The contingent a priori: has it anything to do with indexicals?

Can some contingent truths be known a priori?: when this question is raised in modern philosophy — as, following Kripke, it often has been — it generally introduces a discussion of certain examples which seem to turn on indexical or indexical-like words (Nathan Salmon lists most of the important work on the topic at pp. 77-8 of Reference and Essence, Blackwell 1982; I have also been helped by Graeme ForbesY typescript 'States of Affairs, Bedeutung and the Contingent A Priori1, 1985, and subsequent correspondence). Sometimes the indexicality is quite obvious, as in 'I am here now', sometimes it appears only on analysis, as in 'If anyone uniquely invented the zip, Julius did', where by stipulation 'Julius' rigidly designates the inventor of the zip (if there is one), for the rigidity of 'Julius' can be expressed by means of the 'actually' operator, which is at least formally analogous to an indexical (cf. Martin DaviesMeaning, Quantification, Neceessity,Routledge 1981, pp. 230-8 on the similar metre rod example). Moreover, this indexicality has been exploited in a well known argument that what is known a priori can be contingent only in a superficial way (Gareth Evans 'Reference and Contingency' Monist 62 (1979) pp. 161-89; technical details are omitted, since deep and superficial contingency coincide for sentences, such as those considered below, without 'actually' or similar operators). The purpose of this note is to argue that there are deeply contingent a priori truths which in no way turn on indexicals. I do not deny that there are also contingent a priori truths which do turn on indexicals, but we are unlikely to get clear on the contingent a priori in-general if we take it to be essentially tied up with indexicals when in fact it is not