Academic journal article Linguistic and Philosophical Investigations

Sainsbury and Tye Fail to Solve Frege's Puzzle

Academic journal article Linguistic and Philosophical Investigations

Sainsbury and Tye Fail to Solve Frege's Puzzle

Article excerpt

"Discoveries happen in math too, and Fregeanism is really about discovery."

-Jason Stanley1

1. Introduction

Sainsbury and Tye (2012) argue for their Originalist theory of concepts on the basis that it solves seven well-known puzzles. These puzzles are traditionally seen as puzzles about language, and in particular puzzles concerning semantic content. Sainsbury and Tye argue that they have analogues as applied to the content of thoughts. In this paper I argue that their solution of Frege's Identity Puzzle (Frege 1892) is implausible, because Originalism fails to explain the difference in cognitive significance which Frege took to be the puzzling explanandum. In the next section I briefly review Frege's Puzzle before explaining the putative solution of Sainsbury and Tye in Section 3. The bulk of the paper is Section 4, where I lay out in detail the reason for finding the Originalist solution unsatisfying. In concluding I briefly put the project of this paper in perspective of Sainsbury and Tye's book, and point out a take-home message of the book which remains even if Originalism fails.

2. Frege's Puzzle

Although Frege's puzzle of identity (1892) is familiar, a brief reminder may be important for understanding where Sainsbury and Tye go wrong in their solution. True identity statements of the form 'a = a' are trivial and uninformative, yet true identity statements of the form 'a = b' (where 'b' is a name which refers to the same object as 'a') can be significant and informative. The names 'Phosphorous' and 'Hesperus' were names given to the planet Venus by the Ancient Babylonians, who thought incorrectly that the astronomical bodies which they saw in the morning sky and in the evening sky were two different objects (at least according to philosophical legend).2 This results in the following statements for illustration:

(1) Hesperus is Hesperus.

(2) Hesperus is Phosphorous.

If the meaning of a name is nothing over and above its referent,3 then (1) and (2) have the same meaning, since the names 'Hesperus' and 'Phosphorous' designate the same object. However, (1) and (2) "have different cognitive values," as Frege puts it (p. 78). For instance, someone could believe (1) but not believe (2), as presumably the Babylonians did. Furthermore, (1) is trivial in the sense that it doesn't add to our knowledge, but (2) was an important discovery about the world. Frege's solution was to divide semantic content (which he called "thoughts") into two parts: Besides the referent ("bedeutung") of a name or other expression, there is also the sense ("sinn"), "wherein the mode of presentation [of the referent] is contained" (p. 57).4

Sainsbury and Tye note that this puzzle about sentences and their constituent names has an analogue for thoughts and their constituent concepts (as Fodor and others have previously realized). Putting the puzzle in these terms, they sum it up thus (p. 4):

The concept HESPERUS would seem to represent just what the concept PHOSPHOROUS represents. So how can there be a difference between the thought that Hesperus is Hesperus (trivial) and the thought that Hesperus is Phosphorous (an important empirical discovery)?

3. The Originalist Solution

Sainsbury and Tye present some arguments against Frege's solution to the puzzle (22-26), most of which are familiar from Kripke (1980). For instance, one argument is that differences in knowledge concerning e.g. Aristotle would result in idiosyncrasies of sense associated with the name 'Aristotle,' but we all presumably share the same concept ARISTOTLE when we think thoughts about him (Sainsbury & Tye 2012, p. 17, 24). Since these arguments are well-known and I am sympathetic to them, I won't rehearse them here, but instead turn the focus to their positive view. The key Fregean principle which they reject is that cognitive differences such as that between (1) and (2) are to be explained by a difference in semantic content (p. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.