Our problem is, by what test are we to appraise the adequacy or inadequacy of our ideas to represent the reality which, at the phenomenological level, is taken as confronting us? Is it possible, for example, to hold that Correspondence is the 'test' as well as the 'meaning' of Truth? that to know whether we have got Truth we have to compare what is affirmed in the judgment with a 'reality' somehow independently given? The objections to such a view are very numerous and very obvious, and it is really abundantly clear that Correspondence cannot be the 'test.' But since it would be overbold to suppose that this theory is, even yet, without disciples, and since, further, one does often in practice seem to oneself to be applying the 'Correspondence test,' it will be well to open with some observations upon it. To do so will at least serve to prepare the ground for a more satisfactory theory.

Let us take, then, the kind of instance in which we are accustomed to imagine ourselves to be testing a truth by its 'correspondence' with 'fact,' and let us see what it is exactly that we are doing. Suppose that Smith says to us 'This is a stone pillar,' and that we, regarding the pillar closely and possibly feeling its texture with the hand, think or say, 'Yes, that is true, for the "facts" bear it out. Smith's judgment corresponds with the fact.' Now do we really here compare Smith's judgment with what is just 'fact'?

For the sake of precision we must allude first of all to an elementary point which has to be borne in mind, on whatever view one holds of the test of truth, with reference to the appraisement of the judgments of other persons. Smith's judgment is not, strictly speaking, one of the terms in our comparison. We do not compare Smith's judgment, for the simple reason that we cannot know it. The verbal expression of Smith's judgment in the proposition is what we know, and this is only an indication, capable of all degrees of inexactitude, of the real inner nature of what is being affirmed. What we do compare with the so-called 'fact' is the meaning which the verbal expression of Smith's judgment provokes us to affirm, a meaning which we assume to be identical in all relevant particulars with Smith's meaning. It is quite clear that this assumption is, at least sometimes, profoundly mistaken. The sage may signify assent to the fool's declaration that 'all is vanity,' because the meaning these words stimulate him to affirm is in accord (as he thinks) with 'fact.' But if the sage were cognisant of the actual thought in the fool's mind he would in all probability tell the fool that he was talking nonsense. The assumption, then, that the judgment of another can be one term in our comparison is never strictly true, and is often false in a way which promotes serious fallacies.

This consideration, however, is relatively unimportant. It is the other term of the comparison that is our chief concern. What is the status of this 'fact' with which we suppose the judgment to correspond? And let us henceforth assume, to avoid the difficulties indicated in the last paragraph, that it is our own judgment, not another person's, that we are trying to appraise. The presupposition of the theory is that the 'fact' is a fragment of reality itself, a transcript of the objective world unmodified by any subjective influences. But analysis of the 'fact' very soon discloses to us that such 'facts' are nothing but creatures of theory. A fact exists for us only in the medium of judgment, and judgment is not a transparent window but a subjective function of an exceedingly complex nature. Activities of distinguishing and relating in terms of certain categorial characters are involved in the apprehension of even the meanest 'fact.' And the arguments which have from ancient times been led to establish this are not overthrown by being ignored. If you choose to neglect this subjective aspect of our 'facts,' what you are left with is not a 'given reality' but a particularly vicious abstraction. It is inexcusable to assume that that which results only in connection with a complex process is itself indifferent to the fact of the process.1

Of course, to 'recognise' this subjective aspect must not be taken to mean that we can straightway pronounce the fact as apprehended to be in disagreement with the fact as it is in the real order - the 'epistemological object' (to use Dr Broad's terminology) in disagreement with the 'ontological object.' That is a matter which can only be settled by further investigation. But what it does mean is that in our so-called 'Correspondence test' we are not comparing a judgment with some directly given reality, but simply with another judgment: with another judgment, which, for reasons not yet specified, we take to correspond with the objective reality and to be therefore 'true'.

As these last words indicate, what the Correspondence theory does is merely to throw the problem of the test of Truth back a stage. The 'fact' with which the judgment is to correspond turns out to be itself a judgment, and 'Correspondence 'tells us nothing as to how we are to know when the latter judgment represents reality - when it is, in short, 'true.' It is evident, then, that in order to apply the Correspondence test with any effect we must already be in possession of some independent criterion of the 'true' judgment.

The fundamental vice of the Correspondence test is thus that its application really presupposes some other criterion. We may easily see that this is so even in regard to that class of instances which may fairly be said to represent the stronghold of the Correspondence theory - propositions in which something is predicted which is, apparently, borne out by the subsequent 'facts.' Brown says to us' There will be a thunderstorm to-night.' If (to use the ordinary way of speech) 'a thunderstorm does in fact occur,' we naturally tend to say that Brown's judgment is proved true by the event. It accords with 'fact.' But what, again, is the 'event' or 'fact'? In so far as I can use it as a term in my comparison, it is just a judgment of whose veridical character I entertain no misgivings. What happens is that I compare Brown's judgment (in the measure in which I apprehend it) with the judgment which I find myself making with considerable conviction (and suppose other persons similarly situated to find themselves making with like conviction) 'to-night' with respect to the climatic conditions. And the truth of this latter judgment, the genuineness of its correspondence with reality, is clearly something that demands its own independent justification.

1 Cp. Bradley, Appearance and Reality, pp. 27-8.