Abstract
In his well-known essay on Wittgenstein, Saul Kripke maintains that dispositional analyses of meaning cannot work mainly because whereas the concept of disposition is descriptive, that of meaning is normative. Unfortunately, neither Kripke nor his followers have ever spelled out this “argument from normativity” in full detail. As a result, the argument does not have good press. This paper offers an explicit version of the argument. In particular, (1) I try to explain what the claim that meaning is normative amounts to, (2) I try to clarify what supports it, and (3) I sketch a valid version of the argument.
Table of contents
A number of philosophers maintain that dispositional analyses of meaning (such as those of Dretske 1981, Fodor 1990 and Heil & Martin 1998) cannot work simply because meaning is normative (see, e. g., Blackburn 1984, Kripke 1982 and McDowell 1984). Both the validity of the argument and the truth of its premise have been widely questioned (you can see, on the one hand, Boghossian 1989 and, on the other hand, Boghossian 2005 and Heil & Martin 1998). In my opinion, the limited popularity of this “argument from normativity” is due to the fact that its proponents have not clarified enough either what it means to say that meaning is normative, or what are the consequences of this claim, or what supports it. In what follows, I will try to throw some light on each of these points.
Well, what does it mean to say that meaning is normative?
It should be clear that it means something stronger than:
- (1) A sentence has a meaning only if its utterances can be divided (at least in principle, approximately and for the most part) into correct and incorrect
(both here and in what follows, by “sentence” I mean declarative sentence; in order to deal with non-declarative sentences, some minor adjustments are enough; moreover, most of what I maintain can be easily adapted to the case of subsentential expressions). After all, (1) is rather uncontroversial (Dretske – not exactly a supporter of the normativity of meaning – practically states it explicitly – 1981, p. 190).
Some may think that it means the same as:
- (2) A sentence has a meaning only if there are correctness criteria for its use, that is: only if there is a rule for its use.
Both Blackburn and Boghossian seem to conflate (2) with (1) (Blackburn 1984, p. 281 and Boghossian 1989, p. 517), but, in fact, (2) is stronger than (1). I could grant that if the utterances of, say, “There’s glory for you!” could not be divided (not even in principle, approximately and for the most part) into correct and incorrect, “There’s glory for you!” would have no meaning and still stress that in order to label one of these utterances “correct” or “incorrect”, no correctness criterion for the use of the corresponding sentence is needed: for example, I could maintain that all that is needed is that the speaker attaches to the utterance in question a thought and that a thought is something that, by its very nature, can be labelled “correct” or “incorrect” (alternatively, I could tell some story about notions like those of causal relation and communicative intention, or I could maintain, like Humpty Dumpty, that all that is needed is that the speaker chooses what that occurrence means – Carroll 2000, p. 213). Of course, nothing prevents the friend of (2) from acknowledging that a full-blooded analysis of the concept of meaning calls for some “mentalistic” concept (after all, in order to follow a rule, you have to be a person, that is: a thing with a mind); but instead of saying that for a sentence to have a meaning is for it to be used to convey a thought, the friend of (2) will likely say that it is because a sentence has a meaning that it can be used to convey a thought (see Sellars 1969, p. 523) (this is to say that the expression of thought depends on language, not that thought itself does, nor that the concept of thought depends on that of language – for this latter idea, see Geach 21971, pp. 75-117 and Sellars 1981).
No doubt, (2) links meaning to something more than mere regular behaviour (such as that of a well-trained parrot), but, in spite of what Davidson seems to have thought (1992), this is not to say that it links meaning to explicitly stated rules; (2) links meaning to rule-following, and even if rule-following is something more than mere regular behaviour, this is not to say that one can follow only explicitly stated rules (see Sellars 1954, pp. 204-209). In a certain sense, (2) only says that a sentence has a meaning only if it has a character (see Kaplan 1989, p. 505) (without saying anything about the meaning of utterances and without saying exactly what the nature of the relation between the character of a sentence and its meaning is), and this claim seems to follow straightforwardly from the link meaning-communication: that of meaning is a theoretical concept, and its aim is that of explaining communicative phenomena (entertaining a conversation, obeying an order, reviewing your notes for the talk, etc…); hence, it is a conceptual truth that a sentence has a meaning only if it can be used to communicate; but a sentence can be used to communicate only if it has a character, or so it seems. Be that as it may, also (2) is rather uncontroversial (even though Grice maintained that it links meaning and value – 1989, pp. 297-303 –, Fodor – like Dretske, not exactly a supporter of the normativity of meaning – seems to endorse it without hesitation, albeit only implicitly – 1990). Therefore, it should be clear that saying that meaning is normative means, once again, something stronger.
As far as I can see, the following hint is on the right track:
- (3) A sentence has a meaning only if there are correctness criteria for its use, and something can determine these criteria only if it can motivate the use that a speaker makes of the sentence.
That (3) is strong enough can be seen from the fact that it allows us to build the following argument:
- First premise: dispositional analyses of meaning maintain that what determines the correctness criteria for the use of a sentence is a set of dispositions; different analyses deem relevant different sets, but they all agree that the relevant set must count, among its elements, also some unmanifested dispositions.
- Second premise: something can determine these criteria only if it can motivate the use that a speaker makes of the sentence.
- First lemma: something can determine these criteria only if speakers can have non-inferential knowledge of it.
- Second lemma: dispositional analyses of meaning are committed to the view that speakers can have non-inferential knowledge of unmanifested dispositions.
- Third premise: speakers cannot have non-inferential know ledge of unmanifested dispositions.
- Conclusion: dispositional analyses of meaning cannot work.
Namely, what seems to be a valid version of the argument from normativity (it is worth noting that this version of the argument is somewhat related to Wright’s “epistemological argument” – see, e. g., Wright 1989, pp. 175-176, but also Wittgenstein 1953, § 153).
The derivation of the conclusion and that of the second lemma are trivial, while that of the first lemma is warranted by what seems to be a truism concerning the epistemology of motivations (some may be inclined to see this truism as an hidden premise and the argument as an enthymeme; for present purposes, nothing of importance hinges on this point). Therefore, I believe there is little point in questioning the validity of the argument. But what about the truth of its premises? I do not see a dispositionalist questioning the first one. And the third one seems to rest on a sound argument; even if we restrict our consideration to the speakers’ own past dispositions, it seems clear that (in the sense of “can” relevant here) speakers cannot have non-inferential knowledge of unmanifested dispositions. Philosophers sympathetic to Sellars’ conception of observational knowledge (see, e. g., Sellars 1997, pp. 68-79) may suggest that (roughly speaking) in order to non-inferentially know that at a certain time I had a certain disposition, it is sufficient to non-inferentially know that at that time my brain was in a state that, together with a certain stimulus, causes a certain response; however, as a matter of fact, speakers do not keep track of their own past brain history. Philosophers sympathetic to Ryle’s conception of dispositions (see Ryle 22000) may suggest that (roughly speaking) in order to non-inferentially know that at a certain time I had a certain disposition, it is sufficient to non-inferentially know that at that time I underwent a certain stimulus, to which I gave a certain response; however, it is apparent that such a strategy would not be able to supply non-inferential knowledge of unmanifested dispositions. We are left with the second premise, which is what (3) adds to (2). Therefore, the question is: given that (2) is rather uncontroversial, what supports (3)?
In my opinion, the answer is: the very same evidence that supports (2). If (2) is rightly understood, what (3) “adds” to it is in fact something that is already implicit in (2) itself.
As I said before, (2) links meaning to rule-following, and rule-following must be distinguished from both mere regular behaviour and following a rule explicitly stated. If we identify rule-following with mere regular behaviour, the “theory” of the meaning of utterances that fits (2) best (an utterance has a meaning only if the speaker follows a rule in performing it) forces us to the cumbersome conclusion that the utterances performed by a well-trained parrot have a meaning (note that as applied to utterances, “meaning” means content – see Kaplan 1989, pp. 523-524); if, on the other hand, we identify rule-following with following a rule explicitly stated, (2) leads us to an infinite regress, or to a vicious circle (see Davidson 1992, as well as Sellars 1954, pp. 204-207). So much for what “rule-following” does not mean; we can now turn to what it does mean. Martin and Heil give the following characterization:
An agent who follows a rule acts on the rule, his action is based on or motivated by a commitment to the rule (1998, p. 284).
But what does it mean that the agent’s action is motivated by a commitment to the rule? For present purposes, we can focus on those cases in which the agent’s action is a speaker’s utterance and the rule is a rule for the use of the corresponding sentence. For the sake of simplicity, we can then focus on the case of an utterance of the one-word sentence “Carmine!”, which can be conceived of as an answer to a question about the colour of a certain object (I assume that the character of this sentence is identical with that of its sole subsentential component – the word “carmine”). Finally, it is useful to formulate a possible rule for the use of the word “carmine”; here is something that a dispositionalist should appreciate: an application of “carmine” is correct if and only if it is in accordance with the relevant set of dispositions. And so, the question is: what does it mean that my utterance of “Carmine!” is motivated by a commitment to this rule? Well, saying that this utterance is motivated by a commitment to this rule is saying that it is motivated by “the relevant set of dispositions”. But these dispositions are, according to the dispositionalist, what determines the correctness criteria for the use of “carmine”. As soon as we try to clarify the concept of rule-following, we are forced to acknowledge that it is a conceptual truth that something can determine the correctness criteria for the use of a sentence only if it can motivate the use that a speaker makes of the sentence; it is in this sense that what (3) “adds” to (2) is in fact something that is already implicit in (2) itself.
Well, I have sketched what I believe is a valid version of the argument from normativity. I have also tried to show that it rests on plausible assumptions. I do not want to leave you with the misleading impression that I believe that dispositional concepts should play no role in an account of the concept of meaning. I do not believe that. Still, I do not think that such an account can be reduced to a dispositional analysis of meaning.
Literature
- Blackburn, Simon 1984 “The Individual Strikes Back”, Synthese 58, 281-301.
- Boghossian, Paul A. 1989 “The Rule-Following Considerations”, Mind 98, 507-549.
- Boghossian, Paul A. 2005 “Is Meaning Normative?”, in: Christian Nimtz and Ansgar Beckermann (eds.), Philosophie und/als Wissenschaft – Hauptvorträge und Kolloquiumsbeiträge zu GAP.5, Paderborn: Mentis, 205-218.
- Carroll, Lewis, 2000 “Through the Looking-Glass and What Alice Found there”, in: Lewis Carroll, The Annotated Alice – The Definitive Edition, New York-London: Norton & Company, 129-274.
- Davidson, Donald 1992 “The Second Person”, Midwest Studies in Philosophy 17, 255-267.
- Dretske, Fred I. 1981 Knowledge and the Flow of Information, Oxford: Blackwell.
- Fodor, Jerry A. 1990 “A Theory of Content, II: the Theory”, in: Jerry A. Fodor, A Theory of Content and Other Essays, Cambridge-London: MIT Press, 89-136.
- Geach, Peter 21971 Mental Acts – Their Content and Their Objects, London: Routledge & Kegan Paul.
- Grice, Paul 1989 “Meaning Revisited”, in: Paul Grice, Studies in the Way of Words, Cambridge-London: Harvard University Press, 283-303.
- Heil, John and Martin, C. B. 1998 “Rules and Powers”, Philosophical Perspectives 12, 283-312.
- Kaplan, David 1989 “Demonstratives – An Essay on the Semantics, Logic, Metaphysics, and Epistemology of Demonstratives and Other Indexicals”, in: Joseph Almog, John Perry and Howard Wettstein (eds.), Themes from Kaplan, New York-Oxford: Oxford University Press, 481-563.
- Kripke, Saul 1982 Wittgenstein on Rules and Private Language – An Elementary Exposition, Oxford: Blackwell.
- McDowell, John 1984 “Wittgenstein on Following a Rule”, Synthese 58, 325-363.
- Ryle, Gilbert 22000 The Concept of Mind, Chicago: The University of Chicago Press.
- Sellars, Wilfrid 1954 “Some Reflections on Language Games”, Philosophy of Science 21, 204-228.
- Sellars, Wilfrid 1969 “Language as Thought and as Communication”, Philosophy and Phenomenological Research 29, 506-527.
- Sellars, Wilfrid 1981 “Mental Events”, Philosophical Studies 39, 325-345.
- Sellars, Wilfrid 1997 Empiricism and the Philosophy of Mind, Cambridge-London: Harvard University Press.
- Wittgenstein, Ludwig 1953 Philosophical Investigations, Oxford: Blackwell (parallel German and English text).
- Wright, Crispin 2001 “Wittgenstein’s Rule-Following Considerations and the Central Project of Theoretical Linguistics”, in: Crispin Wright, Rails to Infinity – Essays on Themes from Wittgenstein’s Philosophical Investigations, Cambridge-London: Harvard University Press, 170-213.
Refbacks
- There are currently no refbacks.