Incommensurability: truth or consequences.

• Author: Leiter, Brian
• Position: Response to articles by Eric A.Posner and Frederick Schauer in this issue, p. 1185, 1215 - Symposium Comments: Law and Incommensurability

Frederick Schauer wants to bypass the question of whether the "incommensurability thesis" is true.(1) His question is whether it would be good to believe the thesis along some nonepistemic dimension (such as whether it is utility-maximizing). Eric Posner thinks we can explain why people affirm belief in incommensurability, quite apart from the truth of the thesis.(2) In this sense, both of their articles might just as well have appeared in a symposium on "Norms for Belief" or "Explanations for Normativity," as in one on "Law and Incommensurability." The claims that human beings possess free will, that judges are bound to decide according to law and have little or no room for the exercise of discretion, or that creationism is as much "science" as evolution, are all claims that might be addressed in the spirit of Posner and Schauer. We might, of course, ask, "Are these claims true?"--the standard philosophical question--or we could ask, a la Posner and Schauer, "Is it or would it be useful or advantageous to believe these claims?" What we might call the "truth-norm"--the norm that we should only believe what is true--is ordinarily the dominant norm in academic discourse. Even indirect utilitarians like John Stuart Mill and Henry Sidgwick accept, at bottom, the truth-norm: It is only because utilitarianism is truly the correct normative theory that, as a merely instrumental matter, they think it best if most people conducted their affairs by reference to nonutilitarian norms. By contrast, Schauer thinks that we can, or should, dispense with the truth-norm in thinking about incommensurability, and Posner thinks that we can explain belief in incommensurability without regard to its truth.

1. SCHAUER, SEMANTICS, AND TRUTH

   Let us call "Epistemic Agnosticism" the view that norms for belief need not make reference to the truth-bearing features of propositions. Thus, we might say that one ought to believe certain propositions simply because doing so maximizes good consequences or because doing so satisfies a nonconsequential demand of reason. Let us call "Fictionalism" the view that one ought to believe certain false propositions because doing so maximizes good consequences or because doing so satisfies a nonconsequential demand of reason. Fictionalism is just a subset of Epistemic Agnosticism, though one that highlights the radical character of the position: For by this line of reasoning, even the falsity of a proposition is not an obstacle to the justification of belief.

   I do not mean to impugn Fictionalism because of its radicalism. I am Nietzschean enough to recognize both that truth is a genuine property of propositions and that the value of believing true propositions is an entirely different question.(3) I do think, however, that once we enter this realm of debate, we leave entirely behind philosophical questions distinctive of incommensurability, and we enter the realm of epistemology and speculative empirical social science.(4)

   Schauer's article is clearer than Posner's about its irrelevance to the truth of the incommensurability thesis, and this is a virtue. Schauer speaks, even in the title, of "Instrumental Commensurability," and he is at pains to emphasize that "there is a normormative fact of the matter" about whether or not values are incommensurable.(5) His fundamental point, however, is one about epistemology (what we ought to believe), not semantics (what words mean), and he only confuses issues, I think, by bringing on board certain baggage from the philosophy of language.

   There is, first of all, a certain muddying of the philosophical waters in suggesting that Charles Stevenson's notion that certain predicates have "emotive meaning"(6) and Philippa Foot's notion that certain predicates are "thick concepts"(7) (involving an inextricable bonding of the descriptive and evaluative)(8) show "that one need not reject the fact-value distinction nor subscribe to the central tenets of continental philosophy in order to recognize that descriptive sentences containing seemingly descriptive words arrayed in a seemingly descriptive semantic structure often mask statements and conclusions that are in important ways normative, evaluative, and prescriptive."(9) It is certainly correct to observe that a central theme of twentieth-century philosophy has been that syntax, or grammatical form, is a poor guide to semantics, or meaning. And it is also surely correct that semantics can be parsed from one's metaphysics of facts and values. But it is misleading to embed this point in a discussion of Stevenson and Foot. For Foot does reject the fact-value distinction, and what is more, she thinks her point about the semantics of "thick" concepts undergirds that rejection.(10) Similarly, Stevenson thinks (like R.M. Hare(11) after him) that the fact that one can parse, at the level of meaning, the descriptive from the evaluative shows precisely that the fact-value distinction is a viable one after all.(12) Schauer cautions that there are "important differences" between...