Brandom and fundamental normativism

normativism and Brandom’s (1998) view. Let me therefore elaborate on particular similarities and differences between the two.

hypothetical. One of my reasons for this was that it seems to better capture the sense in which it can be entirely appropriate to violate the relevant norms in the context of a conversation in which one wants, e.g., to speak in metaphor or tell a joke. These are situations in which by saying something that one ought not say one can convey a message to one’s listeners different from what one would convey if adhering strictly to one’s position in entitlement-commitment relations. In doing so one is not doing something wrong but something warranted by the pragmatics of the conversation.

Another reason why I do not follow Brandom in this regard is that I conceive of the implicit normative structure of discursive practices somewhat differently than he. Brandom’s idea, which he draws from Kant and Sellars, is that

normative vocabulary (of which ‘ought’ and ‘should’ are paradigmatic) has the logical expressive function of making explicit in the form of something that can be said (put in the form of a claim) an attitude that otherwise could be implicit only in what is done—namely, the endorsement of a pattern of practical reasoning. (1998, p. 271)

To me, this idea of an explicitating logical function of normative vocabulary as allowing us to say what we do such that we can qualify as speakers to begin with (cf. 2001, p. 89) is too narrow. To see why, consider Glüer and Wikforss’s point that any categorization of things into As and non-As can be used to derive normative consequences if a suitable norm is in force; e.g., by being socially imposed. I take it that normative vocabulary is not just a meta-vocabulary for explicitating normative (reasons-providing) relations implicit in practice. It also (perhaps more importantly) has the pragmatic declarative function of imposing in the form of something that can be said normative (reasons-providing) relations not already implicit in practice, and hence not something the vocabulary is making or can make explicit.

Searle (1995) presents several examples of this sort for social and institutional kind concepts. For instance, a speaker might claim that the

speaker might not be making explicit normative relations implicit in applications of the concepts ‘president’ and ‘commander in chief’ but to make it the case that his audience accepts the commitment-entailment suggested, in which case “X is president” and “X is commander in chief”

will be commitment-preserving. Since, in this case, the concepts were not inferentially related in accordance with norms implicit in their use prior to the declaration, the speaker did not make any such norm explicit. What he did, if successful, was to impose a new commitment-entailment from

‘president’ to ‘commander in chief’ such that their contents are now inferentially related. If, now, someone is demanded a reason for saying “X is commander in chief” he may point out that he is entitled to “X is president”, from which “X is commander in chief’ can be entitlement-derived. What all this means is that normative vocabulary is not in the business merely of explicitating implicit normative relations in use, but also in the business of imposing, in the form of declarations, what normative relations should be in force.

Expanding the pragmatic function of normative vocabulary in this manner is not, as far as I can see, in any way incompatible with Brandom’s view. Nor, of course, does it imply the falsity of his statement in the quoted passage. It does, however, reach out to the anti-normativist suspicion that what we are really saying when we say that one ought or may (not) so-and-so is that normative relations have been declared to be in force. I also-and-so believe that recognizing this richer pragmatic function of normative vocabulary supports Brandom’s general project by implying that the material inferential relations, in which expressions are embedded and that bestow them with content, are essentially contestable.

To see what I mean, consider Brandom’s example (2001, 69-70) of the concept ‘Boche’ (borrowed from Dummett 1973). This example is meant to capture a process of conceptual change and its implication for commitment-entitlement relations. ‘Boche’ is introduced as entitlement-entailed by ‘German’. Thus, if someone is of German nationality one is permitted to say that he is Boche. However, ‘Boche’ also entitlement-entails ‘barbarous’. So, if one is permitted to say that someone is Boche

inference from ‘X is German’ to ‘X is barbarous’ is entitlement-preserving.

Denying that there are Boches is to deny that there are Germans; admitting that there are Boches is to accept that Germans are barbarous. Thus, if one refuses the implication from German nationality to barbarousness then one has to refuse to use ‘Boche’ because of the implicit material commitment-entitlement inferential relations its introduction into one’s vocabulary imply.⁸ Concerning the explicitating function of normative vocabulary, Brandom writes:

The proper question to ask in evaluating the introduction and evolution of a concept is not whether the inference embodied is one that is already endorsed, so that no new content is really involved, but rather whether that inference is one that ought to be endorsed. The problem with ‘Boche’

and ‘nigger’ is not that once we explicitly confront the material inferential commitment that gives the term its content it turns out to be novel, but that it can then be seen to be indefensible—a commitment we cannot become entitled to. We want to be aware of the inferential commitments our concepts involve, to be able to make them explicit, and to be able to justify them. (2001, pp. 71-2)

It is precisely in the context of discussions like these, I argue, that the logical expressive function of normative vocabulary, emphasized by anti-normativists, of declaring the appropriateness of certain material inferential relations as being in force should be recognized. The idea is that many concepts might not commitment-or-entitlement-entail material inferences to certain contents but be declared to do so by means of declaratives like “X is to count as Y” (“Germans are to count as barbarous”) from which it would follow that a consequence of application of ‘X’ obliges an inference to ‘Y’ even if that inference is not already implicit in ‘X’. This means that there may be several concepts whose contents are not determined by material inferential relations in which they are implicitly embedded but whose contents are rather socially imposed. And the importance of this                                                                                                                

8 Similar concepts are ‘nigger’, ‘whore’, ‘faggot’, ‘communist’, etc. The point is that these concepts are caught up in material inferential relations whose circumstances of application are descriptive (e.g., ‘German’) but whose

point is that it might actually not be the case that one can derive from the normative inferential relations in which concepts stand that one ought apply them. If the ‘oughts’ are imposed then they are hypothetical, such that it is only conditional on one’s accepting the material inferences they commit one to that one ought to apply them.

Thus, I agree with Brandom that semantic content conceptually presupposes responsiveness to commitment-entitlement relations. But I disagree with the view that these relations involve categorical obligations for concept application because, first, in some pragmatic contexts, e.g., metaphor, it might actually be precisely by not saying what one ought that the meaning of what one says is determined and, secondly, in many cases commitment-entitlement relations are socially imposed and therefore not such that one ought, in the categorical sense, conform to the norm. Oughts are rather, I’d like to say, essentially contestable. By this I mean that although all conceptual content conceptually presupposes entitlement relation, there is no concept embedded in commitment-entitlement relations that obliges one to certain inferences. Fundamental normativism, in this context, can be expressed as the view that all semantic content conceptually presupposes implicit normative relations of commitment-entitlement in use, and no commitments-entitlement relations are such that one ought categorically to conform with them because all such relations are essentially contestable (recall Brandom’s own example with ‘Boche’).

As I see it, content and meaning is fundamentally normative in that concept possession and application is inevitably a matter of bowing to or turning against, accepting or rejecting—essentially, taking a stance on—

content-constitutive normative relations. Possessing a concept is (implicitly) taking a normative stance in a field of normative material inferential relations; and concept use is a (implicitly) normative action in that field, of committing to and entitling others to material inferences between contents of concepts used. Sometimes normative relations not implicit in possession or use are not evolved through (fieldwork) normative action but imposed from the sidelines, as it were, in attempts to change the

and use as essentially normative stances and actions in a field of material commitment-entitlement inferential relations, where some inferential relations are imposed, shows is that all oughts in play are contingent.

Playing the game is a necessary and sufficient condition for counting as having and using concepts but one cannot directly derive categorical oughts from playing the game because oughts are contestable.

I promised earlier to discuss Glüer and Wikforss’s criticism of Brandom’s understanding of content as normative. Defending his view, with the material added by fundamental normativism, against their criticism is my aim in what remains.