Online research in philosophy

It is hard to exaggerate the extent to which Kripke’s work has changed the way we do philosophy. He has been compared to Socrates and touted as “arguably the world’s greatest living philosopher.” 1 The London Review of Books proclaims that “philosophers are busily rewriting all of semantics (and a good deal of epistemology) in Kripkean terms.” Just what are these terms? The central semantic notion is that of rigid designation, by now a household name. Kripke (Kripke 1972 (henceforth NN)) says that a rigid designator is something that designates the same object in every possible world where it designates at all. Kripke himself has been reluctant to give a theoretical account of rigid designation, preferring instead to talk in terms of a ‘better picture’ which comports with intuition but it seems reasonably clear that Kripke intends this to be a semantic property that some words have (like names and natural kind terms) and that others lack (like definite descriptions). This ‘better picture’ has been developed rigorously by Michael Devitt into a theory (Devitt 1974; Devitt 1981a; Devitt and Sterelny 1999). In the meantime attention has shifted from a general discussion of what rigid designation is to the question of whether and how rigid designation works for natural kind terms like ‘gold,’ and ‘tiger’ (Devitt forthcoming).

What is it for a predicate or a general term to be a rigid designator? Two strategies for answering this question can be found in the literature, but both run into severe difficulties. In this paper, it is suggested that proper names and the usual examples of rigid predicates share a semantic feature which does the theoretical work usually attributed to rigidity. This feature cannot be equated with rigidity, but in the case of singular terms this feature entails their rigidity, as understood in the standard characterisation. Hence, it is appropriate to call this feature proto-rigidity.

Kripke defines a rigid designator as one that designates the same object in every possible world in which that object exists. He argues that proper names are rigid. So also, he claims, are various natural kind terms. But we wonder how they could be. These terms are general and it is not obvious that they designate at all. It has been proposed that these kind terms rigidly designate abstract objects. This proposal has been criticized because all terms then seem to come out rigid, thus trivializing rigidity. The paper starts with further criticisms of this proposal aimed particularly at a recent version given by LaPorte. The paper goes on to develop and defend an alternative proposal presented briefly in Devitt and Sterelny (1999): instead of taking those natural kind terms to rigidly designate an object we take them to rigidly apply to the members of their extensions. Schwartz has rightly insisted that a notion of rigidity must do some theoretical work if it is to be interesting. The paper argues that rigid application for kind terms does the same primary work as rigid designation for singular terms, the work of refuting description theories of some terms; and it does the same secondary work, of explaining certain modal phenomena (with one exception).

The article discusses an idea of how to extend the notion of rigidity to predicates, namely the idea that predicates stand in a certain systematic semantic relation to properties, such that this relation may hold rigidly or nonrigidly. The relation (which I call signification) can be characterised by recourse to canonical property designators which are derived from predicates (or general terms) by means of nominalization: a predicate signifies that property which the derived property designator designates. Whether signification divides into rigid and non-rigid cases will then depend uponwhether canonical property designators divide into rigid and non-rigid ones. But, I shall argue, they do not, and so the only notion of rigidity gained this way is trivial. To show this, I first focus on the kind of canonical property designators which could be thought to be nonrigid, canonical designators such as having the colour of ripe tomatoes which themselves contain non-rigid property designators. An argument to the effect that such complex canonical designators are non-rigid is rebutted, five arguments to the effect that they are rigid are formulated, and finally an explanation of their rigidity based on the general nature of canonical property designators is presented.

There is a considerable sub-literature, stretching back over 35 years, addressed to the question: Precisely which general terms ought to be classified as rigid designators? More fundamentally: What should we take the criterion for rigidity to be, for general terms? The aim of this paper is to give new grounds for the old view that if a general term designates the same kind in all possible worlds, then it should be classified as a rigid designator. The new grounds in question have to do with excavating the connection between rigid designation and semantic structure. Other original contributions of the present work consist in developing responses to some objections to this approach to rigid designation.

What does it mean for a general term to be rigid? It is argued by some that if we take general terms to designate their extensions, then almost no empirical general term will turn out to be rigid; and if we take them to designate some abstract entity, such as a kind, then it turns out that almost all general terms will be rigid. Various authors who pursue this line of reasoning have attempted to capture Kripke’s intent by defining a rigid general term as one that applies to the objects in its extension essentially. I argue that this account is significantly mistaken for various reasons: it conflates a metaphysical notion (essentialism) with a semantic one (rigidity); it fails to countenance the fact that any term can be introduced into a language by stipulating that it be a rigid designator; it limits the extension of rigid terms so much that terms such as ‘meter’, ‘rectangle’, ‘truth’, etc. do not turn out to be rigid, when they obviously are; and it wrongly concentrates on the predicative use of a general term in applying a certain test offered by Kripke to determine whether a term is rigid.

The notion of a rigid designator was originally introduced with respect to a modal semantics in which only one world, the world of evaluation, is shifted. Several philosophical applications employ a modal semantics which shifts not just the world of evaluation, but also the world considered as actual. How should the notion of a rigid designator be generalized in this setting? In this note, I show that there are two options and argue that, for the currently most popular application of two-dimensional modal semantics, proper names ought to be treated as rigid relative to the world considered as actual.