, so that the seeing of it does something even to constitute it.” On the contrary, when we speak of the moon's being as it is, we are referring to what is seen, i.e., to the actual characteristics of the moon, and not at all to anyone's getting to know them or to how he does so. Similarly, when Professor Morris Miller says that we cannot “even affirm the independence of what is real apart from the mind which is called on to assert the fact of the ‘independence’, and which cannot do so out of nothing” (i.e., apparently must do so out of its own resources), the sufficient answer is that there is here a confusion between the conditions of our asserting X and the conditions of X itself.

It may be added that it is the idealist who makes the mind “nothing”, by identifying it with its relations of assertion, just as he makes things nonentities by identifying them with their being asserted. That “the object for us can only exist as related to the experiencing mind” seems plausible only because “existing for us” is a common loose expression for “being known by us to exist”; and this is then translated into “being known by us to exist for us”. As I pointed out in “The Knower and the Known”, out of the identity, “What is known to us is known to us”, idealists construct the highly disputable assertion, “What is known to us is known as being known to us”. No evidence but the identity is ever offered for this view, but, even if it were true, being known to us would be different from being known as being known to us, and likewise from being. The “irrational use of abstraction”, which Professor Morris Miller deprecates, is simply the attempt to get idealists to realise what they are saying, and not “overcome” distinctions by smothering them in ambiguous phrases.

Such ambiguities lie at the root of all monistic theories, i.e., all theories which deny independence, or the “ultimate reality” of differences. The realist answer (stated similarly in “The Non-Existence of Consciousness”note) is that, if we say that differences are comparatively unreal, then “the comparative unreality of differences” is ultimately real.

  ― 48 ―
Yet it is not the ultimately real or Absolute; it must be an aspect or expression of the Absolute. But, in taking this view, we are admitting that it is really different from other aspects or expressions. Again, when we say that the Absolute is self-subsistent and its aspects are relatively existent, we are recognising the independent existence of “the self-subsistence of the Absolute” and “the relative existence of the aspects”; i.e., we are recognising, in spite of ourselves, a single way of being. It is seen, therefore, that Monism is not only a false doctrine but an incoherent one; that it implies a division, which it cannot sustain, between “higher” and “lower” orders of being, i.e., that it is dualistic or rationalistic. The realist has to supplement his assertion of real difference or independence with a rejection of the false distinction of any other way of being from existence, since only among existents can there be real relations.

II. Realism versus Dualism

Realism thus appears in the second place as an empiricist doctrine, or theory of existence as the single way of being; and this brings it into conflict with any theory of “ultimate” or unhistorical entities — things of “higher” reality, because they are above change. As already noted, Monism is merely one particular resort of the rationalistic dualist; the unbridgeable gulf between the “higher” and the “lower” remains, whether we postulate many “ultimates” or only One. The locus classicus of the idealist-rationalist entanglement is Plato's Parmenides, where we are shown the illogicality of both hypotheses, “that there are many” and “that there is one” — the obvious solution being that there are none, i.e., no “ultimates”; which was the conclusion already reached by Gorgias.

The point is that the believer in ultimate or eternal entities is logically bound to deny historical things altogether. This was the position taken up from the beginning by the Eleatics in their criticism of the Pythagoreans. That which has not “real being”, really has not being; i.e., it is nothing at all. Or again, if the “higher” is the reality of the “lower”, then there is really only the “higher”. It was easy for the Eleatics, then, to show that the Pythagoreans could give no account of history in terms of what they regarded as the real; if the real is unhistorical, the historical is unreal. But the sole reality or Absolute of the Eleatics is in no better case; as soon as they say anything about it whatever, they represent it as having “aspects” which are only relatively to it and are nothing in themselves; so that, by the same argument as before, there are no aspects, and there is no One. The second and longer part of the Parmenides shows the overthrow of Eleaticism by means of the same logic as the Eleatics had used against the Pythagoreans; but this is because it is a commonsense logic, a logic of events, that logic, in fact, which is involved in all discussion and criticism. Thus, as soon as the monist says anything at all, he can be refuted; and, of course, if he says nothing, there is no Monism to refute. But he has a dialectical advantage

  ― 49 ―
over the ordinary rationalist, who says a great deal, and openly employs the distinction between the real real and the somewhat real.

What the Parmenides shows, then, is that all doctrines of “ultimates” fall together, because they all have to admit the “relative” but can give no coherent account either of its relation to, or of its distinction from, the “ultimate”. If this relation or distinction is ultimate, then both its terms must be ultimate, and if it is not, then there is, “ultimately”, no relation or distinction. Thus the refutation, in the first part of the dialogue, of the reconstructed Pythagoreanism of Socrates is itself sufficient to establish the doctrine of a single way of being, the being of historical things which are related and distinguished, come to be and cease to be. And if the arguments of Parmenides against Socrates can be made good, we may credit Plato with holding a realistic and empirical theory, though he may not have worked it out so consistently as did the earliest critic of Pythagoreanism, Heraclitus.

The Eleatic criticism of rationalism is equivalent to the rejection of constitutive relations. The early Pythagoreans had held that the real was certain units, and that empirical things were simply arrangements of these units, so that the reality of a thing was simply the units which constituted it. And the Eleatic arguments, which found their clearest form in Zeno's paradoxes, were to the effect that this derivation from the real admitted the reality of something other than the real. Later Pythagorean theory recognised as ultimate certain forms, or types of constitution of things, the things themselves having their reality in being (or in so far as they were) of the nature of the forms. This relation, expressed by Socrates as “participation” of things in forms, is supposed to overcome the logical objections to different ways of being, but actually the later statement of the theory is just as vulnerable, and by the same line of reasoning, as the earlier. The important feature of the theory of Socrates (as we find it expounded particularly in the Phaedo) is that he introduced a definite reference to the proposition, but, as this did not lead him to reject the doctrine of different orders of being, a propositional criticism of his “higher realities” was still required.

As I have argued, a relation can hold only between two things, each having characters of its own, i.e., between two independent existents, not between an “ultimate” and a “relative”, or, for that matter, between two “ultimates” — and it is this which gives point to the Monism of Parmenides, as to that of Spinoza. It is here that I come into conflict with Mr W. A. Merrylees, who recognises “the teleological nature of reality”, and accepts participation as “that relation between anything and its ideal (and between that and its ideal until we reach the first principle) in virtue of which it is intelligible”. In other words, Mr Merrylees holds the Socratic theory that things are constituted by what they tend towards or strive after, that a thing has reality to the extent to which it approximates to its reality or perfection. Not only does he hold this, but he regards the theory as Plato's, and he considers that it is supported, instead of being disposed of, in the Parmenides. In replying previously

  ― 50 ―
(“‘Universals’ and Occurrences”; A.J.P.P., Vol. VII, No. 2note) to Mr Merrylees's criticisms of realism, I asked what he made of the Parmenides. This appeared in his article (which had, in fact, been written earlier) in the following number; he regards it as a logical failure on the part of the Eleatics to weaken Socraticism and to maintain their own position. What I require to show, then, is that the “participation” theory is really overthrown.

The discussion begins from the type of “contradiction” brought out by Zeno in his paradoxes. How can the same things be both like and unlike, as they must be if there are many of them? The importance of this question is that to answer it properly is to give an account of the distinctions and relations which hold among things themselves, and to show that to go beyond them for explanations, besides being inadequate, is unnecessary. But as Socrates tries to evade the difficulty by means of his explanatory “forms”, a demonstration of the inadequacy of his theory has to be made. And this is what Parmenides undertakes.

Mr Merrylees expresses no dissatisfaction with the argument of Socrates on likeness and unlikeness, or, in general, to show that “though one idea cannot be or become its opposite, nevertheless the same subject may participate in opposite ideas”; but, indeed, to recognise its weakness is to recognise the weakness of the whole Socratic theory. To say that the form of unlikeness “is the opposite of likeness”, and that, while no one “could prove the absolute like to become unlike, or the absolute unlike to become like…there is nothing extraordinary in showing that the things which only partake of likeness or unlikeness experience both”, is to take all the meaning from the term “opposite”. If like means “having qualities in common with”, and unlike “not having qualities in common with”, then no two things can be at the same time like and unlike. If, on the other hand, unlike merely means “having qualities not shared by”, then any two things whatever are both like and unlike. But, in that case also, there is no point in saying that likeness and unlikeness are “opposites”, and that, for that reason, they cannot participate in one another.

Thus Socrates has said nothing to show that, if there is such a thing as participation by one form in another, likeness cannot partake of unlikeness. Obviously it might be said that likeness is unlike some other relation, e.g., paternity. The real point is that to say that two “forms” are really opposites is equivalent to saying that no thing can “partake” of both at the same time. Incidentally, the example is a good one as showing the tendency of Socrates to confuse a relation, such as likeness, with a quality of a single thing; just as he confuses the quality good with the relation being pursued, and, in general, any quality a thing has with the thing's striving after an “ideal” of that kind.

It is this which gives point to the doubts of Socrates as to whether all things have forms, and Mr Merrylees misses the point in saying that, in the mild rebuke of Parmenides, “Plato is expressing his own view — that

  ― 51 ―
there is an idea of whatever can be thought”. Why should Socrates, youthful as he was, have had any doubts on the subject? Entirely from the nature of the theory of forms; from the ridiculousness of saying, for example, that a particular piece of mud is striving after the ideal of perfect muddiness. As Burnet suggests, the theory of ideals appears plausible only when there does seem to be some important end to be attained. But even then the criticism still remains, and this is what the “third man” or “two world” argument enforces, that the particular and the ideal must each be regarded as having characters of its own, if the one is to be related in any way to the other.

Thus whether we take the case of mud or any other, whether or not we admit that certain things are themselves “just muddy”, we have to admit that they are just something, or else they are just nothing. The latter is, of course, the Eleatic view which Socrates wished to avoid. And it is only natural that Parmenides should demonstrate his critic's failure. If the particulars have not their character in themselves, they cannot have it out of themselves as an end to strive for, for there is no “they” to have it. This is the objection to the contention in the Phaedo that any actual pair of things that we call “equal” are not just equal, but are only “nearly equal”. If so, then they are just “nearly equal”, and not “nearly nearly equal”; and there is no earthly reason for saying that their reality consists in an equality which they do not have, and not in an approximate equality which they do have. To quote my previous reply to Mr Merrylees: “In short, as the Parmenides shows, we can maintain the doctrine of ideals only by describing things in terms which do not apply to them, but all the time we are using terms which do apply to them, and so are contradicting the doctrine of ideals.”

Socrates would seem to be employing the same device when he says, in the passage quoted, that things which “only partake of” likeness or unlikeness may partake of both; suggesting, i.e., that the things may be not quite like and not quite unlike. But this, besides implying that for idealists the things are “advancing in all directions”, still indicates that the things are to be understood as having characters of their own, just as definite as anything they could be supposed to strive after. Thus, as Parmenides insists in his final argument, if we are to talk about the things at all, we must do so in terms of the characters they really have (in the “sensible world”), and any reference to characters they “ought to” but do not have (to “forms in the intelligible world”) is quite irrelevant. So, in respect of relations, if the things have any real relation, whether we call it “participation” or anything else, it also will be in the sensible world, and so will what is thus related to the things.

Naturally, Parmenides is not satisfied with this empirical conclusion, since he does not believe in the sensible world. But this merely means that for him the conclusion would be that we are not to talk about the things or “many”. Nothing can have (or partake of) being, since if it is being, there is no relation, and if it is not being, there is nothing to have the relation. What the later part of the dialogue shows, however, is that

  ― 52 ―
even Parmenides cannot help talking about things, if he talks at all. And the upshot is that the Eleatic and the Pythagorean-Socratic theories of ideal being are shown to be alike untenable. This proof can only be of an empirical character, but that does not hinder it from being historically true to Eleaticism. It has not been observed that the paradoxes of Zeno, for example, bring out not the “self-contradiction” of Pythagorean theories but a contradiction between their rationalist assumptions and their empirical assumptions; and that this contradiction is demonstrated empirically, however little this may accord with the conclusions Zeno wished to establish. As already noted, the use of the Eleatic method to destroy Eleatic philosophy is credited to Gorgias, and even Parmenides may quite well have been acquainted with his work.

The correct conclusion, then, is the empirical one, that there are no “higher” entities, but everything that concerns particulars is on their own level of existence. Mr Merrylees considers that the plausibility of Parmenides's criticism of the Socratic attempt to bridge the gulf between “higher” and “lower” depends on the choice of a metaphor. The real point is that, no matter how we describe the relation, the description must be such that we can recognise as a single situation “a thing partaking of a form”, and this situation can only have a neutral sort of being (neither “higher” nor “lower”) which must likewise be that of its constituents. And all the proposed relations are intelligible only as perceptible relations between perceptible things, i.e., historical relations between historical things. We all know what is meant by “having a share of”, as when the owl and the panther were sharing a pie; we can see the partakers and the partaken of, and we can see the partaking going on. Similarly, when one thing “comes under” another, or “is copied from” another; we have often come across such complex, existing states of affairs. But none of these experiences helps us to understand the real “participation”; it is something unspeakable — and so is not even something, and not even unspeakable.

Socrates himself puts forms and particulars on the same level, when he admits that they may both “come under” forms, and even under the same forms. We get such examples of this, in the Phaedo, as “Fire is hot” and “This body is hot” (as I pointed out in “‘Universals’ and Occurrences”). And we get the answer to it, in the Parmenides, in the infinite regress of forms. Mr Merrylees meets this argument, according to which, he says, “the idea is (1) a character common to many things which we discover by comparing these things, (2) a character which, when found, can in the same way itself be compared with these things, thereby revealing a further character common to itself and these”, by saying that “Plato's real point” is, that while (l) may be true in a sense, (2) is not. “The idea is of a different order to the things compared, and cannot in any sense be regarded as another thing alongside these with which they can be compared.”

This can only mean that, in spite of Socrates, things and a form cannot both come under the same form, and, as already suggested in the

  ― 53 ―
statement that participation is “that relation between anything and its ideal (and between that and its ideal until we come to the first principle) in virtue of which it is intelligible”, that a thing cannot come under more than one form. So that, if this body is fiery, then, even if fire is hot, this body is not hot; and, if we go on to the “first principle”, this body is not real — a conclusion which would have pleased Parmenides. Again, for one thing to be different from another, it must have a different character; so that every thing has its peculiar “ideal”, and every ideal has its ideal — and when do we ever come to the “first principle”? On the other hand, granting that a thing's ideal is the explanation of that thing, we have to go, in explaining anything, right to the “first principle”, the “self-explanatory”, and we can recognise no distinction between anything and anything else; there is only the self-subsistent One — another conclusion which would have pleased Parmenides.

Thus the refutation of Eleaticism, in the second part of the dialogue, is more than sufficient for the refutation of Socraticism, which appears separately in the first part. It has therefore to be recognised that “This body is fiery”, “This body is hot” and “Fire is hot” are propositions all of the same order, and their terms are all of the same order. So far from the recognition of “forms” settling problems such as that of the like and the unlike, it renders them insoluble; they can be solved only by sticking to things, and recognising that they are sorts of things, i.e., historical situations or occurrences. We have to reject the distinction between being and becoming, and recognise, with Heraclitus, that whatever is, is in process and whatever is in process, is. Thus the realistic rejection of “constitutive relations” develops into the empirical recognition of a singe way of being, that, namely, of observable things — existence; and the position finally appears as that of a positive and pluralistic logic of events. It was fitting that, in the development of such a logic as far as Plato was able to carry it, honourable mention should be made of the names of Parmenides, Zeno and Socrates, since they had all contributed to the working out of the propositional method, though none of them had seen its incompatibility with “ultimates” and all of them had opposed the Heraclitean theory of a single historical order.

III. Realism versus Relativism

Realism appears finally as a positivist doctrine, a logic of propositions or events; and this brings it into conflict with every theory of degrees of truth and reality. It will have been seen that there are natural affinities between the different rationalistic theories; indeed, as Burnet has shown, the Eleatic was simply a heretical Pythagorean. It is characteristic of the instability of the whole position that the extremes between which rationalism fluctuates are the Eleatic doctrine of the One as the sole reality and the doctrine of the super-Eleatic, Gorgias, who held that “there is nothing” (absolute) but all is “relative”. The inconsistency of this Sophistic position, involving, as it does, a hidden Absolute which

  ― 54 ―
appears obscurely as Opinion, does not prevent Relativism from being the most prevalent of “philosophic” views.

The realist answer is that there is something absolute, namely, facts; that even the relativist doctrine itself implies that “the relativity of all” is an absolute fact — not absolute in the sense of being above history, but absolutely historical; so that the doctrine cannot be maintained. Socrates understood that the task of philosophy is to save science from degenerating into scepticism, but he was unable to carry out this task because he was not a realist. The scientist, in so far as he recognises facts and a pluralistic order of events, is in a stronger position than the teleologist. But when he falls short in his logic, divides matters of certainty from matters of uncertainty and makes “probability” the guide of life, when, in fact, in the Pythagorean manner, he separates the rational from the irrational and appears as an unconscious teleologist, his errors are far more difficult to root out. In our own day scientific agnosticism has achieved apotheosis in the doctrine of Relativity.

Rationalist fluctuations are due simply to this, that the rationalist cannot state his doctrine at all without introducing a certain amount of empirical fact, “irrational” as he may call it, and steps have to be taken to conceal the conflict of this fact with the “ultimates”, whatever they may be. And until the recognition of a logic of events has prompted us entirely to “remove hypotheses” of degrees of reality and treat things on a common level, we are prone to fall into dualistic errors and, while imagining that we are conducting a straightforward inquiry, to remove appearances, i.e., deny facts, instead of “saving” them. It is only from the division of the rational from the irrational that “theories of knowledge” have grown up, and that illogical considerations of “certainty” and “probability” have replaced the sole basis of scientific progress, the formulation of propositions which we believe to be true.

The position of Mr E. V. Miller is not that of extreme relativism; like most modern scientific theories, it comes fairly close to the doctrines of the semi-Pythagorean Atomists. In particular, it adopts the notion of “data” or of “that whose nature it is to be given to some mind” — a notion which can no more be supported than that of any other constitutive relation. But, as we have seen in connection with Socrates, there is no limit to the multiplication of “ultimates” and “relatives”, once we have made the fatal division, and Mr Miller's “ultimates”, which he calls “enjoyments”, are quite unable to make contact with “the world of truth”.

Mr Miller develops his views from a criticism of Realism, as I expounded it in “The Non-Existence of Consciousness”. Pure realism, he considers, suffers shipwreck “on the rock of the distinction of truth from error”. As a mere theory of propositions, it provides no criterion of truth. Thus, in any dispute, “the contestants can never get away from propositions, and as there is nothing in the system [i.e., the system of propositions] to cause us to regard one proposition as better than another — no ultimates, as Professor Anderson says — they can never get away from doubt, or even mitigate it in the slightest degree”. It may be said,

  ― 55 ―
however, that this is not true of actual discussion. Though there may be cases in which B goes on denying every premise that A brings forward, it does happen occasionally that A gets back to a proposition which B admits, and which settles the dispute in favour of A.

But, supposing that this does not happen, does this mean that “they can never get away from doubt”? Not at all. As Mr Miller has put it, “A makes a statement which he believes to be true, but B doubts it; what can A do to show that he is right?” A began, it appears, not by doubting but by believing; and, though he has failed to make B believe, this does not give A any reason for ceasing to believe. And, of course, if he had convinced B, that would not be an additional reason for believing. “A supported proposition”, I agree, “is not less doubtful than any other”, since it is supported only by other propositions. But this is merely to say that inference is not the only way of getting to know, that, in fact, it is impossible except on a basis of observation or belief. But what we believe, as Mr Miller has said, we believe to be true — without asking for any “criterion”. And when we doubt, we doubt whether a proposition is true; we do not believe that the proposition is “doubtful”.

The realist position is, then, that there is no criterion of truth, nothing by believing which we believe something else. If the criterion is a proposition, we have not “got away from propositions”, and we still require a criterion to apply to it. If it is not, it cannot settle any dispute. Mr Miller's solution is that certain propositions have “original truth”. They are those “thought-objects” which are the “original counterparts” of an “enjoyment”. Whenever an enjoyment occurs, there is given, in the world of contemplation, a thought-object which, as corresponding to the enjoyment, is true. But these counterparts, which as original are sensed, may later be recollected and manipulated so as to give any number of propositions, which not being original cannot be true, though they may have true consequences, and thus be more or less “probable” or “reliable”. The enjoyments, on the other hand, cannot be contemplated at all; they cannot appear in propositions, since they are not thought-objects, and, as relations are thought-objects, enjoyments cannot even be related to thought-objects.

The main realistic objections to this theory may be briefly stated. If enjoyments cannot be related to thought-objects, they cannot have them as “counterparts”. If, as Mr Miller says, he believes in the existence of enjoyments, he is formulating the proposition, “Enjoyments are existing things”. But, more important still, “original truths” cannot settle disputes in the way required, because people disagree just as vehemently about what is immediately in front of them (what they “sense”) as about anything else. If, in the Sophistic manner, each man's “data” are held to be “true for him”, still that will not give a point of agreement for the settlement of a difference; and it will be impossible to show how such a datum could ever be a verification of a speculative proposition, because the verifying and verified propositions require to have some terms the same. And finally, unless we could find implication as a relation among

  ― 56 ―
propositions, i.e., unless implication were a sensible fact, we should never know how to infer.

It appears, then, that we are entitled to recognise neither degrees of truth nor anything outside “the world of truth”, and, if we did, it would bring us no nearer “certainty”. The theory of “enjoyment” falls with the monistic theory of “consciousness”. Realists will not deny that there is a relation between what a person believes and the state of his mind; but they maintain (a) that this implies that “states of mind” can themselves be discussed and contemplated; (b) that “the fact that I believe X” is not a criterion of the truth of X, nor is “what caused me to believe it”, unless this happened to be inference, at all relevant to the proof of it; (c) that the only propositions that anyone will accept as deciding an issue are such as he himself believes, or regards as facts. If science advances it is because one or more persons believe truths, and not because any authoritative criterion, any “ultimate”, is discovered. The absence of agreement about facts would in no way suggest a possibility of agreement about “ultimates”. And, indeed, the existence of science implies nothing as to general agreement, but only that somebody knows; and what he knows consists of true propositions.

It is most of all in ethics that these scientific requirements are repudiated, and that disagreement and the absence of a “criterion” are taken to imply that there are no independent facts, or that the facts cannot be discovered. But this is a quite incoherent view, as has been shown most particularly by Socrates, in his criticism of Sophistic theories, and by G. E. Moore. Disagreement implies no “doubtfulness”, but is possible only if each disputant believes something to be definitely the case. If subsequent consideration leads him to believe that it is not the case, he is still in the position of asserting something to be an independent fact. And if he is doubtful about the matter, this does not affect any other person who has a definite belief about it. Only if believing X implied (as it does not) believing that everybody believes X, would disagreement have the required effect. Thus if anyone passes a “moral judgment”, he is asserting the occurrence of a certain moral situation. Yet (as I pointed out in “Determinism and Ethics”; A.J.P.P., Vol. VI, No. 4note) there are theorists who find it possible to say that the moral judgments which we pass are the “data of ethics”; that is to say, that ethics has to begin by studying our judging, our acts of “approval” and “disapproval”, instead of with what we judge to be facts. It is further supposed that we promote, or advocate the promotion of, what we approve. And thus the field of ethics is defined by our attitudes of approving, promoting, advocating, etc., and what we take up these attitudes to comes in only relatively.

Having thus, as strongly as I could, repudiated ethical (or any other) relativism, I am surprised to find Professor T. A. Hunter regarding my more recent criticism of it as an “inconsistency” on my part. I have certainly at the same time attacked any theory of an absolute or ultimate,

  ― 57 ―
since “ultimates” and “relatives” hang together; but I have always maintained that true, ethical or other, propositions are absolute, or independent, facts — in other words, that they are actual occurrences. Professor Hunter does not escape Absolutism by saying that “all goods are relative”; he merely omits to state what his absolute is, i.e., to what they are relative, while his language implies that the things he is speaking of are actually or absolutely both good and relative. In fact, his ethical absolute turns out to be no other than the old Sophistic “convention” (??µ??) or, as he puts it, “a code to act upon”. It is the more surprising that he praises Socrates (who developed his ethical views in criticism of the Sophists) for bringing morality to earth, and neglects the real exponents of a “practical morality”.

The fact that Socrates upheld the relativity of goods to “The Good”, did not prevent him from presenting some very forcible criticism of “codes”, and this realistic work was his real contribution to ethics. It appears notably, as I pointed out in my comments (A.J.P.P., Vol. VII, No. 4) on Professor Hunter's original article (“Theory and Practice in Morals”; A.J.P.P., Vol. VII, No. 1), in the Euthyphro and in Book I of the Republic. The argument is that, even if we could divide good acts into religious and social, we cannot say that good religious acts are those which please the gods, because that does not tell us which those are, and because “pleasing” is only a confused way of saying “being regarded as good by”; and similarly we cannot say that good social acts are those which benefit people, because this also does not tell us which those are, and because “benefiting” means “doing good to”. In each case a qualitative distinction is required, and in each case there is a covert reference to the quality good. These, as I indicated, are the fundamental objections to the popular notion of “rendering services”.

These considerations also apply to Professor Hunter's theory of the “divergence” between theory and practice in morals. There are, it appears, codes and codes; codes to act on and codes only to advocate. A person's moral practice is the code he acts on, while his moral theory is the code he advocates; and there is divergence if he does not act on what he advocates. Now we should all doubtless agree that people have habits or ways of behaving, and the realist is at liberty to say that some of these ways of behaving are good and others are bad. Not so the relativist; for him they can only be recognised or not recognised, i.e., supported or opposed. Where, then, does morality come in? On this view, it would seem that the activities of molecules are as moral as those of men; they interact, they aid or hinder one another. What is the difference?

The difference, it appears, is that people do not merely do or refrain from doing certain things, but certain things are “done” and others are “not done”. Or, as Professor Hunter, in his absolutist way, puts it, men act on “principles”. This is nothing but a dualistic attempt to introduce a peculiar kind of causality, the teleological, into human affairs. But, even so, it fails to account for the “divergence”. If the code a man formulates

  ― 58 ―
is a statement of the principles he acts on, and if his actual principles of action are different, he is simply mistaken, or else he is lying — in any case, the code is merely false. But Professor Hunter will not accept this interpretation. For him, therefore, the code is not a statement of the principles on which its formulator acts, but is a declaration of what is the thing to do or what are the principles to act on. We see, then, that there is no divergence whatever, if the man does not “act up to” his code, because his actions are not a code; but secondly, and more important, that the supposed relation “to” conceals a qualitative distinction between “the thing to do” and “the thing not to do”. And this confusedly apprehended quality is goodness; the thing to do is the thing which it is good to do, i.e., that the doing of which is good, i.e., a good activity. This, then, is a moral fact, which the relativist only obscures and distorts with his “codes” and his “principles”.

Now, granted that activities may be good or bad, they still have the ordinary causal relations; they support and oppose one another. In “Determinism and Ethics” I adopted the Socratic view that goods always support one another and oppose bads, while bads may support or oppose other bads. But, as I said, all this cannot be found out merely by observing relations of support and opposition between human activities. On the contrary, unless we take qualities into account, we cannot distinguish an ethical struggle from a merely physical one. The failure to recognise qualities is incidentally the main defect of the utilitarian doctrine of “the greatest good of the greatest number”, or the greatest possible mutual support But even when we do recognise qualities and can thus appreciate the ethical struggle, the carrying on of that struggle is not, I stated, the same as our investigating it. “People in general do not think very much about the goodness of their activities. They are simply to be found trying to make discoveries or to produce works of art, exhibiting love or courage, or, on the other hand, imposing obligations on themselves or others, because they are made that way, i.e., because their character, in relation to their history, has so developed. And these are the conditions, and not any metaphysical ‘freedom’” — and equally, I should add, not any metaphysical “utility” — “on which, if at all, further development is possible.”

That is my answer to Professor Hunter's queries, “how we are to decide whether an activity is good, bad or indifferent, and why we take the trouble to study the moral facts”, and again to the question, “if, as Professor Anderson admits, we must study moral facts, must we not come to some conclusions, and, if we do, may not these conclusions influence or direct conduct?” I have never admitted that “we” must study moral facts; if anyone is of an inquiring turn of mind and is specially interested in ethics, he will do so; and, of course, his interest in the subject may be stimulated by other people's. Regarding the investigatory activity as good, whether it is directed to ethics or any other subject, I conclude that it will assist other goods; but I could not conclude, without circularity, that that is its “purpose”. We learn about goods, as about other

  ― 59 ―
things, by observing them; and we can be assisted by what other people tell us about them and their distinctions from and relations to other things. For example, I recognise a qualitative distinction between investigation and obscurantism, and I say that the former is good, and the latter is bad. And by this I do not mean either that I am an investigator or that I think that other people, who are not that way inclined, “ought to” investigate. But I certainly do not cease to recognise the distinction because someone else does not recognise it. I could do so only if I became acquainted with other facts which implied that there was no distinction of the sort I had attempted to draw.


For Realism, then, as against all the “ultimates”, facts are good enough. It does away with the philosophy of good intentions, of “service” to the idol of social utility or any other, and establishes philosophy as logic, the logic of events. It rejects as unphilosophic the “saving of hypotheses” of discontinuity, whether the hypothesis is only a little one or is the great One. And, rid of “meanings” and “purposes” and other products of “vicious intellectualism”, it proposes as the formal solution of any problem the interaction of complex things.