Re: One bete too many

On 2011-02-20, J.L.Speranza wrote:

> Grice mentions the many (12) beasts (betes noires,

> in French) but adds that they are ALL offspring (as it

> were) from Minimalism. And in "Reply to Richards" he

> provides an 'economical' argument against Minimalism _in

> toto_, rather than, alas, caring to deal with each bete

> noire independently.

Unfortunately PGRICE has been downgraded on Google books to snippet view, so I can't go back and look at this.

> --- I noted in my reply to Jones, elsewhere, that it's

> best to trace Grice's study ("Meaning") to his seminar

> material on Peirce.

I don't suppose there is anything online about this?

I found this:

Grice in the wake of Peirce

Ahti-Veikko Pietarinen

University of Helsinki

In which I discover Grice as an anti-psychologist (according to Pietarinen).

I had supposed Grice's talk about speaker's meaning, which seems to connect meaning with some mental act or event, to be a kind of psychologism.

But JL tells me:

> ... it is

> "Meaning" which attempts to reduce the 'semantic' to the

> psychological (via reductive rather than reductionist

> analysis).

So I am puzzled there about whether Grice's ideas about meaning were psychologistic.

Even if so there is possibly another reconciliation, apart from my attempt to differentiate the kinds of language at stake (ordinary versus formal scientific).

That is in the differences between semantics and meaning,

and between semantics and pragmatics.

In this I am tempted to say that Carnap's interest in semantics is primarily concerned with giving truth conditions to formal languages, and not concerned with explanation or even explication of the concept of "meaning".

And on the second point, that (as seems generally recognised) a large part of Grice's insights belong to pragmatics rather than semantics.

So now we have three differentiators on the table which might disarm possible conflict on semantics.

It seems to me that Grice comes closest to Carnap's concerns when he discusses "central meaning" (in his retrospective epilogue?), and in doing so closest to semantics, rather than pragmatics or in the analysis of "meaning" as a concept in ordinary language.

> It IS complex. Back in the day, Grice was THOUGHT as

> having providing the ultimate reduction of the

> 'semantic' to the psychological.

But perhaps we should not call this "semantic"?

> "This looks like tough ground for a constructive

> conversation but I think that when you think about the

> distinct interests of the two philosophers there is

> reason for the different attitudes, and possibilities

> for reconciliation appear. Frege's interest was

> specifically in the language of mathematics. Subsequent

> philosophers may have applied his ideas more widely, but

> Frege's anti-psychologism is to the best of my knowlege

> specific to the language of mathematics. It is

> mathematical concepts which he seeks to divest of the

> psychological elements which were common among

> mathematicians and philosophers of mathematics in his

> day. And in this he has been successful, almost all

> mathematicians today, and probably most philosophers

> have a completely non-psychological understanding of

> mathematical concepts."

>

> I would tend to think he was perhaps opposing, say, Mill,

> System of Logic. I've never read a more psychologist

> book than that! And of course, Mill thought that 2+2=4

> is syntethic a posteriori!

Mill certainly gets his attention, 18 mentions in the Grundlagen der Arithmetic.

> "Carnap's interest was in formal language for science,

> and the ideals of mathematical formulation and that of

> objectivity in science make the same kind of conception

> of semantics seem appropriate for languages in which

> scientific theories can be formalised. To this we might

> add, that these languages are intended for publication

> rather than discussion, they are not primarily spoken

> but written languages, and cannot depend upon the kind

> of contextual clues which may contribute significantly

> to oral discourse."

>

> This is ironic, in a sort of way, if one is reminded of

> the 'infamous' Vienna Circle. Apparently, they just

> gathered to TALK!

Yes. And Carnap's idea was that all philosophy was analytic and could be formalised.

But to understand this you have to distinguish a particular class of propositions which are the official doctrines of the discipline, and consider the metatheory to be specifically about those, and not about numerous other kinds of fact which the discipline deals with en passant (like who are the experts on particular topics and where they work, or discussions about methods, and a whole load of other things).

This is important for understanding the logicist thesis.

There are a lot of really wierd discussions which take place on FOM (for example) which involve completely unmathematical propositions which some mathematical logicians care about.

Also some mathematicians will tell you (and this may be most or even nearly all), that mathematics is mainly transmitted, taught, learned, discovered in (spoken) discussions between mathematicians, not through published literature or texts.

At the same time, there was at one time an apparent concensus that "the standard of proof" for mathematical propositions is provability in ZFC, even though exhibiting a formal proof is not required. You can sometimes see this kind of very precise criteria for what constitutes and established result in action. For example, Harvey Friedman, who is the principle proponent of the supplementation of ZFC by large cardinal axioms, is very careful (according to his own account) to describe a result which depends on a large cardinal axiom differently to one which can be proven in ZFC without large cardinals, and this he attributes to the editorial policy of the journals he publishes in.

>

> ---- Grice of course loved to TALK too. So there is a

> parallel between the unwritten, conversational,

> doctrines of the Vienna Circle (I never knew WHERE they

> met.

The difference is that he was interested in the informal conversation as subject matter, whereas the Circle were not,

and, at least in Carnap's case, he denied that his formal semantic methods were applicable to natural languages.

> A note about Grice's manoeuvres. I tend to think, and

> like to self-convince me about this, that the whole

> manoeuvre of the 'conversational implicature' was a

> technical point that Grice wanted to make about a

> certain outdated mode of doing philosophy.

Well that's certainly my impression, my very first impression on reading Grice was that he was addressing issues in the philosophy of Austin which I myself had found unsatisfactory.

> So, 'conversational implicature' is meant to

> elucidate an analysis of philosophical theses put

> forward by philosophers. He wasn't really, Grice wasn't,

> on the trivialities of "She hasn't been to prison yet".

> Rather, on say, why 'material implication' doesn't do,

> in Strawson's view, to mirror the behaviour of 'if'.

But surely the reason why one has to go into conversational implicature to elucidate the analysis is that the analysis is itself primarily an analysis of ordinary language?

The problems (from a certain point of view!) with ordinary language philosophy of the kind from which Grice was departing were of (at least) two distinct kinds.

The first kind, which are the ones which Grice was principally addressing were that as an analysis of ordinary language it was just wrong.

Some of the most glaring issues of this kind seem to originate with Wittgenstein, who seemed to think that if something was too obvious for anyone ever to say it then it couldn't be true.

When Grice sorts these out, he is slightly removed from the idiocy, for he really is sorting out philosophical mistakes rather than promulgating falsehoods about ordinary language.

(and of course, less idiotic, because closer to the truth).

But there is no point in doing this if the original philosophical investigations were not just wrong, but misdirected. If the entire enterprise was a misdirection of philosophical talent, then a fine-grained corrective analysis is equally misdirected.

Against this kind of critique Grice responds to defend the enterprise. As you would expect, a detailed critique betokens an acceptance of the general conception of philosophy, though it doesn't mean an agreement on which are the important problems to address within that conception.

> It is LINGUISTS and other scientists (so-called) -- e.g.

> psycholinguists, ethnomethodologists of conversation,

> etc. -- which naturally focus on what Grice's theory is

> obviosuly about. But recall that for Grice, the point

> was in the application. The fun, he said in "Further

> notes", not in the WoW reprint, is when one cashes out

> implicature for its philosophical price, as it were.

And what is its philosophical price?

Austin too writes as if the analysis of language were a preliminary to something more important, but leaves one wondering whether life is long enough to reach these applications.

I do think that Grice does put on a better impression of desiring to address some real philosophical problems.

On the Kripke stuff I think I will come back later, when I have something more substantial to offer.

Roger Jones

One bete too many

Jones writes:

"When last "the conversation" (between Grice and Carnap) through which Speranza and I were approaching the Eternal City was in progress, the principle avenue along which progress was envisaged lead Carnap down the gauntlet provided by Grice's numerous betes noires. My own difficulty in progressing this line was that for many of the betes we seemed to have little material from Grice to clarify exactly what it was he abhorred. This particularly difficult because it seemed that the central thread in these abhorrences was minimalisation, and that positivism, even the relatively giving positivism of Carnap, does involve a lot of this."

Yes. Grice mentions the many (12) beasts (betes noires, in French) but adds that they are ALL offspring (as it were) from Minimalism. And in "Reply to Richards" he provides an 'economical' argument against Minimalism _in toto_, rather than, alas, caring to deal with each bete noire independently.

Jones:

"My own pet thesis was that if one could probe Grice's mind one would find that what he abhorred was various dogmatic minimalisms, and that he might be persuaded to find some value in the more liberal pluralistic minimalisms in which Carnap mostly engaged. Making this thesis stick is not so easy, particularly in relation to minimalisms on which Grice wrote little. I have therefore in my more recently thoughts on this felt that other avenues might be more productive. There are of course the threads identified in the retrospective epilogue, whose primary merit is that they compiled with hindsight to reflect what he actually wrote about. There is one that seems to me at present to be of particular interest.
It is a topic which was very important to both, on which there is at first glance an irreconcilable gulf, but on which it seems to me that a conversation might well show that the gulf is illusory and that resolving the apparent conflict might be illuminating. This is "meaning"."

---- Yes. Jones was posting something on Russell Dale's history of 'meaning' (PhD disseration for New York Univ, under Schiffer) which relates.

--- I noted in my reply to Jones, elsewhere, that it's best to trace Grice's study ("Meaning") to his seminar material on Peirce. This in various pretty basic ways. Jonathan Bennett, on the other hand, in his "Linguistic Behaviour", makes the wrong connections.

He notes

1956 Grice/Strawson. In defense of a dogma
1957 Grice. Meaning

---- Bennett takes Grice's "Meaning" to be a reply to "In defense of a dogma". The latter was involved with the 'vicious' circle of semantic notions. And while G/S provide a behaviourist way out of the circle, it is "Meaning" which attempts to reduce the 'semantic' to the psychological (via reductive rather than reductionist analysis).

On the other hand, with S. R. Chapman, in her _Grice_, I now tend to think (it's good to change one's exegesis every now and then) that Grice was insulted by the overuse of technicisms in Peirce. If you have read Peirce you know what I mean: the icon, the symbol, the index, and what have you.

Grice thought it would be best to reduce all that to "proper" (i.e. ordinary) English, and speaks of

"... means..."

instead. The rest is history!

---- There is the connection with Stevenson, too, which may have bearing with Carnap. From Peirce we get to Morris. And Stevenson was involved in that group (Unity of Science). So, there was more like this 'complex' "research paradigm" springing form Peirce that Grice is considering. Carnap fits well in (vide his early, "Introduction to semantics", and his constant lip-servicing, if that's the word, pragmatics, as well -- and cfr. Jones and my research on Carnap's brief manifesto for pragmatics as the science of utterance, assertion, and belief, elsewhere).

Jones:

"The gulf has already been apparent between Speranza and I, for the kind of semantics which is of primary interest to Carnap, and to myself, and which comes down from Frege, is something which is completely divested of "psychologism", in which meanings have nothing whatever to do with minds or mental entities (unless minds are the subject matter of the language under analysis). Grice on the other hand builds his conception of semantics on speakers intentions (insofar as I understand it, which is not so much!)."

It IS complex. Back in the day, Grice was THOUGHT as having providing the ultimate reduction of the 'semantic' to the psychological. This was important for careful minds. E.g. Fodor (not one) would rather have a Language of Thought installed in one's brains. But a Griceian cannot just postulate a 'meaningful sentence' as the object of one's propositional attitude ("Jones believed that it is raining"). So, Grice would consider things like:

Jones uttered, "It is raining"
By uttering "It is raining", Jones meant that it was raining. I.e. By uttering "It is raining", Jones intended Speranza to believe that Jones believed that it was raining.

The odd thing is that this convinced Speranza thoroughly. ("I was disarmed by the inability of Speranza to check with the facts. He came to believe that I believed,
and intended Speranza to believe that I believed, that it was raining, by my merely having uttered, "It is raining". The odd implicature is that I never said _where_").

Jones:

"This looks like tough ground for a constructive conversation but I think that when you think about the distinct interests of the two philosophers there is reason for the different attitudes, and possibilities for reconciliation appear. Frege's interest was specifically in the language of mathematics. Subsequent philosophers may have applied his ideas more widely, but Frege's anti-psychologism is to the best of my knowlege specific to the language of mathematics. It is mathematical concepts which he seeks to divest of the psychological elements which were common among mathematicians and philosophers of mathematics in his day. And in this he has been successful, almost all mathematicians today, and probably most philosophers have a completely non-psychological understanding of mathematical concepts."

I would tend to think he was perhaps opposing, say, Mill, System of Logic. I've never read a more psychologist book than that! And of course, Mill thought that 2+2=4 is syntethic a posteriori! Grice would often joke with more grice to the mill. I once did a study (whose mimeo I must have somewhere and which I should scan, with H. Arlo Costa) on the mentalism (so-called) in Mill and Grice. I was amazed by the proto-Griceian views in Mill's analysis of syncategorematic particles like "however" and "but". A great author!

----

Jones:

"Carnap's interest was in formal language for science, and the ideals of mathematical formulation and that of objectivity in science make the same kind of conception of semantics seem appropriate for languages in which scientific theories can be formalised. To this we might add, that these languages are intended for publication rather than discussion, they are not primarily spoken but written languages, and cannot depend upon the kind of contextual clues which may contribute significantly to oral discourse."

This is ironic, in a sort of way, if one is reminded of the 'infamous' Vienna Circle. Apparently, they just gathered to TALK! Ayer enjoyed those conversations, even if his German was minimal. And so did who would later came out as a bit of a traitor to the whole Vienna Circle ideals: an American chap with a Manx surname (W. V. Orman).

---- Grice of course loved to TALK too. So there is a parallel between the unwritten, conversational, doctrines of the Vienna Circle (I never knew WHERE they met. I suppose the university. Odd: I would have selected a pub in downtown) and Grice's Play Group meetings.

Grice did meet in pubs. Notably "Lamb and Flag" and the more prestigious "Eagle and Child" (or "Bird and baby" as Grice called it). They are close enough to St. John's. The meetings were systematically on Saturday mornings. I would not have attended if I had been in the Oxford of the day, since, my last tutorial finished on Friday, I would have taken the FIRST TRAIN to London, and stay there till, say, Monday night, or whenever my next tutorial would be. Although perhaps Mondays I would have golfed?).

---

So, while there is this idea of _publication_, etc. (and Warnock recollects in his "Saturday mornings" how unBritish 'publication' was considered then. "We really didn't care to publish our views. We were proud of our parochialism") philosophy is indeed about the conversation -- 'endless conversation' as I think Richard Rorty has it. On the way to the City of Eternal Truth, of course.

Jones goes on:

"The relevant context is the literature, or some narrowly circumscribed part of it. Grice on the other hand, was of course trying to understand how ordinary spoken language works. In this case, it will much less often be the case that a sentence has some definite objective meaning. Add to this, Carnap's explicit denial that one could expect to deliver a formal semantics for natural languages. It is arguable therefore, that the different attitudes toward meaning are not in conflict, that they represent the different realities of distinct kinds of language."

A note about Grice's manoeuvres. I tend to think, and like to self-convince me about this, that the whole manoeuvre of the 'conversational implicature' was a technical point that Grice wanted to make about a certain outdated mode of doing philosophy. Recall he presented WoW:I-VI at Harvard 1967, and the Prolegomena all deal with specimens from:

Ryle
Austin
Wittgenstein
Strawson
Hart
Hare

and so on. So, 'conversational implicature' is meant to elucidate an analysis of philosophical theses put forward by philosophers. He wasn't really, Grice wasn't, on the trivialities of "She hasn't been to prison yet". Rather, on say, why 'material implication' doesn't do, in Strawson's view, to mirror the behaviour of 'if'.

It is LINGUISTS and other scientists (so-called) -- e.g. psycholinguists, ethnomethodologists of conversation, etc. -- which naturally focus on what Grice's theory is obviosuly about. But recall that for Grice, the point was in the application. The fun, he said in "Further notes", not in the WoW reprint, is when one cashes out implicature for its philosophical price, as it were.

---- It is true that in Oxford seminars between 1961 and 1967, on, boringly, "Logic and Conversation" he WAS BUILDING on a rationality-based account of minimal conversational exchanges. So a case can be made for a purely theoretical interest by Grice on this matters. But as a matter of history, it all started -- as he says in WoW:Retr. Ep., strand 6 -- out of Grice's discomfort with "later Wittgensteinian" rejections of sense-data ("The pillar box seems red to me, looks as if it is red; and, by Golly, this should surprise not even Witters, since the 'bloody' thing _is_ red!")

Jones:

"Furthermore, it seems to me now that if Carnap wished to continue his opposition to certain kinds of metaphysics, then a prime target would be the metaphysics of Kripke, and that because Kripke seems to suppose himself to be obtaining objective metaphysical truths in ordinary language without adopting any special terminology, especially without making use of specific philosophical notions of necessity (which is what Carnap does), then one part of an argument against this could be to deny that ordinary language does have any such concepts."

Kripke, oddly, can have a sense of humour. I pity the man after all the complications he was having with the American Philosophical Association (I'm not sure if Kripke does not call that Association a 'bete noire'). In his book, "Seas of Language", Dummett quotes from Kripke. "Seas of language". Kripke's quote refers to

"Socrates is called "Socrates""

--- see how deep the seas of language can go. Ditto, Grice refered to the 'berths' of language (Grice Club, search engine). So on a charitable view, one can see Grice's and Kripke's attempts to be such of reconstructing, say, 'logical form' out of its "pretty good guide" (in Russell's words): grammar.

----

Jones:

"In such an argument, the kind of attention to the realities of ordinary language which we find in Grice would be helpful. There is some similarity here with the contribution of Grice on the analytic/synthetic dichotomy, for Grice's contribution here with Strawson, is to the effect that no failure in precise enunciation of the distinction by philosophers can undermine the practical efficacy of related notions in ordinary language."

Indeed. And they DO provide a criterion: a behavioural one. Since they are working in the context of "ordinary language philosophy" the point is often not stressed strongly enough. But they DO propose:

"I just can't understand what the dickens you are talking about!" -- as an apt conversational reply to an analytically false 'conversational move' ("My neighbour's three-year child is an adult").

Versus:

"Incredible to some, but for some reason, I DO believe you!" (as an apt answer to the synthetically false, "My neighbour's three year child understands Russell's theory of types". Note that in this scenario, G/S do mention the case when such a child IS PRODUCED. "We must come to the conclusion that the proposition uttered WAS true." Whereas in the first case ("... is an adult") this would require a different 'conceptual scheme' or semantic interpreation of the basic predicates involved (e.g. "ADULT"), and so on.

Jones:

"Though this is the opposite way round, ordinary discourse providing support for a philosophical distinction."

Excellent point, that.

Recall that at the time, Grice and Strawson were hosting Quine at Oxford, so perhaps 'analytic' was TOO ordinary to be true!

Jones:

"On Kripkean metaphysics the aim would be to undermine the idea that this kind of metaphysics is other than a philosophical construction. Of course Grice might not care for such a campaign, even if it built on his insights!"

The other day I was checking O. Pooley's semantics page at the Oxford Univ site, and it's amazing the time he dedicates in his seminar to ... Kripke! It's not just Kripke-Kripke, but minor Kripke like Kripke on 'semantic reference' and 'speaker reference'. I have work by Patton and Stampe refuting Kripke's early misuse of Grice. It's amazing how it developed into something Grice also found interest, in "Vacuous Names", when he considers Identificatory versus Non-Identificatory Reference -- Kripke's line being put forward also by Donnellan with his referential/attribute distinction.

So, yes, the Kripke-Grice interface is worth considering! Whatever the results!

Bete's and the conversation

When last "the conversation" (between Grice and Carnap) through which Speranza and I were approaching the Eternal City was in progress, the principle avenue along which progress was envisaged lead Carnap down the gauntlet provided by Grice's numerous betes noires.
My own difficulty in progressing this line was that for many of the betes we seemed to have little material from Grice to clarify exactly what it was he abhorred.

This particularly difficult because it seemed that the central thread in these abhorrences was minimalisation, and that positivism, even the relatively giving positivism of Carnap, does involve a lot of this.
My own pet thesis was that if one could probe Grice's mind one would find that what he abhorred was various dogmatic minimalisms, and that he might be persuaded to find some value in the more liberal pluralistic minimalisms in which Carnap mostly engaged.  Making this thesis stick is not so easy, particularly in relation to minimalisms on which Grice wrote little.

I have therefore in my more recently thoughts on this felt that other avenues might be more productive.  There are of course the threads identified in the retrospective epilogue, whose primary merit is that they compiled with hindsight to reflect what he actually wrote about.

There is one that seems to me at present to be of particular interest.
It is a topic which was very important to both, on which there is at first glance an irreconcilable gulf, but on which it seems to me that a conversation might well show that the gulf is illusory and that resolving the apparent conflict might be illuminating.

This is "meaning".

The gulf has already been apparent between Speranza and I, for the kind of semantics which is of primary interest to Carnap, and to myself, and which comes down from Frege, is something which is completely divested of "psychologism", in which meanings have nothing whatever to do with minds or mental entities (unless minds are the subject matter of the language under analysis).
Grice on the other hand builds his conception of semantics on speakers intentions (insofar as I understand it, which is not so much!).

This looks like tough ground for a constructive conversation but I think that when you think about the distinct interests of the two philosophers there is reason for the different attitudes, and possibilities for reconciliation appear.

Frege's interest was specifically in the language of mathematics.  Subsequent philosophers may have applied his ideas more widely, but Frege's anti-psychologism is to the best of my knowlege specific to the language of mathematics.  It is mathematical concepts which he seeks to divest of the psychological elements which were common among mathematicians and philosophers of mathematics in his day.   And in this he has been successful, almost all mathematicians today, and probably most philosophers have a completely non-psychological understanding of mathematical concepts.
Carnap's interest was in formal language for science, and the ideals of mathematical formulation and that of objectivity in science make the same kind of conception of semantics seem appropriate for languages in which scientific theories can be formalised.  To this we might add, that these languages are intended for publication rather than discussion, they are not primarily spoken but written languages, and cannot depend upon the kind of contextual clues which may contribute significantly to oral discourse. The relevant context is the literature, or some narrowly circumscribed part of it.

Grice on the other hand, was of course trying to understand how ordinary spoken language works.  In this case, it will much less often be the case that a sentence has some definite objective meaning.

Add to this, Carnap's explicit denial that one could expect to deliver a formal semantics for natural languages.

It is arguable therefore, that the different attitudes toward meaning are not in conflict, that they represent the different realities of distinct kinds of language.
Furthermore, it seems to me now that if Carnap wished to continue his opposition to certain kinds of metaphysics, then a prime target would be the metaphysics of Kripke, and that because Kripke seems to suppose himself to be obtaining objective metaphysical truths in ordinary language without adopting any special terminology, especially without making use of specific philosophical notions of necessity (which is what Carnap does), then one part of an argument against this could be to deny that ordinary language does have any such concepts.  In such an argument, the kind of attention to the realities of ordinary language which we find in Grice would be helpful.
There is some similarity here with the contribution of Grice on the analytic/synthetic dichotomy, for Grice's contribution here with Strawson, is to the effect that no failure in precise enunciation of the distinction by philosophers can undermine the practical efficacy of related notions in ordinary language.   Though this is the opposite way round, ordinary discourse providing support for a philosophical distinction.  On Kripkean metaphysics the aim would be to undermine the idea that this kind of metaphysics is other than a philosophical construction.
Of course Grice might not care for such a campaign, even if it built on his insights!

Roger Jones