Questions to Ponder (Gettier Discussion: Nov. 14th)

Greetings fellow philosophers,

In 1963, Gettier shocked analytic epistemology with just 2 pages of philosophy. I think it will be good to get this brilliant short piece under our belt and to start to feel some of the problems in epistemology that it raised. 

So as you read the paper, ponder the following questions; we’ll be sure to touch on these questions in discussion, and perhaps go further… (also, be thinking of any of your own questions you may have):

(a) What motivated the JTB analysis of knowledge?

(b) What is the closure principle?

(c) How, exactly, do Gettier’s examples work?

(a) through (c) are exegetical questions. We’ll have to answer them first. The following two go beyond exegesis:

(d) Did Gettier really succeed? Couldn’t the examples instead be counterexamples to the closure principle?

(e) If knowledge isn’t justified true belief, then what is it? 

 

It doesn’t take much to read Gettier. But the text warrants quite a bit of thought. I look forward to a philosophically good time!

Advertisements

25 thoughts on “Questions to Ponder (Gettier Discussion: Nov. 14th)”

  1. The problem disappears if justification and truth are collapsed into the same concept. Positioning justification and truth as distinct elements of knowledge is the result of a general tendency to project and reify specious differences and relations of meaning within a single lexicon between more or less synonymous terms that have evolved within separate lexical contexts.

    1. Could you explain what you mean by 'single lexicon' and 'lexical contexts'?If 'single lexicon used in a single lexical context' means 'the sort of language a given group of people with a given interest use at a given time', then I'd simply say that the two concepts, 'justification' and 'truth', evolved in the same lexical context. If we're speaking of the concepts (though not necessarily the terms), that context is just common sense. Common sense: Consider the case where your reliable friend tells you that it snowed yesterday in her backyard. It's reasonable to believe that it in fact snowed yesterday in her backyard. However, she simply had a slight memory lapse, it was actually a few days ago – not yesterday – that it snowed in her backyard. So you have good reason to think it snowed in her backyard yesterday but, in fact, it didn't. Moreover, it in fact snowed a few days ago but you don't have a good reason to think that it did.Common sense recognizes a difference between having a good reason to believe something and that something being the way things actually are. Common sense – common sense people thinking in common sense ways – is an example of a single lexicon where both concepts evolved. Analytic philosophers just call those concepts 'justification' and 'truth' respectively. Analytic philosophy – well, actually it goes back to Plato – is an example of a single lexicon where both terms evolved.Do you see no difference between having a good reason to think it snowed in your friend's backyard yesterday and it actually being the case that it snowed in her backyard yesterday? What do you think about the case I described?

      1. See my reply farther down.A note on common sense: It's not as common as you think. Or, rather, it only exists in so far as we temporarily sink our anchors at the same harbor before sailing on toward separate destinations. (See Nietzsche's passage on star friendship.)

  2. What is “knowledge”? That is, Gettier seems to be saying the following,Knowledge = Justified + True + Belief What do justified, true and belief mean according to Gettier, and is there a determined/fixed definition for each?

  3. First of all, hats off to Michael for leading a most excellent discussion group meeting! It was fun.rnrnGoing into the discussion, my intuitions told me that (in Case I, for example) proposition (e) would in reality have two distinct meanings – one which is true and one which Smith is justified in believing is true – and that they can only be called the same proposition when forced into logical notation. It seems a little sneaky.rnrnGiven our discussion, I have interest in examining the Closure Principle more closely. Unfortunately, I haven’t had time to do that this weekend. I would like to express some very general epistemological concerns, however…rnrnCorrect me if I’m wrong, but I gather that analytic epistemology is concerned with what we can rightly claim to say that we know. Immediately I ask, can we know things we cannot say? By that I mean things ineffable – not just things that we’ve yet to figure out how to say (but can). And, if so, how does epistemology address this sort of “knowledge?”rnrnI want to ponder the difference between the phenomenon of knowing and the conditions for knowing. We certainly must talk about the two in different ways, but I’d like to find out how they connect.rnrnPerhaps the difference involves the difference between knowing and the ability to express knowledge. Now I’m thinking of Bergson, who, if I understand him correctly, makes a clear distinction between the inner and outer worlds – claiming that knowing is “immediate” and that talking about (reflecting upon) knowledge dances around what is known. This implies that different rules apply to the communication of knowing and the nature of knowing, and that when we confuse the two, we’re asking for trouble.rnrnI’ll close this rambling mess by quoting the first paragraph of Bergson’s “Introduction to Metaphysics.”rnrn

    “If we compare the various ways of defining metaphysics and of conceiving the absolute, we shall find, despite apparent discrepancies, that philosophers agree in making a deep distinction between two ways of knowing a thing. The first implies going all around it, the second entering into it. The first depends on the viewpoint chosen and the symbols employed, while the second is taken from no viewpoint and rests on no symbol. Of the first kind of knowledge we shall say that it stops at the relative; of the second that, wherever possible, it attains the absolute.

  4. First of all, hats off to Michael for leading a most excellent discussion group meeting! It was fun.Going into the discussion, my intuitions told me that (in Case I, for example) proposition (e) would in reality have two distinct meanings – one which is true and one which Smith is justified in believing is true – and that they can only be called the same proposition when forced into logical notation. It seems a little sneaky.Given our discussion, I have interest in examining the Closure Principle more closely. Unfortunately, I haven't had time to do that this weekend. I would like to express some very general epistemological concerns, however…Correct me if I'm wrong, but I gather that analytic epistemology is concerned with what we can rightly claim to say that we know. Immediately I ask, can we know things we cannot say? By that I mean things ineffable – not just things that we've yet to figure out how to say (but can). And, if so, how does epistemology address this sort of “knowledge?”I want to ponder the difference between the phenomenon of knowing and the conditions for knowing. We certainly must talk about the two in different ways, but I'd like to find out how they connect.Perhaps the difference involves the difference between knowing and the ability to express knowledge. Now I'm thinking of Bergson, who, if I understand him correctly, makes a clear distinction between the inner and outer worlds – claiming that knowing is “immediate” and that talking about (reflecting upon) knowledge dances around what is known. This implies that different rules apply to the communication of knowing and the nature of knowing, and that when we confuse the two, we're asking for trouble.I'll close this rambling mess by quoting the first paragraph of Bergson's “Introduction to Metaphysics.”

    If we compare the various ways of defining metaphysics and of conceiving the absolute, we shall find, despite apparent discrepancies, that philosophers agree in making a deep distinction between two ways of knowing a thing. The first implies going all around it, the second entering into it. The first depends on the viewpoint chosen and the symbols employed, while the second is taken from no viewpoint and rests on no symbol. Of the first kind of knowledge we shall say that it stops at the relative; of the second that, wherever possible, it attains the absolute.

    1. I really like the distinction you draw between two types of knowledge. The 'phenomenon of knowing' and 'knowing things we cannot say' on the one hand and 'conditions for knowing' and 'the ability to express knowledge' on the other. Roughly, I think the distinction maps onto the distinction between acquaintance knowledge and propositional knowledge. A wine taster's knowledge of the taste of a certain wine is acquaintance knowledge, not propositional. But – though he really can't communicate what it tastes like to me – it is knowledge nonetheless. How does analytic epistemology address acquaintance knowledge? Poorly, I think. Notice the analyses of knowledge were analyses of 'knowing that', i.e. propositional knowledge. But – and I'll just assert this – all propositional knowledge ultimately derives from (the building blocks of) acquaintance knowledge.This is one reason why I think analytic epistemology has a lot to learn from phenomenology (a branch of continental philosophy that gives a disciplined description of conscious experience). There's a lot of work to be done on how acquaintance knowledge grounds propositional knowledge…

  5. 'Truth' and 'justification' have separate genealogies that precede their technical (read: uncommon) use by analytic philosophers. Such use, moreover, is not entirely congruent with Plato's use of aletheia and logos, respectively, the latter in fact being translated as “an account.” So, first, there is no necessary structural relation between truth and justification as distinct concepts in the constitution of knowledge, and yet, second, there is enough semantic isomorphy between 'truth' and aletheia on the one hand and 'justification' and logos on the other (if only, perhaps, by means of a series of translations) for analytic philosophers to portray themselves as falling within a unified trajectory of thought stretching back to ancient Greece.The case you describe appeals to a sense of knowledge that exceeds what anyone can achieve as a human being and therefore defies the “common sense” use of the term. Specifically, it requires that I have access from within the example to “the way things actually are” precisely as you stipulate them from outside the example in order to be justified in my claim to knowledge. So, to answer your question (“Do you see no difference between having a good reason to think it snowed in your friend's backyard yesterday and it actually being the case that it snowed in her backyard yesterday?”): Sure, but it is not a difference that makes difference with respect to knowing something.

    1. The example is a case of justified false belief, so I'm confused why you think I require that you have access from within the example to “the way things actually are”. Within the example, you don't. That's the point. You don't and yet, within the example, you're still justified. The belief is justified though false; hence justification does not equal truth.(Your reliable friend, when it was snowing a few days ago – not yesterday – had “access to the way things actually are”. Interpret the phrase however you want as long as you let someone who is watching snow fall have access to the fact that snow is falling. Again, this is common sense.) You say: “Sure, but it is not a difference that makes difference with respect to knowing something?” I ask: What do you think the difference does amount to? BTW, I was not (above) defending knowledge = JTB. All I was saying is that J does not equal T.

      1. Unless you wish to revise your example, my otherwise reliable friend does not have access to “the way things actually are” with respect to when it snowed: you've already stipulated from your God's-eye point of view that her memory is flawed, and common courtesy enjoins me to entertain that fact consistently throughout our discussion. So, sticking to that fact, there is no common sense basis within the example itself for evaluating the belief as false. There is, however, a common sense basis for evaluating the belief as true, namely, the very facts adduced in its justification.But if you still wish to maintain that I don't have knowledge given “the way things actually are,” it's only because you're conflating what counts as common sense outside the example with what counts as common sense inside the example.Right now I'm drinking Jack Daniels. I know this because I just broke the seal of the bottle labeled “Jack Daniels” and without averting my eyes for a second, without even blinking, I poured the bottle's contents into a tumbler and took a hearty quaff. This is plain old common sense. But it's possible that I don't actually exist except in the deranged mind of an undergraduate philosophy student devising weird Gettier-type scenarios, and this student has furthermore stipulated that the bottle's contents have actually been swapped with Jim Beam back at the distillery by a disgruntled bottler. Now if you wish to be consistent between your evaluation of what I believe in your own example, Michael, and what I believe right now concerning what I'm drinking, you have to abandon what we have just established to be common about what I know concerning what I'm drinking.Or do you actually maintain that common sense dictates that I don't know that I'm drinking Jack Daniels despite all the evidence pointing to the fact that I am, since a transcendental undergraduate philosophy student may have stipulated that I'm actually not? Do you maintain that common sense dictates that I don't know that 2 + 2 = 4 since, after all, I might be deceived by an evil demon? Now you can patronize me all you want about common sense, Michael, but it seems to me that you are the one who has actually lost it.

        1. Michael, I apologize if that last one sounded snarky. I'm a little drunk at the moment.The case you describe is purely hypothetical. Within that case itself, however, what is purely hypothetical is the case of my friend having a lapse of memory. But is that purely hypothetical case sufficient to offset the otherwise common sense claim that I know, on the strength of my friend's testimony, that it actually snowed yesterday? Bear in mind that the hypothetical case of memory lapse is rather far-fetched considering that you have already stipulated (outside the example) that my friend is reliable. Are far-fetched hypothetical cases enough to offset common sense claims to know something?

          1. I think your last question is interesting. Given your friend's reliability, I think you ought to believe that you know the relevant fact, and you ought even to utter:”I know that it snowed yesterday in my friends backyard.”Because that's the right thing (the justified thing) to believe in the circumstance.Further, if it in fact had snowed yesterday, then you would indeed know the relevant fact.But you can't know something that isn't true. You'd be right to claim you know. The case is simply one where, though you're right to claim you know, in fact you don't have knowledge (because the belief isn't true). (Sometimes we don't necessarily use 'knowledge' (the word) in this way. Sometimes I say someone has knowledge just in case he has really justified beliefs about a certain discipline – “he knows his stuff”, for example – or something like that (compatible with each belief in fact being false). But we shouldn't let words distract us more than it takes to disambiguate them. Philosophers (at least the analytic ones) aren't concerned with that sense of 'knowledge', it isn't as interesting as the sense that picks out that intriguing state of being rightly in touch with the truth. – We want our beliefs to be true, right? – ) So clearly, since indeed your friend could have briefly forgot the day (or whatever), and since it is clearly possible that she tell you what she did despite the fact that it actually didn't snow in her backyard yesterday…the case remains one where you are justified, but in which your belief, nevertheless, is not true.

        2. “What do you think the difference does amount to?”It amounts to a disjunction between what, from a perspective internal to the example itself, is purely hypothetical and what the evidence (my reliable friend's testimony) actually points to.

        3. Let's try this: Does common sense permit me to state that it is true that I am drinking Jack Daniels given the facts that justify that statement? But how can that be if “being true” is distinct from “being justified”? In order for something truly to be what it is, it only needs to be true to the procedures by which it is disclosed to us as the thing it is.

          1. Yes. Common sense says you should claim the things to be true that you're justified to believe.Doesn't make them true though. I'm curious what you think of The Matrix. When you saw the movie (if you haven't, watch it!), what did you think about the ordinary empirical beliefs Neo likely had even while Matrix-trapped? Were they true?

            1. Alright Badda Being, in spite of my best efforts I ended checking out what you've done here. Love the Jack Daniels thing… great stuff, even made better by the fact that I'm drunk (which is probably why I'm writing this). So 2 things:1) “common sense” is nothing more than failure to actually think (try using “common sense” about what appeals to “common sense” actually ask you to do… wait, better yet just think about what it means to argue from “common sense”). 2) Badda being, just start some other discussion. Hell I'll even join in, but don't, DO NOT do the matrix crap that horrible professors have all been condemned to do because of pop philosophy that would have us substitute actual thought for shit like “common sense.”

            2. Furthermore…Justification evokes the logic of “justifying one's actions,” of aligning it with an original act or “event.” For a belief to be justified it must conform with or “be true to” that event. Justification is truth, truth is justification.Evidence is only a tool to provoke conformity, to provoke truth.

              1. hmm.As a different angle on the question, if 'justification is truth and truth is justification' what explains the fact that we have two terms here and not one? Further, why have a lot of people thought that these two terms apply in different circumstances? (Prima facie, it seems to indicate that there are two ideas floating around that are different…) What do you think explains this?Above, you said that: “It amounts to a disjunction between what, from a perspective internal to the example itself, is purely hypothetical and what the evidence (my reliable friend's testimony) actually points to.”I'm assuming that that which 'is purely hypothetical from the perspective internal to the example' is 'truth' and 'what the evidence actually points to' is 'justification'.I won't here contest this way of making the distinction, but instead ask… Isn't that a real distinction that makes a difference? Isn't it the case that purely hypothetical things matter? Doesn't 'truth' matter?Surely it makes a difference if one of our hypotheses is wrong? (Indeed, don't we revise our beliefs? Why else would we do this unless we have falsified one of our hypotheses?)

  6. I think your last question is interesting. Given your friend's reliability, I think you ought to believe that you know the relevant fact, and you ought even to utter:”I know that it snowed yesterday in my friends backyard.”Because that's the right thing (the justified thing) to believe in the circumstance.Further, if it in fact had snowed yesterday, then you would indeed know the relevant fact.But you can't know something that isn't true. You'd be right to claim you know. The case is simply one where, though you're right to claim you know, in fact you don't have knowledge (because the belief isn't true). (Sometimes we don't necessarily use 'knowledge' (the word) in this way. Sometimes I say someone has knowledge just in case he has really justified beliefs about a certain discipline – “he knows his stuff”, for example – or something like that (compatible with each belief in fact being false). But we shouldn't let words distract us more than it takes to disambiguate them. Philosophers (at least the analytic ones) aren't concerned with that sense of 'knowledge', it isn't as interesting as the sense that picks out that intriguing state of being rightly in touch with the truth. – We want our beliefs to be true, right? – ) So clearly, since indeed your friend could have briefly forgot the day (or whatever), and since it is clearly possible that she tell you what she did despite the fact that it actually didn't snow in her backyard yesterday…the case remains one where you are justified, but in which your belief, nevertheless, is not true.

  7. Re: Michael's latest.Internal to the example, there is no truth in the hypothetical situation even if we posit from our external viewpoint that such is the case: “common sense” therein dictates that we do not become true to it in our actions. In fact, to do so would be insane. It would defy common sense. It would be unjustified.Now you'll recall that at the beginning of this discussion I indicated that 'truth' and 'justification' evolved within “separate lexical contexts.” A contemporary example of this can be found in the field of library and information studies with the terms 'cataloging' and 'metadata analysis'. Cataloging is metadata analysis, metadata analysis is cataloging — which is to say that, like truth and justification, their meanings are isomorphic and for the most part interchangeable. Even so, that hasn't prevented certain theorists from trying to pin down their semantic differences in terms of essences that transcend their separate genealogies, and to establish a systematic relationship between them within a single, overarching project.On the question of “mattering”: it depends on who you ask and with respect to what event you want that person to be true to. A hypothetical situation is not an event. Its proposal, however, is. Thus I've been true to your example this whole time because it matters to me — i.e. it amuses me to have this discussion. But the hypothetical situation internal to that example cannot matter to me in so far as I am to obey common sense, least of all because no one has proposed it to me therein — unless of course you revise and qualify your example without justification, and in that sense defy the truth that has emerged between us.

    1. Thanks for this last post. The middle paragraph especially helped learn more of your perspective.I'd like to try another example. (I want to try to bring out how the distinction between truth and justification matters in practical life.)Suppose that you’re a civil engineer, and your entire job is to verify the strength and safety of a certain bridge. (It’s a massive and important bridge, you run various tests each day, let’s say.) Now, though you’re good at math, the calculations involved are so huge and complex that you rely on certain computer systems to collaborate your tests results. One night, a crazed (but intelligent) madman bent on destroying the bridge puts a virus in your computer system. The virus makes your computers calculate that ‘everything is well’ (e.g., the strength of the steel is of degree X – where X is appropriate strength), even if everything is not well, i.e., even if the steel is actually weak, and far too weak for the safety of the bridge’s operations. Having placed the virus, this genius madman gradually (and imperceptibly, from mere visual observation) weakens the bridge’s strength through application of (say) certain acids at certain steel joints. After a few weeks of this (and throughout this time you work valiantly at your verification work), the bridge (to your total and appropriate surprise) collapses, killing over a hundred people.Now, before the collapse, it seems correct to say1. You (the civil engineer) were justified in believing that the bridge was safe.2. The bridge was not safe. What you were justified in believing was not in fact true.To bring (1.) and (2.) out clearer (in a practical way), the subsequent federal investigation into the bridge’s collapse should absolve you (the civil engineer) from any wrongdoing once they discover the foreign virus on your computer systems. In other words, you had done your best – you had no way to know about the virus – you were justified in believing what you did about the bridge’s safety. However, the very fact that the bridge collapsed due to weak steel joints means that your justified belief was not in fact true. Does truth = justification in a case like this?

      1. Yes. The semantic isomorphy of truth and justification makes no difference whatsoever to my absolution post-investigation. Granting that truth is a matter of fidelity, I am held accountable for the collapse only to the extent that I was not true to my duty as a civil engineer.All you are doing here is taking the form of our external relation to your previous example and inserting it into a new example as part of its internal structure. Here, however, the relation is not between a “real” situation and a hypothetical one but between two points in time inside the example itself.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s