A NEW KIND OF ETHICAL NATURALISM?
6. 1. THERE used to be only one kind of ethical naturalism, namely what we might call semantic ethical naturalism or analytic ethical naturalism. This held that moral words have the same meaning as certain combinations of non-moral expressions (which combinations depended of course on the particular naturalistic theory in question). A crude simple example of such a theory would be one that G. E. Moore considers in his famous 'Refutation of Naturalism' ( 1903), namely the view that 'good' means the same as 'pleasant' or 'pleasure- maximizing'. He attributed a naturalistic view of this general sort, certainly wrongly, to J. S. Mill. His refutation of such views was at first almost universally accepted, and people propounded instead various intuitionistic views about the meaning of 'good' and other moral words. After that the intuitionist theories that they had embraced fell into disfavour because of well-known difficulties (mainly that of saying what the 'non-natural properties' could be that the intuitionists said the moral words 'were the names of', as they sometimes put it).
A way out of the impasse thus reached was suggested by the emotivists. Starting with the Swedish philosopher Axel Hägerström ( 1911)--though emotivism did not really catch on in the English- and German-speaking worlds until the 1920s--the emotivists suggested that the reason for the difficulty in finding what properties moral words connoted was that they did not connote any properties at all, but had a different function altogether. They were not, as it was said, 'descriptive' words. They were used to express emotions or attitudes of the speaker, and to produce such attitudes in hearers.
The emotivists were not the first non-descriptivists; there is a clear statement of a prescriptivist position in the last chapter of J. S. Mill System of Logic ( 1843), in which he echoes Hume's famous pronouncement in Treatise III. 1. i about 'ought' and 'is' and even calls____________________