Leiter Reports: A Philosophy Blog

News and views about philosophy, the academic profession, academic freedom, intellectual culture, and other topics. The world’s most popular philosophy blog, since 2003.

  1. Fool's avatar
  2. Santa Monica's avatar
  3. Charles Bakker's avatar
  4. Matty Silverstein's avatar
  5. Jason's avatar
  6. Nathan Meyvis's avatar
  7. Stefan Sciaraffa's avatar

    The McMaster Department of Philosophy has now put together the following notice commemorating Barry: Barry Allen: A Philosophical Life Barry…

Peter Hacker Did Not Like Timothy Williamson’s Book

Brian Leiter

Via Weatherson, I see that the irascible (and sometimes idiosyncratic and dogmatic [cf. paragraph 7]) Peter Hacker has a rather savage and critical review of Williamson's The Philosophy of Philosophy.  In the hopes of clarifying what's really at issue here, I thought I would single out a substantive criticism from Hacker's review and invite reader comments on its merits.  This is from p. 343 of the review:

Having shown to his satisfaction that philosophical truths are not generally about words or concepts, Williamson queries how philosophy might nevertheless still be an armchair activity that aims at conceptual truths. Since confinement to an armchair does not deprive one of one's linguistic competence, perhaps conceptual truths are those that can be achieved merely through reflection on that competence. This might be so, he writes (pp. 50–1), if all, or all core, philosophical truths were analytic in some sense which imposed no constraints upon the world and hence could be known from the depths of an armchair. Williamson suggests that this view was embraced by those analytic philosophers who believed that philosophical truths are linguistic or conceptual. But this is demonstrably false. Among Oxford philosophers who took 'the linguistic turn', the only significant one who thought that all philosophical propositions are analytic was Ayer (at the age of 26). The manifesto of the Vienna Circle followed Wittgenstein in denying that there are any philosophical propositions. Ryle, Austin, Strawson and others did think there are, but nowhere suggested that they are analytic. All insisted that philosophy is a conceptual investigation, but none held that its task is to disclose analytic truths. It is therefore astonishing that Williamson decides to use 'analytic' and 'conceptual' interchangeably (p. 50). So conceptual truths are analytic, according to Williamson. This is not only historically unwarranted, it is also arguably philosophically misconceived. Such philosophical assertions as 'Idealism and materialism are both answers to an improper question' (Ryle), 'Material objects and persons are the basic particulars of our conceptual scheme' (Strawson), or 'There can be no such thing as a "private language" ' (Wittgenstein), are not analytic, and their proponents did not hold them to be. But they are conceptual truths.

Thoughts from readers?  Critical reactions must be signed with a full name and matching e-mail address.

Leave a Reply

Your email address will not be published. Required fields are marked *

25 responses to “Peter Hacker Did Not Like Timothy Williamson’s Book”

  1. Robert Johnson

    "Conceptual but not analytic". What the hell else could they be?

    Reply
  2. Jon Cogburn

    Re: Robert Johnson- Synthetic a-priori? To be fair, one could (and Friedman does) argue that we saw how that worked out in Germany almost a century ago during the glory days of return-to-Kant.

    I forget which philosophy of mind book I read that had a footnote savaging Hacker on this very point. The author (I just remember that it was a really great book with a blue cover by a Canadian and that it had two chapters on Dennett and fantastic discussion of Higher Order Thought theories) can't understand how the methodology of latter day Wittgensteinians such as Hacker and D.Z. Phillips get out of the good arguments that analyticity cannot do the job of "necessity." The strongest of these arguments have resonance far outside of the Positivists' attempt to do away with the synthetic a priori.

    Has anyone provided at least a prima facie justification of Wittgensteinian appeals to "grammar" in light of the really compelling arguments that the Putnam of "Meaning Holism," and "Representation and Reality" perhaps over-generously attributes to Quine? Or the equally unassailable arguments one can abstract from Mark Wilson's "Predicate Meets Property" and Stephen Stich's "Deconstructing the Mind"?

    I own but have yet to read Hacker and Bennett's mammoth Philosophical Foundations of Neuroscience. From paging through it, it sure doesn't look like in actual practice he stays firmly ensconced in armchair.

    Reply
  3. Sam Rickless

    For most non-Wittgensteinian working philosophers, the phrases "conceptual truth" and "analytic truth" are synonyms. But not for Hacker, whose take on all this is inspired by Wittgenstein's distinction between "grammatical" and "non-grammatical" truths. Hacker thinks that there are two kinds of conceptual truths: logical and mathematical truths on the one hand, and grammatical truths on the other. Grammatical truths can be both analytic and non-analytic. Importantly, grammatical truths are not propositions, but expressions of rules for the use of linguistic expressions. They are therefore in some sense "antecedent" to truth, and so neither true nor false. We can "say" that a grammatical truth (such as "material objects and persons are the basic particulars of our conceptual scheme" or "red is darker than pink") is true, but in saying this, all that we are saying is that the relevant truth is the expression of a rule for the use of a proposition or set of linguistic expressions. When Hacker says that it is "historically unwarranted" and "arguably philosophically misconceived" that all conceptual truths are analytic, his use of the phrase "conceptual truth" is governed by a Wittgensteinian rule for the use of that phrase. I suspect that when Williamson says (or takes for granted) that all conceptual truths are analytic, his use of the phrase "conceptual truth" is not governed by such a rule. So I suspect that Hacker and Williamson are talking past each other.

    Reply
  4. Brian Leiter

    Sam, thank you, that was very helpful, and makes sense of a number of puzzling things in the review essay.

    Reply
  5. Brian Weatherson

    That is very helpful Sam, though I sort of want to stick up for Hacker on one point.

    One of Hacker's points is that Williamson has misinterpreted the mid-century philosophers, and so his refutations of them have no bite. I think whatever we want to say on the conceptual point, he's got to be right on the historical point.

    Saying that the ordinary language philosophers made too few distinctions between modal concepts (because they hadn't read Kripke) seems to me to be a simple historical error. Perhaps after Kripke we have a better idea of what the synthetic a priori has to look like, and perhaps the Kantian and/or Wittgensteinian notions of synthetic a priori aren't ultimately defensible. But that doesn't mean that the mid-century philosopers didn't make the distinction between analytic and a priori; at worst they made it badly.

    Reply
  6. mark lance

    One thing to add to what Sam says above is that for WIttgenstein the distinction between grammatical and non-grammatical is contextual. ("What at one moment is an observed concomitant of experience is at another used to define it" – roughly; don't have the text in front of me at the moment.) That is, as I argued in a couple papers and developed in my book with Hawthorne, his view is that there's a distinction between principles that determine how language is to be used, and claims within the language, but that as we learn new facts about the world, this line changes. That is, the grammatical/empirical distinction is empirical. So while there is plenty of room for "investigation of concepts" or better investigation of the rules of language, this is not at all "arm-chair" in the sense of a priori. It isn't something one can think intelligently about in abstraction from empirical knowledge.

    There are many views about the nature of language that allow for a distinction between truths about the function of language and empirical truths about the world without buying into the classical notion of analyticity.

    Reply
  7. Jon Cogburn

    The book I mentioned above that critiques the appeal to "grammar" as an attempt to dissolve the mind-body problem is William Seager's "Theories of Consciousness," (Taylor and Francis, 2007).

    A really first-rate critique of D.Z. Phillips' use of "grammar" to motivate a non-cognitivist construal of religious discourse is the paper, "Religious Language Games," by Graham Oppy and Nick Trakakis (essay 7 in Andrew Moore's and Michael Scott's "Realism and Religion: Philosophical and Theological Perspectives," (Ashgate, 2007))

    I haven't read that book of Williamson's so I don't know how justified Hacker's criticisms are, but to extend Weatherson's point in a different direction, we probably should be wary of too much Kripkean triumphalism. (1) Even if the problems of necessity are taken to be wrapped up, Hume's wakeup call about modality is only one part of what motivated Kant. With the notable exceptions of John MacDowell and Graham Priest, contemporary analytic philosophy hasn't seemed that interested in addressing the other parts. One might worry that the narrative of logical positivists as linguistic neo-Kantians who were ultimately defeated by Kripke is a bad narrative because we end up not worrying about all the other reasons (starting with the arguments of the Dialectic) not having to do with necessity per se that lead to forms of transcendental idealism. [Note: Dummett's whole "realism/anti-realism" brief was motivated by worry about this other stuff, and he is best understood as trying to extend the machinery of logical positivism to answer them as well. And from MacDowell you can interpret the late Davidson and Kripke-Wittgenstein as motivated by the other issues too.] (2) In standard possible worlds you interpret the model theory as holding the meaning of the non-logical terms fixed and change the worlds to get truth assignments [the fact that two-dimensionalism adds another axis does not alter this]. But the Quinean arguments (cited above) against positivism render the notion of "holding meaning fixed" in modal semantics very problematic in any case. Of course there is no problem as far as doing things to get soundness and completeness proofs, but appealing to such frameworks to justify a priori philosophical methodology seems completely unsupportable to me. I think the best book containing relevant discussion of this issue is still Gendler and Hawthorne's anthology, "Conceivability and Possibility," (Oxford, 2002).

    Reply
  8. Mark Silcox

    One of the things that I've noticed over my past fifteen years or so in the biz is that the expression "conceptual truth" is often used as _code_ by philosophers who lack self-confidence about their background in the Phil. of Language and want to call something an analytic truth, but have heard vague news that Quine somehow forbade us from doing that. So maybe that's the trend that Williamson is tapping into. If so, then Hacker makes a perfectly legitimate criticism by suggesting that Williamson is using the term in a historically illegitimate way.

    Reply
  9. robert allen

    'Material objects and persons are the basic particulars of our conceptual scheme' (Strawson)

    If philosophy is merely the study of our conceptual scheme, rather than mind-independent reality, then Plato, Aristotle, their Medieval followers, and even Descartes, who took themselves to be articulating deep and important truths about how things are objectively, were simply benighted. I find that VERY hard to believe, not to mention quite a letdown. Yeah, yeah I suppose someone has to address the concerns of the Prussian genuis and the Akronite- but for my part I am going to proceed as if they were (profoundly) mistaken. I wasted enough time already writing a dissertation on that other destroyer of philosophy, the later Wittgenstein.

    Reply
  10. David Papineau

    Backing up Mark Lance, isn't the basic point that someone who says that non-analytic claims can be known a priori owes some account of how this is possible? True, there are various plausible accounts of this kind for special subject matters. But what doesn't look like a plausible account is that a priori knowledge can fall out of the grammatical structure of our language. If our language commits us to substantial non-analytic claims about the world, then the obvious query is what warrants us in adopting this language rather than one with different commitments. To ignore this worry just looks like conservative complacency about what may well be historically accidental features of some particular language.

    Reply
  11. Jonathan Ichikawa

    Williamson seems aware that he's tapping into the trend described by Mark Silcox. He writes (p. 50):

    'Notoriously, the idea of analyticity has been under a cloud ever since Quine argued that "a boundary between analytic and synthetic statements simply has not been drawn". Nevertheless, the idea is still active in contemporary philosophy, often under the less provocative guise of "conceptual truth." The terms "analytic" and "conceptual" will henceforth be used interchangeably.'

    While it may be that this use ignores important distinctions in the history of analytic philosophy, I'm not sure that it's a fair criticism of Williamson to complain that this is so. This for the relatively banal reason that Williamson is not, and is not pretending, to produce a faithful contribution to the history of philosophy. He's describing a view that he thinks is currently active.

    Reply
  12. N. N.

    David Papineau asks,

    "If our language commits us to substantial non-analytic claims about the world, then the obvious query is what warrants us in adopting this language rather than one with different commitments."

    For Wittgenstein (as Hacker explains him), grammatical rules cannot be justified by reference to reality. Why not? Hacker gives several reasons in Wittgenstein: Mind and Will:

    "(ii) Justifying grammar by reference to the facts leads to an infinite regress. Any attempt to justify grammatical rules by reference to how things are in reality must employ a language in giving that justification. The form of words that purports to justify a rule of grammar must itself have a grammar. If it has the same grammar as that which it purports to justify, the justification begs the question. If it has a different grammar, then (a) it determines different concepts and so cannot be about the same thing, hence is irrelevant; and (b) that grammar too will stand in need of justification. So any attempt at grounding grammar is reality will launch us upon an infinite regress of justifications.
    […]
    (iv) No description of reality can justify grammar. Any attempt to justify grammar by reference to reality must take the form of a grammatically licit description of how things are. Such a description is given by a proposition with a sense. Consequently its negation too must make sense, for the negation of a proposition with sense, which describes how things are, is itself a proposition with sense. But for such a proposition to justify a grammatical rule which delimits the sense of sentences and excludes nonsensical forms of words, the negation of the justifying description would have to be nonsense, not a falsehood. This has two corollaries, both of which were discussed by Wittgenstein.
    (a) If it were possible to justify grammatical rules by reference to reality, those rules would be superfluous. One cannot say that a grammatical rule is made necessary by certain properties of things: e.g. the rule that excludes the words 'transparent white' or 'flashing black' cannot be justified by saying that white is not transparent or that black is not radiant. For if one could say this, then it would make sense, even though it would be false, to say that this white glass is transparent or that the traffic lights flashed black. But then the grammatical rule would be superfluous, since what it does is precisely to exclude such forms of words as nonsense.
    (b) Any justification of grammar by reference to reality requires the possibility of describing a reality that would not justifify that grammar. A justifcation of our grammar by reference to reality should, it seems, take the form of saying that since reality is thus-and-so, the rules of grammar must be such-and-such. But one must then also be able to say that if reality were otherwise, then the rules of grammar would have to be different. However, one cannot sensibly say how reality would have to be in order for a different grammar to be justified. For in order to describe such a different reality, one would have to use the very combinations of words which our existing grammar excludes, i.e. one would have to talk nonsense. But if something counts as nonsense in the grammar which is to be justified, it cannot at the same time pass for sense in the grammar of the propositions purporting to justify it." (p.221-2)

    Reply
  13. John Koethe

    I'm puzzled by the assumption that many seem to be making that when Hacker speaks of conceptual questions whose answers aren't analytic he must have Wittgenstein's "grammatical" questions in mind, so that this is really a dispute over what one thinks of Wittgenstein. Hacker is surely right about conceptual vs. analytic claims. Just to take two recent topics of our faculty reading groups–topics whose connection with Wittgenstein is remote–as examples, the question of whether there can be absolute generality, i.e. quantification over everything, is obviously a conceptual one (what empirical investigation could possibly bear on it?) whose answer isn't analytic in any recognizable sense; and the same is true of the question of whether Mary, in Jackson's well-known argument, learns about some new nonphysical property when she first sees a rose, instead of some physical property she already knew about under a new guise. Indeed, though I haven't read Williamson's book and don't know how accurately Hacker represents it, it is hard to see how Williamson himself could actually hold the view Hacker attributes to him, that all conceptual truths are analytic, given his well-know and controversial position on vagueness, which he arrives at via a conceptual investigation, but which he surely doesn't regard as analytic in any familiar sense.

    Reply
  14. mark lance

    I'm not too taken with the arguments NN produces here on Hacker's behalf. re ii: the fact that grammar g is used in making the claims that justify the claim that "it is best to use grammar g" is not in any way viciously circular. And the second "horn" would only imply triviality if the goal was a justification of all aspects of grammar. But it certainly does not follow that one cannot justify any given bit of grammar in any particular case in which there is a reason to ask for one. (Wittgenstein is certainly not a classical foundationalist.)

    iv: What is meant by "justifying a rule of grammar – a rule that delimits sense from nonsense". Clearly the issue is justifying the claim that "grammar g is the best – in whatever sense of best one thinks relevant – grammar". But there is no reason at all to think that the negation of a claim justifying this would be nonsense.
    iva: really the same point – to "justify the rules" is to justify the claim that these are the right (or best) rules for the task at hand, say empirical description of a given phenomenon. The fact that reality is certain ways is obviously relevant. for a nice concrete example, consider Nancy Cartwright's discussion of specific conditions of systems that need to be in place for it to be possible to find useful laws governing the functioning of those systems. The rules that define the role "law-like" statement and institute it as part of the language one uses to describe the system, are justified by appeal to the existence of conditions that will make this work out – that is will make this lead to a successful theoretical endeavor.
    ivb: again, this simply assumes that if we are going to justify any bit of grammar, we are going to justify it all, at once. (Remember: 'grammar' here is a technical term that is much broader than standard linguistic uses of 'grammar'.) So, to pick one example, think of the question of whether "race" is a biological or an ethnographic concept. That is paradigmatic of the sort of thing LW means by a grammatical fact – the structuring rules that give sense to claims about 'race'. It is, I would claim, not biological in our language, but it is certainly easy to describe a world in which there would be a genuinely biological concept of 'race' – there are actually definable sub-species of humans, all rational, but with relatively stable sets of relevantly distinct characteristics, marked by external physiology, along with clear genetic autonomy of the groups – and the fact that our world isn't like this is what justifies us in treating 'race' as a non-biological, term with a particular sort of performative force in constituting a social distinction.

    Reply
  15. Constantine Sandis

    Hacker’s point, surely, is that for someone who champions rigour and precision much of this work was surprisingly shoddy, especially the history of philosophy which dominates the opening chapters. Hacker and Williamson are not merely talking past one another. It is simply false to think that there is such a thing as THE analytic/synthetic distinction. Frege’s conception of analyticity, for example, is not extensionally equivalent to Carnap’s or Bolzano’s (let alone Kant’s). One does not need to agree with Hacker’s own views on grammatical propositions to see this. Indeed, Hacker does not even mention (let alone appeal to) these views in his review. Williamson, by contrast, mischaracterizes ‘Wittgensteinians and others’ for (allegedly) holding that definitions, stipulations, and rules grammar are ‘analytic truths’ (p. 54). Yet, as Hacker has noted, in 18,000 pages of manuscripts Wittgenstein only uses the term 'analytic' once, and that in order to suggest that mathematical propositions are not analytic. Nor do Wittgensteinians maintain that conceptual truths such as ‘there can be no such thing as a private language’ are analytic, anymore than ‘red is darker than pink’ or ‘2 + 2 =4” is analytic (on any understanding of the term).
    When such mischaracterizations are repeatedly appealed to (either implicitly or explicitly) in arguments against a variety of ways of doing philosophy, poor history of philosophy leads to irrelevant arguments (against straw men) that prove very little. Williamson ambiguously describes certain conceptions of correct philosophical enquiry as maintaining that it is ‘about’ thought, concepts, or language (rather than about the world) and that all philosophical truths are either linguistic or conceptual. But none of the philosophers which Williamson mentions thought any such thing (not that Williamson takes the time to clarify what he means by ‘about’ here – perhaps because that would be devoting time and energy to conceptual clarification). When, for the briefest moment, he considers the view that linguistic usage might tell us something about the world he quickly dismisses it merely by objecting to Dummett’s highly idiosyncratic account of the philosophical role of semantic function (p.20). The more commonplace idea that clarifying a particular concept could help one to better understand the nature of whatever it is a concept of is not even mentioned.

    Reply
  16. N. N.

    Mark Lance: "What is meant by 'justifying a rule of grammar – a rule that delimits sense from nonsense.'

    For Hacker's Wittgenstein, a rule of grammar – e.g., 'red is darker than pink' – determines what it makes sense to say. A justication of 'red is darker than pink' by reference to reality amounts to the claim that the rule is true. However, if it makes sense to say that it is true, it must also make sense to say that it is false. That is, it must make sense to say 'red is not darker than pink.' The latter sentence violates the grammatical rule, and therefore, by Wittgenstein's lights, is not false but nonsensical.

    Constantine Sandis: "Nor do Wittgensteinians maintain that conceptual truths such as 'there can be no such thing as a private language' are analytic, anymore than 'red is darker than pink' or '2 + 2 = 4' is analytic (on any understanding of the term)."

    I would add that even standard examples of analytic truths (e.g., 'Bachelors are unmarried men') are not considered analytic by Wittgenstein, if by 'analytic' is meant 'true in virtue of the meanings of the terms.' As Hacker interprets him, Wittgenstein takes such statements as partly constituting the meaning of the terms. Like mathematical 'propositions,' they are transformation rules for empirical propositions. The rule 'Bachelors are unmarried men' licenses the transition from 'John is a bachelor' to 'John is an unmarried man.'

    Reply
  17. mark lance

    NN: A justication of 'red is darker than pink' by reference to reality amounts to the claim that the rule is true.

    N.N.:
    It is easy to refute a view if one caricatures it. By saying that "red is darker than pink" is a rule, you are already ruling out the idea that it can be true. If the claim is only that "we can't coherently treat this as no different from a standard empirical claim and at the same time as a rule structuring language" then one hardly needs to go through long arguments. That much is obvious and something that no one ever denied.

    The only interesting question is whether we can know a priori that a given rule is the correct one to use in structuring our language so as to make it maximally empirically adequate. Is it just arbitrary decisions, or maybe decisions subject to non-empirical "pragmatic" standards, or is the choice of such rules itself subject to empirical criticism? So the question is, as I said in the earlier post, whether 'rule R is the best one to use in structuring language' can be empirically justified. That it can be is not merely true, it is something that Wittgenstein insists on. He says very clearly in numerous places, cf PI 560-70 for example, that it is not an arbitrary or undebatable matter what the rules of the language-game are.

    Reply
  18. Robert Allen

    "If our language commits us to substantial non-analytic claims about the world, then the obvious query is what warrants us in adopting this language rather than one with different commitments."
    David Papineau

    I used to push Wittgenstein’s neo-Kantian line about grammar until I discovered the following Aristotelian methodology. I begin with an intuition, say that persons are real and endure or choose freely. (If you ask me to justify this intuition, all I can say is something like, ‘Look, that is just what every sane person believes and those who profess otherwise at least live as if it’s true’. A fellow scholar once got me to re-think my view on endurance simply by remarking that it was a “crazy metaphysic.”) I canvass the various possible definitions of it: Locke’s, Hobbes’, Geach’s, and Chisolm’s (e.g.) in the case of the former, Hobbes’, Kane’s, and Chisolm’s (e.g.) in the case of the latter. I then select and defend the one that leaves the intuition intact. (Chisolm being the current winner in both cases.) It seems to me that Kant’s entire philosophy is based on conflating distortion and representation.

    Reply
  19. N. N.

    Mark Lance,

    It's unclear (to me) what is meant by 'empirically adequate,' 'subject to empirical criticism' and 'empirically justified.' Presumably (given your comment about the difference between grammatical rules and "standard empirical claims"), you don't mean that the rule is known to be true a posteriori.

    Can a rule be 'justified' by "non-empirical 'pragmatic' standards"? For Hacker, there is something that could be called 'pragmatic justification.' As he puts it (in a passage just after the one I quoted), "a system of rules which is simple, convenient, and easily taken in is pragmatically justified. One can give reasons why it is preferable to alternatives. But the justifying reasons do not make the rules right or correct. They make them useful."

    In §§ 560-70 of the Investigations, Wittgenstein does say that some rules 'have a point' while others do not. Some rules are 'essential' parts of the game while others are not. But what this means is that some rules are logically related to others while others are not. Whether we use the king or the queen in the practice of deciding who gets to move first in a game of chess has no bearing on the rest of the rules of the game, and therefore, has no bearing on the rest of the practice of playing chess. On the other hand, whether the king can move one space or eight does have a bearing on the rest of the practice of the game. In this sense, a rule can be 'essential' — to play a game in which the king can move eight spaces is to play a game other than chess.

    Consider your example concerning the concept 'race.' If our concept of 'race' is essentially (in the above sense) not biological, then I would deny that "it is certainly easy to describe a world in which there would be a genuinely biological concept of 'race.'" We could play a different game with the word 'race,' but it's not obvious what relations the different game would have to the one we actually play.

    You write, "the fact that our world isn't like this is what justifies us in [adopting a grammatical rule]." This appears to be a justification by reference to reality. It seems to me that all facts 'about the world' are, in the relevant sense, grammatical facts. That is, the concepts that we employ in any description of the world are given by the rules of a particular game of describing. We can attempt to 'justify' a form of description by relating it to another form of description in a different game, but in doing so we have not exited language.

    Reply
  20. Constantine Sandis

    Certainly Wittgenstein would not (under any interpretation) call 'Bachelors are unmarried men' an analytic truth. I agree with Mark Lance that many rules of language, on Wittgenstein's view, are not accidental (and that we can make empirical enquiries here). But one need not intepret Wittgenstein this way or that here (or related issues such as whether or not grammatical rules can be justified by reference to reality or whether it is meaningless to claim that a rule is true or false) in order to evaluate Williamson's book or Hacker's review of it. Nothing Hacker says in his review relies on his view of such matters, his claim is that Williamson caricatures a host of philosophers including Wittgenstein, Ryle, Strawson, and Austin when he attrributes to them the view that conceptual truths are analytic. This is false on any received sense of analyticity and incompatible with numerous competing interpretations of the conceptual claims made by the aforementioned philosopher(interpretations which Williamson ignores…indeed he does not consider any).

    Reply
  21. Robert Allen

    "You write, "the fact that our world isn't like this is what justifies us in [adopting a grammatical rule]." This appears to be a justification by reference to reality. It seems to me that all facts 'about the world' are, in the relevant sense, grammatical facts. That is, the concepts that we employ in any description of the world are given by the rules of a particular game of describing. We can attempt to 'justify' a form of description by relating it to another form of description in a different game, but in doing so we have not exited language."

    No, when the researchers involved in the Human Genome Project discovered that there were no 'race genes' that established that the concept of race does not represent reality, but distorts it. That form of description is now unjustified, for it required genetic differences that we now know do not exist. This is not a linguistic fact; the researchers "exited" language.

    Reply
  22. mark lance

    NN: REsponses throughout in caps
    It's unclear (to me) what is meant by 'empirically adequate,' 'subject to empirical criticism' and 'empirically justified.'
    WELL, I CERTAINLY DON'T THINK ANYTHING I'M SAYING HERE REQUIRES A THEORY OF EMPIRICAL JUSTIFICATION. FOR PURPOSES OF THIS ARGUMENT, I MEAN WHATEVER ANY REASONABLE PERSON MEANS BY THIS. NOTHING I SAY RELIES ON ANY CONTENTIOUS VIEWS OF WHAT EMPIRICAL JUSTIFICATION IS, JUST THAT IT EXISTS.
    Presumably (given your comment about the difference between grammatical rules and "standard empirical claims"), you don't mean that the rule is known to be true a posteriori.
    WHY EVER WOULD YOU SUPPOSE THAT? THERE ARE MANY VIEWS ACCORDING TO WHICH DIFFERENT BITS OF LANGUAGE CAN PERFORM DIFFERENT FUNCTIONS, YET THE WHOLE BE SUBJECT TO EMPIRICAL REFUTATION. QUINE IS PROBABLY THE MOST FAMILIAR HERE. I HAVE MY OWN SPECIFIC VIEWS ON THIS, IN MY RECENT BOOK WITH REBECCA KUKLA, AND HAD SLIGHTLY DIFFERENT VIEWS IN AN EARLIER BOOK WITH JOHN HAWTHORNE, BUT NONE OF THAT IS REQUIRED HERE. THINKING THAT RULES PLAY A DIFFERENT ROLE FROM STANDARD EMPIRICAL DESCRIPTIONS JUST DOESN'T IMPLY THAT THERE CAN'T BE EMPIRICAL REASONS FOR ACCEPTING OR REJECTING THEM.

    Can a rule be 'justified' by "non-empirical 'pragmatic' standards"? For Hacker, there is something that could be called 'pragmatic justification.' As he puts it (in a passage just after the one I quoted), "a system of rules which is simple, convenient, and easily taken in is pragmatically justified. One can give reasons why it is preferable to alternatives. But the justifying reasons do not make the rules right or correct. They make them useful."
    YES, OF COURSE. THIS VIEW IS FAMILIAR. CARNAP HELD IT FOR A SHORT WHILE. THAT'S WHY I MENTIONED IT AS AN OPTION. THE QUESTION IS WHAT ARGUMENT THERE IS FOR CLAIMING THIS SEPARATION – THE RULES ARE ONLY "USEFUL" BUT THE OTHER CLAIMS ARE "EMPIRICALLY JUSTIFIED". MY CLAIM IS THAT THERE IS NO REASONABLE ACCOUNT OF THE EMPIRICAL JUSTIFICATION/MERELY USEFUL DISTINCTION THAT MAKES THIS COME OUT TRUE.

    In §§ 560-70 of the Investigations, Wittgenstein does say that some rules 'have a point' while others do not. Some rules are 'essential' parts of the game while others are not. But what this means is that some rules are logically related to others while others are not. Whether we use the king or the queen in the practice of deciding who gets to move first in a game of chess has no bearing on the rest of the rules of the game, and therefore, has no bearing on the rest of the practice of playing chess. On the other hand, whether the king can move one space or eight does have a bearing on the rest of the practice of the game. In this sense, a rule can be 'essential' — to play a game in which the king can move eight spaces is to play a game other than chess.
    BUT THAT IS NOT ALL HE SAYS IN THE PASSAGE. "569: LANGUAGE IS AN INSTRUMENT, ITS CONCEPTS ARE INSTRUMENTS. NOW PERHAPS ONE THINKS IT MAKES NO GREAT DIFFERENCE WHICH CONCEPTS WE EMPLOY. AS AFTER ALL IT IS POSSIBLE TO DO PHYSICS IN FEET AND INCHES AS WELL AS IN METRES AND CENTIMETRES. BUT EVEN THIS IS NOT TRUE [OBVIOUSLY IMPLYING THAT THE IMPORT IS GREATER IN MOST CASES] IF FOR INSTANCE CALCULATION IN SOME SYSTEM OF MEASUREMENT DEMAND MORE TIME AND TROUBLE THAN IT IS POSSIBLE TO GIVE." IT IS POSSIBLE TO FOCUS JUST ON THE EXTREMELY TRIVIAL EXAMPLE AND THINK HE MEANS USEFULNESS IN SOME WAY THAT IS SEPARABLE FROM EMPIRICAL IMPLICATIONS, BUT THOUGH I'M NOT GOING TO GO THROUGH THE WHOLE PI, I DON'T THINK THAT READING IS DEFENSIBLE OVER-ALL.

    Consider your example concerning the concept 'race.' If our concept of 'race' is essentially (in the above sense) not biological,
    I DON'T KNOW WHAT YOU MEAN BY THE CONCEPT BEING ESSENTIALLY NON-BIOLOGICAL. THE QUESTION IS WHETHER WE SHOULD ACCEPT A GIVEN RULE GOVERNING THE USE OF THE WORD 'RACE'. SURE, THERE ARE LOTS OF INTERESTING QUESTIONS ABOUT HOW A DIFFERENT RULE, HENCE POSSIBLY DIFFERENT GAME (BUT WHY ASSUME THIS RULE IS ESSENTIAL?) WOULD RELATE TO THE CURRENT ONE. BUT NONE OF THOSE QUESTIONS NEED TO BE ANSWERED TO NOTE THAT EMPIRICAL DISCOVERY IS RELEVANT TO THE QUESTION OF WHICH RULE WE SHOULD ADOPT. then I would deny that "it is certainly easy to describe a world in which there would be a genuinely biological concept of 'race.'" We could play a different game with the word 'race,' but it's not obvious what relations the different game would have to the one we actually play.

    You write, "the fact that our world isn't like this is what justifies us in [adopting a grammatical rule]." This appears to be a justification by reference to reality. YES, SURE IS. It seems to me that all facts 'about the world' are, in the relevant sense, grammatical facts. That is, the concepts that we employ in any description of the world are given by the rules of a particular game of describing.
    THAT JUST SEEMED TO COMMIT A USE/MENTION FALLACY. IN MAKING CLAIMS WE USE CONCEPTS, WHICH ARE WHAT THEY ARE IN VIRTUE OF GRAMMATICAL RULES, THEREFORE WE ARE TALKING ABOUT GRAMMAR. We can attempt to 'justify' a form of description by relating it to another form of description in a different game, but in doing so we have not exited language. NO ONE IN THIS DEBATE EVER CLAIMED THAT WE CAN EXIT LANGUAGE, IF BY THAT YOU MEAN THINK, ARGUE, EXPLAIN, ETC. WITHOUT USING LANGUAGE. THAT'S A COMPLETE NON SEQUITOR.

    FINALLY, RE ROBERT ALLEN: I WOULD NOT SAY THAT IT IS QUITE SO STRAIGHTFORWARD THAT THE CONCEPT OF RACE DOES NOT REPRESENT REALITY. THE CONCEPT IN FACT INSTITUTES REALITY BY ASSIGNING PEOPLE SOCIAL ROLES ON THE BASIS OF FALSE BIOLOGICAL THEORIES, AND THEN ONCE THE DISTINCTION IS UP AND RUNNING, OUR TALK OF RACE CERTAINLY REFERS TO GENUINE SOCIAL CLASSES. (SLOGAN: RACE IS A FICTION, BUT RACISM IS REAL, AND BECAUSE IT IS, RACE BECOMES REAL.) THIS IS ACTUALLY A CLASSIC CASE OF "CONSTITUTIVE MISRECOGNITION" – THE INSTITUTION OF A SOCIAL KIND VIA THE WIDESPREAD ACCEPTANCE AND USE OF FALSE BELIEFS. BUT I CERTAINLY AGREE THAT THE WORLD IS INVOLVED.

    Reply
  23. Robert Allen

    Mr. Lance,

    Racial concepts were thought by those employing them not to mark social distinctions, or not merely to reflect them, but to correspond to NATURAL KINDS. They turned out not to- that makes them false. The false makers here are mind/language independent facts. You said that we cannot justify a form of description by exiting language. I have given a counterexample to that claim: I can now justify my description of human beings as homogeneous by citing the FINDINGS of the Human Genome Project.

    Reply
  24. N. N.

    Mark Lance,

    "THERE ARE MANY VIEWS ACCORDING TO WHICH DIFFERENT BITS OF LANGUAGE CAN PERFORM DIFFERENT FUNCTIONS, YET THE WHOLE BE SUBJECT TO EMPIRICAL REFUTATION. QUINE IS PROBABLY THE MOST FAMILIAR HERE."

    The Davidsonian reply (which, in my opinion, is broadly Wittgensteinian) is also familiar. But this is probably not the forum for a wider discussion of the coherence of scheme/content dualism.

    "IT IS POSSIBLE TO FOCUS JUST ON THE EXTREMELY TRIVIAL EXAMPLE AND THINK HE MEANS USEFULNESS IN SOME WAY THAT IS SEPARABLE FROM EMPIRICAL IMPLICATIONS."

    In the passages you cite, there simply isn't anything that suggests that Wittgenstein is thinking of usefulness in terms of 'empirical adequacy.'

    "THAT JUST SEEMED TO COMMIT A USE/MENTION FALLACY. IN MAKING CLAIMS WE USE CONCEPTS, WHICH ARE WHAT THEY ARE IN VIRTUE OF GRAMMATICAL RULES, THEREFORE WE ARE TALKING ABOUT GRAMMAR."

    It's not a use/mention fallacy (though I can see why you might think it is). To describe something by employing concepts is not to 'talk about grammar.' For example, it's not a grammatical fact that the flower is yellow. Nevertheless, the concepts employed, the terms of description (e.g., 'yellow'), are given by grammar. It is a grammatical fact that that color is yellow (i.e., is called 'yellow'). In this way, the sense of any description is determined by grammar. This is what I meant by claiming that all facts are *in the relevant sense* grammatical (in retrospect, it was a poor choice of words).

    I think I'll leave it at that. Thanks for the discussion.

    By the way, I agree with Constantine Sandis that such questions need not be raised, much less answered, to pass judgment on Williamson's book.

    Reply
  25. mark lance

    Robert:
    I didn't say any such thing about "exiting language." That was NN who was disagreeing with me. I reject the whole idea of "exiting language" as used by NN, as based on the use/mention confusion I explained. But if the point is to give an example of a claim that LW would consider to be grammatical getting refuted empirically, I've been arguing for that possibility in every post.

    And of course racialists thought they were claiming to be picking out a natural kind. But they also derived various normative conclusions from it. They took one's being of a certain presumed natural kind to imply a social role. The point is that although they were wrong about that, they nonetheless instituted a social role, that people still exist in. It is not just false to say that most of my neighbors are black. they are, and it is a particular social role. That it arose because of a mistaken claim that they are of a biological kind makes it a different role than any that could have arisen any other way.

    NN: How could Davidson help you? You are claiming that there is a scheme-content dualism: some claims that are language defining that are exlusively "pragmatically" justifiable, and the rest of the claims, that exist within the system and are "empirically" justifiable. I'm denying that there is any such way to carve things up.

    I disagree about LW, and think that if you read all this in context it is pretty clearly implying what I said, but that is a longer discussion.

    On Use/mention: if "all facts are grammatical facts" means only that the sense of a description is determined by the concepts involved in it, we certainly don't disagree.

    I too agree with Constantine that none of this is really relevant to the Hacker-Williamson issue. As I have not read Williamson's book, I have no views on that.

    Reply

Designed with WordPress