Travis' Saying and Understanding offers the foundations for what he hopes will be a generative theory of illocutions. I want to criticize the “generative” character of the theory.

Note first that I am not going to criticize the theory because — as is true of Saying and Understanding — there are no rules for its generation, nothing to let us see how it will work. The book is only a discussion of the foundations of such a theory. My criticism is that I do not see how it can be generative, given what Travis expects it to do. So in this critique I shall outline what I think he expects it to do, how he expects it to work, and why I think it cannot.

This interesting volume (published by Yale University Press, 1974) offers sustenance to almost all who have an appetite for the problems of practical reason. As the Proceedings of the First Bristol Conference on Critical Philosophy, it contains the five papers presented there, with the remarks of commentators and, in three cases, replies. First, the bill of fare.

Roderick Chisholm opens with “Practical Reason and the Logic of Requirement”, about which I shall say no more than to quote the beginning of G.E.M. Anscombe's comment: “It is characteristic of Professor Chisholm to carry one always by a succession of small harmless-looking moves which then suddenly enable him to pull out some large kicking rabbits like a conjurer operating with a hat. This time he has produced the large kicking rabbits, but the initial movements do not seem as unsuspicious as usual.” (pp 17–18) Those who like what Chisholm does, will find fine detail in his paper, the comments, and his reply; unfortunately, for me Professor Chisholm's rabbits have the disconcerting invisibility of Harvey.

Frege: Philosophy of Language (Duckworth, London: 1973; xxv, 698) has been heralded as Michael Dummett's long-awaited magnum opus on Frege. Actually, however, as the author himself tells us, it is only the first of a two-volume series devoted to Frege's philosophy of language and his philosophy of mathematics respectively.

The book itself has been long in preparation, the writing of it having been interrupted for several years. This fact could not help but leave some marks on the organization and content of the various chapters. Still, all in all, it presents a remarkably interesting and provocative whole. It is topical, rather than purely genetic in approach, dealing with various fundamental features of Frege's philosophy of language in some 19 chapters; e.g., with his sense-reference distinction, his theory of assertions, his conception of the nature and place of truth-values, propositions and thoughts, his views on quantification and identity, the nature of proper names, his polemic about explicit vs. implicit (contextual) definitions, his theory of incomplete (predicative) expressions, and the like. However, the historical needs of the reader are also well-served.