“There's one thing certain,” said a historian of my acquaintance when he heard the title of this paper, “that's a problem which would never perturb a working-historian.” He was wrong: a working-historian first drew it to my attention; and in one form or another it raises its head whenever historians discuss the nature of their own inquiries. Yet in a way he was right. His mind had turned to the controversies of epistemologists, controversies about “the possibility of knowledge”; historians, he rightly felt, do not trouble their-heads about such matters.

Among the manifold tendencies that contributed to the philosophical revolution of the seventeenth century was a revival of Greek atomism. The ancient particulate theories of nature had been rediscovered by way of Lucretius and Diogenes Laertius in the fifteenth century and later scholars explored the principles of Democritus and Epicurus.

Justification of beliefs can be spoken of from two points of view. In ordinary life it consists in giving reasons of a kind which will be accepted as good under normal circumstances. I know that my friend is still in the garden, because I was talking with him there a few moments ago, and I have not seen him come out. We can sometimes give reasons for our reasons, but there is a limit to this. For the most part we do not go beyond common-sense considerations.

IN the right circumstances and the right frame of mind we are prepared to make aesthetic appraisals of almost anything, from hills, cottages and cars, to symphonies, people and poems. My problem is to try and set a boundary in at least one direction to the catholicity of this kind of judgement. I want to argue that when we use a word from our aesthetic vocabulary for appraising a theory in science or a proof in mathematics we are not properly to be described as making an aesthetic appraisal, though there are close parallels with our normal use of these words.

It is one of the objects of what is sometimes called “general theory of value” to study all sorts of value judgments or (what would be a better name for them) evaluational judgments. But what the sorts of evaluational judgment are is a question that has so far by no means been settled. There are only two kinds of evaluational judgment that are universally recognized and that have well–established names, the ethical or moral, and the aesthetic. Another pair that have sometimes been mentioned are the prudential and the economic. The object of this paper is to direct attention to still another, quite distinct, sort of evaluational judgment which we commonly make, to give judgments of this sort a name, and to characterize them in a provisional way.

If we ask what are the problems which have to be dealt with when the subject of art is being discussed in a philosophical manner, we shall no doubt receive a variety of answers: but there will be in one respect a considerable measure of agreement, viz. that the main problem consists in discovering (a) what is the common property in all works of art which distinguishes them from things that are not works of art, and (b) what is the common property in all good works of art which distinguishes them from bad or mediocre ones. Certainly an immense amount of labour has in the past been devoted to this purpose. Recent philosophical practice, however, has implied some dissatisfaction with this type of approach, and the purpose of this article is to try to show that the dissatisfaction is justified.

IN “What Morality is Not”, Philosophy, XXXII (1957), Mr. Alasdair Maclntyre argues against the view, now common, “that universal–izability is of the essence of moral valuation”. On page 331 he uses an argument which is an adaptation and extension of Moore's naturalistic fallacy argument, and which is generalizable. As Moore's argument, if cogent, holds against all definitions of “good”, “right”, etc., so Maclntyre's argument, if good, holds against all definitions of “moral” and “morality”. For this reason I shall examine his argument, as I once examined Moore's.1 I wish to do this partly because I should like to go on looking for definitions or elucidations of words like “moral” such as are contained in the uni–versalizability view, and partly in order to air some general questions which are raised by what Maclntyre says.