1 ‘What is right (or wrong) for one person must be right (or wrong) for any (relevantly) similar person in (relevantly) similar circumstances.’ Discussed in Generalization in Ethics (Singer, 1961), esp. chs. 1 & 2. This corresponds in a number of respects, but not in all, with what R. M. Hare and others have called ‘universalisability'. I have discussed these differences and some confusions about the Generalisation Principle in ‘Universalisability and the Generalization Principle’ in Potter and Timmons, 1985, 47–73; in ‘Imperfect duty situations, moral freedom, and universalizability’ in Starr and Taylor, 1989, 145–169; and in ‘Universalizability', forthcoming in The Cambridge Dictionary of Philosophy, ed. Robert Audi.
2 Rawls, 1955, 3n; 1958, 164K. Pincoffs, (1966, ch. 3, sec. 2, pp. 53–56), provides an especially intriguing analysis of the distinction between practices and institutions. Although in the end I was not persuaded by Pincoffs's arguments to shift my use of the term ‘institution', these arguments are not only worth considering, they need to be considered.
3 I presented a preliminary and much abbreviated account of this conception of institutions in Morals and Values (1977), 341 ff.
4 William Greider, The New Yorker, 16 November 1987, p.68, published since (1989), as Secrets of the Temple: How the Federal Reserve Runs the Country; wherein the ideas in the passage quoted may be found in and around pp. 226–229. It is interesting that the book does not, in its later accounts, adhere consistently to the idea of money as an illusion. Cf., e.g., pp. 242, 265, 453, 620, 673, 685, 688, and 714. Discussion of the consistency of Greider's ideas, however, is more appropriately reserved for a discourse explicitly on money. I have quoted his striking account of money because it is an observation on an especially interesting and certainly abstract institution, which may have application to other institutions as well.
5 I owe this point, or the thought behind it, to my colleague Claudia Card.
6 At least I find it difficult to characterize. An interesting discussion, almost the first word on the topic, is provided by Winch (1958). But of course there are valuable discussions by Durkheim and Weber and others.
7 The following statement by A. J. Liebling helps illustrate the complexity of these terms: ‘The function of the press in society is to inform, but its role is to make money’ (Liebling, 1964, Foreword, p. 7, par. 22). Suppose ‘role’ and ‘function’ were interchanged, to give: ‘The role of the press in society is to inform, but its function is to make money.’ This doesn't fit as well, and certainly would not say what Liebling wanted to say. Why not, and how are ‘role’ and ‘function’ to be defined?
8 The New Republic, vol. 208, no. 9, issue no. 4076, 1 March 1993, p. 5.
* An earlier and abbreviated version of this paper was presented at the XII InterAmerican Congress of Philosophy in Buenos Aires, 28 July 1989, and was published in Spanish, translated by Roberto de Michele, technical revisions by Eugenio Bulygin, under the title ‘Etica Institucional', in Analisis Filosofico, vol. 10, no. 2 (November 1990), pp. 123—138. This is its first appearance in its author's native language; it has been considerably revised for its appearance here, as a result of discussion on several occasions at different places.