Ep. 344: Gettier and Goldman on Justified True Belief (Part One)

0:07 Mark: You're listening to the Partially Examined Life, a podcast by some guys who at one point set on doing philosophy for a living, but then thought better of it. Our Questions for episode 344 are what is knowledge and what is justification when it comes to beliefs? And we read some articles Is Justified True Belief Knowledge by Edmund Gettier from 1963, what is justified belief by Alvin Goldman from 1979, and the the Inescapability of Gettier Problems by Linda Zagzebski from 1994. We also read the Stanford Encyclopedia article entry on Reliabilism by Alvin Goldman and Bob Beddor from 2021. More information about these articles and the podcast please see partiallyexaminedlife.com this is Mark Linsenmayer self presenting in Madison, Wisconsin.

0:53 Seth: This is Seth Paskin, sure that he is in Austin, Texas via his perfectly reliable clairvoyance faculty.

1:00 Wes: This is Wes Alwan correct only by chance in Cambridge, Massachusetts.

1:05 Dylan: This is Dylan Casey wondering where the degrees of certainty are in Madison, Wisconsin.

1:11 Mark: All right, so we had had this Gettier article which is three pages long, by the way.

1:16 Wes: Do we know is it Getty A or Gettier?

1:19 Mark: Just on videos people were calling him Getty.

1:21 Dylan: Okay.

1:21 Mark: Unless I'm misremembering that.

1:23 Wes: Oh wait, that's a reliable process, Mark.

1:29 Dylan: How do you know?

1:30 Wes: If we count on that, then what can we count on?

1:32 Mark: How about we. I'm googling Gettier pronunciation. I'm gonna listen.

1:36 Wes: I think we have to leave this in just given the pronounced names doc. Given the relevance to this episode.

1:41 Mark: All right, it's supposedly Getty A according to this random YouTube video. Hey, and this is Mark during the editing phase. It is Gettier. It's normally Gettier. It's a French name. But he was an American, he went by Gettier. So we mispronounce it for the rest of the episode. But keep in mind I did have the right idea initially.

1:59 Wes: See, I thought I knew that.

2:00 Mark: So yes, this famous famous paper we had had requested that we covered on the show before, but it just seemed too insubstantial, I guess. I don't know. This came up again, I think in a close reads context and certainly the literature inspired by it is plentiful enough. There are many YouTube videos that I looked at.

2:20 Wes: You're really skimming over that Close reads context where we were fumblingly trying to remember what reliabilism was all about.

2:29 Mark: Well, I was. So you picked though, once we determined we're going to do this three page paper, what else we were going to do and found this Alvin Goldman paper. Some of the videos had recommended the Zagzebski paper but it's not exactly a completely linear thing from the Gettier to the Goldman. And I'm wondering what like. In fact I was looking online at just epistemology class syllabi and they're. They're like in different sections. They're related in that Gettier is asking as the title signifies is justified true Belief knowledge. This goes back to Plato and Aristotle and he comes up with cases where a belief seems justified, it seems true, it certainly believed, but yet we would not consider it knowledge.

We'll get into that but then the movement to well, we just have to change our notion of justification. We have to. We have to work on that notion. So I guess that was the direction that Goldman was headed. So it was at least inspired by this initial thing. Though most of the videos I was looking at that were talking about the Gettier and the literature surrounding it. We're not talking about Goldman specifically. Did you have any thoughts, Wes? Was it just the Stanford article that led you there by Goldman, the author

3:38 Wes: of that Stanford article?

3:40 Mark: He's one of the authors.

3:40 Wes: That's hilarious. It's another evidence gathering problem there. If the guy writing the Stanford article recommends his own paper, have we chosen the right paper? I forget.

3:53 Seth: Yes. When we read Tomasello I remember like half of the citations in Tomasello were Tomasello. See Tomasello 2000. See Tomasello 2004.

4:04 Wes: He's written a thousand books and he just repeats himself so much. But he's great. But anyway, I forget I do probably Stanford played a part in that. But I think this is the big sea change as Goldman describes his own paper. This reflects a sea change in epistemology. So yeah, the Goldman paper isn't directly addressing Ghettier, but it is an attempt to sketch out a new epistemology in light of the fact of the Gettier problem. It's just kind of implied, which is just that yeah, we can no longer say that justified true belief is the same thing as knowledge. We need to either add something to justification or we need a different account of justification, which is what the Goldman paper does in painstaking detail.

4:55 Dylan: In order to get to knowledge or

4:58 Wes: in order to say that true belief is knowledge, we need either to revise concept of justification that didn't exist before or we need to say it's justified true belief plus X. This is the way Zagzebski puts it. Is that right? Zagzebski.

5:14 Mark: Zagzebski that is a phonetic pronunciation.

5:17 Wes: Take them apart. Zagzebski. Okay.

5:20 Dylan: Zagzebski. So I found myself feeling in the weeds in a sort of analytical philosophy forest in this, where I really wanted to understand what explicitly everybody meant by each of these terms, like understanding the problem. And it took me a little while, honestly, for instance, to get to that knowledge. So this may sound stupid, but to get to that knowledge has to be true when you know something. So that was part of the deal. Right?

5:55 Seth: Right.

5:56 Dylan: Okay. And so that's the whole.

5:59 Wes: The belief has to be true.

6:00 Dylan: Yeah, well, for it to be knowledge, whatever it is, when you say it

6:04 Wes: is knowledge, truth is a property of the belief and that's part of what confers knowledge. What's part of what makes it knowledge.

6:10 Mark: But I think Dylan's right semantically that if knowledge consists of true beliefs that you have, so you can say this knowledge is true. I mean, unless that's just. Just redundant.

6:22 Wes: Because it's redundant.

6:22 Seth: That is, that is redundant.

6:24 Wes: If we're going to be analytic philosophers just for one episode, we got to behave in a nitpicking way like analytic philosophers.

6:31 Mark: You can't say these tautologies are true. I mean that's just as.

6:35 Wes: I mean, maybe it's trivial.

6:36 Mark: Trivially true.

6:37 Wes: Okay.

6:38 Dylan: One equals one has got to be part of analytic philosophy.

6:41 Seth: Right? Go ahead, Seth. So I have strong memories of reading this Gettier paper back in the day and hearing the stories about how he managed to get tenure off this four page paper and never really wrote much of anything else ever again. I don't know if those are apocryphal or if that's actually true, but this is like the best and worst of all analytic philosophy. Little like Eddies to get caught in because it perfectly demonstrates the enterprise of undertaking painstaking parsing, linguistic, yada, yada kind of analysis on a problem that should not have been a problem to begin with.

That is just something that everybody was taking for granted because that's really the way this starts. Right. So it starts with according to Gettier and really in the literature there's this idea that there are to be able to say that you know something. There are essentially three conditions that need to be met and that this thematically started with Plato and has been sort of just assumed in the history of philosophy. Whether that's true or not, I don't know. But. And those three things are, as Dylan was pointing out, one. So we'll use the language in the article.

S knows that P if and only if P is True, S believes that P and S is justified in believing that P. There are different formulations that he brings up, but essentially the general problem statement here is that to say that somebody knows something, the thing has to be true. You can't know something that's false, you have to believe it. So there's a truth component, a belief component, and then a justification component, however you want to characterize it. Meaning you had to have come to that belief that in that thing, through some reasonable myth, you have to be justified.

8:36 Dylan: Whatever justified means. Whatever it means is the thing that you did.

8:40 Seth: Yes, exactly. So that's the problem statement.

8:43 Mark: We got that explicitly in Plato, right? That if you merely know how to make something, if you're doing it, have a techne, then it's not real knowledge because you're just sort of doing things out of habit and they happen to work out. And so you have to have the reasons for things. This was actually a big finding. It is considered in the history of philosophy to be one of the, I think two things Plato contributed, that justification is actually necessary. It seemed like, Dylan, you were asking, where did the true part come from? Like, is that something you wanted to pick up?

9:12 Dylan: Well, I just feel. I felt like I was stupid and I like the way that Seth articulated it or explicated what was just said in the Gettier article is that the knowledge part requires the true part. Because I found myself falling over in the Zagzebski article. And I can get into this a little bit more later because there's something I want to unpack this. There's something about this that just feels just totally weird to me because at the end it seems like we're making a big deal about, well, we can't be certain about things. There always be a way in which you're not 100% certain. And knowledge requires certainty. And I'm like, are you surprised by this?

9:54 Wes: Yeah, I don't think they're concluding, I mean, there's no conclusion that knowledge requires certainty. I mean, I think you're right. We could discard the concept of knowledge altogether and agree with, I think probably many continental philosophers that maybe there is no such thing as knowledge, strictly speaking, but we might want to say that some things are closer to knowledge than not. And in that case we want to define ideal limits. We want to say, well, if something were knowledge, what would it be? And that way we can say what counts as being closer to it and farther away from it so the problem doesn't go away even if we become or skeptical and Say that strictly speaking, we don't know anything, which I'm a very sympathetic to that idea.

But the other part of this, with saying that P is true, false beliefs can be justified. That's really important. So you have to specify truth because it's a very, very important thing that we can be justified in believing things that are untrue.

10:49 Seth: No doubt. But there's a similar sort of sense of disquiet in my soul as what Dylan articulated. Because I think what this really shows is. So imagine you're talking about something like the rationalist era skepticism, right? So go back to idealists versus realists, all that kind of stuff. Your skepticism and your doubt about whether you can be certain in your knowledge has to do with your epistemic relation to the world. And by the time you get to the 20th century, people are sort of beyond that a little bit, and they're

11:21 Wes: like, okay, well, can I just jump in? None of these authors are saying knowledge requires certainty.

11:26 Seth: No, no. Well, I'm not talking about certainty. I'm talking about skepticism.

11:29 Wes: I just want to. Because you use the word certain again, and Dylan used it, and make clear that. That. That is not the thesis here.

11:36 Seth: Okay, so then the point that I'm trying to make is the historical roots of skepticism are skepticism about two things. One is truth at the level of truth, and then skepticism about the acquisition of knowledge, like how you come to know things. And this casting of it in terms of saying, let's just assume, let's give that. That P is true epistemology and knowledge is really about understanding beliefs and how we come to those beliefs, not so much about perceptual systems or the way in which. So when Gettier blows this whole thing up, it freaks everybody out, I think, because it's as though we had settled into a paradigm that people felt comfortable with with respect to epistemology that had kind of, if not solved, at least put aside certain kinds of questions.

And then now we're right back to the same place where we have all of this, let's just call it skepticism about how we can know things and how we can know that we know things and so on. And so I think that's why this was such a big deal.

12:40 Dylan: Can I replay that back to you a little bit, Seth, or just rephrase it? So when I read the Gettier paper, I thought, okay, so he's being clever. And the reason I focused on certainty, and maybe it's the wrong word to do, is that he constructs examples in which you have justified true belief. But it turns out that there is some exception that in a very understandable and clear way shows that you don't actually have knowledge. You were wrong. It's not true. So what your belief is is not true. And so therefore you don't have knowledge. And so that's clever and all that stuff.

And it makes sense. The way you articulated this up to me sort of seems to make the statement that well, this rocked the world because the space for the kinds of uncertainty about that truth exploded to be much larger because of Gettier. Because like the examples that he gives, is it this is the problem that there is one, or is it that there's effectively an infinite number, there's an infinite number of ways in which your justified true belief is not knowledge. You should basically show.

13:49 Wes: Why don't I sketch this out real quick? Because we just keep talking about it without saying what it is for the listeners. Sure, we haven't explained the Gettier problem, but we're critiquing it.

13:58 Seth: Yeah, okay. All I did was try to lay out the problem.

14:00 Wes: So yeah, you could have justification for belief and yet and it turns out that the thing you believe is true, but it's just that the justification you have doesn't line up with why it's actually true. You're just accidentally right. So the example he gives will just give one of them. Smith has good evidence that Jones is going to get a job because the president of the company told him that. And we're going to treat testimonial evidence as evidence as good evidence. Jones has 10 coins in his pocket and Smith has seen that. So we get two pieces of evidence that are direct in a way, and then we get an inference.

And this is where the problem comes in. When you start making inferences, Smith infers that the person who gets the job will have 10 coins in his pocket. Well, it turns out that he, unbeknownst to himself, has 10 coins in his pocket and that the president was lying, that it's actually he, Smith who's going to get the job. So he was justified in believing his inference that the person who is going to get the job is 10 coins in his pocket. But that doesn't make it knowledge because his justification is actually unrelated to why it is that that proposition is true.

So that's a quick and dirty explanation of what the Gettier problem is.

15:19 Dylan: Thanks for that, Wes. And so as you articulated, the key is there he believes something that's true. But the process of justifying it involves things that weren't true. So it's not justified properly. Yeah, it's.

15:34 Wes: He was justified in believing those things even though they weren't true.

15:37 Mark: Yeah. Getty does not give an analysis of why I think what you just said is reasonable. But it's. He just ends after three pages. He's given his two examples, that being one, I'm sure we'll get to the other one, and. And then he's out. So it's this vast literature of like, well, okay, what was wrong, even the way you put it, Wes, that the things that made it true were unrelated to his justification. I don't know if he even says that much. Does he actually say why these don't work? Or he just says, I'm showing this. No, doesn't work. Bam.

16:09 Wes: He doesn't. But I think. Does he use the word accidental or is that just in the zag. Zebsky. She harps on this whole idea of the accidental. And that's the way I read Gettier.

16:21 Mark: Yep.

16:21 Seth: So he says right after the intro, where he talks about various formalizations. I shall argue that A, which is the first formulation of the traditional view, is false in that the conditions stated do not constitute a sufficient condition for the truth of the proposition. The same argument will show that B and C fail if has adequate evidence for. The point is, if he's saying that those three conditions, it's true, you believe it, and you're justified in believing it are not sufficient. So essentially, I think the way you characterized it, Wes, is one way that it can fail. But he has in mind that being justified in believing it is not sufficient because there's going to be other conditions that need to be applied to ensure that that justification is appropriately lined up.

17:10 Wes: It is justified. True belief. For Getier, he's not saying it's not justified because the premises turned out to be false or because the premises were false, but they were justified. And one of his important premises is B. Right. Or the second thing he mentions, which is that if your initial propositions are justified, you're allowed to believe in the entailments. The entailment is justified. And that's what we have here. We have, like I said, an inference. And yeah, he is justified in believing the President. He is justified in his believing that Jones has the 10 coins in his pocket. And he does. You know, as Zagzebski shows, you could have true. Both of the premises could be true, and you could still get a Gettier problem, which we'll get to, but so you don't have to have A false piece of evidence.

17:57 Seth: Yeah. So Smith has 10 coins in his pocket and is applying for the job. Jones has 10 coins in his pocket and is applying for the job. Smith was told by the president of the company Jones will get the job. So Smith thinks whoever gets the job will have 10 coins in his pocket. He knows the other guy has 10 coins. He doesn't know that he has 10 coins. And unbeknownst to him, the president is picking him for the job. So the point is, the conclusion that he says the guy who gets the job will have 10 coins in his pocket is correct.

18:28 Wes: And justified.

18:29 Seth: And justified. So it's true, it's justified and Smith believes it. The only thing is it's a different guy.

18:36 Dylan: So it's not knowledge.

18:38 Seth: Well, it's definitely not knowledge, but it is, it does satisfy all three of those conditions and that's, it's just justified true belief.

18:45 Dylan: But Gettier will say it's not knowledge.

18:47 Mark: Yeah. Do you think, does anybody want to argue that actually it is knowledge he does know? In fact, I feel like a sort of de dicto de re thing sense reference thing might help us. I don't know if that's exactly the right word, but you know, he knows de dicto those words. The guy that has 10 coins in his pocket will get the job. He knows that claim. He knows it is true what he thinks it means, what it actually points to, that it points to him and not to Jones. That is where the mistake comes in. So it's like it depends how you characterize what the belief is, but that

19:22 Wes: just identifies knowledge with true belief irregardless of the justification.

19:27 Mark: Well, the question is whether his, what you called principle B, just that logical entailment works. So I mean we can think about this.

19:37 Wes: Justification transfers with the entailment.

19:39 Mark: Right. Justification transfers with logical entailment. So we already know that if A is true and then A, then B is true, then B will be true. But if I believe that A is true and I'm justified in believing that, I'm justified in general in believing the modus ponens. Because that's just basic. Of course, but those are two different kinds of justification. And so can you then combine, actually say I know the conclusion, I'm justified in believing conclusion based on those two. I mean, Gettier thinks there's no problem here. Like I had legitimate evidence for one. Obviously a logical truth is going to be self evident. So therefore I definitely have evidence, I have justification for the conclusion. I, I, I kind of want to push back on that I'm not sure that the conclusion is clear enough.

20:26 Wes: Then no truth that we know that depends on inference would be justified in that case.

20:32 Seth: Yeah. There's a couple things that are tricky, I think, about this particular example. First of all, I have to question whether something in the future can be considered true, like, Jones doesn't have the job. Smith was told by the President, Jones will get the job. And so he's saying, oh, he knows Jones is going to get the job. Like, that's a weird thing.

20:52 Wes: He doesn't know Jones is going to get the job.

20:55 Seth: Well, he believes he's justified in believing that Jones is going to get the job because the President told him so. So.

21:02 Wes: Yeah. So testimonial evidence. One could question the value of testimonial evidence. That would be a.

21:07 Seth: Or of not just testimonial evidence, but future states versus current states. There's a modal component here because this is something that hasn't happened that is theoretically going to happen.

21:19 Wes: Yeah. Evidence always involves time. This is one thing that Goldman is going to point out. It happens over time.

21:25 Mark: Couldn't it involve, like, who's been poisoned? I believe that the person with 10 coins in his pocket has been poisoned. So it doesn't have to be a future thing. Like, I don't, I don't think that making in the future significantly changes it because you'd have exactly the same issue.

21:39 Seth: That's fine.

21:40 Mark: I actually was poisoned. Oh.

21:41 Seth: You know, but remember, the other thing he says is he doesn't say, I have justified belief that Jones is going to get the job. He says, the thing is, I have justified belief that the person who gets the job has 10 coins in his pocket. And then I think there were other examples, either in the literature or the second Gettier thing where he starts using logical operators like, or, you know, where you have like truth values and conjoined sentences and things like that. That gets kind of weird. I'm like, to me, that's a bit of a sketchy kind of counter example. I look for something a little more concrete. Do we grant the claim of the paper that justified true belief is insufficient for knowledge? Because if not.

22:23 Wes: Yes.

22:23 Seth: There's no point in going on.

22:25 Mark: Well, it certainly raises the issue of, like, well, what do you mean by justification? And that's, you know, what we're going to chase for the rest of.

22:31 Seth: Right. Which is really the rest of the podcast. Yeah, I think is more or less

22:35 Mark: about that topic, because you can't just say he's correct if you don't Know, well, if you don't have. You could have. You could question the theory of truth. We did multiple episodes on what truth entails. I thought maybe Dylan was starting to pick at this. Like, what if you're, you know, a pragmatist. So in fact this is one of the things that comes up way later in, in like the. The Stanford article. But these evil demon cases, if, you know, I'm in a situation, if I'm in the matrix and I say I ate a sandwich and, and I'm pretty justified in thinking I ate a sandwich, but actually I didn't eat a sandwich because there are no sandwiches and you know, I'm getting nutrients pumped into the back of my neck.

A pragmatist is still want to say, well, it's actually true that you ate a sandwich. Because truth is dependent on a sort of frame of reference and it actually is dependent on epistemic norms. It is internalist in to some degree. It is not purely externalist. So we might want to say, well, in a sense it is true they ate a sandwich. But ultimately, you know, from the frame of reference of the people outside the matrix, it is not true that it's a sandwich. So we can pick at that. That's not what we're picking at in this episode. Or we could pick at the justification.

23:41 Wes: Let's pick it things.

23:44 Mark: Can we give the second example? Because that is a present tense one.

23:48 Wes: Do we need that?

23:49 Mark: It's at least interesting in that it's not modus ponens, you know, an if then statement. It is.

23:54 Wes: Well, it's still inference. Right.

23:56 Mark: It's a disjunction though. Yeah, it's an inference to a disjunction.

23:59 Wes: So, yeah, so we can make these inferences. You know, we can take something that we are justified in believing. In this case, it's Jones owns a Ford and Smith has lots of reasons for believing that Jones owns a Ford, including perceptual evidence and blah, blah, blah. And then he can make an inference from that proposition to either Jones owns a Ford or Brown is in Boston. You could just add these things that, you know, get any propositions you want these, you know, it's just a logical truth that these or propositions and you know, conjoining things with ORs, you know, you'll get a true. If one of them is true, then the whole conjunction is true.

24:39 Mark: So disjunction, not conjunction, disjunction.

24:41 Wes: Sorry.

24:41 Mark: Yeah. Which is a very weird thing for people who have not taken formal logic

24:44 Wes: that it is how the OR operator

24:47 Mark: It doesn't mean one of these alternatives must be true. I mean, it just means. Well, it means at least one.

24:53 Wes: Well, if P is true, then P, or Q or. Or R, that whole disjunctive statement is necessarily true is entailed by P being true. Right. Modus ponens is a different. And these are. They're actually related. Right. Because these things transform into each other. So it's not completely unrelated to modus ponens. So that's a weird. Yeah, it's going to look weird to someone who hasn't done formal logic, but it's important in formal logic that this is the case. So, yeah, once you are justified in believing that Jones owns a Ford, you're justified in believing that either he owns a Ford or Brown is in Boston or he's in Barcelona or in. He's in Brest Litovsk. Or you can do a bunch of different disjunctive statements. In other words, by coming up with a random city and placing Brown in that city, even though you have no idea whether he actually is in that city.

25:44 Mark: So it turns out that Brown does not own a Ford. So actually the initial thing was false, even though you were justified in believing it. But it happens that Brown is in one of those cities. So the disjunction between Brown owns a Ford and the other guy is in this city ends up being true.

25:59 Wes: So the inference is justified. Right, Because Smith is justified in believing Jones has owns a Ford even though he doesn't. And Smith made a good inference. The entailment is there, so the conclusion is justified. But it turns out, yeah, he doesn't own a Ford and Jones happens to be in one of those cities. So it's just luck that he got. He's not right for the right reasons, he's right for the wrong reasons, which are just completely accidental. So he's justified, but the justification isn't what he thought it was.

26:32 Mark: Right. Gettier says he's justified because of, again, those two things. He was justified in the first, actually false proposition that the guy owns a Ford and then he was justified just in that. It's another rule of logic. And so obviously that's justified. And belief justification is supposed to pour through these rules of logic. And I think this is what's being questioned is not, you know, I think the people arguing against Gettye are not saying, you're right, there needs to be something in addition to justified true belief. They're just saying, well, that's not real justification.

Real justification means that the belief actually has to Be causally tied, for instance, to the fact of the matter. It has to be the fact that Jones is in Barcelona that caused me to believe this. Just junction something or Jones is in Barcelona. The fact that nothing to do with it means is why it's not justification at all.

27:23 Wes: Yeah, well, no, this is a semantic decision.

27:25 Seth: Now you use the magic word, Mark, so.

27:27 Wes: Cause you can say, right, this is just semantics. We can say we're going to revise a concept of justification or we're going to take the simple concept of justification that Getty is using and add X as Zagzebski puts it. So, right, she's. It's either knowledge equals JTB or justified true belief plus X or knowledge equals true belief plus justification reconceived as. Reconceived as warrant or well founded. So it doesn't matter. It doesn't matter which way we say it.

27:58 Dylan: JTB +x or j prime TB, JTB

28:03 Wes: +x or TB +x where x is J reconstructs J prime. Either way it's the same thing. But yes, we need to revise our notion of justification or add something to this to Getier's notion of justification, which is just the simple accepted version at the time.

28:23 Dylan: I want to articulate a little more clearly, at least for myself, what would success look like? As Zagzebski noted, the problem in the Gettier problems is that there's an accidental string of causes that leads to true belief. And so what we believe is true,

28:41 Wes: well, there's bad luck, then there's good luck.

28:44 Dylan: Okay, what we believe is true, but we don't want to call it knowledge because the sequence of justifications, the causal sequence, whatever you want to call it, is unsatisfying in some way. And in particular in the case with the Gettier problems, it involves things that weren't true or accidental. Obviously they weren't justified belief. So there's a, a string there and is what good looks like that we have every time we have something justified as true belief that it is actually knowledge that's actually true. So we basically have a formula, a process to always be certain. Or is it that we make it so that whenever there's. That justification doesn't have any accidents in it?

29:36 Wes: I don't think this doesn't operationalize anything so that we can be certain of things. None of this confers certainty and it doesn't give us a formula for making sure things are justified. In part, right, because justification is comparative in a matter of degree and we can be justified in believing things that are false, which immediately eliminates the relationship between justification and certainty. And as Zagzebski points out, you can't ever tighten those up completely. You can't ever get completely from justification to truth. So. Or at least you'll say that we could.

30:10 Mark: You can if you cheat.

30:11 Wes: But hold on, hold on. So the point is that the goal of this is simply to understand the concept of justification. What Gettier reveals is just that. We don't fully understand the concept of justification. We don't fully understand the concept of knowledge, and we don't fully understand their relationship. It was intuitively plausible that they're related, and Plato got us that. But if you dig further, you see that it's not enough and it's just a matter of curiosity about what knowledge is, what justification is. There's no practical upshot this to this, unfortunately.

30:43 Dylan: Maybe I'm just being completely boneheaded, but what does good look like then?

30:49 Wes: What do you mean? What does that question mean? What does good. What look like a good explanation of knowledge?

30:54 Dylan: What would be successful?

30:56 Wes: Maybe there is no successful. Okay, how do you do where you

30:59 Mark: can't come up with counter examples? That's. We did the same thing in our Grice episode last summer where it was. This is the analytic. It's really infuriating. And the only good thing about the Gettier paper is he just stops really quickly. But he could easily write something like Goldman did. Or what we saw in that Grice paper, where you just keep bringing up more and more examples and just refining your wording and saying, oh, well, here's another example that doesn't work for this new wording. Or is this brings out some counter. But this is just the game of analytic philosophy. Is. So.

31:30 Wes: Yeah, I'm just as annoyed by analytic philosophy as everyone. All right, so you can't beat me on that count. But I'm trying to get what's positive about this. Sure, it's some reflection on the concept of knowledge and the concept of justification. So all we get in the beginning is, oh, it's an aporia. I don't really fully understand knowledge. I don't fully understand justification. And this formula that people have been repeating for centuries. Or have they? It's another historical question, but is not adequate. Okay, so success looks like resolving an aporia.

32:02 Mark: Okay, A word that I don't know, that we even talked about in our aerosol episode about this.

32:06 Wes: It's a puzzle.

32:07 Mark: It's a puzzle.

32:08 Dylan: Okay, it's a puzzle.

32:09 Mark: What is distinctive about an aporia as

32:11 Wes: a type of puzzle, well, it's kind of arresting. It blocks up the flow of theory onto more productive terrain. It's not just out on the sidelines.

32:21 Dylan: It's.

32:21 Wes: Yeah, it's a bottleneck I just want

32:23 Mark: to dispel because I'm wondering like, did these guys read Frege? So a Fregean example would be like, I believe Superman is strong and Superman is in fact Clark Kent. I don't know that it doesn't follow then that I believe Clark Kent is strong. Like that's Frege's point, that belief doesn't flow through. Unless I believe Superman is Clark Kent, then yes, I will believe Clark Kent is strong as well. And is there anything about this distinction between Superman is in fact Clark Kent, I believe Superman is Clark Kent. That would be, that would shed light on this coins example.

For instance, you know, I believe that, that Jones has 10 coins. I believe Jones got the job, therefore I believe that. I mean, I would want to say I believe that Jones, who has 10 coins got the job. But what he wants to say is I believe that the guy who has 10 coins or the recipient of the job has 10 coins. So we've introduced some opacity. We've introduced, you know, a sort of Clark Kent, Superman situation here. And I'm just not sure if this is legitimate.

33:26 Dylan: Well, what it is is that it's not that it's illegitimate. It is an inference that you can make. It just turns out that it's a gap that you can't make. A Like the example when I was immediately brought to was like the example given here is a perfect example how all kinds of mistakes are made. You know, from. It's the way in which a sitcom works and it's also the way in which bugs arise in code and the way in which conclusions are. The wrong conclusion is made in a physical experiment is the inference has a gap in it. And so you end up just being wrong. And so.

34:05 Wes: Well, it's not a bad inference, right?

34:07 Dylan: Yes, it's not a bad inference. And you get down to the quality, you get down to thinking about, well, what are all the things I'm going to do in order to make the degree of the alignment between my justified true belief and my inference. Because like the problem in the Gettier one is, seems to me is that if you go on to make other conclusions based upon that gap, you're going to be wrong on other things, right?

34:35 Wes: Yeah. I don't think reliabilism, you know, as a theory helps us avoid being in the spot of someone like.

34:42 Dylan: No it doesn't.

34:43 Wes: So it's not going to increase certainty. It just tells us because this lead, because G leads to reliabilism. I think it's quite interesting. It's sheds some light on how we ought to think about justification because we will, you know, as we move forward we'll see that we that we're going to change our conception of justification from simply involving these internally accessible belief states to processes that lie outside of the awareness of the believer. And those are right, those are evidence gathering processes processes, belief formation processes which themselves we don't necessarily have access to. So that's really. There are important consequences for that. Right. Including Cartesian foundationalism not working out as an epistemological theory as Goldman will get into. There are lots of important things that flow from this problem and a reliabilist analysis of justification.

35:48 Mark: Yeah there's this distinction between internalist justifications. We only have to look at what the person actually thinks that he believes that versus externalist versus you know, that there could be actually your belief is justified even though you don't know it. Just like you don't know whether your belief is true. So that truth is an externalist thing. But this is the intuition that I'm running against in trying to say actually Smith does not believe that the guy who has 10 coins got the job. He doesn't have that belief at all. Because I want to say a belief is something that I understand the contents of and what he thinks his contents is. Jones who has 10 coins, he doesn't

36:29 Wes: just making an assumption about what someone is likely to believe under those circumstances. And it really applies to this specific case that doesn't apply to every Gettier problem. So yes, in that circumstance a few of us are going to say oh, the person who is going to get the job has 10 coins in their pocket. It's not irrelevant. You're just going to think Jones is going to get the job. But there would be circumstances in which we might think about that.

36:54 Mark: If somebody asks you the question do you assent to the fact that the person, that the person, whoever it might be, has 10 coins in his pocket.

37:03 Wes: You've been primed in some way, right?

37:05 Mark: Yeah.

37:06 Wes: A fortune tellers the person who gets the job is going to have 10 coins or something stupid like that. So there are circumstances in which which it might become relevant to think about to form the belief that the person who gets the job is going to have 10 coins in his B. You know, it would have behooved maybe Gettier to come up with that kind of example where there's an obvious. Where the belief seems like a likely belief to have formed. But I don't think it matters because that's not the point. And we could come up with examples. I can't on the spot, but I think it's in the zag Zebsky, where, you know, it's.

It would be perfectly ordinary belief to have formed and it's justified and true, but it's not knowledge. And we do have to accept intuitions. I mean, I think this is part of what Dylan, Mark, you're getting at. I mean, we do have to. There is a fundamental intuition about what knowledge is. That's at work here. That's not that we can't justify any further.

37:57 Dylan: Can I just want to ask a question about knowledge that was bugging me when we talk about internal, external. The way all of this talks about knowledge involves, I'll call it facts about the world. And there's another way in which we use knowledge in where we might say something is true. So if I say Mark knows how to play guitar, but if Mark can play guitar but doesn't know how to explain to anybody how to play guitar, is that just an exception? Or we just. We would say, well, he actually doesn't know how to play guitar. He plays guitar, but he doesn't know how to play guitar.

38:32 Mark: That's the Plato's question. Yeah, we might say, well, it's a kind of knowledge. It's just not explicit knowledge or something.

38:39 Wes: Well, this is a different. This is procedural knowledge. This isn't propositional knowledge. So nothing that we say about propositional knowledge implies anything about procedural knowledge necessarily. Maybe it does.

38:51 Mark: But yeah, I. I kind of want to. I like that. Because then we could say, you know, do I really have the belief that either Jones owns a Ford or Smith

39:02 Seth: Brown lives in Barcelona?

39:04 Mark: Brown lives in Barcelona. Do I really have that? What actions would I take? Like, if I want to see really what the belief is, you might say it's not just given the proposition. And the proposition has a certain amount of vagueness because the. The guy that. Or whatever. Because there's these opaque parts of it. But because if I go to find the third guy, do I go to Barcelona or not? If I don't do any of the things that are accompanying that disjunction, then I don't actually believe the disjunction.

39:33 Dylan: I don't know.

39:34 Mark: Because a disjunction is like grue and bleen. Right. Just a disjunction is an Unnatural category that if you want to say I believe this thing, I believe this proposition, but then you disjoin it to something else. I believe this or this other thing. Unless they sort of fit together. The grue-bleen thing just means like there's a natural category of green, but there's an unnatural category of green before yesterday. And then blue.

39:58 Wes: You're focused on things that are just specific to this particular example. They don't come up in all Gettier cases.

40:05 Mark: That's why pick the other one anyway.

40:07 Seth: Yeah.

40:08 Wes: So it doesn't say anything about the Gettier problem if you don't like the example.

40:12 Seth: Yeah. Now the disjunction is even worse. It's like having a false premise as your fp. Then Q, you know, if P is false or whatever, then anything's true. Right. Or I'm sorry, anything can be inferred from a false premise. I think that's just not helpful in this circumstance. Like saying you can append any kind of truth, you know, proposition in a disjunction to something that you're trying to define knowledge is not helpful, I believe

40:38 Mark: so it was not in one of the ones we read, but one of the initial. What you just said in the videos I was watching on this is one of the suggestions that was initially made right after Gettier poised. This is. Well, justification doesn't flow through if there are false lemmas, which I forget exactly what a lemma plays in mathematics, you

40:56 Wes: know, but it's like a sub. Sub theorem.

40:58 Mark: Yeah, yeah. So there are false beliefs involved. The fact that you falsely believes that Jones is going to get the job or falsely believes that the guy owns a Ford. The fact that those are involved, that is what makes the justification not flow through. That it's not a truly justified anyway. Or in other words, knowledge is justified. True belief plus no false lemmas involved. This is at least one of the. One of the attempts.

41:21 Dylan: Well, that's the way troubleshooting works. When you find out that you don't have knowledge because you made a mistake, you end up taking it apart and making the chain correct. Do you find the.

41:34 Wes: All right, so let's discuss Goldman. Get into at least so we can get into reliabilism in the next part.

41:40 Dylan: Yep.

41:41 Wes: And you can get that because I think we're. We milked, you know, Gettier. I mean, it would have been nice to discuss the Zagzebski more. But part of this is supposed to. We're supposed to try to understand reliabilism.

41:53 Mark: So I have no regrets about our treatment of actually taking this seriously and applying our own critical faculties to it?

42:00 Dylan: Yes.

42:01 Mark: Come back next week, get part two or become a Partially Examined Life supporter. A citizen, as we sometimes say. You can do that in a few different ways. You can read about all of them at partially examined life.com support.

42:11 Wes: Thanks, thanks, thanks.