Preface and Acknowledgements

Extract I started grappling with the philosophical challenges presented by vagueness in the early 1970s. At that time, I think it fair to say, almost nothing of real significance had been written on the topic since the contributions of Eubulides of Megara.1 In the modern era, in particular, philosophers of language from Frege on had been for the most part content to theorize in ways that marginalized vagueness, or to focus on idealized languages in which there was none. No one writing before 1970 seemed fully to have taken the measure of the awkwardness of the Sorites paradox,2 or the depth of its roots, as usually formulated, in our intuitive thinking about what kind of ability mastery of a language is. My curiosity about the topic was originally piqued by conversations with my friend the mathematician Aidan Sudbury and with Michael Dummett, then my colleague at All Souls, who around that time was working on the stunning lecture that he later published as 'Wang's Paradox' (Dummett 1975). My own interest initially stemmed from concerns in the philosophy of mathematics: I was drawn to the thought that the apparent open-endedness of the extension of a vague predicate might provide a fruitful model for the manner in which a finitist should think about the putatively infinite extension of natural number, and that a correct logic of vagueness might accordingly be appropriate for a finitist number theory. My subsequent paper 'Strict Finitism' (Wright 1982) was the upshot of my reflections in that direction. But while thinking about finitism I became preoccupied with the Sorites paradox itself. Dummett's paper argued, inter alia, that vague expressions do indeed affect natural language with inconsistency—that is, that the paradox shows that our use of vague expressions is governed by rules that are actually inconsistent. That struck me then as an incredible conclusion,3 but one that was nevertheless forced by a certain conception of the significance of the kind of theory of meaning for a natural language to which philosophers of the time aspired, at least in Oxford in the mid-1970s, in the throes of the then reverberating 'Davidsonic boom'. This conception is what I punningly dubbed the 'governing view'—crudely, that understanding a natural language is, through and through, a rule-governed competence. The idea, crudely, was that we are able to parse a novel sentence by (in some sense) working out the conjoint implications of the rules of syntax relevant to its mode of construction and the semantical rules governing its occurrent primitive expressions. My first two chapters4 elaborate and critique that thought, and indeed my efforts to refine it resurface in several places in this volume. But at that time I attempted no specific resolution of the paradox other than, in this way, to try to undercut one kind of motivation for (one form of) its major premise.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.