Sortal Predicates
Feldman (Fred)
Source: Nous, 7.3, Sept. 1973, pp. 268-282
Paper - Abstract

Paper SummaryNotes Citing this PaperText Colour-Conventions

Philosophers Index Abstract

  1. John Wallace, David Wiggins, Robert Ackermann and others have appealed to alleged features of sortal1 predicates in their attempts to solve philosophical puzzles about identity, quantification, and lawlikeness. They have offered various criteria of sortalhood2.
  2. I present twelve such criteria, and attempt to show that on some, almost every predicate is a sortal3, and on others almost none is.
  3. No two of the remaining criteria are equivalent.
  4. My conclusion is that we cannot evaluate any substantive thesis stated in terms of sortals4 until we have a clearer notion of what a sortal5 is supposed to be.

Author’s Introduction
  1. A number of philosophers have claimed that an important distinction may be drawn between predicates that are "sortals6" and predicates that are "non-sortals7". In recent years, John Wallace, David Wiggins, and Robert Ackermann, among others, have written specifically on this distinction. It has also been claimed that important doctrines in Aristotle, Frege, Strawson, and Quine turn on the notion of the "sortal8 predicate". (See ["Wallace (John R.) - Sortal Predicates and Quantification"], p. 9 and ["Wiggins (David) - Identity & Spatio-temporal Continuity"], p. 28.) Furthermore, the closely related concept of the "substantival general term" in Geach is believed by some to be of considerable philosophical significance (see his Reference and Generality).
  2. Wallace (p. 12) claims that an interesting and novel form of quantification theory may be obtained by adding a position for sortals9 into the quantifiers. The suggestion is that this form of quantification theory has important advantages over the standard form. Wallace generously suggests that Geach presented the rudiments of this "sortalized10" quantification theory in Reference and Generality.
  3. Wiggins suggests that sortals11 play a decisive role in the logic of identity. He apparently accepts the thesis that every identity statement stands in radical need of supplementation by a sortal12. "If anyone tells you a = b, you should always ask them 'the same what as b ?'" (p. 1). This thesis also seems to echo a Geachean doctrine.
  4. Ackermann claims that sortals13 "have important but hitherto unremarked consequences for the notion of lawlikeness" (["Ackermann (Robert) - Sortal Predicates and Confirmation"], p. 2). He goes on to suggest, among other things, that the raven paradox may be dealt with by requiring that the antecedent predicate in any instance of a law be a sortal14. 'Non-black thing' is not a sortal15, and so 'all non-black things are non-ravens' is not a law and so is not confirmed by observations of non-black non-ravens.
  5. In light of the fact that it may have all these applications, the concept of the "sortal16 predicate" is clearly of philosophical importance. However, 'sortal17' is not an expression in ordinary, non-technical use. To determine what it means, we must rely on the criteria of sortalhood18 proposed and suggested by the philosophers who make use of it. These criteria fall into three main categories:
    • I. Counting Criteria: These are based on the idea that only a sortal19 predicate can individuate things in such a way as to make counting possible. Wallace puts this view by saying that sortal20 predicates "provide a criterion for counting" the things to which they truly apply (p. 9). He considers this idea to be part of the "traditional wisdom" about sortals21. Ackermann and Strawson seem to accept something like this view, too.
    • II. Mereological Criteria: These criteria are based on the idea, roughly, that if a sortal22 predicate truly applies to a thing, then it does not also apply to the parts of the thing. Ackermann, Wallace, and perhaps Frege and Quine have suggested mereological criteria for sortalhood23.
    • III. Essence Criteria: Wiggins claims that "Strawson's notion of a sortal24 predicate descends directly from Aristotle's notion of second substance" (p. 28). Wiggins evidently feels that his own use of 'sortal25' is in the same tradition (p. 65, n. 2). A sortal26 predicate, on this view, is one that gives a suitably "substantial" answer to a question of the form 'what is x?' A sortal27 expresses the "nature" or "essence" of the things to which it truly applies.
  6. My purpose in the present paper is to show that the suggested criteria of sortalhood28 are non-equivalent. What is a sortal29 on one criterion is not a sortal30 on others. Since 'sortal31' is a technical term, and our only access to its meaning is through the criteria suggested by these philosophers, we are in the unfortunate situation of not being able to tell precisely what it means. It may even be the case that the word 'sortal32' is being used in different ways by different philosophers and thus has come to express a number of distinct concepts ambiguously. Until this confusion is cleared up, there is little point in trying to evaluate any substantive thesis stated in terms of sortals33.

Text Colour Conventions (see disclaimer)

  1. Blue: Text by me; © Theo Todman, 2018
  2. Mauve: Text by correspondent(s) or other author(s); © the author(s)

© Theo Todman, June 2007 - May 2018. Please address any comments on this page to File output:
Website Maintenance Dashboard
Return to Top of this Page Return to Theo Todman's Philosophy Page Return to Theo Todman's Home Page