«DAVIDSON BETWEEN WITTGENSTEIN AND TARSKI RICHARD RORTY University of Virginia The intellectual movement called “analytic philosophy” has teetered ...»
CRITICA, Revista Hispanoamericana de Filosof´a
Vol. XXX, No. 88 (abril 1997): 49–71
DAVIDSON BETWEEN WITTGENSTEIN AND
University of Virginia
The intellectual movement called “analytic philosophy” has
teetered back and forth between philosophy as therapy and
philosophy as system-building. There has been a certain amount of tension between the analytic philosophers who are interested simply in getting rid of pseudo-problems and those who want to give systematic explanations of the pseudo-ness of these problems in the form of analyses of the concepts used in their formulation. Donald Davidson’s work spans this gap. Different parts of his work appeal to therapeutically inclined pragmatists like myself and to systematizing conceptual analysts.
Much of Davidson’s work follows up on that of such systematizers such as Frege, Carnap and Tarski. The so-called “Davidsonian program”, in which extensionalism is salient, belongs to this side of the analytic tradition. When in 1975 Michael Dummett rejoiced that the disciples of Carnap had expelled those of Austin from the holy places, he was celebrating the fact that Oxford had become a hotbed of Davidsonianism.1 See M. Dummett, “Can Analytic Philosophy Be Systematic, and Ought It To Be”, in his Truth and Other Enigmas, Harvard University Press, Cambridge, Mass., 1978, pp. 437–458.
On the other hand, a lot of the doctrines for which Davidson became famous in the 1980s are more akin to those of the later Wittgenstein than to any views held by Carnap or Tarski. Consider his claims that most of our beliefs must be true, and that there is no distinction between knowing a language and knowing our way around in the world. You can grasp his arguments for these claims even if you have no interest whatever in what Dummett called “a systematic theory of meaning —the sort of theory which the later Wittgenstein thought implausible and unnecessary, but of which a Tarskian truth-theory for a natural language is paradigmatic. You can be excited by these arguments even if you are the kind of philosopher who, like me, still can’t figure out why symbolic logic is supposed to be an essential propaedeutic to the study of philosophy, and who still quails at the sight of a quantifier.
Philosophers of this sort think they get the point of the Philosophical Investigations, but are still unsure whether they get the point of Tarski. They tend to be dubious about two features of Davidson’s work. The first is his continuing agreement with Quine that there is no matter of fact about translation —an agreement which seems a hangover of the scienticism and reductionism which permeated logical positivism. The second is Davidson’s propensity to connect all his arguments to the project of constructing T-theories for natural languages. We therapists tend to think that we can keep most of the arguments while ignoring the project.
In the first part of this paper, I shall talk about indeterminacy. I shall expound an interpretation of what Davidson says on this topic which has been put forward in an unpublished paper by Bjorn Ramberg, and which I have found very illuminating. In the second half, I shall discuss the question: is there any more reason to think of linguistic behavior as illuminated by a systematic theory of meaning than to think of bicycle-riding behavior as illuminated by a systematic theory of the relations between moving bicycles and their environment?
Part I In 197l my philosophical views were shaken up, and began to be transformed.2 That was the year in which Davidson let me see the text of his 1970 Locke Lectures, which included an early draft of his “On the Very Idea of a Conceptual Scheme”. That paper still strikes me as epoch-making.
It will, I think, be ranked with “Two Dogmas of Empiricism” and “Empiricism and the Philosophy of Mind” as one of the turning-points in the history of analytic philosophy.
In 1972 I published an article called “The World Well Lost”3 which owed its central argument to Davidson’s lectures. In that article, as in much that I have written since, I attempted to synthesize Davidson and Dewey. I pointed out that both philosophers were attacking the Kantian distinction between receptive sense and spontaneous intellect, and doing so for similar reasons. Furthermore, both were suggesting that all the links between mind and world are causal, and none representational. These suggestions, I claimed, dissolved a great many of the problems about the relation of mind or language to the world —problems which had been bequeathed to analytic philosophy by Russell, C.I. Lewis and Carnap. They thereby advanced Wittgenstein’s therapeutic project.
The next dozen paragraphs or so of this paper are taken from my “Davidson on the Mental-Physical Distinction”, which will appear in The Philosophy of Donald Davidson, a volume in the Library of the Living Philosophers. Prof. Lewis Hahn, editor of the Library, has kindly given me permission to publish this fragment of my paper here.
“The World Well Lost”, Journal of Philosophy, 69, 26 Oct 1972, pp. 649–665 (reprinted in my Consequences of Pragmatism, University of Minessota Press, Minneapolis, 1982).
But in 1972 I also published an article criticizing Quine’s claim that the indeterminacy of translation was different from the ordinary underdetermination of empirical theories.4 I argued that Quine had never given a satisfactory sense to the term “fact of the matter”, and that the contrast he invoked between the factual and the non-factual seemed to be the same contrast that he had been concerned to blur in the concluding paragraphs of “Two Dogmas of Empiricism”.5 I had expected Davidson to concur on this point, and I was taken aback when it turned out that he heartily agreed with Quine about the indeterminacy of translation.6 During the intervening quarter-century, we have continued to disagree over this point. I have persisted in thinking that the anti-dualist line of thought developed in “Two Dogmas”, a line of thought which had led Davidson to reject the spontaneity-receptivity and scheme-content distinctions, should also lead him to reject the very idea of a distinction between the presence and absence of what Quine called “a fact of the matter”.
Quine’s invidious distinction between the “baselessness of intentional idioms” and the better “based” idiom of “Indeterminacy of Translation and of Truth”, Synth` se, 23, e March 1972, pp. 443–462.
My arguments in that article paralleled, and borrowed from, Hilary Putnam’s “The Refutation of Conventionalism”. Putnam had said there that Quine seemed to be making “an essentialist maneuver” in arguing for indeterminacy. (See Putnam, Mind, Language and Reality, Cambridge University Press, Cambridge, 1975, p. 174.) Davidson had made his agreement with Quine explicit in “Mental Events” (1970). There he says, for example, “The heteronomic character of general statements linking the mental and the physical traces back to the central role of translation in the description of all propositional attitudes, and to the indeterminacy of translation [... ]” (Davidson, “Mental Events”, in Essays on Actions and Events, Oxford University Press, Oxford, 1980, p. 222.) physical science strikes pragmatists like me as a residue of the unfortunate positivist idea that we can divide culture into the part in which there is an attempt to correspond to reality and the part in which there is not. If you drop the idea that some of our sentences are distinguished by such correspondence, as Davidson has, it seems natural to say, as Dewey and Wittgenstein did, that all our idioms are tools for coping with the world. This means that there can be no philosophical interest in reducing one idiom to another, nor in asking whether and how a non-extensional language might be replaced with an extensional one.7 As pragmatists see it, we are equally in touch with reality when we describe a hunk of space-time in atomic, molecular, cellular, physiological, behavioral, intentional, political, or religious terms. Looking for an ontological or epistemological gap between such idioms strikes pragmatists as like looking for such gaps between a small Phillips screwdriver and a large crescent wrench; there are all sorts of similarities and differences, but none of them have onAt p. 176 in Inquires into Truth and Interpretation, Davidson says “It seems to be the case, though the matter is not entirely simple or clear, that a theory of truth that satisfies anything like [Tarski’s] Convention T cannot allow an intentional semantics, and this has prompted me to show how an extensional semantics can handle what is special about belief sentences, indirect discourse, and other such sentences.” It is not clear that Davidson is still interested in this latter attempt —the so-called “Davidsonian Program”. I do not see an easy way to combine his earlier claim that a learnable language must have a recursive structure, the sort of structure captured by a truth theory for the language, with his more recent view that since “we have discovered [... ] no portable interpreting machine set to grind out the meaning of an arbitrary utterance”, “we may say that linguistic ability is the ability
to converge on a passing theory from time to time”. (“A Nice Derangement of Epitaphs”, in Ernest LePore (ed.), Truth and Interpretation:
Perspectives on the Philosophy of Donald Davidson, Blackwell, New York and Oxford, 1986, p. 445.) I should be happy to find that Davidson had lost interest in both recursivity and extensionality, but I am not sure whether he has or not.
tological or epistemological import. Quine, however, propounds a physicalist ontology when he explicates his claim that translation is not simply underdetermined, but indeterminate, by saying that two incompatible translations, unlike two physical theories, are “compatible with all the same distributions of states and relations over elementary particles”.8 What seems puzzling about Davidson’s agreement with Quine about indeterminacy is that Davidson has no particular interest in elementary particles, nor does he share Quine’s attachment to nerve endings. He thinks it best for translators to go straight from speakers’ behavior to distal objects, and straight back again, skipping over neural stimulations. Davidson has displayed no interest in epistemology, nor in a physicalistic ontology, nor, for that matter, in any other kind of ontology.9 I see this lack of interest in ontological and epistemological questions as another of Davidson’s laudable resemblances to Dewey.
Presumably Davidson would grant that Quine had several different motives for advancing his doctrine of the indeterminacy of translation. He might also grant that one or more of those motives (for example, what he once called “an adventitious philosophical puritanism”) may have been misguided. But he holds that the doctrine itself is nevertheless sound and important. He is willing to restate it, but not to dismiss it. Davidson thinks that even antirepresentationalists, people who believe (as he and I do) Quine, Theories and Things, Harvard University Press, Cambridge, Mass., 1981, p. 23.
In “The Method of Truth in Metaphysics” (1977), Davidson says that his metaphysics is that of events, persons, things, etc. —whatever you have to talk about to state a truth theory for English. But this is not the invidious sort of ontology which Quine has in mind when he says that he does not want there to be more things in his ontology than there are in heaven and earth.
that our links with the world are merely causal, rather than representational,10 ought to recognize the importance of the difference between intentional and other idioms.
For my part, I am happy to grant that attributions of intentional states are far more holistic than any other attributions we make, but that is as far as I want to go. I cannot see why this holism is supposed to have ontological implications. I think of holism in the ascription of predicates as a matter of degree: some predicates signify, to use Putnam’s terminology, “single-criterion concepts”, others “multi-criterion concepts”, and still others have conditions of correct application so incredibly complex that there seems little point in using the term “criterion” at all.
But I do not see why greatly increased complexity of conditions of application of predicates should be a symptom of the absence of a “fact of the matter”.
A year or so ago I wrote a paper summarizing my doubts about Davidson’s defense of the non-factuality of the intentional, and sent it for comment to Bjorn Ramberg. Ramberg responded with a paper of his own, one which seemed to me to put the matter in a whole new light.11 Ramberg’s paper has caused me to rethink my own criticisms, and to read Davidson differently.
John McDowell says, correctly, that Davidson is “blandly confident” that “empirical content can be intelligibily in the picture even though we carefully stipulate that the world’s impacts on our sense have nothing to do with justification”. (McDowell, Mind and World, Harvard University Press, Cambridge, Mass., 1994, p. 15.) As McDowell also says, “Rorty singles out for commendation precisely the aspects of Davidson’s thinking that I have objected to” (ibid., p. 146). What I commend is the idea that, in McDowell’s terminology, the world has no “rational” control over our beliefs, but only causal control.
This paper, “Post-Ontological Philosophy of Mind”, a comparison between my views and Davidson’s, will appear in Rorty and his Critics, ed. Robert Brandom, Blackwell, Oxford and New York, 2000.
Ramberg suggests that the famous Brentanian irreducibility of the intentional is an unfortunate distraction from the inescapability of the normative. Ramberg points out that the mind-body distinction is intertwined with the person-thing distinction, and that Davidson, by combining a theory of action with a theory of truth and meaning, has illuminated the relation between the two distinctions.