John Maynard Keynes, answered from the texts and cited to the page.
Probability is a logical relation between one proposition and another — between a conclusion and the evidence bearing on it — and the degree of belief it is rational to entertain is fixed by that relation, not by the believer's temperament.12 Many such relations admit of no numerical measure, and some pairs cannot be compared at all. Distinct from the balance of the evidence is its weight: the sheer amount of relevant evidence in hand.345
We come next to the distinction between that part of our rational belief which is certain and that part which is only probable. Belief, whether rational or not, is capable of degree. The highest degree of rational belief, or rational certainty of belief, and its relation to knowledge have been introduced above.A Treatise on Probability, pp. 10–11
The first writer I know of to notice this was Ancillon in Doutes sur les him; it is certain that we also intend to convey our belief that we are acting rationally in regarding his theory as probable.A Treatise on Probability, pp. 5–6
As soon as mathematical probability ceases to be the merest algebra or pretends to guide our decisions, it immediately meets with problems against which its own weapons are quite powerless. And even if we wish later on to use probability in a narrow sense, it will be well to know first what it means in the widest. 7.A Treatise on Probability, pp. 6–7
While taking pains, therefore, to avoid any divergence between the substance of this chapter and of those which succeed it, and to employ only such periphrases as could be translated, if desired, into perfectly exact language, I shall not cut myself offfrom the convenient, but looser, expressions, which have been habitually employed by previous FUNDAMENTAL IDEAS writers and have the advantage of being, in a…A Treatise on Probability, pp. 19–20
But the first equality can exist in other cases also. If, for instance, a and b are members respectively of different sets of three equally probable exclusive and exhaustive alternatives, then a/h = b/h; but these arguments may have very different weights. If, however, a and b can each, relatively to h, be inferred from the other, i.e. if a/bh = 1 and b/ah = 1, then V(a/h) = V(b/h).A Treatise on Probability, pp. 81–82
Ramsey's criticism is generally held to have won the technical argument, and the logical relation has few defenders in that form — though the specialist literature disputes how far Keynes actually retreated, and O'Donnell argues the assimilation of Keynes to subjectivism is a misreading. The non-numerical and non-comparable strand, and the concept of weight, have had a substantial second life in economics.