Saturday, October 29, 2005

Know Show

It's commonly assumed that in order to know something, we must be in a position to show or establish its truth — or at least be aware of the reasons which ground our knowledge. This is what most skeptical arguments depend upon — a brain in a vat might have experiences subjectively indistinguishable from my own, so there's no way to establish that I'm not a BIV, and so I must not know it either. Externalists simply deny this last step. You can have 'knowing' without 'showing'. Perhaps all that really matters for knowledge is that my true belief is formed by a reliable process, or that it 'tracks the truth' through close possible worlds -- regardless of whether I'm aware of this fact. Now, sometimes people's immediate reaction is to think that such externalism simply begs the question against the skeptic. "After all, you can't really show that you're not a BIV. The externalist just assumes that we're not, and argues from there. But what entitles him to such an assumption?" In this post, I want to explain why such complaints are misguided.

(I'll be drawing upon the responses I've previously made to Blar and Rex Hubbard when they offered these objections.)

First, note that the skeptic's challenge is whether we can have actual knowledge in light of the counterfactual possibility of our being BIVs. The question is not whether our knowledge would survive were we actually massively deceived -- obviously it would not. False beliefs cannot constitute knowledge. So the only cases worth considering are ones where we have true beliefs.

The skeptic accepts this, but argues that even if we have true beliefs about the external world, these cannot constitute knowledge because our subjective uncertainties mean that our beliefs lack justification. The externalist response is to deny that knowledge requires subjective awareness of justification. Instead, they suggest, our beliefs must be reliable or 'justified' in a more objective sense. There must be an appropriate connection between the truth of the belief and our believing of it. What matters for knowledge is that this connection holds as a matter of fact. It does not matter whether we are aware of the connection.

To reiterate: the skeptic is willing to grant that my belief that "I am not a BIV" is true. So the externalist is not begging the question when he assumes that I'm not a BIV. (Besides, we can always discharge the assumption via conditional introduction, if you insist. That is, we conclude that if I am in fact not a BIV then I in fact have knowledge. Since it's certainly possible that I'm not a BIV, it follows that we can, possibly, have knowledge.) And the externalist will certainly agree that we cannot show that we're not BIVs. He grants the skeptic this much. The dispute is simply over whether I can know that I am not a BIV, given that my belief is true but undemonstrable. On an externalist conception of knowledge, I can have such knowledge.

(Of course, that doesn't show that externalism is the true theory of knowledge. Perhaps it isn't. But that's another issue. The current question is whether the truth of externalism would defeat skepticism. And the answer is clear: it would.)

Another side-issue is whether it is in fact true that these objective conditions required for knowledge hold. For example, is it true that I'm not a BIV? I obviously believe so, but this isn't something I can show. But the whole point of this post is to explain why this doesn't matter. I can't show that I have knowledge, because I can't show that I satisfy the objective conditions required for knowledge (which at a minimum include truth). But, contrary to the skeptic, this doesn't mean that I lack knowledge. A belief might be true even if we can't show it. Similarly, it might constitute knowledge, even if we can't show it.

Anyway, this all ties in rather nicely with a brilliant anti-skeptical move by Thomas Reid, recently quoted on the common sense philosophy blog:
Reason, says the skeptic, is the only judge of truth, and you ought to throw off every opinion and every belief that is not grounded on reason. Why, Sir, should I believe the faculty of reason more than that of perception; they came both out of the same shop, and were made by the same artist; and if he puts one piece of false ware into my hands, what should hinder him from putting another.

In other words, shouldn't the consistent skeptic also doubt the dictates of logic, trusting nothing — not even the conclusions of their own arguments? Such an uber-skeptic would have no chance of showing that we lack knowledge. At best, he might insist that we can't show that we have knowledge (though Reid would presumably question his certainty even of this). But even if we conclude that we can't show anything at all, that still leaves upon the possibility that we know all sorts of things. Because, as already explained, knowing does not require showing. (The latter might be sufficient for knowledge, but it isn't necessary.)

(Aside, I must say there seems something incredibly odd about being skeptical of the rational faculty. I've suggested such a skeptical scenario before, but I don't know quite what to make of it all. I guess it just goes to show how misguided is the Cartesian ideal of casting off "all assumptions" — as if it would still be possible to think at all after doing so! Hmph. Actually, this ties in with the stuff in my previous post about the need for innate constraints, the impossibility of extreme empiricism, etc. Anyway, I'll stop rambling now. Maybe the comments will shed more light on the matter...)

14 comments:

  1. "The current question is whether the truth of externalism would defeat skepticism. And the answer is clear: it would."

    Not so. Skeptics claim that we do not have the kind of knowledge which requires subjective certainty. Externalists claim that we do have a kind of knowledge which does not require subjective certainty. These two claims are compatible.

    ReplyDelete
  2. Well, I think the skeptic wants to make a stronger claim than that. Everyone already agrees that we can't attain subjective certainty. So if that's all the skeptic claims, they're not saying anything very interesting or controversial at all. But they're normally understood as making the much stronger claim that we have no justification whatsoever for discarding the BIV scenario and accepting our everyday beliefs. (I explain this in more detail in my original skepticism post.) It is this stronger form of skepticism that I meant to be discussing.

    ReplyDelete
  3. I'm not sure if my original comment got swallowed up or what, but here goes again.

    Discussions like these are always disappointing to me because all the participants seem to uncritically accept the same premise that's screwing them all up. Very crudely, it's asserted that knowledge=true belief. (Or sometimes "justified true belief" since some would argue that being correct completely by accident doesn't constitute knowledge.) But whence does this assertion come from? It's just taken as obvious, but why should it be?

    The way to cut the gordian knot here is to discard this assumption and concede forthrightly that all of our knowledge is conjectural. The search for unshakeable foundations of knowledge that will hold come what may is an ultimately fruitless quest for the philosopher's stone, as countless volumes full of unsuccessful attempts can attest to. Neither so-called "pure reason" nor so-called "sense data" can provide justifications for any conjecture because both are fallible (though of course both can be very helpful). Which is what you'd expect in the light of what we've learned from cognitive psychology and neuroscience.

    This of course leads us to a very Neurath-Duhem-Quine view of knowledge, and some people don't like this, but frankly I'm baffled at the attachment to justificationism/foundationalism after its given rise to so many dead ends and pointless arguments such as the BIV cottage industry. Taking a non-justificationist view, BIV hypotheticals are of no more interest or consequence than Dennett's epiphenomenal gremlins:

    "My example is 'did you know that there's seven epiphenomenal gremlins in every cylinder of every internal combustion engine', they have no mass, they have no weight, you can't see them, they don’t make any difference, it doesn't change the horse power, or the acceleration or anything. Now that’s silly, that's completely silly, it is not worth anybody’s time. It is however logically conceivable."
    -- Dan Dennett

    Non-justificationism allows us to be universal skeptics without tangling ourselves up in any absurdities -- we're perfectly allowed to have critical preferences for this or that theory (or "belief" if you like) even if it has no justification, as long as it's held up well against criticism. All sources of criticism (empricial, logical, whatever) are permitted, but none has any privileged status.

    "We are like seafarers, who must rebuild their ship in open sea, without being able to take it apart in a dock and build it up of its best constituents from the bottom up."
    -- W.V.O. Quine

    ReplyDelete
  4. "Everyone already agrees that we can't attain subjective certainty."

    This is false! The intelligent are full of doubt, but the stupid are cocksure. That's the trouble with the world.

    That's (pretty much) all the skeptic claims, but it's an important claim. Good skeptics adopt common-sense realism as an overarching scientific hypothesis, and, within that context, externalism about knowledge is a strong contender.

    ReplyDelete
  5. How can you have multiple levels of analyss in the same ypothetical - I meanif you define knowledge as being somthing which you know and is true surely for consistancy you mut asses it from the big picture point of view (ie you are either in a vat or not in a vat (from this perspective there is no "possibly in a vat")
    Or if you define knowledge as THINKING you reflect the outside world then it doesn't mattter if you are in a vat or not.
    surely we can think of a million examples of this and cannot gain a meaningful resolution to any of them (surely we can all see that?).

    I might be agreeing/converging with matt here in a sense.

    ReplyDelete
  6. S.E. - by "everyone" I mean "all philosophers" :) And note that the "philosophical skeptic" that I'm talking about is a very different kettle of fish from the people you call "skeptics" in everyday life. Nobody (or hardly anyone) is a skeptic in the stronger sense that I'm talking about. Still, it's a challenge we must face up to, which is why I've written about how they can be refuted.

    Genius - I can't understand what you're trying to say.

    Matt - Do you really think that false beliefs could constitute knowledge? Suppose you think you know P, but then I show you that P is actually false. Would you respond: "Doesn't matter, I still know it anyway"? Or would you instead say, "Oops, I guess I didn't have knowledge in this case after all"? Surely the latter is the appropriate response. False beliefs can't constitute real knowledge. That's why we all say that knowledge = justified true belief (or whatever).

    Now, this doesn't actually lead to the problems you think it does. What's all this about "unshakeable foundations"? Philosophers gave up on that nonsense a century ago. I've explicitly stated that we cannot attain certainty. That's no problem. You seem to be assuming that if knowledge requires truth, then we must be able to show that our beliefs are true before they may count as knowledge. But that assumption is precisely what this post was arguing against! All we need is for our beliefs to be true as a matter of fact; whether we can show this (or otherwise be aware of it ourselves), is an entirely separate issue.

    "Neither so-called "pure reason" nor so-called "sense data" can provide justifications for any conjecture because both are fallible"

    So what? You're the only one assuming that justification must be infallible. Why believe that?

    "justificationism/foundationalism"

    They're not the same thing. Foundationalism and Coherentism are rival theories about the structure of justification. They're not really relevant to the present post. 'Justificationism', as I understand it, is the view that you generally shouldn't go around believing things without adequate justification (no-one's asking for absolute certainty though!). Non-justificationism is silly.

    "we're perfectly allowed to have critical preferences for this or that theory (or "belief" if you like) even if it has no justification, as long as it's held up well against criticism."

    If it's held up well against criticism then that might well constitute justification for the belief. See here. (Probably better to continue the discussion in that thread too, since it'd be more relevant to that post.)

    ReplyDelete
  7. Richard,

    Perhaps I should backtrack a bit here; that post wasn't so much directed specifically against you as it was a general observation of what I see as the central problem with this kind of discussion. From what I can tell, the "skeptical" position you're arguing against holds that knowledge=justified true belief, whereas your "externalist" position holds ony that knowledge=true belief, regardless of whether or not we can justify it. (Correct me if I've got this wrong.) I think both positions are misguided, but the former more than the latter. I do agree with you that there's no necessariy relationship between the truth of a belief and our ability to demonstrate it, but I would go further.

    According to your view, even if someone happens to be right about something entirely by accident, he still has "knowledge." Example: Dave is a crazy fellow who believes,based solely on a dream he had, that there is buried treasure in his backyard. He goes out there with a shovel, and as luck would have it, it turns out there really is buried treasure out there. By your lights, Dave had knowledge of that treasure. This, however, seems to run afoul of our intuitions about "knowledge" as much as the example you use on me:

    "Suppose you think you know P, but then I show you that P is actually false. Would you respond: "Doesn't matter, I still know it anyway"? Or would you instead say, "Oops, I guess I didn't have knowledge in this case after all"? Surely the latter is the appropriate response."

    This is circular. When you start out defining knowledge as "true belief," of course it becomes total nonsense to speak about "conjectural knowledge," as I do. But I'm explicitly dropping that premise and saying that there's a cleavage between truth and knowledge (and between belief and knowledge for that matter, see below). If I make an assertion and you give it the modus tollens treatment, I can either say "okay, my conjecture has been refuted" if I can't find fault with your argument, or I can turn around and counter-criticize your criticism.

    Here's another quick and dirty argument as to why "true belief" doesn't accurately capture our intuitive notion of "knowledge": I have knowledge of the phlogiston theory of combustion, but I don't believe in it, nor is it true. Is my understanding of it somehow not knowledge? I think not. I'm sure you can think of plenty of other examples of having knowledge of something that you don't believe, and/or that isn't true. So yes, I'm saying that knowledge of the false is still knowledge -- just not in the sense that you're using the word.

    "What's all this about "unshakeable foundations"? Philosophers gave up on that nonsense a century ago."

    No, they didn't. Rudy Carnap, to name one, continued searching for it until he puttered out in the early 1950s. Nowadays most pay lip service to the idea, but implicitly slide back into that sort of thinking without acknowledging it, as can be seen in these kinds of discussions.

    "So what? You're the only one assuming that justification must be infallible. Why believe that?"

    I would submit that I'm not the only one, or else this entire discussion wouldn't be going on. After all, if a supposed source of knowledge is not entirely reliable, then where does the "justification" come from? (The situation isn't any better if you try to retreat from certainties to probabilities, but I won't go into that unless I have to.) If there's no general criterion for truth, justificationism fails.

    "They're not the same thing. Foundationalism and Coherentism are rival theories about the structure of justification. They're not really relevant to the present post. 'Justificationism', as I understand it, is the view that you generally shouldn't go around believing things without adequate justification (no-one's asking for absolute certainty though!). Non-justificationism is silly."

    There's nothing silly about it, though I think we're using "justificationism" in slightly different ways. The way you use it above seems vacuous -- after all, there's not even the slightest hint about what would count as "adequate justification". In any case, I'll be bold and say that there is never justification for our beliefs. It's possible to obtain true knowledge about the world, but there's no way we can ever be sure we have it.

    ReplyDelete
  8. I think you've misunderstood me on a couple of points, but I'll try to explain myself more clearly.

    I don't think that knowledge is mere 'true belief'. That would be clearly mistaken, for just the reasons you point out. Clearly something more is required. The internalism/externalism debate is over what that "something more" is. Internalists suggest it is an internal recognition or awareness of subjective justification. (So when someone asks "what reason do you have to belief that?" you can answer them.)

    Externalists, by contrast, ask for a more objective sense of 'justification'. As I (too briefly) described in the main post: "There must be an appropriate connection between the truth of the belief and our believing of it." That is, our belief must in some sense be caused by the truth. Possible worlds accounts capture this idea by requiring knowledge to satisfy various counterfactuals, e.g. that if P had been false then I would not have believed P. (Compare this to your buried treasure example: Because Dave believed it on the basis of a random dream, he would still have had this belief even if there wasn't really any treasure in his back yard. The truth of his belief -- the fact that there really is treasure in his back yard -- played no role in causing him to acquire this belief. That's why it fails to count as knowledge.)

    It's worth noting that externalism is not always more lenient than internalism. In Gettier cases, for example, people have subjective justification but lack the objective 'truth connection'. (Even though their beliefs happen to be true, the truth is not properly connected to their believing of it. Their belief was based on some mistake instead.)

    "I'm sure you can think of plenty of other examples of having knowledge of something that you don't believe, and/or that isn't true."

    There are different senses of knowledge. I've been talking about propositional knowledge, or "knowledge that [something is the case]". There is also procedural knowledge, or "know how [to do something]". And you've mentioned the knowledge of acquiantance, or "knowledge of [some object]". These are three very different things, and I've only been talking about the first sense of the term.

    You certainly don't know that phlogiston exists, because it doesn't, and false beliefs can't constitute knowledge. But of course you can still be acquianted with false beliefs. You know that the phlogiston theory is false. (Your belief that 'the phlogiston theory is false' is itself a justified true belief, after all.) I hope that's all clear. Anyway, for now we may simply note that these other senses of the word 'knowledge' are quite irrelevant to discussions of propositional knowledge.

    "I can either say "okay, my conjecture has been refuted" if I can't find fault with your argument, or I can turn around and counter-criticize your criticism."

    Of course, who would deny this? But you're merely talking about what's going from a subjective standpoint. I agree with all of that. What I'm saying is that if your belief really is false (regardless of whether you admit it), then it cannot constitute knowledge. One way to illustrate this is, as I suggested, to consider our attitudes towards our revised beliefs. Suppose you agree: "okay, my conjecture has been refuted". Will you still consider your past, mistaken belief to have been genuine knowledge? Obvious not! You will say to yourself, "I thought I knew it, but I was mistaken. I never knew it after all."

    Such an admission betrays your commitment that knowledge requires truth. If false beliefs could constitute knowledge as you suggest, then you could instead respond, "Yup, I was wrong. I admit that. I don't believe that P anymore. But I used to, and when I did believe it, I really knew it too. It doesn't matter that it was false, it still counts as knowledge!" That's clearly incoherent. It is conceptually impossible for us to simultaneously recognize a belief as being false and yet as nevertheless constituting knowledge. Deep down, we all know that knowledge entails truth. (So if it ain't true, we don't really know it. We may think we do, but then we're simply mistaken about what we do and don't know.)

    "Nowadays most pay lip service to the idea, but implicitly slide back into that sort of thinking without acknowledging it, as can be seen in these kinds of discussions."

    Where? Point to anything I've written which betrays an implicit commitment to infallibilism. (Or any other contemporary philosopher, for that matter.) You're simply making things up! Indeed, as previously noted, YOU are the one making all the misguided infallibilist assumptions here.

    "if a supposed source of knowledge is not entirely reliable, then where does the "justification" come from?"

    Justifying reasons are indicators of truth. Just like smoke is an indicator of fire. Neither is infallible. Neither needs to be. You'll need to explain your objection to probabilistic justifications.

    "It's possible to obtain true knowledge about the world, but there's no way we can ever be sure we have it."

    This is mildly irritating, because it makes me think you haven't really read anything else I've written, either in the main post or subsequent comments. Here's the thing: I am 100% in agreement with the quoted sentence. That's exactly what I've been saying all along. We can obtain knowledge in fact, even though we can't show that we have it with absolute certainty. But then, in a massive leap of (il)logic, you use this as a basis to conclude that "there is never justification for our beliefs"!? What's the link?

    To clarify my position: all our claims to knowledge are fallible. Sometimes we have it, sometimes we don't, and we can't tell either way for sure. But, nevertheless, whether we have it or not is a matter of objective fact (just like truth is). It depends upon our belief being both true and having that "something extra" described earlier in this comment. If our belief has those two features -- and it certainly can -- then it constitutes knowledge. If it doesn't, then it is NOT really knowledge, no matter what we happen to think about it. If I make an informed "conjecture", it might be justified, or it might not. (Evolutionary theory is justified, creationism isn't.) It might be true, or not. It might constitute knowledge, or not. The mere fact that we've made a conjecture doesn't suffice to guarantee that it really constitutes knowledge. We can make mistakes, after all.

    ReplyDelete
  9. seems to me your issue would always findimentally lie with the exact definition of knowledge - applying your definition is not engaging with the argument you wish to engage with (even if it is an apear to popularity).

    so precisely
    Externalists believe that dictionaries should call knowledge "justified truth"
    and Sceptics believe that dictionaries should change that to somthing like "certain justified truth" (ok dont ask me to summarize other peoples beliefs...). There is no logical way to determine which is "true" so it is just a brute political debate in a sense a debate with mr webster and Mr collins as opposed to with eachother.

    worse yet what if the person you are talking to doesnt use common usage? Then again you are just talking right past eachother even if you do win the debate with the wider english establishment.

    ReplyDelete
  10. Not really, it isn't a merely terminological debate. (Philosophers try not to waste their time on those.) Rather, as explained in the main post, the skeptic argues that we can't even be slightly justified in our beliefs. (He might point out that, subjectively speaking, we have no evidence at all to favour common-sense realism over the BIV scenario.) There are three broad responses that can be made to this, as explained in my skepticism overview, of which 'externalising' the justification is one.

    "what if the person you are talking to doesnt use common usage?"

    Stop talking to them, unless they can explain why their non-standard sense of the term is worth caring about.

    ReplyDelete
  11. Richard,

    I'm willing to step back and let this drop since I'm beginning to wonder if we haven't just been mostly misunderstanding eachother. Besides that, my head hurts. I'll just say that my position is pretty much David Miller's here (PDF). Read it if you like and judge for yourself if we're in agreement or not.

    ReplyDelete
  12. I still find it very hard to see how it could be anything other than a terminological debate.

    the three solutions you propose seem to be
    a) change definition of words to being internal (coherentism)
    b) change the definition of the words (to be "tree in the matrix" or "tree in real life")
    c) change the definition of "justified" to be somthing that can happen by accident.

    Your usage of the word jsutification in your post above leaves me with two simultanious interpretations
    1) an perfectionists view
    "of course you cant be justified in coming to any firm conclusions because you dont have enough information!"
    2) of course you are justified in doing things -
    "you have goals and you probably feel you achieve them on a daily basis so clearly many of your views are justified!"

    I see no contradiction at all there except the debate over the word.

    ReplyDelete
  13. No, the word doesn't matter. I could talk about "schmustifaction" instead, and claim that this is what's really important for our beliefs to have. Internalists and externalists would still have substantive disagreements about which concept really matters, even after we clearly delineate the words.

    I'll write a new post about terminological disputes to clear this up.

    ReplyDelete
  14. It could be argued that you are a brain in a vat; the vat being your cranium and that without it and your physical senses you would have no connection with the physical world. When I look out of the window to get to know what the weather is doing I am more or less knowingly making an observation of the weather which is different to just happening to notice it is raining or it is sunny and different again to a meteorological observation which might be made by a machine. The deliverance's of our senses are not events isolated from ideas we have regarding those deliverance's, the two comprise our experience. As for truth; I do not regard the weather as being something that is either true or not true, as to thoughts I might have regarding the weather the position is not so simple. Since my whole being is designed to keep me and others not merely existing but happy, I can say that I am more than merely a brain in a vat and that this is most certainly not a merley terminological dispute.

    ReplyDelete

Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)

Note: only a member of this blog may post a comment.