Individualism and Interpretation
Interpretational accounts of meaning, which understand what a speaker means in terms of how his linguistic (and other) behavior would be best interpreted, have frequently been taken to be incompatible with accounts of meaning which stress language's "social" character. Indeed, Donald Davidson, probably the most prominent interpretationalist, has criticized writers such as Tyler Burge precisely for claiming that social usage can determine the contents of our thoughts and utterances. However, one can reconcile interpretational accounts of meaning with language's social character if one distinguishes "methodological" from "ascriptional" individualism. Ascriptional individualism requires that what one means by one's own terms be independent of how others use theirs. Methodological individualism, on the other hand, requires only that the criteria for the correct application of one's terms ultimately be grounded in facts about oneself. Davidson's hostility to Burge's work results at least partially from his running these two types of individualism together. One can be a methodological individualist without being an ascriptional one, and while interpretational accounts of meaning are committed to methodological individualism, arguments for languages social character are incompatible only with ascriptional individualism.
We can see this by considering Burge's well known discussion of Bert and his claim that he has "arthritis" in his thigh (especially since it is Burge's conclusions about just this case that Davidson criticizes). The scenario should be familiar enough, and will be rehearsed just briefly here. Bert has a number of correct beliefs about arthritis such as that he has had arthritis in his knuckles for years, that older people frequently have it, that cold weather aggravates it, etc., but he also believes that his arthritis has spread to his thigh. When he tells this to his doctor, he is told that arthritis only affects the joints. As a result, Bert gives up his belief, and goes on to wonder what could be wrong with his thigh. In spite of his idiosyncratic belief, Bert is still taken to refer to arthritis by "arthritis." Now consider a counterfactual world in which Bert "proceeds from birth through the same course of physical events that he actually does, right to and including the time at which he first reports his fear to his doctor," but in which Bert's society uses the word "arthritis" to pick out inflammations outside of the joints as well (this cluster of inflammations will be here referred to as "tharthritis"). Burge argues that, in such a scenario, Bert would express a true belief when he said "I have arthritis in my thigh," and that is because his use of "arthritis" would refer to tharthritis. What we take Bert to mean by the word "arthritis" is thus determined, at least in part, by the usage of the term in the society to which he belongs.
The ascriptional individualist must say that this practice of ours is mistaken. If Bert's behavior and physical environment are the same in both cases, the ascriptional individualist must claim that what he means will be the same as well. The ascriptional individualist is thus committed to treating Bert as meaning tharthritis by "arthritis" in both cases, and thus as having correctly applied his term in the actual world as well.
It is far less clear that the methodological individualist has any such
commitment. All that methodological individualism requires is that 'external'
factors (such as the usage of the surrounding community) be relevant to
what the interpretee means only if they are taken to be so by the interpretee.
(In its crudest form, he takes, say, "arthritis" to mean "what
the experts mean by 'arthritis.'") Methodological individualism is
thus only incompatible with positions requiring that what a speaker means
be determined by social usage regardless of the speaker's attitude
towards such usage. Davidson puts such methodologically individualistic
restrictions on the relevance of social usage in the following passage.
Under usual circumstances a speaker knows he is most apt to be understood if he speaks as his listeners would, and so intends to speak as he thinks they would. He will then fail in one of his intentions if he does not speak as others do. This simple fact helps explain, I think, why many philosophers have tied the meaning of a speaker's utterance to what others mean by the same words .... On my account, this tie is neither essential nor direct; it comes into play only when the speaker intends to be interpreted as (certain) others would be. When this intention is absent, the correct understanding of a speaker is unaffected by usage beyond the intended reach of his voice.
This claim about the indirectness of the connection between social usage and utterance content is extremely plausible. While we are usually willing to consider ourselves subject to the norms of the surrounding community, if we decide to deliberately flout communal usage, it is hard to see why we should still be understood in terms of it. If I insist on calling stools "chairs" in spite of having the non-standardness of my usage pointed out to me (and I make my idiosyncrasies clear to those unfamiliar with them), then it would be perverse to attribute to me a false belief when I call a stool a "chair." I would simply mean something different by "chair" than the rest of my community.
Still, while the tie between the meaning of a speaker's utterance and what others mean by the same words may be neither essential nor direct, it is often there nonetheless. One can agree with Davidson that if one lacks the intention to be interpreted as others would be, then one shouldn't be so interpreted, and still argue that in many (if not most) cases, we do intend to be interpreted as others would be. Claiming that what speakers mean by their words is determined by how they intend to use them does not entail ascriptional individualism if how they intend to use them is often tied to the linguistic practices of their communities. In particular, we can plausibly assume that Bert intended to be interpreted as meaning by "arthritis" what his doctor meant by the term.
This intention need not be explicitly formulated. It can be manifested in the fact that, when faced with the incompatibility of various beliefs he has about "arthritis" (that the doctor is talking about "arthritis", that he has "arthritis" in his thigh, that he heard a news item about "arthritis" that afternoon, etc.), Bert chooses to preserve a consistent set that is in line with social usage. We generally intend to be interpreted as our fellows are, and while such an intention is defeasible (if, for instance, our usage strays too far from the public norm), minor differences between ours and the public usage will not be enough to make us give it up. This phenomenon is especially pronounced in more 'disciplined' types of discourse where there is a concerted effort to ensure that there is a clearly defined accepted usage to which speakers adhere when using a specified technical vocabulary. (It is more acceptable to use "bug" idiosyncratically than it is "insect.")
Indeed, the fact that Bert is speaking to a doctor and not, say,
his brother is relevant here. When we are speaking to experts in contexts
where their expertise is being appealed to, the likelihood that we intend
to follow 'expert usage' will be considerably higher than when we are talking
within a possibly idiosyncratic sub-community such as our family. It may
be the case that, when Bert complains to his brother, "I hate this
cold weather, it always makes my arthritis worse," he should be interpreted
as meaning tharthritis by "arthritis," but the case Burge
describes is not like this at all. Bert is going to see a doctor and in
this context it seems quite plausible to think that Bert intends to be following
expert usage. His utterance of "I think that I have arthritis in my
thigh" would have (very roughly) the gist of "I know there is
this disease you doctors call "arthritis" and I think that I have
it in my thigh." Such a 'metalinguistic' reworking of his
communicative intention may have little plausibility in the case of his
complaint to his brother (which may have roughly the gist of "I hate
this cold weather, it always makes the pain in my knuckles and thigh worse").
Burge's choice of context has a lot to do with the plausibility of his example.
Methodological individualism is thus compatible with our at least occasionally
ascribing contents in a fashion that is ascriptionally non-individualistic.
Davidson actually endorses such a possibility in passages such as the following:
I do not doubt the existence of [the linguistic division of labor], or even its importance.... we can take it to be part of the meaning of an expression that its reference is to be determined by expert opinion.... So for the words "elm" and "beech" to pick out the appropriate trees there would have to be experts, but we cannot conclude that the meaningful use of these words demands a social setting.... It is obvious that the linguistic division of labor can't be essential to verbal communication. If everyone meant by "elm" "what others mean by elm," the word would have no reference. The linguistic division of labor is a device that can come into play only after the basic linguistic skills that tie words directly to things are already in place. So no matter how universal the linguistic division of labor is in practice, it cannot constitute the essential social element in language. We could get along without it.
Davidson here admits that experts determine what most of us refer to by terms like "elm," so if the experts had referred to something different, we would have too. These 'non-individualistic' content ascriptions are, however, justified on methodologically individualistic grounds. Accepted usage is relevant only because the individual takes it to be. Davidson seems only to be denying that, for any given word, this can be true of all of us, and thus that the division of linguistic labor is essential to language as such. His real target is, then, not the claim that the truth-conditions of our thoughts and utterances may be in part determined by the usage of our community, but rather the stronger claim that such dependence is a necessary feature of language use.
Indeed, a recognition of the social character of actual languages pervades much of Davidson's writings. Davidson not only claims that "a social theory of interpretation" is made possible by belief taking up the slack between sentences held true by individuals and sentences held true (or false) by public standards, but also argues that states of mind "are identified in part by the social and historical context in which they are acquired." As a matter of fact, Davidson's describes a case very much like Burge's when he claims that (because of the environment in which he learned the word) he would mean porcupine by "porcupine" even if he applied the term to both porcupines and echidnas, and that this phenomenon is "ubiquitous, since it is inseparable from the social character of language."
However, in spite of his occasional recognition of our language's social
character and its compatibility with his methodological individualism, Davidson
is often quite dismissive of ascriptional non-individualism:
I am not impressed by [Dummett's] or Burge's or Putnam's insistence that words may have a meaning of which both speaker and hearer are ignorant. I don't doubt that we sometimes say this, and it's fairly clear what we have in mind: speaker and hearer are ignorant of what would be found in some dictionary, or how people with a better or different education or a higher income use the words. This ... imports into the theory of meaning an elitist norm by implying that people not in the right social swim don't really know what they mean.
Davidson here not only questions whether the participants in a conversation can fail to know the correct usage of their terms, but also suggests that 'expert usage' can only have a sort of prestige value and is irrelevant to what non-experts mean by their terms. Given that it isn't required by his methodological individualism and is actually incompatible with some of his other commitments, how can we explain Davidson's official hostility to ascriptional non-individualism?
Perhaps the main reason for this hostility is Davidson's failure to always distinguish the two types of individualism. As a result, he interprets an ascriptional non-individualist like Burge as promoting a position that is incompatible with even methodological individualism. Because of this, he naturally feels that Burge's position is incompatible with his own, and thus that Burge's arguments need to be criticized. Unfortunately, much of Burge's argument rests on his intuitions about how Bert's utterance to his doctor should be interpreted, so Davidson's criticisms of Burge often involve contesting the correctness and importance of these intuitions. Davidson suggests instead that Bert be understood as meaning something different than the doctor by "arthritis," and that his deference be understood as his changing what he means in order to communicate more easily with his fellows However, Bert's interaction with his doctor is fairly close to a best-case scenario for endorsing ascriptionally non-individualistic attributions on methodologically individualistic grounds. Consequently, if we interpret Bert individualistically here it seems quite plausible that we will be committed to interpreting everyone individualistically everywhere. Davidson's strategy for defending methodological individualism thus implicitly commits him to the ascriptional variety.
As stated earlier, Davidson may understand Burge as attacking methodological individualism because of his interest in "the essential nature of the skills" that are "basic to linguistic communication" (and not "usual, though contingent" features of such communication). Such an interest may lead him to overlook the fact that what is at issue with cases like Bert's use of "arthritis" is not what language must be like for anything, or even what it must be like for us, but rather how it is, in fact, for Bert. Burge's claims about our language need not be understood as applying to language as such. As a result, even if it turned out that language was not essentially social, this would not affect the claim that contents actually ascribed to us should often be individuated non-individualistically.
If one wants to defend methodological individualism, then, one need not quibble about whether or not Bert has made a false claim about arthritis. Methodological individualists can freely admit that Burge is right about Bert's utterance to his doctor. All they need to do is argue that (1) such ascriptions are compatible with methodological individualism since they can be explained by Bert's reliance on the doctor's usage in that context, and that (2) in contexts where no such reliance is present, there is no reason to ascribe contents in a 'social-externalist' fashion. Methodological individualists need not challenge the correctness of Burge's characterization of Bert's utterance, they need only challenge the extent of its generalizability.
Davidson thus need not understand Burge's work as the kind of threat
to his position that he frequently treats it as being. Once we distinguish
methodological individualism (which clearly is an essential part of the
Davidsonian program) from ascriptional individualism (which may even be
incompatible with aspects of Davidson's position), we can accommodate ascriptionally
non-individualistic attributions within a methodologically individualistic
framework. As a result, nothing prevent one from accommodating language's
social character within an interpretational account of meaning.
I'd like to thank Akeel Bilgrami, Robert Brandom, Joe Camp, Jim Conant, Mark McCullagh, John McDowell, Ram Neta, James Shelton Asa Wikforss, and members of the 1997 meeting of the Southwest Philosophical Society for comments on earlier versions of this paper.
1 See Burge, T, "Individualism and the Mental", in French, Uehling & Wettstein. Midwest Studies in Philosophy IV: Studies in Metaphysics, (Minneapolis: University of Minnesota Press, 1979). Davidson's criticisms can be found in his "Knowing One's Own Mind", Proceedings and Addresses of the American Philosophical Association. (1987) and "The Social Aspect of Language", in McGuinness and Oliveri (eds.). The Philosophy of Michael Dummett, (Dordrecht: Kluwer, 1994). The case is extensively criticized from a Davidsonian point of view in Bilgrami's Belief and Meaning (Oxford: Blackwell, 1992).
2 Burge "Individualism and the Mental," p. 77.
3 Both Davidson and Bilgrami make this claim about Bert. Of course, the ascriptional individualist could also try to argue that in both cases Bert means arthritis by "arthritis", but, to my knowledge, no defenders of ascriptional individualism do so.
4 Davidson, "The Second Person", in French Uehling and Wettstein, Midwest Studies in Philosophy XVII: The Wittgenstein Legacy, (Notre Dame, University of Notre Dame Press, 1992) p. 261.
5 Such cases would, presumably, be limited to those words referring to items for which we feel confident that we could set up adequate criteria without anyone's help (I may be able to do this with "chair" but not with, say, "electron," or, for that matter, "elm").
6 The attribution of standard meanings to Bert may thus simply result from the application of Wilson's Principle of Charity to his usage. (See Wilson, "Substance Without Substrata", Review of Metaphysics 12 (1959) p. 532.) Wilson's original formulation of the principle involves assigning referents to a speaker's terms that maximize the number of truths among his beliefs involving those terms. Allowances are made, of course, for some beliefs being more important than others, and which beliefs are most important to the speaker are determined by how he reacts when he discovers that they conflict. What a speaker means by a term can vary from context to context, because the importance of the associated beliefs may vary from context to context as well.
7 What counts as 'minor' or 'too far' may be context-relative and partially determined by the speaker.
8 For a discussion of such cases, see Kay, P. "Linguistic Competence and Folk Theories of Language", Proceedings of the Berkeley Linguistics Society, 1983.
9 Such metalinguistic formulations of their intentions may not be available to most speakers, but such intentions can, nevertheless, be manifested in their behavior.
10 And this is not necessarily because of the relative expertise of the participants involved. Bert and his brother may both use the term idiosyncratically and yet both be interpreted according to social usage (say, they are discussing a recent news report about arthritis).
11 Davidson, "The Social Aspect of Language" 5-6.
12 Davidson "Belief and the Basis of Meaning", in his Inquiries into Truth and Interpretation, (Oxford: OUP, 1984), p. 153. He also claims that belief is a private attitude that "is not intelligible except as an adjustment to the public norm provided by language" (Davidson "Thought and Talk", in his Inquiries into Truth and Interpretation, p. 170).
13 Davidson, "The Myth of the Subjective", in Krausz, M. (ed.) Relativism: Interpretation and Confrontation, (Notre Dame: Notre Dame University Press, 1989), p. 170.
14 Davidson "The Myth of the Subjective", 167. This environment Davidson has in mind cannot just his physical environment, since Davidson cannot be merely suggesting that he refers to porcupines because he learnt the word "porcupine" when he was in contact with members of the natural kind porcupine. Not only has he expressed skepticism about the idea of natural kinds ("Knowing One's Own Mind", 450), but this would also involve an extremely implausible claim about our acquisition of the word.
15 Davidson, "The Social Character of Language", 12.
16 Other reasons include the following two: First, Davidson comes out of a tradition that is clearly committed to ascriptional individualism. Quine's original discussion of "radical translation" in Word and Object, for instance, explicitly ruled out the relevance of interactions with "native kibitzers" whether the informant was disposed to rely on them or not (Quine, W.V.O. Word and Object, (Cambridge: MIT, 1960), chapter 2). Second, Davidson may occasionally run together cases where we deliberately flout the societal norms with those where we are merely ignorant of them. In the former case we lack the intention to be interpreted as others are, while in the latter case such an intention (or at least the assumption/presupposition that we will be so interpreted) is probably present. This difference explains why, even if the first type of speaker should be interpreted individualistically, the latter type usually should not. While there are occasions where we are willing to be interpreted idiosyncratically if our usage departs from the social norm, the intention to be interpreted as others are usually is the default case.
17 There is good reason to think that Burge's position is, in fact, methodologically individualistic. After all, Burge does stress the importance of "the individual's intentions or attitudes toward communal conventions" ("Individualism and the Mental" p.114), and later papers such as "Wherein is Language Social" (in George, A. (ed.). Reflections on Chomsky, (New York: Blackwell, 1989)) only reinforce the impression that he takes the non-individualistic character of our thoughts to be ultimately justified by our own attitudes towards and reliance upon the usage of others.
18 "Burge's evidence for this claim seems to rest on his conviction that this is the way anyone (unspoiled by philosophy) would report ... I doubt that Burge is right about this, but even if he is, I don't think it proves his claim. Ordinary attributions of meanings and attitudes rest on vast and vague assumptions about what is and is not shared (linguistically and otherwise) by the attributer, the person to whom the attribution is made, and the attributer's intended audience. When some of these assumptions prove false, we may alter the words we use to make the report, often in substantial ways. When nothing much hinges on it, we tend to choose the lazy way: we take someone at his word, even if this does not quite reflect some aspect of the speaker's thought or meaning." (Davidson, "Knowing One's Own Mind" 449.)
19 Such accounts of what is going on when we defer made most explicit in Bilgrami's Belief and Meaning.
20 Davidson "Communication and Convention," in his Inquiries into Truth and Interpretation, pp. 278, 280 (italics mine).
21 Burge does, of course suggest that his points
are true of all human languages, but this is not because he thinks that
the social character is essential to language as such, but rather because
he thinks that our tendency to look to others to help set the standards
for our own words "derives from psychological necessities for human
beings." (Burge, "Wherein is Language Social, 187.)