VRG: Maddy on Quine
The chapter on Quine may seem like in-fighting amongst naturalists, but it has the virtue of delineating further, through the criticisms of Quine, the SP's position. If what I have to say today is unclear, that's because my thoughts on the issues are still fuzzy.
The criticisms of Quine appear to come from two directions: on the one hand he is said to incorporate non-naturlistic elements into his naturalism, on the other the exact character of his naturalism is taken to be problematic. In actual fact, however, the two strands of criticism are closely related. For, on Maddy's view, Quine's failing to be naturalistic enough - his taking on board too great a portion of the goals and concepts of the traditional philosophical (metaphysical and epistemological) project - is really one of the major sources of the problems with his position.
Of central importance here is the way in which Maddy interprets Quine as responding to (Stroudian) scepticism. Quine admits that there is a certain sort of circularity involved in the naturalist's response to scepticism (see fn. 4, p.84), in that the results of scientific investigation are relied upon in offering an account of knowledge. However, Maddy argues, Quine's own conception of what scientific investigation amounts to appears to undercut this move. For Quine thinks that the job of the naturalised epistemologist is to offer an account of "the relation between the meager input and the torrential output" (Quoted on p.89). Quine's 'input' is the 'stimulation of his sensory reseptors', and the 'output' is his theory. As is well known, Quine thinks that the former underdetermines the latter.
But, argues Maddy, thinking of things in this way is problematic. For our acceptance of one theory over another is, in a certain sense, arbitrary, and so it is unclear what justifies us in relying on this particular one in our response to the sceptic - in our account of the world and our knowledge of it. The problem, as Maddy sees it, is to conceive of naturalised epistemology as, roughly speaking, a sophisticated empiricism pursuing much the same goals as traditional epistemology (i.e. to account for the legitimacy of our 'output 'on the basis of our 'input')
SP sees the project of naturalised epistemology differently from Quine. She refuses to give the (Stroudian) sceptic the foothold he requires, because she declines the challenge of responding to the sceptic on his own terms. To the question of how we know, say, about medium sized physcal objects, she simply cites the relevant scientific findings (about neurophysiology, visual perception, etc.). SP finds no use for Quine's input/output distinction and thus doesn't think of the one as underdetermining the other. If so, then the Stroudian sceptic can't mount the same challenge as he did to Quine.
Early in the chapter, Maddy says of SP that, "she doesn't denounce as 'unscientific' the desire of Stroud's skeptic to find an independent justification for her methods; she simply sees no way to do this." (p.85). This way of putting things, however, seems to me to be rather precarious. For it sounds rather like accepting scepticism. Quine openly denies that science is "answerable to any supra-scientific tribunal" (quoted on p.85). SP, in keeping with her refusal to delimit science, doesn't do this. But, so it seems to me, she needs to have some other way of answering the objection that her methods are problematically circular. If all SP says to the Stroudian sceptic is that his objections are unanswerable, I think that she is in trouble. Maddy sometimes gives the impression that the difference, between SP and Quine, makes the difference between the legitimacy or otherwise of employing the apparently circular reasoning involved in the naturalistic response to scepticism. At no point does SP rely on any proposition such as 'beliefs formed scientifically are justified', for she has no conception of 'the scientific' per se, rather she simply uses the science itself. Does this sort of move help? We need to think about exactly the sort of circularity involved in both Quine's and SP's views.
Consider two arguments, A and B:
(A1) Beliefs formed scientifically are justified
(A2) Science (psychology perhaps) says that we see medium sized physical objects
So,
(A3) We see/there are medium sized physical objects
(B1) The human visual system works in such-and-such a way
(B2) That the human visual system works in this way allows us to see medium sized physical objects
So,
(B3) We see/there are medium sized physical objects
Roughly speaking, A is supposed to characterise a Quinian naturalist's reasoning (ignoring other aspects of Quine's view, such as ontological relativity, and concentrating on the naturalism), and B SP's reasoning. I hope I'm not mischaracterising here, although I admit that I'm pretty uncertain.
My question is whether A is problematically circular in a way in which B avoids. Obviously neither are explicitly circular (in neither case does the conclusion appear as a premise). But both look somehow problematic. The problem with A looks to be that our reasons for accepting (A1) are dependent on our accepting (A3). That is, if we didn't already accept (A3) we wouldn't accept (A1). With B, it looks as though the same can be said for both (B1) and (B2), that our reasons for accepting them rely on our already accepting (B3).
The question whether this sort of circularity is (always) a bad thing, making the arguments useless, is I think a very difficult one. For it depends on our giving a satisfying gloss on "our reasons for our accepting___ are depndent on our accepting___". But my point is that both arguments seem to be in the same boat. So, I don't see how SP's strategy is an improvement in the Quinian one (ignoring the comments about input/output, and the later discussion of holism which I leave to one side).
So, just admitting to the Stroudian sceptic that you can't answer his objection is not good enough. You need to be able to account for the reasonableness of your own procedures. If this is done by relying on the results of those procedures, the Stroudian can say that you're arguing in a circular way. But, as I see it, Quine and SP face the same challenge here.
The criticisms of Quine appear to come from two directions: on the one hand he is said to incorporate non-naturlistic elements into his naturalism, on the other the exact character of his naturalism is taken to be problematic. In actual fact, however, the two strands of criticism are closely related. For, on Maddy's view, Quine's failing to be naturalistic enough - his taking on board too great a portion of the goals and concepts of the traditional philosophical (metaphysical and epistemological) project - is really one of the major sources of the problems with his position.
Of central importance here is the way in which Maddy interprets Quine as responding to (Stroudian) scepticism. Quine admits that there is a certain sort of circularity involved in the naturalist's response to scepticism (see fn. 4, p.84), in that the results of scientific investigation are relied upon in offering an account of knowledge. However, Maddy argues, Quine's own conception of what scientific investigation amounts to appears to undercut this move. For Quine thinks that the job of the naturalised epistemologist is to offer an account of "the relation between the meager input and the torrential output" (Quoted on p.89). Quine's 'input' is the 'stimulation of his sensory reseptors', and the 'output' is his theory. As is well known, Quine thinks that the former underdetermines the latter.
But, argues Maddy, thinking of things in this way is problematic. For our acceptance of one theory over another is, in a certain sense, arbitrary, and so it is unclear what justifies us in relying on this particular one in our response to the sceptic - in our account of the world and our knowledge of it. The problem, as Maddy sees it, is to conceive of naturalised epistemology as, roughly speaking, a sophisticated empiricism pursuing much the same goals as traditional epistemology (i.e. to account for the legitimacy of our 'output 'on the basis of our 'input')
SP sees the project of naturalised epistemology differently from Quine. She refuses to give the (Stroudian) sceptic the foothold he requires, because she declines the challenge of responding to the sceptic on his own terms. To the question of how we know, say, about medium sized physcal objects, she simply cites the relevant scientific findings (about neurophysiology, visual perception, etc.). SP finds no use for Quine's input/output distinction and thus doesn't think of the one as underdetermining the other. If so, then the Stroudian sceptic can't mount the same challenge as he did to Quine.
Early in the chapter, Maddy says of SP that, "she doesn't denounce as 'unscientific' the desire of Stroud's skeptic to find an independent justification for her methods; she simply sees no way to do this." (p.85). This way of putting things, however, seems to me to be rather precarious. For it sounds rather like accepting scepticism. Quine openly denies that science is "answerable to any supra-scientific tribunal" (quoted on p.85). SP, in keeping with her refusal to delimit science, doesn't do this. But, so it seems to me, she needs to have some other way of answering the objection that her methods are problematically circular. If all SP says to the Stroudian sceptic is that his objections are unanswerable, I think that she is in trouble. Maddy sometimes gives the impression that the difference, between SP and Quine, makes the difference between the legitimacy or otherwise of employing the apparently circular reasoning involved in the naturalistic response to scepticism. At no point does SP rely on any proposition such as 'beliefs formed scientifically are justified', for she has no conception of 'the scientific' per se, rather she simply uses the science itself. Does this sort of move help? We need to think about exactly the sort of circularity involved in both Quine's and SP's views.
Consider two arguments, A and B:
(A1) Beliefs formed scientifically are justified
(A2) Science (psychology perhaps) says that we see medium sized physical objects
So,
(A3) We see/there are medium sized physical objects
(B1) The human visual system works in such-and-such a way
(B2) That the human visual system works in this way allows us to see medium sized physical objects
So,
(B3) We see/there are medium sized physical objects
Roughly speaking, A is supposed to characterise a Quinian naturalist's reasoning (ignoring other aspects of Quine's view, such as ontological relativity, and concentrating on the naturalism), and B SP's reasoning. I hope I'm not mischaracterising here, although I admit that I'm pretty uncertain.
My question is whether A is problematically circular in a way in which B avoids. Obviously neither are explicitly circular (in neither case does the conclusion appear as a premise). But both look somehow problematic. The problem with A looks to be that our reasons for accepting (A1) are dependent on our accepting (A3). That is, if we didn't already accept (A3) we wouldn't accept (A1). With B, it looks as though the same can be said for both (B1) and (B2), that our reasons for accepting them rely on our already accepting (B3).
The question whether this sort of circularity is (always) a bad thing, making the arguments useless, is I think a very difficult one. For it depends on our giving a satisfying gloss on "our reasons for our accepting___ are depndent on our accepting___". But my point is that both arguments seem to be in the same boat. So, I don't see how SP's strategy is an improvement in the Quinian one (ignoring the comments about input/output, and the later discussion of holism which I leave to one side).
So, just admitting to the Stroudian sceptic that you can't answer his objection is not good enough. You need to be able to account for the reasonableness of your own procedures. If this is done by relying on the results of those procedures, the Stroudian can say that you're arguing in a circular way. But, as I see it, Quine and SP face the same challenge here.
Labels: Circularity, Naturalism, Quine
posted by Joel at 7:03 am
3 Comments:
I don’t have much to add. By exposing a similar circularity in the SP’s and Quine’s respective arguments, you’ve highlighted a central problem of this chapter, and I agree that the SP is in trouble.
Maybe, however, there is still a way to distinguish Quine’s circle from the SP’s circle. I’m not sure if that affects your conclusion -- that they are in the same boat as far as the Stroudian skeptic is concerned. But there seems to be a difference in just how the skeptic’s question is unanswerable for the Quinean naturalist and the SP. Maybe this difference makes it less important whether the skeptic sees them in the same spot.
In Maddy’s view, Quine’s philosophical naturalism is based on an explicit negation of first philosophy. Therefore, she argues, it remains stuck in some of the epistemological and metaphysical games that first philosophy plays. Quine presupposes that beliefs formed scientifically are justified. Maddy claims that this premise is vulnerable to Stroudian skepticism because it still has a conventional epistemological flavour (it operates with concepts like belief and justification). So it might still be asked: “how do you justify your beliefs otherwise, extra-scientifically?”. Dismissing this question as unanswerable will involve the damaging admission that Quinean naturalism results of its own positing against the background of first philosophy (p. 89). It is due to this positing that Stroudian skepticism becomes unanswerable, and the Stroudian skeptic will question this positing.
In contrast to this, the SP’s first premise is that perception functions in this and that way. The idea seems to be that those contesting her premise end up asking “does it really function this way?” or “couldn’t it function differently?”, but cannot sensibly ask “does it function at all?”. The reasons for this are those mounted earlier on against extraordinary dreaming. There is plentiful evidence for the functioning of perception and very little for extraordinary dreaming. Consequently, the skeptical question does not become unanswerable due to the SP’s own positing. In a sense, her methods do not allow her to posit herself against the background of first philosophy. They just allow her to assess whether other theories (and her own) satisfy the demands of empirical theorising widely construed.
What ultimately absolves the SP from skeptical probing, Maddy appears to think, is that once the skeptical question becomes unanswerable for empirical theorising, it also ceases to be an objection to what the SP calls her methods. It does not make sense to explain empirically how perception would not function at all. So there is nothing to concede to the skeptic for the SP, and she can move on.
This still leaves unanswered your important question about circularity in general. But the respective circular arguments appear to behave differently under skeptical probing.
To follow on from Christian's comment. I think that the issue of good/bad circularity is closely connected to Quine's empiricism and his confirmation holism. As Christian points out, Quine still plays the epistemological game of traditional philosophy: he gives a *general account* of how observation and the hypethetico-deductive method operate in experience. This account points out that there is a gap between sensory input and theoretical output. That Quine feels under philosophical pressure because of this gap betrays that he is still beholden to traditional epistemological concerns, that is, he sees the work of philosophy within naturalised epistemology as drawing perfectly general epistemological conclusions about science that follow from a perfectly general epistemological analysis of our cognitive situation.
The conclusions he draws lead to confirmational holism and thus a kind of pragmatism about scientific practice that is in tension with the 'robust realism' of his naturalism. The pragmatist strand in his thinking is that theories are posited as true partly on the basis of various theoretical virtues, rather than *simply* because they are true descriptions of the empirical facts. But more than this, given his holism such positing infects the latter - descriptions of facts are in some sense relativised to theoretical choices. This relativisation is what lets in the Stroudian sceptic who points out Quine has brought sceptical pressure on science from 'within'.
The sceptic can ask: Why this total theory rather than that?
Quine will reply: There are no questions about 'reality' external to our total theory given confirmational holism, so I don't need to respond to your challenge.
But the sceptic will reply: Surely you mean no standards of assessment external to some total theory or other, for the very formulation of confirmation holism has gone through the idea that scientific theories per se are generated in the face of an underdetermination of theory building by observation and hypethetico-deductive method. That is, you have given a 'scientific' account of how (1) the process of forming perceptually-based beliefs and the truth of those beliefs can come apart, and (2) how there could be different total theories based on different theoretical choices. It follows from the very nature of this account then, that you cannot simply shrug your shoulders at my challenge.
I presume Maddy's problem with Quine is that rather than give a genuinely scientific account of how perceptually-based beliefs can be true, he gives an account still beholden to traditional epistemology - an empiricist one in terms of sensory stimulation and hypothetico-deductive method. And he further betrays such beholdenness by assessing this model in terms of whether the two latter constitutive elements of scientific practice can justifiably be said to ground our total theory as a true representation of reality (that they can't is what generates underdetermination). Presumably SP's position would be that there is no such general epistemological account of how our perceptually-based beliefs truly represent the world. We trust such belief-forming mechanisms because actual science (cognitive science, psychology etc.) give us good reasons to (it *explains* scientifically how they work). Thus SP has not undermined science from within vis a vis the sceptic as Quine has. (Whether there are other reasons to take SP's trust in science to be circular is a good question.)
I think then, that there is space between Quinean naturalism and SP. But something else troubles me. Maddy says on p85 that the SP goes along with Descartes when he tries to put science on a secure footing but peels away when the project fails. But why take an interest at all if it is mistaken to try and answer traditional epistemological questions (to account for how beliefs about the world can be true outside of actual scientific investigation)? Similarly, Maddy says on the same page that SP does not denounce Stroud's philosophical scepticism as 'unscientific'. But surely she does, for she rejects it for the same reason she distinguishes herself from Quine: she simply doesn't feel the heat of traditonal epistemological questions due to her being already enlisted in the Neurathian conception of scientific enquiry (one where only actual scientific practice counts as proper to justifying knowledge). In other words, if SP is just a more thoroughgoing naturaliser of epistemology than Quine, then purporting to take Descartes and Stroud seriously seems disingenuous.
Two points:
1. On circularity: SP will of course want to say that there is no threat of vicious circularity in her position. She might start by saying that Joel’s B argument foists on her an appearance of circularity that she need not go for. She might reformulate the content of B as a simple laboratory statement: the workings of the human perceptual system enable us to perceive medium sized physical objects (see p. 91). Is there anything worryingly circular here? Well, she says, of course we are investigating the perceptual system of one medium sized physical object located among others, but there is nothing worrying about that – we can do so unproblematically precisely because we do have that perceptual system. If we didn’t have it, we would not be able to perceive physical objects, and so would not be able to detect, and a fortiori investigate, human beings; but we do, and so we can.
2. Even if something along the above lines does end the vicious circularity worry, there is surely an air of fragile hopefulness here if this is thought to be an adequate resting point for SP. Here I want to go back – apologies – to the worries about simply shunning two-level theories. The response to the circularity worry above is made from the heart of the laboratory. “Of course, the Second Philosopher is born native to the laboratory”, Maddy says in this context (p. 91). But how does being born native to the laboratory ensure that one stays there? Of course one might as a matter of temperament stay in the lab where one was born; one might be the village naturalist. But without further work, any normative claim that one should so remain, that when it comes to doing philosophy that is one’s proper place, will be without justification. In different ways, Hume, Kant, Carnap, Quine, all offer a more resilient understanding and constitution of a naturalist orientation. But they achieve this by resorting to some sort of two-level theoretical structure. Without nesting in some such structure, SP’s adherence to the laboratory culture seems to be grounded merely in personal preference and habit, offering no non-dogmatic resistance in the face of any change in individual temperament, whether motivated by reflection or otherwise.
Of course, there are problems with two-level structures, but it is not clear where SP could stand – consistent with her own brief – in order to argue against the coherence of them. (And even if she did manage to do so, it is not clear that what we would be left with would still count as a form of naturalism, in the sense of an empirical realism that falls short of traditional metaphysical aspirations and claims; but that is another matter.)
Post a Comment
<< Home