Tom Swann Hylton On Quine

Hylton - Quine

##Introduction

Quine more systematic than often acknowledged. Looking at one part obscures constitutive relations to the rest, threatens appearance of arbitrariness.

At the core is naturalism. Rejection of any knowledge other than common sense and science. Not different in kind.

What is the status of the naturalistic claim itself? Based on science. Accepts the circularity. Denial of first philosophy in the sense of dictating terms to science.

Theories and Things P 21 naturalism is “the recognition that it is within science itself, and not in some prior philosophy, that reality is to be identified and described”.

P2 “No standards which transcend our most successful science.”

circular definition?<<<

reconcieving old problems. often this is not clear.

what counts as science? why focus on it to the exclusion of all else?

main task: how to study human cognition in this way?

epistemological and metaphysical naturalistic projects.

denies the view of Quine as saying what we can do without. but the accepts the doing without, saying it makes sense only against his positive aims. can give an account (of what he takes to be) language use in (what he takes to be) purely naturalistic terms.

indeterminacy of translation a relatively minor point.

##1 Quine’s Naturalism Carnap rejects the idea that philosophy gives us knowledge of the world. Philosophy is second order knowledge about first order. Quine thinks philosophy is concerned with the world. What Carnap concieves as second order knowledge is really just a matter of contributing to first order knowledge about the world. The search for canonical notion is the search for ultimate categories.

Naturalism. No vantage point on knowledge other than knowledge. Word and Object: Neurath’s raft. Last paragraph: no cosmic exile. Cannot study and revise a conceptual scheme from outside of all conceptual schemes. “can scrutinize and improve the system from within” WO 276 >>>The total system, but leaves open whether one can work on parts of one from outside these parts.<<<

Why science over common knowledge? ‘Science’ is broad => even sociology and history. common sense is extended by science, and often improved.

often unarticulated in Quine’s work.

knowledge fine for the vulgar, replaced with ‘the system of the world’ for technical use. this avoids the problem of whether I can know if I could be wrong.

seamlessness of knowledge. no split into a priori and a posteriori. so philosophy cannot focus on the a priori part, even ‘from within’.

closely connected with his attitude to the analytic-synthetic. doesn’t deny any such distinction, but the one he accepts will do “no serious philosophical work”.

Hylton puts naturalism before analyticity in explanatorty order

Quine can allow but is not concerned to explain the differences between kinds of knowledge. exception with observation sentences. so what does it rule out? a priori knowledge. can accept differences, but no simple distinction.

so what differences?<<<

The Scope and Language of Science : “I am a physical object…” assuming truth, what relevance? why drawing on that part of theory? science is continuous with common sense. Philosophy should aspire to the standards of clarity and explanatoriness of the sciences. WP P235 “we, concerned to distill the essence of scientific discourse, can profitably purify the language of science beyond what might reasonably be urged upon the practicing scientist.”

how is purifying language not second order work?<<<

TT P185 experience is worthy of analysis, too obscure to help analysis P184 sentences and dispositions to assent. linking observables to observables as the basis for causal understanding of belief.

only problems posable in sufficiently clear terms will be pursued by Quine.

II Stimulations and Science “Emitted noises are about the world in virtue of their relations to such stimulations”P 12

how could ‘aboutness’ get into the picture?<<<

PT 19 “finding of science itself… that our information about the world comes only through impacts on our sensory receptors.” starting with the stimuli is empiricist. “as thorough-going an empiricist as he could be.” P13

not it’s not! it’s empirical-sciencist. presupposing knowledge of the world as being some way. fine. but the problem of empiricism is how any such knowledge is possible.<<<

how stimuli grounding thought if the stimuli are absent? highly interconnected system of knowledge. observation sentences correlated with sensory stimulations. (They are not, however, about sensory stimulation). Other sentences connected with O sentences, indirectly. becomes increasingly difficult to say what O sentences are at stake. bringing order to the system as a whole.

not predicting stimulus, predicting the O-sentences correlated with them.

PT 20 science has its “checkpoints” in prediction

basis for science being refined common sense. all knowledge has the same standards of succes - prediction of observation sentences. scientist has the same standards of evidence as the layman, is just more careful

knowledge as biological. continuous with other animals

all necessarily in the business of gathering knowledge. P16 attempts to do it as carefully as possible end up with science.

circularity: science from wihtin science? but we know this in the same way we know everything else. but this wont allay sceptical thoughts. TT22 “Our overall scientific theory demans of the world only that it be so structured as to assure the sequences of stimulations that our theory gives us to expect.”

no sense to the idea of some extra-theoretical reality.

Realism, instrumentalism, pragmatism

undermines realism? sometimes explicitly instrumentalist about objects. TT p1 “our very notion of things”. ! But also insists that he is a realist. TT P 21

instrumentalism presupposes distinction between kinds of entities - real and merely useful (for getting to knowledge of the real).

for Quine, no contrast. no sense in denying the senses—leaves the concept of evidence unintelligible.

thinks ordinary objects are from an inferior conceptual scheme, but not inferior just because from any conceptual scheme. there is only ordinary knowledge. no objects given than in a stronger sense than ordinary objects. these also are posits.

WO 22 “To call a posit a posit is not to patronize it… Everything to which we concede existence is a posit from the standpoint of a description of the theory-building process, and simultaneously real from the standpoint of the theory that is being built. Nor let us look down on the standpoint of the theory as make believe; for we can never do better than occupy the standpoint of some theory or other, the best we can muster at the time.”

begging questions about what kind of theory?<<<

epistemological v.s. ontological standpoints - outside and inside.

his naturalism means that saying that our knowledge is “just a conceptual apparatus” doesn’t conflict with realism. inquiry into the theory-building process takes for granted the theory built.

P22 the epistemological standpoint presupposes the ontological standpoint: “our explanation of how we come by our theory of the world is itself part of that theory”. but epistemology contains ontology: “study of how we come to know that very theory.”

EN P 83 reciprocal containment. studying how humans commit themselves to things, thereby committing us to things.

“in doing epistemology we draw on our theory of the world; our possession of the very theory that we draw upon, however, is to be examined and explained by the epistemological enterprise.”

according reality to whatever fits into a sufficiently explanatory theory.

cognitively pragmatist.

IV “Our Theory”

a set of sentences. public observable.

makes him not an eliminativist about language-expressibility of knowledge states

no fact theory distinction >>>must mean: no facts not statable in some theory. formal dependence without existential dependence<<<

the whole as a regulative ideal. WO 251 “let the reconciliations proceed”

‘our’ = best theory best understood in terms of prediction of sensation, clarity, simplicity all knowledge answerable to the same standards.

V Tasks for philosophy

What is there for Quineans to do?

epistemological: our relation to the world. meaning and knowledge ontological: regimentation of our theory. nature of the world.

epistemological approached in later Quine by way of development. but how do we get from stimulus to science?

concerned to show that there is a naturalistic account of the process of language learning that doesn’t take for granted ideas of meaning and understanding.

P27 genetic explanation gives us the constitution of the thing. >>>this is a terrible argument!<<< allows us RR3-4 “to see whatever there is to see about the evidence relation, the relation borne by theory to the observations that support it” >>>so it consists just in the right relations to the O sentences.<<<

Quinean metaphysics is clarification of our theory. not a rival to our science. not a second order enterprise. clarifying a theory is an advance in knowledge

um… wont it still have the same commitments?<<<

ideally our theory implies its evidence. regimenting our theory into first order logic. canonical notation requires decisions about what entities to quantify over.

what can talk of ‘best’ reformulations come to for Quine? not leaving everything as it is.

P30 “we might have reason to think that certain standards are the best available, perhas because paradigmatic cases of out knowledge employ those standards, or perhaps for more abstract theoretical reasons… the appropriate standards are those which will best enable the system of our knowledge as a whole to do what we want of it”

doesn;t think canonical notation is the only meaningful language, nor taht we oughtto speak it.

what is the status of philosophy itself? Quine’s philosophy is clearly not itself the result of lab or field work. its justification, by its own lights, must be in how it contributes to our overall system of the world. justification even if very remote from prediction, just fits neatly or simplifies or…

##2 Philosophical Background

II Carnap: Aufbau

Gestalt to sense-data

reduction is complete translation, elimination later it becomes a justification relation without elimination

sections 126 - 127 first step out of the solipsistic into the intersubjective offers desiderata, acknowledging they can’t all be fulfilled at the same time. no unique satisfaction of the desiderata. deals with complete assignments of colours to space-time points all at once, not atomistically. needed to deal with hallucination strongly suggests that no eliminative definition of ‘is at’ is possible >>>how?<<< never a completed total assignment. later experience may lead us to change our minds about assignments made on the basis of earlier experience. a definition of ‘is at’ would but an end to this process. >>>don’t see the last bit<<<

Quine’s lessons: looseness of fit between experience and theory. TDE P 41 “our statements about the world face the tribunal of sense experience not individually but only as a corporate body.” holism is the reason for the failure of the project. underdetermination of theory by evidence.

endorses the relation of meaning and experience. Vienna circle didn’t take verificationism seriously enough. Quine as ur-positivist. undermines notion of meaning

III Disposing of metaphysics

Carnap’s conventionalism. metaphysical disputes are actually about language. practical decision, not theoretical. all needed for science is the acceptance of realistic language.

doesn’t imply that there isn’t a dispute. people do speak different languages.

Principle of Tolerance “It is not our business to set up prohibitions, but to arrive at conventions…”

language and logic constitutive of possibility of knowledge. so without them, no disagreement. which logic and language to adopt is a practical decision - efficacy and convenience.

epistemic notions make sense only inside particular languages.

ESO P 207 “To be real in the scientific sense means to be an element of the system; hence this concept cannot be meaningfully applied to the system itself.”

Except it must! for we must talk about the framework, adjudicate the pragmatic credentials of different ones, etc. pragmatics cannot be without content<<<

if we disagree about the language, there are no rules about which to use. bad philosophy is treating external questions as if they were internal.

must be a definite fact about whether disagreement is internal or external

IV the analytic-synthetic distinction

the rules of the language have a special status. analytic implicit definition. take these sentences to be true. H says that truth of the terms comes from the meaning of the words. >>>so we stipulate the sentences as true, and this fixes meaning, and this explains why the sentences are true?<<<

analyticity to explain necessary truth and a priori knowledge. account for logic and math

logical form. finding the analytics. continuous with deflating metaphysics: analysing the language to find the questions at stake

isn’t this compatible with perfectly good questions being asked? the metaphysics language…<<<

for Carnap philosophy is in no sense empirical. pure syntax and semantics from descriptive. pure = construction independently. comparable with math.

general syntax and semantics. general theory from pure but applicable to actual.

##3 The Analytic-Synthetic Distinction

Doubts about meanings

requirement of scientific respectability critics say Quine treats meanings as things. but this is fine so long as we know what the idenity criteria are. for meaning: synonymy sets of synonymous sentences? but would not explain synonymy, rather presuppose it.

objection mainly to ‘mentalism’. must begin with the use of language, not with alleged mental underpinnings. public observables.

either Carnap or (Quine -> ~ anyone!)?

synonymy v.s. significance. distinct notions. accepts the latter.

synonymy comes in for most criticism. because this matters if we are trying to define analyticity in terms of meaning.

verificationist? in a sense. but for him the link between meaning and evidence leads him to deny that we can make clear sense of the meaning of a sentence. holism. isolated sentences don’t have implications. P57 “A body of sentences taken togetther may make claims on experience, but it may be impossible to parcel those claims out among the individual sentences.”

scientists don’t put all sentences up for grabs in one go. some treated as backgrounded.

Quine equates cognitive meaning with evidence. so no such thing as the cognitive meaning of one sentence in isoaltion.

Carnap sets out to defend synonmy and analyticity (intensional concepts) in behaviouristic terms. conditions for correct translation. tests for intension independent of existence - asking about imaginary cases. but Quine will deny that this is a good method, even if it works in easy cases. empirical criteria do not determine a uniquie translation

if all that matters is the structure of the behavioural patterns, then how can there be indeterminacy?<<<

indeterminacy as a conjecture. >>>e.g. Katz thinking intensions necessary for linguistics, Quine’s response<<<

allows the lexiographer to use the notion in recording dictionary definitions. doesn’t explain it, but doesn’t it show that there is a good notion here? H: no real concession for the purposes of argument. Quine can reinterpret as ‘indicating correct use’.

II Artificial Languages

Quine thinks these are irrelevant to the dispute.

Carnap claims to pick out logico-math truths in syntax only. show it is true only by talking about the language. but what is allowed in talking about the language? but Godel’s problem. truths that aren’t provable. >>>well I don’t like this interpretation of it, but anyway…<<<

Carnap puts logic and maths in the meta-language. But we can put any class of truths in the meta-language. no help to go to the necessary truths, because analyticity was meant to explain this.

for formal languages? but if we don’t have a general notion, then we have no reason to take technical notions defined in formal languages to be a definition of anything with wider application.

thinks significance is clearer. dispositional understanding of language. “what are wanted as significant sequences include not just those uttered but also those which could be uttered without reactions suggesting bizarreness of idiom.” Meaning in Linguistics P 53 (“could” understood in terms of the simplest laws codifying actual behaviour.)

Two Dogmas looks like no notion of analyticity available. later “analyticity intuitions” W0 P 67. suggestion of understanding analyticity in terms of use in Two Dogmas.

Quine thinks setting down rules for a language is not to choose some sentences whose truth is constitutive of the language. C&LT VIII rules have a special status in the description of language use. but that is all. P 65 “While I may use linguistic rules to convey what language I am imagining, the rules are simply a narrative device… to say that they also have a role in the story is to take further step, one taht is by no means justified simply by the fact that we use rules to set up an artifical language.”

if one doubts the distinction, setting up artifical languages wont dispell the doubt.

but surely must understand oneself as following rules in order to use language?<<<

III Quinean analyticity and the issue of scope

for meaning look to use. C&LT 113-114 “Any acceptable evidence of usage or meaning of words must reside surely in the observable circumstances under which the words are uttered… or in the affirmation and denial of sentences in which the words occur.”

holding true?<<<

which sentences in which a word appears determines its meaning? how to disrciminate? those sentences the learning of which give you most of the other sentences.

similar to Putnam’s view in The Analytic and the Synthetic. P reads Q as denying any distinction. P argues there must be some such distinction, but that it can do no epistemological work. only works for single criterion concepts. little theoretical import. interesting concepts are law-cluster concepts. all answerable to their applicability.

RR 79 “A sentence is analytic if everybody learns that it is true by learning its words”. gives us only pretty trivial analytics. But later allows certain inference patterns. logic. can change bits, but this is to change meanings.

not for maths. not deducible from analytics by analytic steps.

gives us some a priori knowledge. but only epistemologically uninteresting, e.g. that Tuesday is before Wednesday.

IV An Epistemological distinction?

later thinks the issue is not explicating the concept but that it has limited epistemological signficance.

Carnap allows revision of analytics, but this is to change the language. internal / external revision question of correctness in the former case, but not in the latter. no notion of justification in terms of which the change in language can be measured. except: expedience. >>>measurable!<<<

Quine agrees about the factors that decide adoption of a language. but these are also factors of internal questions.

Quine folds frameworks into each other into one large web. But not for the reasons I’m getting at?<<<

P70 “in evaluating a proposed internal revision we may have to take into account the impact of the change on the theory as a whle. This sort of holistic justification is far too complex a matter to be settled by predetermined rules.”

well… ‘rules’.<<<

same factors involved in changing language as well.

no rules setting up tight relations between theory and evidence.

“I am impressed… with how baffling the problem has always been of arriving at any explicit theory of the empirical confirmation of a synthetic statement.” TDE 41-42 a unique formally specifiable justification relation holds only once we take a background theory for granted. holism secures this result.

sentence to be evaluated by appeal to its role in contributing to the best available theory.

no epistemological difference between both kinds of revision.

what is at issue is the Principle of Tolerance. the crucial thing is whether the whole theory helps us deal best with observations. even if we can distinguish between theory and language, we have no reason to be tolerant in either case.

legislative definition but this is a “passing trait”, doesn’t affect the way things come to bear on it. opposite of Carnap’s view that conventionality is imposed on the scientist’s activity. for Quine, true by convention is just historical speculation about how it can to be accepted as true.

P74 “The sorts of considerations thatmight lead us to make a change from one theory within a language to another are not in principle different from the sorts of considerations that might lead us to make a change from one theory within a language to another within the same language”. the theory/ language distinction ceases to be important.

V The putatively a priori

how to explicate logic and math without appeal to analyticity?

no epistemological distinction !!

we cannot deny a role for expedience in what Carnap thought of a synthetic.

but isn’t there a difference in what these disciplines /do/? WP 121 “some such remote confrontation with experience may be claimed even for pure mathematics and elementary logic.”

holism undermines the a posteriori / a priori distinction as well. P77 logic and math;s ultimate justification lies in the role in our wider system of beliefs

but most of maths and logic plays exactly no role in our wider system!<<<

incorrigibility of maths and logic to be explained by the fact that they are implied in every actual use of empirical content.

maxim of minimal mutilation PL 7

room to doubt whether logic could be overthrown. not just from the inconcievability of the alternative as better, or even there being an alternative, but the inconcievability of how we could go about looking for alternatives.<<<

inapplicable maths.? no way to demarcate. but doesn’t say anything about justification. we don’t justify by way of contributions to overall theory. Quinean inspired response: justify the system as a whole.

but the system as a whole justifies its internals by its own means.<<<

Reconceiving Epistemology

EN need to find adistinction between relevant and irrelevant noises made by subjects

epistemology in this sense is concerned not only with how we can know anything but how we can say anything true or false about the world. P82

i.e. is concerned with content<<<

FSS P 16 “It is rational reconstruction of the individual’s and/or the race’s actual acquisition of the responsible theory of the external world. It would address the question how we, physical denizens of the physical world, can have projected our scientific theory of that whole world from our meager contacts with it…”

that rational reconstruction is genetic.

“responsible”? also, analogous antiskeptical move here as with Davidson<<<

repudiates the Cartesian dream.

“The Humean predicament is the human predicament”. impossible to justify our knowledge as a whole by means of induction.

if the aim were to justify science, naturalism would be circular. if the aim is jsut to understand, not vicious.

not globally normative. P84 “Those who think that this is all that is really at stake in talk of the normativity of epistemology will hold that Quine’s view gives up completely on normativity. If we take a less absolutist line, however, we will find, with Quine himself, that some normative elements survive. Quine takes the idea that prediction is the test of an hypothesis”as defining a particular language game … the game of science" (PT P 20…) The goal of knowledge is thus given; epistemology is normative because it tells how we should best act to achieve this goal:

“For me normative epistemology is a branch of engineering. It is the technology of truth-seeking, or, in a more cautiously epistemological term, prediction. Like any technology, it makes free use of whatever scientific findings may suit its purpose… The normative here, as elsewhere in engineering, becomes descriptive when the terminal parameter is expressed.” H&S P 665 …

Epistemology for Quine is normative in the same sort of way: it tells us how we should act so as to obtain successful theories.

all normativity is instrumental.<<<

*I Input: observations, evidence and stimulations

Sense data is meant to be both evident and prior. But we seem to first know things and people.

Quine argues that Vas Frassen’s position collapses into general instrumentalism