Author Information: Michael P. Wolf, Washington and Jefferson College, firstname.lastname@example.org
Wolf, Michael P. 2012. “Rigid Designation and Natural Kind Terms, Pittsburgh Style.” Social Epistemology Review and Reply Collective 2 (1): 133-142.
This paper addresses recent literature on rigid designation and natural kind terms that draws on the inferentialist approaches of Sellars and Brandom, among others. Much of the orthodox literature on rigidity may be seen as appealing, more or less explicitly, to a semantic form of “the given” in Sellars’s terms. However, the important insights of that literature may be reconstructed and articulated in terms more congenial to the Pittsburgh school of normative functionalism.
I. The issues
Reflection on Frege’s (1892/1980) account of sense led many analytic philosophers of language to view proper names and other apparently simple singular terms as covert clusters of descriptions. Thus, a proper name such as “Aristotle” had as its meaning some set of predicates and descriptions like “the author of Nicomachean Ethics, the teacher of Alexander the Great, the husband of Pythias…” The name (or other singular term) then refers to that object which satisfies all of the descriptions, or some suitable subset of them. Kripke (1980) criticizes this approach extensively, arguing that the meanings of proper names and natural kind terms could not be understood in such terms. While such descriptions might be associated with proper names and natural kind terms, those associations did not amount to synonymy. We might imagine other possible worlds — or discover in the actual world — that some or all of the descriptions in such a cluster were not true of the expression’s referent. We might also imagine such instances where something else better fit the descriptions. If Aristotle forged the credentials listed above, he does not cease to be Aristotle; if someone else did all of those things, he does not thereby become Aristotle. The proper name refers to that person and only that person, whether the associated descriptions are true of him or not.
The notion was initially introduced in modal terms by Kripke (1980, 48): “Let’s call something a rigid designator if in every possible world it designates the same object,” while non-rigid designators would not do so. Such a designator may fail to designate at all (i.e., in some possible worlds, Aristotle never exists), but it does not designate something else if that thing happens to have other properties. Nor does something else (say, Aristotle’s brother) turn out to be Aristotle if it meets more of the descriptions in the cluster. When we use the proper name “Aristotle,” its reference is fixed for us in the actual world, and that reference remains fixed no matter what we may discover about this world or stipulate about others. Putnam (1973, 1975b) contemporaneously  extended many of these ideas to natural kind terms, while others have extended them to include still more classes of expressions such as indexicals and pronouns. 
Why should any of this trouble a normative functionalist? Many defenders of rigid designation’s place in a theory of meaning have stated its nature in terms of the “object-involving” character of truth conditions for sentences in which they appear. That is, the object itself would figure in the truth conditions, rather than some means — functions, inference licenses, possible worlds, whatever the tools of the theory may be — for reidentifying and differentiating between potential referents. On a first pass, this would seem to involve an appeal to non-semantic items as elements of the semantic order; a more sophisticated reading would appear to involve appeal to a direct, unmediated word-world relation as an explainer in a theory of meaning. In Sellarsian dialects of Pittsburghese, this would amount to invoking the given in semantic form. Outside those circles, most analytic philosophers after Kripke would say that inferentialist accounts seem to preclude an appropriately direct involvement of objects in their semantic contents. Putnam and Sellars spoke directly to one another on this point, in fact. In 1974 at the APA Eastern meetings, Sellars presented his “Meaning as Functional Classification,” a touchstone of normative functionalism in recent theory of meaning, and comments were offered by none other than Hilary Putnam. Sellars’s views are discussed extensively elsewhere in this edition, but Putnam expressed significant reservations:
The normal-form description of the meaning of a word does include a description of the rules for the use of the word. … The “rule” component gives the full specification of individual competence, or even of collective social competence at a given time; it fails to completely capture what is ordinarily called the meaning of the word and in particular it fails to capture the extension. The battery of rules, whether you take that in an individual or a social sense, does not determine the extension of the word. … The determination of extension is a social matter, and also more than a social matter; it depends also on the contribution of the environment. (1974, 453-454).