0:00

Please consider purchasing the new Partially

0:02

Examined Life book, a breezy read,

0:04

an excellent memento, the ideal gift.

0:07

Learn more at

0:09

partiallyexaminedlife.com/book. You're

0:18

listening to the Partially Examined Life, a podcast by some

0:20

guys who at one point set on doing philosophy for

0:22

a living, but then thought better of it. Our

0:25

questions for episode 344 are what

0:27

is knowledge and what is

0:30

justification when it comes to beliefs? And we

0:32

read some articles, Is Justified

0:34

True Belief Knowledge by Edmund Gettier from

0:36

1963, What

0:38

is Justified Belief by Alvin Goldman

0:41

from 1979, and The Inescapability of

0:43

Gettier Problems by Linda Zagzebski

0:45

from 1994. We

0:48

also read the Stanford Encyclopedia article entry

0:50

on Reliabilism by Alvin Goldman and Bob

0:52

Bedore from 2021. More

0:55

information about these articles and

0:57

the podcast, please see partiallyexaminedlife.com.

1:00

This is Mark Lintemeyer self-presenting in

1:02

Madison, Wisconsin. This is

1:04

Seth Paskin, sure that he is

1:06

in Austin, Texas via his perfectly

1:08

reliable clairvoyance faculty. This

1:10

is Wes Alwyn, correct only

1:12

by chance in Cambridge,

1:15

Massachusetts. This is Dylan Casey

1:17

wondering where the degrees of certainty are

1:19

in Madison, Wisconsin. All

1:21

right, so we had had this Gettier article,

1:23

which is three pages long. By the way,

1:25

do we know is it Gettier

1:28

or Gettier? Just on videos, people were

1:30

calling him Gettier. OK, unless I'm misremembering

1:32

that. Hey,

1:36

that's a reliable process. How

1:39

do you know? We can't count on that. Then what

1:41

can we count on? How about we, I'm Googling Gettier

1:43

pronunciation. I'm going to listen to it here. I think

1:46

we have to leave this in just given the. Pronounced

1:49

names, not given the relevance to this episode.

1:52

All right. It's supposedly Gettier, according to this

1:54

random YouTube video. Hey, and this is Mark

1:56

during the editing phase. It is Gettier. It's

1:58

normally Gettier. It's a. but he

2:00

was an American, he went by Gettier. So we

2:02

mispronounce it for the rest of the episode. But

2:05

keep in mind, I did have the

2:07

right idea initially. So yeah,

2:09

I thought I knew that. So yes, this

2:11

famous, famous paper we had had requested that

2:13

we cover it on the show before, but

2:16

it just seemed too insubstantial, I guess. I

2:18

don't know. This came up again, I think

2:20

in a close reads context. And

2:23

certainly the literature inspired by

2:25

it is plentiful enough.

2:27

There are many YouTube videos that I looked

2:30

at. You're really skimming over that

2:32

close reads context where

2:35

we were fumblingly trying to remember

2:38

what reliableism was all about. Well, I was,

2:40

so you picked though, once we determined we're

2:42

gonna do this three page paper, what else

2:44

we were gonna do and

2:46

found this Alvin Goldman paper, some

2:49

of the videos had recommended the Zzegzewski

2:51

paper, but it's not

2:53

exactly a completely linear thing from

2:55

the Gettier to the Goldman. And

2:58

I'm wondering why, like, in fact, I was

3:00

looking online at just epistemology class syllabi. They're

3:03

like in different sections. They're

3:05

related in that Gettier is asking

3:07

as the title signifies is justified true

3:10

belief knowledge. This goes back to Plato

3:12

and Aristotle. And he

3:14

comes up with cases where a belief

3:16

seems justified, it seems true. It

3:19

certainly believed, but yet we

3:21

would not consider it knowledge. We'll get into that. But

3:24

then the movement to, well, we

3:26

just have to change our notion of justification. We

3:28

have to work on that notion. So I guess

3:30

that was the direction that Goldman was headed. So

3:33

it was at least inspired by this initial thing,

3:35

though most of the videos I was looking at

3:37

that we're talking about the Gettier and the literature

3:39

surrounding it, we're not talking about Goldman specifically. But

3:41

did you have any thoughts, Wes? Was it just

3:44

the Stanford article that led you there by

3:46

Goldman? Was he the

3:48

author of that Stanford article? He's one of the

3:50

authors. That's hilarious. It's

3:53

another evidence gathering problem there. If

3:57

the guy writing the Stanford article recommends his

3:59

own paper. Have we chosen the right paper?

4:01

I forget When

4:05

we read Thomas Elo, I remember like half of

4:07

the citations in Thomas Elo were Thomas Elo See

4:10

Thomas Elo 2000 see Thomas Elo 2004 He's

4:14

written a thousand books and he just repeats

4:16

himself so much, but he's great. But anyway,

4:19

I forget I do Probably

4:21

Stanford played a part in that

4:24

but I think this is the big sea

4:26

change As Goldman describes

4:29

his own paper this reflects a sea

4:31

change in epistemology So yeah,

4:34

the Goldman paper isn't directly Addressing

4:37

Getty a but it is an

4:39

attempt to sketch out a new

4:42

epistemology in light of the fact of The

4:45

Getty a problem. It's just kind of implied which

4:47

is just that yeah we can no longer say

4:49

that justified true belief is the same thing as

4:53

Knowledge we need to either add something

4:55

to justification or we need a different

4:58

account of justification Which

5:01

is what the Goldman paper

5:03

does in painstaking detail in

5:05

order to get to knowledge or? In

5:09

order to say that true belief is

5:11

knowledge We need either to revised concept

5:13

of justification that didn't exist before or

5:15

we need to say it's justified true

5:18

belief Plus X this is the way

5:20

zag Zeb ski puts it is that

5:23

right? Zeb Zeb ski Is that Zag

5:25

Zeb ski that is a phonetic pronunciation

5:27

take them apart Zag Zeb ski, okay,

5:30

Zeb ski So I

5:32

found myself feeling in

5:34

the weeds in a

5:36

sort of analytical philosophy Forest

5:39

in this where I

5:42

really wanted to understand what explicitly everybody

5:44

meant by each of these terms like

5:48

understanding the problem And

5:50

it took me a little while honestly for

5:52

instance to get to that knowledge so

5:56

This may sound stupid, but to get to that

5:59

knowledge has to be true

6:01

when you know something. So

6:04

that was part of the deal, right?

6:06

Right. Okay. And so that's the

6:08

whole- The belief has to be true. Yeah. Well,

6:11

for it to be knowledge, whatever it is,

6:13

when you say it is knowledge- Truth is

6:16

a property of the belief and that's part

6:18

of what confers knowledge. What's part of what

6:20

makes it knowledge. But I think Dylan's right,

6:22

semantically, that if knowledge

6:25

consists of true beliefs that you have,

6:27

so you can say this knowledge is

6:29

true. I mean, unless that's

6:31

just redundant because- It's redundant. That is

6:33

redundant. If we're going

6:35

to be analytic philosophers just for one episode, we

6:38

got to behave in a nitpicking way

6:40

like analytic philosophy. You can't say these tautologies

6:42

are true. I mean, that's just

6:45

as- I mean, maybe. It's trivial. Trivially true. Point

6:47

to me. One equals one has got to be

6:49

part of analytic philosophy, right?

6:53

Go ahead, Seth. So I

6:56

have strong memories of reading

6:58

this Getty paper back in the day

7:01

and hearing the stories

7:03

about how he managed to get tenure off

7:05

this four-page paper and never really

7:08

wrote much of anything else ever. Again, I

7:10

don't know if those are apocryphal or if

7:12

that's actually true, but this is like the

7:14

best and worst of all analytic philosophy,

7:17

little like eddies to

7:19

get caught in because it

7:21

perfectly demonstrates the enterprise

7:23

of undertaking, painstaking, parsing,

7:26

linguistic, yadda-dee-yah kind of analysis on a

7:28

problem that should not have been a

7:30

problem to begin with. That is just

7:33

something that everybody was taking for granted

7:35

because that's really the way this starts,

7:37

right? So it starts with, according

7:39

to Getty and really in the

7:41

literature, there's this idea that there

7:44

are- to be able to say

7:46

that you know something, there are essentially three

7:48

conditions that need to be met and

7:51

that this thematically started with Plato

7:53

and has been sort of just

7:55

assumed in the history of philosophy. Whether that's true or

7:57

not, I don't know, but- And those- three

8:00

things are as Dylan was pointing out. One, so we'll

8:03

use the language in the article,

8:05

S knows that P if and

8:07

only if P is true,

8:10

S believes that P and

8:12

S is justified in believing that

8:15

P. There are different

8:17

formulations that he brings up, but

8:19

essentially the general problem statement here

8:21

is that to say

8:23

that somebody knows something, the

8:26

thing has to be true. You can't know something

8:28

that's false. You have to believe it. So

8:30

there's a truth component, a belief

8:32

component, and then a justification component.

8:34

However you want to characterize it,

8:36

meaning you had to have come

8:38

to that belief that

8:41

in that thing through some

8:43

reasonable myth, you have to be

8:46

justified. Yeah, whatever justified means is

8:48

the thing that you did. Yes,

8:50

exactly. So that's the problem statement.

8:52

We got that explicitly in Plato,

8:55

right? That if you merely know

8:57

how to make something, if you're doing it,

9:00

have a techne, then it's not

9:02

real knowledge because you're just sort of doing things

9:04

out of habit and they happen to work out.

9:06

And so you have to have the reasons for

9:08

things. This was actually a big finding. It is

9:11

considered in the history of philosophy to be one

9:13

of the, I think, chief things Plato contributed,

9:15

that justification is actually necessary. It seemed like Dylan,

9:17

you were asking, where did the true part come

9:19

from? Like, is that something you wanted to pick

9:22

up? Well, I just feel, I felt like I

9:24

was stupid and I liked the way that Seth

9:26

articulated it or explicated what

9:28

was just said in the Getty article

9:30

is that the knowledge part requires the

9:33

true part because I found myself falling

9:35

over in the Zagresky article. And I

9:37

can get into this a little bit

9:40

more later because there's something, I want

9:42

to unpack this. There's something about this

9:44

that just feels just totally

9:46

weird to me because at

9:49

the end it seems like we're making

9:51

a big deal about, well, we can't

9:53

be certain about things. There always be

9:55

a way in which you're not a

9:57

hundred percent certain and knowledge,

9:59

requires certainty. And I'm like,

10:02

are you surprised by this? Yeah, I don't

10:04

think they're concluding. I mean, there's no conclusion

10:06

that knowledge requires certainty. I mean, I think

10:08

you're right. We could discard the

10:11

concept of knowledge altogether and agree

10:13

with, I think, probably many continental philosophers that

10:15

maybe there is no such thing as knowledge

10:17

strictly speaking. But we might want to say

10:20

that some things are closer to knowledge

10:23

than not. And in that case, we

10:25

want to define ideal limits. We want

10:27

to say, well, something were knowledge,

10:29

what would it be? And that

10:31

way we can say what counts

10:34

as being closer to it and farther

10:36

away from it. So the problem doesn't

10:38

go away, even if we become more

10:40

skeptical and say that strictly speaking, we

10:42

don't know anything, which I'm a very

10:44

sympathetic to that idea. But the other

10:46

part of this with saying that P

10:48

is true, false beliefs can be justified.

10:50

That's really important. So you have to

10:53

specify truth because it's a very, very

10:55

important thing that we can be justified

10:57

in believing things that are untrue. No

10:59

doubt. But there's a similar

11:01

sort of sense of disquiet in my soul

11:04

as what Dylan articulated, because I think what

11:07

this really shows is, so imagine

11:10

you're talking about something like the

11:12

rationalist era skepticism, right? So go

11:15

back to idealists versus realists,

11:17

all that kind of stuff. Your skepticism

11:19

and your doubt about whether you can

11:21

be certain in your knowledge has

11:24

to do with your epistemic

11:26

relation to the world. By

11:28

the time you get to the 20th century, people

11:31

are sort of beyond that a little bit and

11:33

they're like, okay, well- Can I just jump in?

11:35

None of these authors are saying knowledge requires certainty.

11:37

No, no, no. Well, I'm not talking about certainty.

11:39

I'm talking about skepticism. I just want to, because

11:42

you used the word certain again and Dylan used

11:44

it and I'm going to make clear that that

11:46

is not the thesis here. Okay. So then the

11:48

point that I'm trying to make is the historical

11:51

roots of skepticism are skepticism

11:53

about two things. One is

11:56

truth at the level of truth and

11:58

then skepticism. about the

12:01

acquisition of knowledge, like how

12:03

you come to know things. And

12:05

this casting of it in terms of

12:08

saying, let's just assume, let's give that

12:10

P is true, epistemology and knowledge

12:13

is really about understanding beliefs and how

12:15

we come to those beliefs,

12:17

not so much about perceptual systems or

12:19

the way in which. So when Gettier

12:21

blows this whole thing up, it

12:24

freaks everybody out, I think, because as

12:27

though we had settled into a paradigm

12:29

that people felt comfortable with, with respect

12:31

to epistemology that had kind of, if

12:33

not solved, at least put aside certain

12:36

kinds of questions. And then now we're

12:38

right back to the same place where

12:40

we have all of this, let's just call it skepticism

12:43

about how we can know things and how we

12:45

can know that we know things and so on.

12:48

And so I think that's why this was such a big deal. Can

12:51

I replay that back to you a little bit, Seth,

12:53

or just rephrase it? So when

12:55

I read the Gettier paper, I

12:58

thought, okay, so he's being clever. And the

13:00

reason I focused on certainty, and maybe it's

13:02

the wrong word to do, is that he

13:04

constructs examples in which you

13:07

have justified true belief, but it turns out

13:09

that there is some

13:11

exception that in a

13:13

very understandable and clear

13:16

way shows that you don't

13:18

actually have knowledge. You were wrong. It's not

13:20

true. What your belief is, is not true.

13:22

And so therefore you don't have knowledge. And

13:25

so that's clever and all that stuff. And

13:28

it makes sense. The way you

13:30

articulated it, Seth, to me, sort of seems

13:32

to make the statement that, well, this

13:35

rocked the world because the space

13:38

for the kinds of

13:40

uncertainty about that truth exploded

13:43

to be much larger because of Gettier.

13:46

Because the examples that he gives is

13:49

the problem that there is one or is

13:51

it that there's effectively an infinite number. There's

13:53

an infinite number of ways in which your

13:56

justified true belief is not knowledge. He should

13:58

basically show you through it. Both

18:00

of the premises could be true and you could still get a Getty A

18:02

problem, which we'll get to. But, so you

18:05

don't have to have a false piece of evidence. Yeah,

18:08

so Smith has 10 coins in his pocket

18:10

and is applying for the job. Jones

18:12

has 10 coins in his pocket and

18:14

is applying for the job. Smith was told

18:16

by the president of the company, Jones will get the job.

18:19

So Smith thinks whoever gets the

18:21

job will have 10 coins in his pocket.

18:24

He knows the other guy has 10 coins. He doesn't know that

18:26

he has 10 coins. And unbeknownst to

18:28

him, the president is picking him for the

18:30

job. So the point is the

18:33

conclusion that he says the guy who gets the

18:35

job will have 10 coins in his pocket is

18:38

correct. And justified. And justified.

18:40

So it's true, it's justified.

18:43

And Smith believes it. The only

18:45

thing is it's a different guy. So it's not

18:47

knowledge. Well, it's definitely not knowledge,

18:50

but it does satisfy all three of

18:52

those conditions. And that's- Yeah, it's just

18:54

justified true belief, but Getty will say

18:56

it's not knowledge. Does

18:58

anybody wanna argue that actually it

19:00

is knowledge? He does know. In

19:03

fact, I feel like a sort of Daedicto de

19:05

Rey thing, sense of reference thing

19:07

might help us. I don't know if that's exactly the

19:09

right word, but he knows Daedicto,

19:11

those words, the guy that has

19:14

10 coins in his pocket will

19:16

get the job. He knows that claim. He knows

19:18

it is true. What he thinks it means, what

19:21

it actually points to, that it

19:23

points to him and not to Jones, that

19:26

is where the mistake comes in. It

19:29

depends how you characterize what the belief

19:31

is. But that just

19:33

identifies knowledge with true belief, irregardless of

19:35

the justification. Well, the question

19:38

is whether his, what you

19:40

called a principle B, just

19:42

that logical entailment works. So, I

19:45

mean, we can think about this- Justification transfers

19:48

with the entailment. Right, justification transfers

19:50

with logical entailment. So we already

19:52

know that if A is true,

19:55

and then A then B is true,

19:57

then B will be true. But

19:59

if I believe- that A is true and I'm justified

20:01

in believing that, I'm justified in general in believing the

20:05

modus ponens because that's just basic

20:07

of course, but those are two

20:09

different kinds of justification and so

20:11

can you then combine actually say

20:13

I know the conclusion, I'm

20:15

justified in believing a conclusion based on

20:17

those two, I mean, Gebbia thinks there's

20:20

no problem here, you know, like I

20:22

had legitimate evidence for one, obviously a

20:24

logical truth is going to

20:26

be self-evident, so therefore I

20:28

definitely have evidence, I have

20:30

justification for the conclusion, I

20:32

kind of want to push back on that, I'm not

20:34

sure that the conclusion is clear enough. Then

20:37

no truth that we know that depends

20:39

on inference would be justified in that case. Yeah,

20:42

there's a couple things that are tricky I

20:44

think about this particular example. First

20:46

of all, I have to question

20:49

whether something in the future can

20:51

be considered true, like Jones's doesn't

20:53

have the job, Smith was

20:55

told by the President that Jones will get the

20:57

job and so he's saying, oh, he knows Jones

20:59

is going to get the job, like

21:01

that's a weird thing. He doesn't know Jones is

21:04

going to get the job. Well, he believes,

21:06

he's justified in believing that Jones is going

21:08

to get the job because the

21:10

President told him so. So,

21:12

yeah, so testimonial evidence, one could question

21:14

the value of testimonial evidence that would

21:17

be a... Or of not

21:19

just testimonial evidence but future

21:21

states versus current

21:23

states, there's a modal component here because

21:26

this is something that hasn't happened that

21:28

is theoretically going to happen. Yeah, evidence

21:30

always involves time. This is one thing

21:32

that Goldman is going to point out,

21:34

that happens over time. Couldn't it involve

21:36

like who's been poisoned? I believe

21:39

that the person with 10 coins in his pocket has

21:41

been poisoned. So, it doesn't have to be a future

21:43

thing. Like I don't think that making it in the

21:45

future significantly changes it because you'd

21:47

have exactly the same issue. That's

21:50

fine. I actually was poisoned. Oh, you know. But

21:52

remember the other thing he says is he doesn't

21:54

say, I have justified belief

21:56

that Jones is going to get the job.

21:58

He says the... as I have justified

22:00

belief that the person who gets the job has 10 coins

22:03

in his pocket. And then

22:05

I think there were other examples, either in

22:07

the literature or the second Gettier thing where

22:09

he starts using logical operators

22:11

like or, you know, where you have like

22:14

truth values and conjoined sentences and things like

22:16

that. That gets kind of weird. I'm like,

22:18

to me, that's a bit of a sketchy

22:20

kind of counter example. I look for something

22:22

a little more concrete. Take control of your

22:25

growing company's finances with Bill's Spend and Expense,

22:27

the free to use spend management software powered

22:29

by the Bill Divey Corporate Card. With

22:31

credit lines from $50,000 to $5 million,

22:34

Bill's Spend and Expense is here to

22:36

help you thrive. Say goodbye to manual

22:38

expense reports and juggling multiple cards and

22:41

say hello to saving time and eliminating

22:43

wasteful spend. Try Bill's Spend and Expense

22:45

today at bill.com/promo 500 and get $500.

22:49

Terms and conditions apply. Advertised credit ranges are not guaranteed

22:51

and are determined at approval. Card issued by Cross River

22:53

Bank, member FDIC. Heat pumps are

22:55

going mainstream with tax credits of up to

22:57

$2,000. The Federal Inflation Reduction Act can make

22:59

it more affordable than ever to upgrade your

23:01

HVAC system and lower your energy bills. Now's

23:03

the time to say goodbye to your old

23:05

fashioned fossil fuel burning system and switch to

23:08

an all electric all climate heat pump from

23:10

Mitsubishi Electric. Learn more about Mitsubishi Electric products

23:12

at patriotair.com. Federal income tax credit available to

23:14

US homeowners only for qualifying heat pump models.

23:16

Terms and conditions for tax credit are dependent

23:18

on federal and state laws and regulations and

23:20

are subject to change. Contact your tax advisor

23:22

for further requirements and to confirm eligibility for

23:24

tax credits. Do

23:26

we grant the claim of

23:29

the paper that justified true belief is

23:32

insufficient for knowledge?

23:34

Because if not, there's no point in

23:36

going on. Well, it certainly raises the

23:38

issue of like, well, what do you

23:40

mean by justification? And that's what we're

23:42

going to chase for the rest of.

23:44

Right. Which is really the rest of

23:46

the podcast. Yeah, I think is more

23:48

or less about that topic. Because you

23:50

can't just say he's correct if you

23:52

don't know. Well, if you

23:54

don't have you could have a question, the theory

23:56

of truth. We did multiple episodes on what truth

23:58

entails. I. I thought maybe Dylan was

24:01

starting to pick at this like, what

24:03

if you're a pragmatist? So in

24:05

fact, this is one of the things that comes

24:07

way later in like the Stanford article, but these

24:09

evil demon cases, if I'm in

24:11

a situation, if I'm in the matrix and

24:14

I say, I ate a sandwich and I'm

24:16

pretty justified in thinking I ate a sandwich,

24:18

but actually I didn't eat a sandwich because

24:20

there are no sandwiches and I'm getting nutrients

24:23

pumped into the back of my neck. A

24:25

pragmatist is still wanna say, well, it's actually

24:27

true that you ate a sandwich because truth

24:29

is dependent on a sort of frame

24:31

of reference and it actually is dependent

24:34

on epistemic norms. It is

24:36

internalist and to some degree, it

24:38

is not purely externalist. So we might wanna

24:40

say, well, in a sense it is true

24:42

that you ate a sandwich, but ultimately, from

24:45

the frame of reference of the people outside the matrix,

24:47

it is not true that you ate a sandwich. So

24:49

we could pick at that. That's not what we're picking

24:51

out in this episode, or we could pick at the

24:53

justification. That's picket things. Can

24:57

we give the second example? Because that is

24:59

a present tense one. Do we need that?

25:01

It's at least interesting in that it's not

25:03

modus ponens, an if then

25:06

statement, it is. Well, it's still inference, right?

25:08

It's a disjunction though. Yeah, it's an inference

25:10

to a disjunction. So yeah, so we can

25:12

make these inferences, we

25:15

can take something that we are justified

25:17

in believing. In this case, it's Jones

25:19

owns a Ford and Smith

25:22

has lots of reasons for believing that Jones

25:24

owns a Ford, including perceptual evidence and blah,

25:27

blah, blah. And then he can make

25:29

an inference from that proposition

25:32

to either Jones owns a Ford

25:35

or Brown is in Boston. You could

25:37

just add these things that, yeah, any

25:40

propositions you want. These, it's

25:42

just a logical truth that these or

25:44

propositions and conjoining things with the wars,

25:46

you'll get a true, if one of

25:49

them is true, then the whole conjunction

25:51

is true. Disjunction, not conjunct. Which

25:54

is a very weird thing for people who have

25:56

not taken formal logic. That it is how the

25:58

operator, it doesn't. I mean, one of

26:01

these alternatives must be true. I mean, it just

26:03

means, well, it means at least

26:05

one. Well, if P is true, then P

26:07

or Q or R, that

26:10

whole disjunctive statement

26:12

is necessarily true, is entailed

26:14

by P being true,

26:17

right? Modus ponens is a different, and

26:19

they're actually related, right? Because these things transform

26:22

into each other. So it's not

26:24

completely unrelated to modus ponens. So that's the

26:26

weird, yeah, it's gonna look weird to someone

26:28

who hasn't done formal logic, but it's important

26:30

in formal logic that this is the case.

26:33

So yeah, once you are justified in believing

26:35

that Jones owns a Ford, you're justified in

26:37

believing that either he owns a Ford or

26:39

Brown is in Boston, or he's

26:42

in Barcelona, or he's in Brest, Lutovsk,

26:45

or you can do a bunch of different

26:47

disjunctive statements, in other words, by coming up

26:49

with a random city and placing Brown in

26:51

that city, even though you have no idea

26:54

whether he actually is in that city. So

26:56

it turns out that Brown does not own

26:59

a Ford. So actually the initial thing was

27:01

false, even though you were justified in believing

27:03

it, but it happens that Brown is in

27:05

one of those cities. So the disjunction between

27:07

Brown and the Ford, and the other guy

27:10

is in this city ends up being true.

27:12

So the inference is justified, right? Because Smith

27:15

is justified in believing Jones owns a Ford, even

27:18

though he doesn't. And Smith

27:20

made a good inference. The

27:22

entailment is there. So the conclusion

27:24

is justified, but it turns out,

27:26

yep, he doesn't own a Ford, and Jones

27:29

happens to be in one of those cities. So

27:31

it's just luck that he got, he's not right

27:33

for the right reasons. He's right

27:36

for the wrong reasons, which are

27:38

just completely accidental. So he's justified,

27:41

but the justification isn't what he

27:43

thought it was. Right, Gettier says

27:45

he's justified because of, again,

27:47

those two things. He was justified in the

27:49

first actually false proposition that the guy owns

27:52

a Ford, and then he

27:54

was justified just in that it's another

27:56

rule of logic. And so obviously that's

27:58

justified and belief just... is supposed

28:00

to pour through these rules of logic.

28:03

And I think this is what's being questioned is not, you

28:06

know, I think the people arguing against Gettier are not

28:08

saying, you're right, there needs to be something in addition

28:10

to justify true belief. They're just saying,

28:12

well, that's not real justification. Real

28:15

justification means that the belief actually

28:17

has to be causally tied, for

28:20

instance, to the fact of the matter. It

28:22

has to be the fact that Jones

28:25

is in Barcelona that caused me to

28:27

believe this disjunction something or Jones is

28:29

in Barcelona. The fact that

28:31

nothing to do with it means, is why

28:33

it's not justification at all. Yeah,

28:35

well, now this is a semantic distinction. Now use

28:38

the magic word mark. So you

28:40

can say, right, this

28:42

is just semantics. We can say we're going

28:44

to revise a concept of justification or we're

28:46

going to take the simple concept of justification

28:49

that Gettier is using and add x as

28:51

Zebzki puts it. So,

28:54

right, it's either knowledge equals

28:56

JTB or justified true

28:58

belief plus x or knowledge

29:01

equals true belief plus

29:03

justification reconceived as reconceived as

29:05

warrant or well-founded. Yeah, J prime. So it

29:07

doesn't matter. It doesn't matter which way we

29:10

say it. JTB plus x

29:12

or J prime TB? JTB

29:14

plus x or TB plus x

29:16

where x is J recons... J

29:20

prime. So forget about it. J prime.

29:23

Either way, it's the same thing. But yes,

29:25

we need to revise our notion of justification

29:28

or add something to this, to

29:30

Gettier's notion of justification, which is just

29:32

the simple accepted version at the time.

29:34

I want to articulate a

29:37

little more clearly, at least for myself, what

29:39

would success look like? As

29:42

Zebzki noted, the problem and

29:44

the Gettier problems is that

29:46

there's an accidental string of

29:49

causes that leads to true

29:51

belief. And so what we

29:53

believe is true... Well, there's bad luck, then there's

29:55

good luck. Okay. What

29:57

we believe is true, but we... you

30:00

don't want to call it

30:02

knowledge because the sequence

30:04

of justifications, the causal sequence, whatever you

30:07

want to call it is

30:09

unsatisfying in some way. And in particular

30:11

in the case with the Gettier problems,

30:13

it involves things that weren't true or

30:16

accidental. I mean, excuse me, that weren't

30:18

justified belief. So there's a string

30:21

there and is what good

30:23

looks like that we have every

30:26

time we have something

30:28

justified as true belief that it

30:32

is actually knowledge that's

30:34

actually true. So we basically have

30:36

a formula, a process to always

30:39

be certain, or is it

30:41

that we make

30:43

it so that whenever there's that justification doesn't

30:46

have any accidents in it? I

30:49

don't think this doesn't operationalize anything so

30:51

that we can be certain of things.

30:53

None of this confers certainty and it

30:55

doesn't give us a formula for making

30:57

sure things are justified in

30:59

part, right? Because justification is comparative and a

31:02

matter of degree and we can

31:04

be justified in believing things that are false, which

31:06

immediately eliminates the

31:08

relationship between justification

31:10

and certainty. And as Zegzewski points

31:12

out, you can't ever tighten

31:15

those up completely. You can't ever

31:18

get completely from justification to truth.

31:21

Or at least you'll say that. You can

31:24

if you cheat. But hold on, hold on.

31:26

So the point is that the goal of

31:28

this is simply to understand the concept of

31:30

justification. What Gettier reveals is just that we

31:33

don't fully understand the concept of justification. We

31:35

don't fully understand the concept of knowledge and

31:38

we don't fully understand their relationship. It

31:41

was intuitively possible that they're related and

31:43

Plato got us that.

31:45

But if you dig further, you see

31:47

that it's not enough. And it's just

31:49

a matter of curiosity about what knowledge

31:51

is, what justification is. There's no practical

31:54

upshot to this, unfortunately. Maybe

31:56

I'm just being completely boneheaded,

31:58

but what does good look

32:00

like then? What do you mean?

32:02

What does that question mean? What does good what

32:04

look like? A good explanation of knowledge? What

32:07

would be successful? Maybe there

32:09

is no successful. Okay. How do you

32:11

do? One where you can't come up

32:13

with counter examples. We did the same

32:15

thing in our Grice episode last summer

32:17

where it was, this is the analytic,

32:19

it's really infuriating. And the only good

32:21

thing about the Getty paper is he

32:23

just stops really quickly, but he could

32:25

easily write something like Goldman did or

32:27

what we saw in that Grice paper

32:30

where you just keep bringing up more

32:32

and more examples and just refining your

32:34

wording and saying, oh, well, here's another

32:36

example that doesn't work for this new

32:38

wording or is this, this brings out some

32:40

counter, but this is just the game of

32:42

analytic philosophy is so. Yeah. I'm just as

32:44

annoyed by analytic philosophy as everyone. All right.

32:47

So you can't beat me on

32:49

that count, but I'm trying to get what's

32:51

positive about this. Sure. Some

32:53

reflection on the concept of knowledge and the

32:55

concept of justification. So all we get in

32:57

the beginning is, oh, it's an apareia. I

33:00

don't really fully understand knowledge.

33:02

I don't fully understand justification and this

33:04

formula that people have been repeating for

33:06

centuries or have they, it's another historical

33:09

question, but is not adequate. Okay.

33:11

So success looks like resolving an apareia.

33:13

Okay. A word that I don't know

33:16

that we even talked about in our

33:18

Aristotle episode about this. Just a puzzle.

33:20

It's a puzzle. Okay. It's a puzzle.

33:22

What is distinctive about an apareia as

33:25

a type of puzzle? Well, it's kind

33:27

of arresting. It blocks up the flow

33:29

of theory onto more productive terrain. It's

33:31

not just out on the sidelines. It's

33:33

yeah, it's a bottleneck. I just want

33:36

to dispel, because I'm wondering

33:38

like, did these guys read Frege? So

33:40

a Fregean example would be like, I

33:42

believe Superman is strong and Superman

33:44

is in fact Clark Kent. I

33:46

don't know that. It doesn't follow then

33:49

that I believe Clark Kent is

33:51

strong. Like that's Frege's point that

33:53

belief doesn't flow through unless I

33:55

believe Superman is Clark Kent. Then yes, I

33:57

will believe Clark Kent is strong as well.

34:00

Is there anything about this distinction between

34:02

Superman is in fact Clark Kent. I

34:04

believe Superman is Clark Kent that would

34:06

be that would shed light on this

34:08

coins example, for instance, you know, I

34:10

believe that that Jones has 10

34:12

coins. I believe Joan got the job.

34:15

Therefore, I believe that

34:18

I mean, I would want to say I believe

34:20

that Jones who has 10 coins got the job.

34:23

But what he wants to say is I believe

34:25

that the guy who has 10 coins or

34:28

the recipient of the job has 10 coins. So

34:31

we've introduced some opacity. We've introduced, you

34:33

know, a sort of Clark Kent Superman

34:35

situation here. And I'm just not sure

34:37

if this is legitimate. Well, what it

34:39

is is that it's not that it's

34:41

illegitimate. It is an inference that

34:44

you can make. It just

34:46

turns out that it's a gap that you

34:48

can't make. Like the example,

34:51

what I was immediately brought to is like

34:53

the example given here is a perfect example

34:55

how all kinds of mistakes are made, you

34:57

know, from it's the way in

35:00

which a sitcom works. And it's

35:02

also the way in which bugs arise in

35:04

code and the way

35:06

in which conclusions are, the

35:08

wrong conclusion is made in a physical

35:11

experiment is the inference has a

35:13

gap in it. And so you

35:15

end up just being wrong. And so, Well,

35:17

it's not a bad inference, right? Yes,

35:20

it's not a bad inference. And you get down to

35:22

the quality, you get down to thinking about, well, what

35:25

are all the things I'm going to do in

35:27

order to make the

35:30

degree of the alignment between

35:32

my justified true belief and

35:35

my inference? Because like

35:37

the problem in the Gettier one is,

35:39

seems to me, is that if

35:41

you go on to make other conclusions based upon

35:43

that gap, you're going to be

35:45

wrong on other things, right? Yeah,

35:47

I don't think reliableism, you know, as

35:50

a theory helps us avoid being

35:52

in the spot of someone like Smith. No,

35:55

it doesn't. So it's not going to

35:57

increase certainty. It just tells us because

35:59

this leads to a gap, because Gettier

36:01

leads to Reliableism, I think it's quite

36:03

interesting. It sheds some light on how

36:05

we ought to think about justification

36:07

because we will, you know, as

36:10

we move forward, we'll see that

36:12

we, that we're going to

36:14

change our conception of justification from

36:17

simply involving these

36:19

internally accessible belief

36:21

states to processes

36:24

that lie outside of

36:26

the awareness of

36:29

the believer. And those are,

36:32

right, those are evidence

36:34

gathering processes, processes,

36:36

belief formation processes,

36:38

which themselves we don't

36:41

necessarily have access to.

36:43

So that's really, they're important

36:45

consequences for that, right? Including

36:47

Cartesian foundationalism not working out as an

36:50

epistemological theory, as Goldman will get into.

36:52

There are lots of important things that

36:54

flow from this problem

36:56

and a Reliableist analysis

36:59

of justification. Yeah,

37:01

this distinction between internalist justifications, we

37:03

only have to look at what

37:05

the person actually thinks that

37:07

he believes that versus externalist versus, you know, that

37:09

there could be actually your belief is justified even

37:12

though you don't know it, just like you don't

37:14

know whether your belief is true. So that truth

37:16

is an externalist thing,

37:18

but this is the

37:20

intuition that I'm running against in

37:22

trying to say, actually, Smith does not

37:25

believe that the guy who has

37:27

10 coins got the job. He doesn't have

37:29

that belief at all, because I want to

37:31

say a belief is something that I understand

37:33

the contents of and what

37:36

he thinks his contents is Jones

37:38

who has 10 coins. He doesn't think.

37:40

I think you're just making an assumption

37:43

about what someone is likely to believe

37:45

under those circumstances. And it really applies

37:47

to this specific case. It doesn't apply

37:49

to every Getty a problem.

37:51

So yes, in that circumstance, a few

37:54

of us are going to say, Oh, the person who is

37:57

going to get the job has 10 coins in their

37:59

pocket. relevant, you're just going to think Jones

38:01

is going to get the job. But there

38:03

would be circumstances in which we might

38:06

think about that. If somebody asks you the question,

38:09

do you assent to the fact that the person,

38:11

whoever it may be, has 10

38:15

coins in his pocket? You've been primed

38:17

in some way, right? Yeah. A

38:19

fortune teller, the person who gets the job is

38:21

going to have 10 coins or

38:23

something stupid like that. So there are circumstances

38:26

in which it might become relevant

38:28

to think about to form the belief that

38:30

the person who gets the job is

38:32

going to have 10 coins in his pocket. It

38:35

would behoot maybe Gettier to come up with that

38:37

kind of example where there's an obvious, where the

38:39

belief seems like a likely belief to have formed.

38:41

But I don't think it matters because that's not

38:43

the point. And we could come up

38:45

with examples. I can't on the spot, but I think

38:48

it's in the Zag Zepsky where it would

38:51

be perfectly ordinary belief to have formed

38:53

and it's justified and true, but it's not

38:56

knowledge. And we do have to accept intuitions. I

38:58

mean, I think this is part of what Dylan,

39:00

Mark, you're getting at. I mean, we do have

39:03

to, there is a fundamental intuition about what knowledge

39:05

is that's at work here. That's not, that we

39:07

can't justify any further.

39:09

Can I just want to ask a question

39:11

about knowledge that was bugging me when we

39:13

talked about internal and external. The

39:16

way all of this talks about knowledge

39:18

involves, I'll call it facts

39:20

about the world. And

39:22

there's another way in which we use

39:24

knowledge and where we might say something

39:26

is true. So if I say Mark

39:29

knows how to play guitar, but

39:31

if Mark can play guitar, but

39:33

doesn't know how to explain to anybody

39:36

how to play guitar, is that

39:38

just an exception? Or we just, we would say,

39:40

well, he actually doesn't know how to play guitar.

39:42

He plays guitar, but he doesn't know how to

39:44

play guitar. That's the Plato's question. We

39:47

might say, well, it's a kind

39:49

of knowledge. It's just not explicit

39:51

knowledge or something. Well, this is

39:53

a different, this is procedural knowledge.

39:55

This isn't propositional knowledge. So nothing

39:57

that we say about propositional knowledge.

40:00

anything about procedural knowledge necessarily, maybe it

40:02

does, but- Yeah, I kind of want

40:04

to, I like that because then we

40:06

could say, you know, do I

40:08

really have the belief that either

40:11

Jones owns a Ford or Smith-

40:15

Brown lives in Barcelona. Brown lives in

40:17

Barcelona. Do I really have that? What

40:20

actions would I take? Like if I want to see

40:22

really what the belief is, you might

40:24

say it's not just given the proposition

40:26

and the proposition has a certain amount of vagueness

40:29

because the guy that or

40:31

whatever, because there's these opaque parts of

40:33

it, but because if I go to

40:36

find the third guy, do I go

40:38

to Barcelona or not? If

40:40

I don't do any of the things

40:42

that are accompanying that disjunction, then I

40:44

don't actually believe the disjunction. I

40:46

don't know, because a disjunction is like grew

40:48

and bleen, right? Just to- a

40:50

disjunction is an unnatural category that if

40:52

you want to say, I believe this

40:55

thing, I believe this proposition, but then

40:57

you disjoin it to something else. I

40:59

believe this or this other thing, unless

41:01

they sort of fit together, the grew

41:03

and bleen thing just means like there's

41:05

a natural category of green, but there's

41:07

an unnatural category of green before yesterday

41:09

and then blue. You're focused on

41:12

things that are just specific to this particular

41:14

example. They don't come up in

41:16

all Getty A cases. That's why I picked

41:18

the other one anyway. Yeah. So

41:21

it doesn't say anything about the Getty A problem

41:23

if you don't like the example. Yeah. Now the

41:26

disjunction is even worse. It's like having a false

41:28

premise as your, if P then

41:30

Q, if P is false or

41:32

whatever, then anything's true, right? Or

41:34

I'm sorry, anything can be inferred

41:36

from a false premise. That's just

41:38

not helpful in this circumstance, like

41:41

saying you can append any kind of truth

41:43

proposition in a disjunction to

41:46

something that you're trying to

41:48

define. Knowledge is not helpful. I

41:50

believe so. It was not in one of the ones we read, but one

41:52

of the initial things, what you just said in the

41:54

videos I was watching on this is one of the

41:57

suggestions that was initially made right after Getty A poised

41:59

this is. Well justification

42:01

doesn't flow through if there are

42:03

false lemmas, which I forget exactly

42:05

what a lemma plays in mathematics,

42:07

you know, but it's like a

42:09

sub sub theorem. Yeah. Yeah. So

42:11

there are false beliefs involved.

42:13

The fact that you falsely believes that Jones

42:15

is going to get the job or falsely

42:18

believes that the guy owns a Ford. The

42:20

fact that those are involved. That is what

42:22

makes the justification not flow through that. It's

42:24

not a truly justified anyway,

42:26

or it owers knowledge is justified. True

42:29

belief plus no false lemmas involved. This

42:31

is at least one of the one

42:34

of the attempts. Well, that's the way

42:36

troubleshooting works. When you find out that

42:38

you don't have knowledge because you made

42:40

a mistake you end up taking it

42:43

apart and making the chain correct. Do

42:45

you find the all right? So let's discuss

42:47

Goldman get into at least so we

42:49

can get into reliabilism in the

42:52

next part. Yep, and you can get

42:54

that because I think we've

42:56

milked. Get

42:59

it. I mean, it would have been nice to

43:01

discuss the zag Zab ski more, but part

43:03

of this is supposed to we're supposed to

43:05

try to understand reliabilism. So I have no

43:08

regrets on our about our treatment of actually

43:10

taking this seriously and applying our own critical

43:12

faculties to it. Yes, come back next week

43:14

at part two or become a partially examined

43:16

life supporter a citizen as we sometimes say

43:18

you can do that in a few different

43:20

ways. You can read about all of them

43:22

at partially examined life. Comm slash support. Thanks.

43:24

Thanks. Thanks. Yeah.