A1 There is something profoundly unsatisfactory about the contributions of 'analytic' philosophy to the study of ethics. Partly this might be attributed to the fact that an analytic philosopher, qua philosopher, can only concern himself with trying to understand moral systems or moral language; the construction and advocacy of moral systems is not a part of his mandate. But even given this preoccupation with analysis rather than advocacy, their contributions seem lacking. N3
A2 An 'analytic' philosopher concerns himself with analytic truths. In this limiting his attention he makes himself vulnerable. He is vulnerable because he may some day discover that there are no analytic truths.
A3 The notion of an analytic truth is dependent upon meaning. A necessary (and perhaps sufficient) condition of a sentence expressing an analytic truth is that there is some relationship between the meanings of the words in the sentence. A necessary condition of a word having a meaning, in any straightforward sense, is that there is something which can be truly said of all proper uses of that word. It does not seem to be a necessary condition of a word filling some role in language that we can say anything much at all about all its uses. (Some things of course we could find to say, what part of speech it is for example, and thus in what sort of way it is combined with other words. This sort of thing is not very helpful in allowing us to discover analytic truths using this word.) So it is not necessarily the case that there are very many analytic truths.
A4 The notion of analyticity, and analytic truths themselves are indubitably of great value in philosophy, but they are tools which must be used with caution. If we try to understand language simply by inspecting the analytic truths which it contains we may expect to have difficulty. In studying those bits of our language which are of special relevance to ethics analytic truths are hard to come by. If we are concerned to understand how morals works, then we must understand how morals work, and a list of analytic moral truths will not get us very far in that direction. N4
A5 The first problem in moral philosophy is that of settling our area of study. For it is commonly held to be concerned with the use of certain words (e.g. good, bad), but with their use in a certain way (morally). This is to prejudge an important issue. We should first consider the use of these words generally, and if in so doing we find that there is a special way of using these words which can be profitably isolated for closer study, then we may do so. An important part of understanding the moral uses of these words is understanding the relationships between these uses and whatever other uses there might be. N5
A6 The preference for studying analytic truths is derived from the desire to know what is possible rather than just what is. But that a sentence is analytically true is itself a contingent truth, might we not want to escape from the chains of that contingency also? If we accept language as it is without considering how it might be, do we not blinker ourselves in the same way as we would if we were to accept the laws of physics as they are and refuse to consider what they might be. The language of morals is a particularly bendable part of language. The way in which we use evaluative concepts, whether or not we have a special subclass of evaluative judgements which we call 'moral', and the ways in which we employ language in relation to our value system, these things are all beyond the jurisdiction of precedent. We may adopt someone elses morals, and their ways of expressing their morals and of coercing others into their moral scheme, but there again we may not. If an analytic philosopher enunciates the analytic truths of morals he is saying of certain schemes that they misuse moral language, or that they are simply not moral schemes at all, and in so doing he unwittingly becomes a moral agent rather than the disinterested observer he might have hoped to be. If I chose to use moral concepts rather oddly, perhaps if I declined to use them prescriptively, then to castigate me for my oddity is a distinctly unphilosophical approach, better to try and understand my oddities. The worst we can say of someones use of language is that we can find no rhyme or reason in it, less severely we might be concerned to point out how unusual it is, but 'correctness' and 'propriety' are epithets which might bring laughter on our heads in the anarchic present. N6