The criticism which has to be passed upon the doctrine of 'Degrees 'from the side of Supra-rationalism is, however, so obvious in character that one's chief difficulty lies in understanding how Bradley can have persuaded himself that the doctrine is, for him, tenable. For it is of the essence of Bradley's position to hold that differences are united in Reality in a manner intrinsically different from the mode of union which is characteristic of, and inseparable from, the finite intellect; and that of this ultimate mode of union the intellect can know nothing. How then can it be possible to grade thought-products according to the degree in which they manifest the nature of this unknown and unknowable Reality? The reply might be, and is, given, that at least we know that the nature of Reality is to be 'self-consistent,' and that we can, and do, grade our judgments in a hierarchy of 'self-consistency.' But this, on Bradleian presuppositions, is merely to play with words. For - and here we have the crux of the whole situation - 'self-consistency' means something different in the two usages. The true self-consistency, that which applies to ultimate reality, Bradley has discovered to lie 'beyond any relational arrangement' (to use his own expression), and to be incapable of envisagement by the intellect. Obviously, then, the ideal of self-consistency which the intellect does admittedly employ in practice in grading judgments cannot be the 'pure' ideal. In so far as it is concretely envisaged, and thus able to be used as a positive criterion, 'self-consistency' can only be a defective or degraded expression of the pure ideal. This is the direct implication of Bradley's fundamental contention that the intellect attempts to advance towards the self-consistent after a manner which, although characteristic of it, is not in fact capable of leading to the truly self-consistent. What such a doctrine implies is surely that the ideal as concretely operative in intellect appears in a bastard form. How otherwise could it be guiding the intellect along a path which does not lead to true self-consistency?

But if 'making no significant difference' means (as it seems to mean in Mr Hocking's subsequent illustration of a ship not ceasing to be the same ship when it reaches a new port) merely that we can abstract from the difference without introducing into our calculations noticeable confusion or error in the ordinary contacts of experience - why, this is something which no responsible defender of internal relations has ever challenged. Bradley has himself frequently stated that we do right to regard some relations as merely external for all practical purposes. The issue is not this, but whether such relations can or cannot be regarded as ultimately external. No Realist can accept the negative answer, no Monistic Idealist can accept the affirmative. The issue is quite vital; and the attempt to mitigate the rigour of the opposition is not, I think, serviceable to the promotion of a right decision.

In short, we have got to recognise that there are two quite distinct forms of the ideal of self-consistency unmistakably posited in the Bradleian epistemology, whether or not they be always kept clearly in mind by Bradley himself. There is the 'pure' self-consistency which a regress upon the conditions of cognitive experience compels us to posit as characterising ultimate reality, but of whose concrete character we know nothing. This we may call the 'noumenal' ideal. And there is the 'empirical' self-consistency which is the criterion and positive guide in our actual intellectual operations. This we may call the 'phenomenal' ideal. The contradiction in Bradley's doctrine of 'Degrees' is that he uses the second form to apportion degrees of the first.

Since I reject the doctrine of 'Degrees' primarily because I accept the basic principles of Bradley's epistemology, it will be necessary for me to deal with the relation of these two forms of the ideal at some length in the sequel. Here all that I am anxious to make clear is that in some sense these two distinct forms must be admitted as components of the Bradleian scheme, and that this admission is fatal to the doctrine of 'Degrees.' It may, however, be advisable to anticipate subsequent explanations to the extent of pointing out that, although the 'phenomenal' ideal is here spoken of as being, as it were, a 'transcript' of the 'pure' ideal, it is not to be supposed that any conscious process is involved in the 'transcription.' The mind does not first of all possess the pure ideal and then proceed to construe it in terms of a relational arrangement. The ideal before the mind is from the first envisaged in its phenomenal aspect. We get at the conception of the 'pure' ideal only by a critical regress which discloses to us that the mind cannot attain to the harmony which it itself demands for Reality by following that form of the ideal which is its positive criterion. It seems a valid interpretation of this unsatisfiable demand to say that there is an ultimate ideal possessing the mind (the 'pure' ideal) which is radically distinct from the operative ideal which controls its concrete procedure (the 'phenomenal' ideal).

It is true, however, that Bradley has, in response to criticism, offered to explain the paradox of his doctrine of 'Degrees of Truth.' To notice his rejoinder will, I think, but add confirmation to our view that the doctrine is, on his premises, untenable. On p. 597 of Appearance and Reality Bradley tells us, in answer to criticism similar to the above, that, so far as he understands the objection, it 'offers no serious difficulty.' No doubt (he urges) from one point of view all judgments are equally false. But there is another no less valid point of view from which it is proper to apportion to them 'degrees of truth.' In theology, he points out, we are not conscious of any incompatibility in maintaining that 'before God,' measured by His Perfection, all men are equally sinners, while also maintaining with like assurance that some men are certainly 'better' than others. Both points of view have their own validity. So too, Bradley suggests, with this apparently incompatible combination of philosophical doctrines - 'Degrees of Truth' and the 'Supra-rational Absolute.' The worst fault that can be alleged against his position, Bradley thinks, is that of 'two proper and indispensable points of view,' he may have unduly emphasised one.