To begin with, it is obvious that normativity is a trivial consequence of conceptual content. If the concept x is the concept WATER, then the concept x is correctly applied, and ought to be applied, to all and only instances of water: the notion of the extension of a concept just is the notion of what it is correct to apply the concept to. Now, it can be asked whether such a trivial truth requires an explanation in terms of an underlying nature of normativity. For example, P. Horwich (Meaning, Oxford: OUP, 1998) sustains a deflationary answer: there is nothing what the referential power of a concept consists in and thus there is nothing what its normativity consists in. On the contrary, a realist thesis of concept-application entails a robust notion of normativity: the natural kind whose instances define the extension of a concept provides the norm according to which the concept-application can be evaluated as correct. The underlying nature of normativity is the natural kind referred to by a concept. It seems that the realist does not distinguish the trivial truth from the robust notion if she says: the realist thesis about concept-application is that the standard of correctness for each concept is precisely the kind of things it keeps track of; in other words, a concept is applied correctly to the kind of things it is the concept of. Here the fault is in “in other words”. The second claim is merely an expression of the trivial truth: that a concept is applied correctly to the kind of things it is the concept of does not entail that such a kind of things is the norm according to which the concept-application is correct. Only the first claim is the one with that import. E. Lalumera (“A Simple Realist Account of the Normativity of Concepts”, Disputatio 1/19 (2005), pp. 205-221) says that a simple realist thesis about concept-application is “that a concept is applied correctly when it is applied to the kind of things it is the concept of” (p. 207). But she also says in a robust style: “Concepts, as trackers of natural kinds, are norm-governed in that they are subject to standards of correctness provided by the kinds themselves” (p. 219).
The normativity requirement of the realist thesis would be something like the following. If x is the concept WATER, then the natural kind water governs the concept-application: x correctly applies to a thing (if and only if the thing is water) because of the thing’s being water. Now, from my point of view a theory of concepts has to be a theory of concept-possession: the theory has to specify the possession-facts constitutive of having a particular concept. Lalumera introduces the Kripke’s paradox (Wittgenstein on Rules and Private Language, Cambridge Mass.: Harvard UP, 1982) but she does not take account of that kripkean scepticism starts from a concept-possession premise. Usually, the premise is interpreted: if x is the concept WATER, then there are possession-facts that govern the concept-application, that is to say, there are possession-facts that constitute a standard of correctness for the concept-application. For example, a fact-possession could be a disposition to apply the concept. It is known that the sceptic establishes that a disposition cannot constitute a standard of correctness: to be disposed to apply a concept to instances of water implies at most that one will apply, not that one should. From this the sceptic would conclude that there are not facts constitutive of normativity. But it would seem that thus constructed kripkean scepticism fails to get grips with the realist thesis, where the facts constitutive of normativity are not possession-facts like a disposition but they are natural kind-facts: the actuality of the instances of a natural kind. The realist facts constitutive of normativity are left untouched.
However, kripkean scepticism can be formulated (G. Wilson, “Kripke on Wittgenstein and Normativity”; “Semantic Realism and Kripke’s Wittgenstein”) so that there are not facts constitutive of normativity at all. Assume the normativity requirement of the realist thesis and a concept-possession thesis. Then, the concept-possession thesis entails the following requirement: if x is the concept WATER, then there are possession-facts that constitute the natural kind water as the condition that governs the concept-application (in the terms specified in the normativity requirement: x correctly applies to a thing because of the thing’s being water). It introduces a new reading of the sceptical point about dispositions: that to be disposed to apply a concept to instances of water does not imply that one should apply means that one will apply the concept not because of their being instances of water. Imagine a Twin Earth context where a disposition to apply a concept to instances of water is cogently a disposition to apply the concept to instances of twater, the twin liquid: it is obvious that one will apply the concept to instances of twater only because of their being like the water, so that one will apply the concept to instances of water only because of their being like the water. Thus, the sceptic concludes that there are not possession-facts constitutive of a natural kind as a condition that governs a concept. It follows that there are not natural kind-facts constitutive of normativity.
A justification requirement is the requirement that norms must guide or justify the concept-application. It seems to be a truism. Any normativity requirement is itself a justification requirement: something is the condition that governs (guides or justifies) a concept-application if and only if it is the condition why the concept is correctly applied. Thus, if a natural kind provides the norm according to which a concept-application is right or wrong, such a natural kind guides or justifies the concept-application. The objection examined by a realist would be that kinds cannot justify the application of concepts because the extension of a natural kind is generally beyond the one’s knowledge. Lalumera does not discuss the explicit premise in the objection: that the extension of a natural kind is generally beyond the one’s knowledge. She does discuss the implicit premise that justification implies epistemic access, the Accessibility Condition: some content c constitutes a subject’s justification for believing p if and only if c is conceptually accessible to the subject (p. 211). Her idea is that the application of a concept can be said to be justified by facts unknown or unknowable to one under an externalist (or reliabilist) account of justification. Then, given the explicit premise the realist thesis results in the following picture: “on the hypothesis that concepts track kinds, correctness of concept-application is an objective matter, independently on one’s knowledge of the extension of the kind” (p. 210).
Lalumera quotes Kripke with respect to the justification requirement: “An answer to the sceptic must satisfy two conditions. First, it must give an account of what fact it is (about my mental state) that constitutes my meaning… But further, there is a condition that any putative candidate for such a fact must satisfy. It must, in some sense, show how I am justifies in giving the answer…” (pp. 210-211). Here the giving the answer is the applying a concept. On the one hand, the quotation says that a fact constitutive of conceptual content and thus of normativity, must show how one is justified in applying the concept. According to a realist reading where Accessibility is not appealed to in the quotation, what the quotation says is a truism (as I have pointed out above): a fact constitutive of normativity, for example a natural kind-fact for the realist thesis, is itself a fact that justifies the concept-application. If the concept WATER correctly applies to a thing because it is a fact that the thing is water, then the fact that the thing is water justifies the concept-application. On the other hand, the quotation says (in the terms of concept-possession I have been using) that a possession-fact constitutive of a natural kind as the condition why a concept is correctly applied must show how one is justified in applying the concept. But according to realist reading what the quotation says would be: a possession-fact constitutive of a natural kind as the condition why a concept is correctly applied must be a fact that constitutes the natural kind as the condition why one is justified in applying the concept. Of course, it is also a truism in the same sense. But then the justification requirement by Kripke would be fully trivial. Thus, contrary to what a realist would sustain the expression “show” has precisely to appeal to an Accessibility Condition. From my point of view, what Kripke intends to say is: if x is the concept WATER, then the possession-facts that constitute the natural kind water as the condition why the concept is correctly applied must be facts that constitute (establish, determine) a knowledge of the natural kind as the condition why one is justified in applying the concept.
Now, it sounds like an epistemic formulation of the justification requirement as opposed to the metaphysical formulation that for a realist is in the quotation, and thus the point of the realist could be withdraw the epistemic one. But whatever Kripke in fact says, what I have just attributed to Kripke is not an epistemic formulation of the justification requirement, one detached from the metaphysical formulation: an Accessibility Condition is not sufficient in order to have an epistemic formulation. Such an epistemic formulation would be something like (S. Soames, “Skepticism about Meaning: Indeterminacy, Normativity, and the Rule-Following Paradox”; “Facts, Truth-Conditions and the Skeptical Solution to the Rule-Following Paradox”): if x is the concept WATER, then the knowledge of the possession-facts that constitute the natural kind water as the condition why the concept is correctly applied must be a knowledge that constitutes (establishes, determines) the knowledge of the natural kind as the condition why one is justified in applying the concept. Here the knowledge of the natural kind-facts as normative is an a priori knowable consequence of the possession-facts, in the sense that such a knowledge follows from the knowledge of the possession-facts. But according to my formulation above, the knowledge of the natural kind-facts as normative is a necessary metaphysical consequence of the possession-facts, in the sense that such a knowledge follows from the possession-facts themselves. Thus, Accessibility is not a feature of a justification to be explained by the epistemology of justification; it is a feature to be explained by the metaphysics of justification. One could say: it is a fact that we usually ignore the intrinsic properties of most natural kinds, so we miss a qualitative grasp of their extensions. The Accessibility objection emerging from this is easy to see in connection with the Twin Earth context. That to be disposed to apply a concept to instances of water does not imply that one should apply means that one will apply the concept not because of one’s knowing that they are instances of water. Precisely the ignorance of the intrinsic properties or the hidden nature of water is the reason why one is also disposed to apply the concept to instances of twater. But my point is the following. To be disposed to apply a concept to instances of water does not imply that one should apply but it does not mean that one will apply the concept not because of one’s a priori knowing that they are instances of water, as an a priori consequence of one’s knowing her disposition. It does mean that one will apply the concept not because of one’s necessarily knowing that they are instances of water, as a necessary consequence of one’s having such a disposition. Another thing is the metaphysical requirement that the natural kind-facts as normative themselves, not a knowledge of them, must be a necessary metaphysical consequence of the possession-facts. But it is not a metaphysical formulation of the justification requirement; it is an independent metaphysical requirement.
Lalumera mainly deals with concept-possession when she introduces the which-question: “which kinds of things sets the standard of correctness in a given case of concept-application, of the many possible ones that can be identified?” (pp. 212-213). Her idea is that the realist thesis would underdetermine the question of correctness because there are many kinds out there for them to count as norms for the application of concepts in a purely objective sense. Then, she appeals to the cognitive limits of concept-possession in terms of saving constraints in order to answer the which-question: the realist strategy “concentrates on placing constraints on which kinds of things there are for us to conceptualize, that is, which are the kinds of things that our cognitive system would afford to represent” (p. 214). Now, it cannot be part of the answer to the which-question what is the starting-point of the question. The kripkean paradox is precisely that there are many kinds out there for them to count as norms for the application of concepts because the cognitive limits of concept-possession do not discriminate between kinds in a purely objective sense. Surely, there are limits on which kinds of things there are for us to conceptualize but the paradox’s lesson is that such kinds cannot be objective. Say that the idea of an objective kind is the idea of a kind whose nature is independent (‘concept-independent’ according to Lalumera) of one person’s cognitive system. It technically means that for any given person there is no necessary function from her cognitive system to the nature of the kind (T. Burge, “Cartesian Error and the Objectivity of Perception”). It does not follow that the question of which are the kinds of things that our cognitive system would afford to represent is a wrong placed question. It does follow that there is not a question of which are the objective kinds (for example, natural kinds in the case of water as opposed to twater) that our cognitive system would afford to represent.
Lalumera thinks that the which-question is the kernel of Kripke’s paradox. However, she is aware of that the relevance of the which-question is dependent on a being-question: “if we admit that for any single case of concept-application the possible standards of correctness are infinite, and all of them are on par, then it is like having no standard of correctness at all” (p. 214). The actual challenge bears on there being a standard of correctness at all. But the underlying being-question is not as Lalumera says: is there any standard of correctness at all, given that there are many or even infinite possible ones? The actual being-question is: is there any standard of correctness at all, given that there are incompatible possible ones? Suppose that the possible standards for that a thought can be evaluated as correct are just the natural kinds water and twater, and thus it is not a challenging thing that there are many or even infinite possible ones. What is a challenging thing is that if the thought is evaluated according to the natural kind water and the salience is that the thought is right, it follows that if the thought is evaluated according to the natural kind twater then the salience is that the thought is wrong. That an occasion of concept-application is right and wrong according to different standards of correctness (all of them on par) is what undermines the question of correctness understood as the intelligibility of there being basis for a distinction between error and correctness at all. Lalumera also places the which-question in the empirical context of standards of correctness that can be identified for the occasion, what tends to eliminate the paradox of inconsistency. Consider the scenario where one sees at the same time a geranium, a plant, an object, and a living object (p. 218). It is obvious that there is not any incompatibility here: if that occasion of concept-application is evaluated as correct according to a standard of correctness, then it could be evaluated as correct according to all of them. I do not see how it can affect mainly to the being-question.
*Este post está parcialmente basado en el comentario que hice a “Norms Out There” de E. Lalumera (X Coloquio Hispano-Italiano de Filosofía Analítica, Universidad de Navarra, España, 2003), una versión temprana de su artículo “A Simple Realist Account of the Normativity of Concepts”. He conservado el idioma original, pero agradezco cualquier comentario referente a un mal uso del inglés.