Mainz, Germany

Dear Kalociński,

Dummett certainly seems to be alluding to Davidson, since Davidson had a similar conception of what it is to have a theory of meaning for a language. There are other signs of his concern with Davidson on meaning further along in the paper.

Of course Dummett also had his differences with Davidson.

H.G. Callaway

Dummett refers to Donald Davidson, cf. DOI: 10.1111/j.0966-8373.2004.00208.x

I think that Dummett refers to Gottlob Frege, "although I should not feel capable of giving a demonstration that this was so to" anybody.

Mainz, Germany

Dear Kalociński,

Dummett certainly seems to be alluding to Davidson, since Davidson had a similar conception of what it is to have a theory of meaning for a language. There are other signs of his concern with Davidson on meaning further along in the paper.

Of course Dummett also had his differences with Davidson.

H.G. Callaway

Mainz, Germany

Dear all,

Readers of this thread may find the following paper of interest:

https://www.researchgate.net/publication/227085210_Semantic_competence_and_truth-conditional_semantics

Its an older paper of mine addressing Davidsonian semantics. At the time I published it, and long afterward, I was also fairly deeply concerned with Dummett and his writings on Frege and semantics. I got quite a lot of mileage and attention for this work, going around to make presentations at European universities in particular.

H.G. Callaway

Article Semantic competence and truth-conditional semantics

Dear Dariusz

Could this not be a generalised statement, referring to any one who denied this approach as fruitful, suggesting that his arguments are not strong enough to sway such people; well it is difficult to wake up people feigning sleep.

Narayanan

The answer is given by Dummett himself in the Preface of “The Seas of Language”:

 «Before this, however, Davidson had, with the greatest mildness, voiced, again in private conversation, a complaint I had not anticipated, that it had been he who introduced the notion of a theory of meaning for a specific language, whereas, in the lecture, I had treated it as a notion in common currency, on which he had put a particular interpretation. The phrase "theory of meaning" is, of course, quite old, in that use in which it denotes a branch of philosophy, in analogy with "theory of knowledge"; this differs sharply from the conception of a detailed theory relating to a particular language and giving the meanings of all words and sentences of the language. Davidson has operated with that conception since the earliest of his published papers on the philosophy of language (that is, since 1967). If I have borrowed it and used it without proper acknowledgement, I hereby express my thanks and my apologies. I find it impossible to be sure whether this is so, or whether there is merely a partial convergence of ideas. Davidson's earliest relevant essays antedate my first book, Frege: Philosophy of Language (London, 1973), in the preface to which I acknowledged Davidson's general influence on me; so an examination of it does not settle the matter. Looking through it, I find the phrase "theory of meaning" used repeatedly, and with no care to distinguish different senses of it. Sometimes it is used in the old sense; more often as denoting a comprehensive theory describing the workings of 'language', with all its component expressions, and thereby explaining what constitutes the possession by those expressions of the meanings that they have, but as if there was only a single human language; but sometimes (e.g. pp. 106, 397, 413-17, 452-6, 460-1) as denoting a theory of this

kind relating to a single specific language. So I certainly had that concept then, without keeping it at all sharply in focus; but that does not show that I did not derive it from Davidson. For any injustice or ungraciousness of which I may have been unintentionally guilty, I hereby sincerely apologize.» The Seas of Language, Preface , p. ix.

In my first comment I presumed that the question was about the proposer of the idea “that this is the most fruitful approach”. If, on the other hand, the question is about “someone who denied it”, then there are many philosophers who would deny it. The most important would be Ludwig Wittgenstein.

Dear H.G., Cesaro

The extent and depth of your knowledge and memory is amazing. I salute you as an amateur linguist

Narayanan

Mainz, Germany

Dear Cozzo & readers,

Interesting that you should enlist Wittgenstein as denying the concept of giving a theory of meaning for a language. It seems a bit anachronistic to make this claim flatly, though. Wittgenstein died long before the concept generated much or perhaps any  interest.

Still the denial suggest the possibility of making comparisons. Wittgenstein's work is a paradigm of post-war "ordinary language" philosophy, though I think he goes very deep; while Davidson, working on the idea of what a theory of meaning for a language would look like, favors formal methods, predominantly.  "Don't look for the meaning," wrote Wittgenstein, "look for the use." That is good advice, and especially in contrast to Russell's logical atomism and associated views. But on the other hand, one might hold that usage is the evidence of meaning.

The lexicographer collects examples of word usage and then summarizes (theorizes?) them into definitions or other general characterizations of word meaning. But the question in Davidson, and in Dummett, seems different: Given that we have the lexicon assembled, how might that enter into accounts of sentence meanings, or semantic competence of speakers and hearers? 

I can imagine that later Wittgenstein would tend to be skeptical about any overly formalist account of how the meanings of words enter into the meanings of sentences & etc. But surely, he doesn't deny compositionality. Wouldn't you agree? That idea would put a large burden on the interpretation of Wittgenstein. 

H.G. Callaway

Dear Callaway,

"The lexicographer collects examples of word usage and then summarizes (theorizes?) them into definitions or other general characterizations of word meaning. But the question in Davidson, and in Dummett, seems different: Given that we have the lexicon assembled, how might that enter into accounts of sentence meanings, or semantic competence of speakers and hearers?"

Do you know any explicit statements by Dummett, Davidson or others, according to which the lexicon and its semantics is taken for granted and assumed to be shared by language community?

I have also another (off topic) question concerning Davidson. I am quite confused by his approach to a theory of meaning for a language. In "Nice derangement of epitaphs" he explicitly denies the conventionality of a language. In particualr - one might think - there are no semantic conventions which govern the lexicon. Would you agree with this interpretation? However, if there are no semantic conventions which govern the lexicon, taking the (shared) lexicon as a starting point of a theory is extremely suspicious! Personally, I think Davidson goes to far with his denial of conventions.

 Dear Dr Callaway,

The question was about Dummett. Whom did Dummett refer to? I think the answer is: Wittgenstein is the most important and influential philosopher Dummett had in mind. For Dummett “we all stand in the shadow of Wittgenstein”, and in many places Dummett writes that Wittgenstein rejects the idea of a systematic theory of meaning.

For example in “Can Analytical Philosophy be Systematic, and Ought it to Be?” we can read what follows.

«It is difficult to see how a systematic theory of meaning for a language is possible without acknowledging the distinction between sense and force, or one closely similar. […] It is, however, just this conception which Wittgenstein attacks. He does not stop at rejecting the claim that all assertoric sentences form a single category, of which a uniform account can be given: he denies that any surveyable list of types of linguistic act can be arrived at. This is precisely to deny that the distinction between sense and force is available to simplify the task of explaining the meanings of sentences by distinguishing two different components of their meanings: our theory of meaning must, for each individual sentence, issue in a direct account of the conventional significance of an utterance of that sentence, rather than one derived from a general description of the use of sentences of some general category to which

it belongs. Not only do we not know in the least how to set about devising a theory of meaning in conformity with this maxim, but it leads to that neglect of the difference between semantic and pragmatic considerations which I noted in the practice of the 'ordinary language' philosophers. Wittgenstein's deliberately unsystematic philosophical method makes it difficult to be certain what his intention was. Did he have in mind some theory of meaning of a completely different kind from that proposed by Frege? Or did he reject the whole idea of a systematic theory of meaning?» “Can Analytical Philosophy be Systematic, and Ought it to Be?”, in Truth and Other Enigmas, p. 450.    I think here Wittgenstein’s answer is «We may not advance any kind of theory» (Philosophical Investigations  §109).

Best regards,

Cesare Cozzo

Mainz, Germany

Dear Cozzo,

Many thanks for taking up the challenge and entering into the comparison. It seems to me that you have been reading Dummett rather closely of late. I should say, it has been a long time, since I read any of Dummett, though I did read him closely when I studied and wrote on Davidson.

If as you have it on Dummett's account, Wittgenstein would reject any general account of assertions, then, I suppose that would leave Davidson out in the cold. Generally, I agree with Dummett on the point that we "all stand in the shadow of Wittgenstein." In spite of that I find little that is deeply problematic in a unified account of assertions. I notice that you neglect to comment on my attention to the relationship between usage and the practice of empirical lexicography--and that you say nothing about the question of compositionality. Instead you restrict yourself to a long quotation of Dummett on Wittgenstein. Though your interpretation of Dummett be ever so admirable, we may still go on to question Wittgenstein in trying to understand more systematic approaches to meaning.

Dummett himself seems to cast some doubts on Wittgenstein in the passage you quote, and it is well known that he favored a Fregean theory of "sense." In general terms, it is the kinship of Dummett's approach and Davidson's which is alluded to in the opening passage quoted in the question at the head of this thread--or so it seems to me.

Let me turn briefly to part of the Dummett which you quote:

Not only do we not know in the least how to set about devising a theory of meaning in conformity with this [Wittgensteinian] maxim, but it leads to that neglect of the difference between semantic and pragmatic considerations which I noted in the practice of the 'ordinary language' philosophers. Wittgenstein's deliberately unsystematic philosophical method makes it difficult to be certain what his intention was. Did he have in mind some theory of meaning of a completely different kind from that proposed by Frege? Or did he reject the whole idea of a systematic theory of meaning?

--end re-quotation

I have no doubt that "ordinary language philosophers" often neglected the needed distinction between semantic and pragmatic considerations. "What we say," and "what we don't say" depend on many factors outside the proper scope of semantics. That is reason for thinking, in fact, that the reasonable practice of empirical lexicography must be deeply informed by such a distinction; and that suggests, as well, that formulating definitions or descriptive accounts of word meanings is a theoretical enterprise which must abstract relevant usage from a good deal of concomitant "noise."  That would be to suggest that formulations of word meanings function to explain relevant usage. The account of the evidence available in usage would then be somewhat "theory-laden" and we begin, at that point to come in sight of more systematic approaches to linguistic meaning.

We have to question Wittgenstein's "deliberately unsystematic philosophical method" to go where Dummett (and Davidson) want us to go. On the other hand, however, I think that few have disputed, say, the private language argument to any deeper effect. 

H.G. Callaway