Tom Swann Sosa Nature Unmirrored Epistemology Naturlized

Sosa - Nature Unmirrored, Epistemology Naturalized 1983

A Knowledge and Justification

not just any confidence is relevant. the athlete might be better off deluding themselves. etc.

“trust, faith, and assurance are not made knowledge by such backing of moral, religious, or prudential reasons. Only epistemic justification will serve for that.” P 50

epistemic justification is - supervenient — because it must be made so by “more basic” properties of the belief - universalisable — “any belief similar to that belief in all relevant respects would be equally epistemically justified.” - governed by principles — “by making it consequent on antecedent properties of belief”

so basically it’s just to say that the properties of the belief explain in some intelligible way whether and why the belief is justified<<<

##B Understanding and Validation

“The Project of Understanding is that of finding as simple, general and convincing an account as we are able to find of the conditions within which our belief of a proposition has the kind of epistemic authority… required for one to know its truth.” Validation is changing our situation so as to meet the understanding project’s conditions

odd way of putting things, but ok<<<

##C Epistemic Rationalism

Descartes as a foil logicism and phenomenalism as further examples to skepticism?

ordinary talk of knowledge is of approximations to superknowledge. but even this calls for spelling out the dimension of superknowledge.

##D Reliabilism

rationalism as a special case of reliabilism?

“Reliabilism requires for the epistemic justification of belief that it be formed by a process reliable in an environment normal for the justification of the belief.”P52 >>>“normal”? normal for the justification? isn’t that massively circular?<<<

Reliabilism is still a form of foundationalism. just that the building requirements for pyramids are less stringent. “A belief may now join the base not only through perfectly reliable rational intuition but also through introspection, perception, or memory.” and ‘erect the superstructure’ by induction.

##E Foundationalism Assailed.

Rorty. Pat Mo N

justification is social. makes it holistic.

Rorty’s attack on foundationalism rests on the confusion of causation and justification. one might define justification in terms of something having been justified (status) and reasons actually given (process).

but argument will often have ultimate premises. ones for which no reasons need be given. are all unjustified. >>>not justified. not justified badly.<<<

there must be some for any argument that is not regressive or circular. >>>difference between saying that argument must level out in propositions that don’t have anything more to be said for them, and propositions that can’t have anything more said for them.<<<

Sosa now starts talking about epistemic authority. (justification has been taken up by Rorty’s conversation.) authority must either come from reasons in its favour, or from somewhere else.

quotes some Rorty

“we can think of … justification as a relation between the propositions in question and the other propositions from which the former may be inferred. Or we may think of both knowledge and justification as privileged relations to the objects those propositions are about. If we think in the first way, we will see no need to end the potentially infinite regress [except until we are satisfied for our purposes]… if we think of knowledge in the second way, we will want to get behind reasons to causes, beyond argument to compulsion from the object known, to a situation in which argument would be not just silly but impossible.” P 159

argues that Sellars and Quine get their holism from “their commitment to the thesis that justification is not a matter of a special relation between ideas (or words) and objects, but of conversation, of social practice.” P170

“For Sellars, the certainty of”I have a pain" is a reflection of the fact that nobody cares to question it, not conversely. Just so, for Quine… Explaining rationality and epistemic authority by reference to what society lets us say, rather than the latter by the former, is the essence of what I shall call “epistemological behaviourism”." P174

Sosa boils this down to five principles

  1. Inferential conception of justification
  2. normative character of inference
  3. social foundation of inferential norms
  4. confusion of causes with reasons No belief is justified because of a special causal relation to its object.
  5. Two sources of epistemic authority from inference, and thus indirectly from social legislation, or directly from social legislation

Taken together, these make Rorty a foundationalist.

C1) All assertion with authority gets it in/directly C2) Indirect authority is justified by inference from assertions which have direct authority C3) direct authority only if approved without prior argument C4) What makes argument permissable is relation to legislated norms

conventionalist justification confuses direct social approval with justification?

how is it by its own account founded? directly or indirectly? surely indirectly. but 3 and 5 lack authority, direct or indirect.

but ditto for the mirror?<<<

##G Reliabilist Foundationalism

possibility of wacky reliable mechanisms. >>>but we can come to see that it is reliable…<<< problem of clairvoyance is simply that it fails to meet the challenge of “doxastic ascent” >>>does he meen assent?<<<

machine and animal ‘reliables’ = a kind of ‘knowing’. human knowing is on higher plane because of its sophistication and coherence and capacity to satisfy self-reflective curiosity. a problem for relibalism that these are left out

second problem for reliabilist foundationalism. need a list of reliable faculties. but how would we find them? by using them. but we would need to use all of them in an ongoing process…

##H Coherence

reliabilism leads to a kind of coherence.

simple coherence isn’t plausible enough. for any set of n beliefs B with any amount of coherent, there are 2^n equally coherent other sets with n belief that are equally coherent. pairwise logical equivalence too strong. >>>what does this mean?<<<

comprehensiveness can also be spelled out in terms of intergration. but no simple theory of this, many dimensions

high degree of consistency a plausible requirement

I Perspectival Coherence

epistemic perspective, explaining how the holder of the beliefs knows about the things covered by their beliefs.

“logical point of origin” for the historico-geographic framework of one’s beliefs. indexed to an “I” and to a time and place. through systematic replacement we can index the beliefs to others at other times. but would be absurd to think we would be as justified to believe this new set. so the perspective must be important.

a form of foundationalism. foundationalism doesn’t reject coherence as one constraint.

##J Epistemology of Epistemology

How do we know that knowledge requires such conditions?

rational intuition for some basic logical principles governing coherence

Rorty has a go at the Greek analogy between reason and vision. we have a faculty of sight. why not a faculty of reason? analogous in explaining why we think logical truths are unassailable.

open to scientific study.

a coherentist must justify their view by appeal to the fact that their view coheres well with the rest of their views.

##K How to Naturalize Epistemology

Reliabilism. “would then fall to the studies of cognition to specify just which processes reliably engender true belief and how, thus giving substance and strength to our overall theory of knowledge.” P64

this involves us in coherentism: “in allowing each faculty included to rest on use of others for support of its inclusion.”

a more radical way to do it in Quine’s EN

World and Object “we can investigate the world, and man as part of it, and thus find out what cues he could have of what goes on around him. Subtracting his cues from his world view, we get man’s net contribution as the difference. This difference marks the extent of man’s conceptual sovereignty — the domain within which he can revise theory while saving the data.”

but of course we can never subtract the cues! the net contribution to what? to meaning / content. see Mc Dowell’s Afterword I <<<

science is a “free creation” RR

even crucial logical features are in the same boat.

Nature of Natural Knowledge P 80 “The steps by which the child was seen to progress from observational language to relative clauses and categorials and quantification had the arbitrary character of historical accident and cultural heritage; there was no hint of inevitability.”

still not clear to me what

Sosa takes the central planks of Quine’s epistemology to be these: P1) epistemology is now psychology, investigating causal relations between sensory input and theoretical output P2) doing this leads to the conclusion that the sensory input does not determine the theoretical output.

if science is free creation, logic is arbitrary, what reason do we have for not thinking that it’s all an illusion?

Quine will reject this via his naturalism. epistemology presupposes science.

Quine’s P2 reinvigorates Russell’s reductio R1) if the scientific account of the world is wrong, then we lack scientific knowledge. R2) if the scientific account is true, that rules out any possibility that we really know such a world even in broad outline. R3) therefore we lack scientific knowledge even in broad outline.

where does Russell discuss this?<<<

but R2? Quine’s own account of how we come to science seems to confirm

surely Quine’s reasons are totally different to Russells?<<<

Quine rejects R2. we are owed an explanation. “the evidential relation is virtually enacted, it would seem, in the learning.” NNK P 74

Sosa claims this response is incoherent. Quine is saying that science is justified by a process of learning that is a ‘free creation’.

totally misunderstands Quine here<<<

“surely the threat is not met simply with the thought that the channels by which theoretical language is learned themselves automatically determine the relation of evidential support.”

this is Quine’s picture, but he will insist that this is all any language and any theories formulated in them mean.<<<

Reply to Stroud in Midwest Studies in Philosophy VI “Yet people, sticks, stones, electrons, and molecules are real indeed, on my view, and it is these and no dim proxies that science is all about. Now how is such robust realism to be reconciled with what we have just been through? The answer is naturalism: the recognition that it is within science itself, and not some prior philosophy, that reality is properly to be identified and descrbed.” P 474

but couldn’t naturalism be replaced with anything, say “fictional realism” P68

how would Quine respond? commitment to science. prior, but also can be backed up by what science lets us do.

<<<

Inconsistent triad

Q1) Our overall scientific theory claims “only that [the world] is somehow so structured as to assure the sequences of stimulation that our theory gives us to expect.” P69 Q2) physical objects are real Q3) naturalism but no first philosophy

Sosa’s complaint. “If it is within science that we settle, to the extent possible for us, the contours of reality; and if science really claims regarding the world only that it is so structured as to assure certain sequences of stimulation; then how can we possibly think reality to assume the contours of people, sticks, stones, and so on”. P69

I think Quine didn’t really want to claim Q1. Science isn’t (all) about stimulation. But it is stimulation that ‘makes possible’ meaning.

This lets Quine respond that truth simply consists in getting the predictions right.

Still doesn’t look very realist though.

At best this makes his naturalism a kind of transcendental idealism. So it would be strictly speaking inconsistent, because the psychology now is not transcendental but empirical; Quine has to start with the stuff about stimulus-meaning.

<<<

“What is more, if science is really the measure of reality, it cannot undercut itself by saying that it really isn’t, that it is only convenient”manners of speaking" to guide us reliably from stimulation to stimulation."

Contrast this way with modest reliabilism. lead to comprehensive coherentism. doesn’t like this…

Doesn’t like Quine’s way.

other options?

##L Epistemology Natualized

Arbitrary not to include the evolutionary epistemology of Campbell, Popper et al genetic epistemology of Piaget; innatism of Chomsky; HPS.

P70 “it is not perfectly obvious how scientific study of the causal conditions of knowledge can bear on philosophical study of the nature of knowledge and the conditions on which it (epistemic justification) supervenes”.

recall the two main projects for epistemology “why not co-opt the studies of cognition as important elements of our Project of Validation.” P71

But would not all of inquiry increase coherence? no. cognition more central. study of the perspectival element

pretty weak<<<