There can be no absolute certainty, then, in terms of the Coherence test, although an intelligible meaning can be attached to the phrase 'degrees of certainty.' No intelligible meaning can, I think, be attached to the phrase 'degrees of Truth.' It is easy to understand how we can have 'degrees of Truth' if Coherence is not only the 'test' but the 'nature' of Truth. Degrees of coherence or system would then be, quite directly, 'degrees of Truth.' But if Coherence is only the test of Truth, and the nature of Truth is correspondence with Reality, the matter is otherwise. Truth becomes primarily a matter of' 'yes' or 'no.' No doubt there is a sense in which we can speak of an ideal representation corresponding 'more' or 'less' closely with the reality. But even so, we could not possibly hold that 'degrees of coherence' show a constant proportion to 'degrees of correspondence.' On the contrary, the course of experience teaches us that often, with the advent of a single fresh item of relevant information, a judgment supported by a high degree of coherence has to be modified in a way which leads to a most radical revision of the mental 'picture' of the real event. The degree of correspondence attaching to the former judgment bore no kind of proportion, we have to say, to its high degree of coherence.

The constant probability, therefore, of having to revise one's mental picture of any reality must be accepted even at the Phenomenological level; with certain reservations, again, shortly to be discussed. 'Hard facts' are just mental pictures which the pressure of experience has not as yet obliged us to modify (though it may well be that it ought to have obliged us). It is sometimes maintained, indeed, that the development of science does not really tend to overturn the ordinary perceptive judgments of daily experience - 'the sun is shining' and so on - but merely places these judgments in a wider orbit of explanatory connection. The scientist, it is said, may make many other judgments when he attends to the phenomena in question, but he also makes the common-sense judgment 'the sun is shining.' Such propositions, it is supposed, have so far resisted modification, and there is no reason to expect that further scientific development will be any more disastrous in its effect. But this view, I think, rests on misunderstanding. What the astronomer is concretely meaning when he judges that the sun is shining is something which bears very little resemblance indeed to the meaning of the plain man. Without venturing upon a description of what the astronomer's mental picture is, it is safe to say that it is not the picture of a yellow ball shooting out light very much as a gun shoots out bullets. The plain man's picture is not as a rule a great deal more refined than this. And the astronomer would certainly say that it is a sheer distortion of the reality. When we attend, therefore, to the inner meaning of the judgment, to what is really being affirmed, it seems plain that even from the Phenomenological point of view we must agree that a constant process of modification of our ideal representation of reality is inevitable.

At the same time I am convinced that it is possible, and that it is one of the defects in the usual statement of the Coherence theory, to push too far this doctrine of the judgment's susceptibility to modification. Consider again the simple perceptive judgment 'the sun is shining' When we make this judgment we are ordinarily aware of certain feelings - e.g. the feeling of warmth - and in so far as we affirm, i.e. make the judgment, that we are 'enjoying' such feelings, what we affirm does not seem to be open to modification after the manner of the content of the 'objective' judgment. And, in certain conditions, our apprehension of 'yellowness' and 'brightness' in this situation will fall under the same category. For in so far as we affirm only that we have or enjoy a certain sensum, and do not assign to it an objective interpretation, our affirmation seems to rest on unchallengeable ground. Of course we do, as a rule, affirm much more. We do, as a rule, affirm that yellowness and brightness belong to the spherical object overhead, not just that we 'enjoy' certain sensa. And at this point - where the judgment refers to 'objective' reality not to subjective feelings - criticism of the judgment's validity may begin at once. But it seems to me of the first importance to see that, with respect to subjective feelings, the person feeling is the ultimate authority as to what is felt. A feeling which he is definitely conscious of enjoying is a feeling which no amount of science or philosophy will ever be able to persuade him that he did not enjoy. This does not imply that every man is an infallible introspective psychologist. Without doubt practice, and knowing what to look for, are indispensable for the accurate observation of our subjective sensations. But the point is that the task of the psychologist in this connection is no more than to use every artifice to induce the integrity of a pure remembrance, unclouded by presuppositions. The ultimate authority is the experiencing subject himself. His testimony, once the way is cleared for its untrammelled deliverance, must be acknowledged to be final. If he insists that he experienced a sensation of yellow, then it is entirely ultra vires for the psychologist (as indeed he well knows) to tell him that he believes this only because of his ignorance of psychology.1 In short, we do have here, in these judgments which report our own subjective sensations, assertions not legitimately modifiable by the progress of science, assertions which may therefore reasonably claim to be 'intellectually incorrigible'.