Proper Names, Contingency A Priori and Necessity A Posteriori

Abstract

After a brief review of the notions of necessity and a priority, this paper scrutinizes Kripke's arguments for supposedly contingent a priori propositions and necessary a posteriori propositions involving proper names, and reaches a negative conclusion, i.e. there are no such propositions, or at least the propositions Kripke gives as examples are not such propositions. All of us, including Kripke himself, still have to face the old question raised by Hume, i.e. how can we justify the necessity and universality of general statements on the basis of sensory or empirical evidence?

Acknowledgements

My profound thanks go to Professor Susan Haack, who read the manuscript of my paper two times, offered her help to revise my English, and gave me valuable comments and advices. My thanks also go to two anonymous referees of the journal for valuable comments on earlier draft of this paper. Needless to say, all possible mistakes are mine.

Notes

¹Kripke also mentions certainty and uncertainty, which belong to epistemology too, but does not focus his attention on them (see Kripke 1980, 39).

²This paper will not distinguish between proposition and statement. Also, In order to escape unnecessary complication, it will not concern problems about empty proper names. When it talks about a name, the name must be assumed to have a referent, that is, not to be empty; when it talks an object or a thing, the object or thing must be assumed to be real or actual, that is, not to be fictional. So, this paper will not use sentences such as ‘if N exist, then N is the F’; instead, it simply use sentences like ‘N is the F’.

³In Donnellan's terminology, in D2, the description ‘the length of S at time t ₀’ has an attributive use, while ‘one meter’ keeps to designate the length of stick S at time t ₀, no matter what length it is; in D4, the description ‘the length of stick S at time t ₀’ has a referential use, while ‘one meter’ denotes the length of stick S at a certain time and under a specific circumstance. See Donnellan, 1972, 356–379.

⁴I maintain that modal logic does not require Kripke's notion of rigid designator, conversely the latter is a philosophical extension of the former. I have four reasons for this: (i) modal logic and its semantics have been established long before Kripke gave his notion of rigid designator. (ii) I myself have argued that all talk about rigid designators can be reduced to talk about actual objects, which has nothing to do with how these objects are named. See Chen Bo: ‘Refuting Kripke's and Soames's Arguments against Descriptivism’, submitted, THEORIA (iii) I consent to Hintikka's claim:

More generally, it is important to realize that if the criteria of cross-identification are specified, quantification into modal and intensional contexts makes perfect sense completely independently of what one may think of names and other singular terms, including their relation to the individuals they stand for. There is no need to assume any particular class of ‘rigid designators’. If a singular term ‘b’ is a ‘rigid designator’ as far as the given class of possible worlds is concerned, this can be expressed in the language by means of quantifiers as

• (8) (∃ x)(b=x) (Hintikka and Sandu 1995, 251)

(iv) Arif Ahmed also argues that even if we do not mention rigid designators, but only mention the relevant objects themselves, we can still talk about those matters about which rigid designators were used to talk (Ahmed, 2007, 26–28).

⁵The original number is kept in order to let readers understand the citation.

⁶Chen Bo: ‘Refuting Kripke's and Soames's Arguments against Descriptivism’, forthcoming.

⁷For example, Hume says, ‘[t]his therefore is the essence of necessity. Upon the whole, necessity is something that exists in the mind, not in objects, nor is it possible for us ever to form the most distant idea of it, considered as a quality in bodies.’(Hume 2000, 1.3.14.23)

⁸It is interesting that Woleñski distinguishes absolute and relative necessity, and also absolute and relative apriority. According to these distinctions, he asserts that ‘Hesperus is Hesperus’ is absolutely necessary by logic (logical truths and their instantiations are always necessary), but that ‘Hesperus is Phosphorus’ is only relatively via additional semantic referential clauses (Woleński 2004, 824–825). It is said that Kripke's view that necessity does not imply apriority becomes plausible if we distinguish relative and absolute apriority. Since this paper has been quite long, I have no further space to consider the possibility carefully. Perhaps I will do it in another paper.