Podcast: Play in new window | Download (Duration: 52:57 — 72.8MB)
Continuing from part one on ch. 1, “Knowledge Sources,” of the Matthew Dasti/Stephen Phillips presentation of the Nyaya-Sutra: Selections with Early Commentaries.
We continue discussing modalities of perception (why no love for a smell-based philosophy?) and further consider the Sutra’s dictum that perception must be inerrant and how this dovetails with the overall Nyaya approach: Consider a claim that someone makes based upon a knowledge source “established” unless someone then has a legitimate reason to challenge it, i.e. can also refer to a knowledge source to claim the contrary. So perception is a legitimate knowledge source, but I can legitimately ask you when you claim to have seen something if you really got a good look at it. What I can’t do is act the skeptic and ask you to doubt all your perceptions. There is no reason to do this; I have no standing — no practical worry — to demand this more global re-examination.
We then move on to inference, which is exemplified in these texts as a matter of induction: I have seen smoke and fire consistently together in the past, so when I see smoke (the inferential mark) now, I conclude there must be fire. However, another example given is the inference that there is a self based on the presence of desire and other mental activities. Descartes’ “I think, therefore I am” is certainly not induction, i.e. not based on past experiences of seeing my thoughts and my self always together, as any past experience would require the same proof of a self. We’ll get into this argument about self more in the second part of our ep. 347, but the point here is just that inference must refer to any kind of reasoning: an inference to the best explanation for something we perceive. Desire seems to be a property, and we know from past experiences that properties have substances, so a desire must inhere in something, and we call that something a self.
But is there a clean line between perception and inference? When I perceive an apple and recognize it as an apple, I am inferring that this thing I now see must be an apple because it looks like previous objects that I and others coupled with the word “apple.” Or if we stick to the requirement that perception is not in itself verbal, does any use of memory necessarily involve an inference?
This may be why the text actually gives “comparison” as a separate knowledge source. For instance, if someone has told me about a bison being “like a cow but bigger, furrier, and brown,” then when I see a bison for the first time, I might use this comparison to attach the word to the animal. This is surely a form of inference, and it’s a specific enough kind of case that it’s unclear that it helps us decide whether recognizing a particular fruit that we’ve seen a lot but which no one’s ever told us the name of counts as mere perception or perception plus some kind of inference.
Finally, we briefly consider testimony, which was especially important and independent of the others because the authors wanted to leave room for the inerrancy of the holy Vedic scriptures. So testimony is not reducible to perception, at least not of the ordinary kind; the scriptural authors had supernatural powers of perception and are reporting those perceptions to us. In ordinary cases, we are relying on the perception or valid reasoning of someone who knows what they’re talking about in the relevant sense, whether because they have first-hand perceptual knowledge of the thing or have the expertise to make valid inferences in that domain (as when we listen to scientists or other experts). Is the decision to accept someone’s testimony itself an inference, such that we should really consider testimony a flavor of inference? If I have often see a certain politician lie, it’s reasonable for me to infer that the thing he’s saying now is also a lie. However, as with all the knowledge sources, a lie wouldn’t actually be considered testimony at all, just pseudo-testimony. These analyses are about classifying the sources of legitimate knowledge, when this is present, and do not actually help us as knowers to distinguish the real thing from an imitation in particular cases. To gain these skills, we’d have to use our perceptions over time and the advice (testimony) of experienced teachers. We’d assume (reasonably, on the Nyaya view) that the information from these sources is usually reliable. Truth is the default, and doubt is introduced only when there’s some reason to challenge an assertion.
Next Time: We’re reading more of this book, covering chapters 3 (“In Defense of the Real”) and 4 (“Self”).
Transcript
0:07 Mark: Hey, this is Partially Examined Life Episode 346 Part 2 We've been talking about the Nyaya Sutra and the various Naiyayikas, the Nyaya philosophers. That's apparently a word, the Naiyayikas. But I think we had just concluded at the end of the last discussion that there are no secondary qualities in this view, that the taste, the color, all these things are actually in the object. And if you don't get it because your senses, because you have jaundice and you see everything is yellow or whatever the normal western philosophical examples, you're just misperceiving. It's not that you have a different, equally valid in your head perception, it's just not a perception at all, at least of that property.
0:53 Seth: Yeah, I mean you can say that the equivalent articulation in our so called tradition would be under normal circumstances or given normal conditions. And that's essentially what is assumed. But I do think it's interesting that they have this special case or they're calling out a special case for sound and sort of neglect to even mention the other senses. I've always thought that sight has been sort of overinflated in the philosophical tradition of the west as a sense and that the others have been somewhat denigrated. But I don't know what it would mean. Like if we couldn't use metaphors based on sight, how would we even talk about philosophy? Like what if you just had to recast the history of Western philosophy using the sense of smell? You'd have a really hard time.
1:48 Wes: Yeah, it's hard to talk about formality
1:50 Seth: with smell and structure and direction, distance,
1:53 Mark: canine philosophy, the dog. Write a sci fi story about the dog people in their epistemology.
1:59 Seth: No, you know what it'd be? It would be like something like an insect or an ant epistemology or an ant ontology.
2:06 Wes: Right.
2:06 Seth: These animals that can communicate with pheromone or with sense and there's a lot of information that gets, but changes everything.
2:14 Dylan: It's really interesting, but I think you're articulating why it's philosophically denigrated, or at least less paid attention to. I think we call it denigrated. You're sort of pointing to the absence of lots of philosophical discussion about smell. Right, but your point, you're also articulating why that's the case is that I haven't thought it through a lot, but at first blush our power of sight is our most finely. Well, it's probably a combination of the one that's Most used. And then the one that's the most finely grained in resolution. Now, I say that knowing that smells you can. We can distinguish just so many kinds of smells and in such small quantities. We don't have stereoscopy with them. Guess you have direction, but it's. But only in the sense of turning around, not in the sense of direction and depth that you have with sight.
3:13 Mark: Yes. Why do we have two nostrils anyway? Is it just redundancy my allergies make? One will be constricted more than the other.
3:22 Dylan: We don't have two noses, though. It just happens to be two holes pointed at the same note.
3:26 Wes: The entire body is symmetrical.
3:29 Dylan: Yeah.
3:29 Wes: And yes, it's an interesting scientific question as to why that is.
3:33 Dylan: But mostly symmetrical.
3:34 Wes: Right.
3:34 Dylan: Your heart's on one side.
3:36 Wes: Yeah. That's where God messed up just a little bit. It's supposed to be right in the center. But yeah, I mean, evolutionarily, our sense of smell has diminished and we do things, important things with vision that other primates do. Smell. Right. Interaction is much more governed by smell with primates. And it's much more. There are visual signals, of course, but with human beings, sexual selection has evolved us to greatly increase the importance of the visual. And I suspect in various other ways as well.
4:08 Dylan: You just gave me the strangest thought, Wes. I was just imagining like an episode of Bridgerton in which all of the romantic interactions and all of those balls were really driven by the way in which dogs interact with each other for smells. It's like just everybody's walking around sniffing each other's butts and stuff like that to figure out who's ready for what.
4:31 Wes: And if we had the same resolution of smell, it would be quite romantic. We could understand if we could read the smells anyway.
4:38 Mark: How would mass communication work? How would we have to figure out how to reproduce smells? The smell o vision. Or it would be the number one priority of the entertainment.
4:48 Seth: Now, dogs. Dogs wouldn't use smell for mass communication, but ants do. Yeah. It just means you have to get your audience in proximity to your message. The medium is the message.
5:02 Wes: Literally. All right, where are we?
5:05 Mark: We were up to. We were still on the first of the pramanas, the knowledge sources. We were on perception and we were still talking about the varieties of perception and inerrancy we had gotten to. Had we said enough about. I thought that saying, well, legitimate knowledge source, the legitimate perception is one that isn't wrong. That this was an unhelpful circular definition. But perhaps it's just a. This is Just how the semantics of the word works. I don't know. Was there something wrong with him? Just stipulating that, as we were just saying, if there's an illusion of some sort, then we want to just say, well, it's.
It actually is not using that perceptive source at all. It just, at least in a practical circumstance, you would say, how do you know that there's a snake in the room? Well, I just looked in the room and there's a snake. And you'd want to say, are you sure you got a good look at that snake? Look again, it was just a rope. You were too hasty. So you did not actually. You thought you were justifying using knowledge source, but in fact it was not a knowledge source at all.
6:12 Wes: If we say that perception is one of our primary bases for knowledge, then we get all kinds of skeptical problems, which I think Western traditions we're familiar with are fine with that. And trying to parse out those skeptical problems and say how it is perception can be a basis of knowledge when it is often wrong. But here I think it just sidesteps all that by saying we're only interested in veridical perception as a knowledge source and then we're going to go from there.
6:43 Seth: Yeah. Although that example, the way you characterized it, Mark, actually I think is. Brings out a nuance that maybe is interesting to talk about. So is there a snake in the room? Yeah, I just was in the room and there was a snake. And what you said, Mark, was, oh, are you sure it wasn't a rope?
7:03 Mark: That's the classical example they gave. But like maybe it was the cat toy, that's a robot snake and it was not.
7:09 Seth: Yeah, it doesn't matter. The point is, if you think about our category of the six different ways you can be conjoined or inherence and whatever, there'd be a difference between mistaking the fact that there's a snake in the room and mistaking the substance. So in other words, thinking that the rope is a snake is an inferential error. Seeing a snake where there is no snake would be a perceptual error error. I think those are two different. I understand Wes's point that they're only concerned with veridical. Essentially talking about the process of veridical statements or claims.
But what would that error mean? What would it mean for you to misidentify a substance through perception versus misidentifying or mischaracterizing your perceptual experience itself? Like I was in the room and there was an elephant Like, I got that wrong. That's just totally wrong versus, oh, I thought it was an elephant, but it was actually a crocodile.
8:15 Mark: Inference is interesting because inference, we're supposed to talk about that in a minute here. And I can't remember if it's some of the objectors that are very briefly brought up or saying, well, actually perception always involves inference because it seems, yes, like this is a topic that should have come up during this naming section, that it's a common Western thing of like, your sensations are never false, like you really did perceive a shape of something and it's only when you brought in the judgment that, oh, this was the fact that there is a snake in the room, that's when you got confused.
It's when you tried to attach the name that you were making a substantial claim as opposed to just reporting your phenomenal states. But this is a whole different view, as Wes was just saying, of perception, whereby even the raw perception that doesn't have a name, if we want to call it a perception at all, can't be misleading in this way. Or something like, or like you could say even your raw perception, oh, I just experienced blue. How could I be wrong about that? Well, this is not an internalist account, this is an externalist account. So if your brain had been, if I had done brain surgery on you to switch your blue receptors and your red receptors around, then actually your raw phenomenal qualia were wrong.
That's something that's like, we don't have space for that in our system. You thought you experienced red, but in fact you experienced blue.
9:40 Seth: And I suspect that they're going to fall back on like a pragmatic resolution where you would say something like, I think this is happening. Dylan, would you go in and take a look, see what you see? And if Wes and Dylan go in and they say, no, it's not a snake, it's a rope, then I'd be like, okay, well, looks like I'm hallucinating, or it looks like I've made the mistake like that. Ultimately, same thing with the blue and red. I mean, ultimately, if you're going to have an implicit or explicit criteria of quote, unquote, normal circumstances, then what's going to constitute normality is going to be something like consensus.
10:21 Dylan: Yeah, I mean, it's brought up in the translators commentary or whatever. One of them is the preoccupation in traditional Western epistemology with error and ways in which to avoid error or get to a single unified path to certainty that usually doesn't involve, often doesn't involve some kind of process. But here I think you're right, Seth, that there's always going to be a turn to. Well, let me check a different way or let me bounce this off of. Consider this other way. There's a triangulation that's always going to be used for any kind of epistemology. And what does that mean? That means that there's a way you're never going to trust an instance of one. Of anything. You're going to always have.
11:11 Seth: That's a really good way to put it.
11:12 Wes: Yeah, yeah.
11:14 Dylan: You're just not going to.
11:15 Seth: Right.
11:16 Dylan: I mean, and I would start thinking about things that we rely on a lot, like when we like basic arithmetic that relies on certain amounts of basic logic and stuff like that, two plus two equals four and all that stuff. But a lot of that you see what you're actually doing when you try to do some arithmetic with bigger numbers, where you end up maybe doing the calculation a couple different ways or a couple different times, you pull out a calculator because you believe in the authority and testimony of the calculator to confirm things.
11:46 Seth: You just put me in mind of, I don't want to say scientific method, but more of a scientific approach. And that's exactly right. And this, by the way, the translators make it clear that this system is the foundation of not just metaphysical inquiry, but scientific inquiry as far as this tradition is concerned. And there's no such thing as. Or the evidence of a single example. Right. I mean you just wouldn't, you would never, you would always look for repeatability or predictability or consistency. All these things are hallmarks of scientific approach to knowledge.
12:23 Dylan: And the only qualification I'd make are there that you would consider cases of single instances, but you'd have to come at them in multiple ways. So you say, well there's only one Earth and you start talking about what's the origin of the earth, how did it come to be right. You're going to end up going at that question in multiple ways and via other analogies that are repeatable and that kind of thing.
12:48 Wes: Should we get into this relation between inference and perception?
12:52 Mark: Do it?
12:53 Wes: Well, first we get the whole thing about it needs to be definitive and then we get an example of looking at someone at a distance. Also an example in the Theaetetus. Another way we might misidentify. So the Buddhists think that perception is theory laden in the sense that it's also that it's inferential or maybe, let's put it this way, experience is that way. And then what lies underneath that is raw, unconceptualized experience. Just looking at the. What the. The editors say here, which. Right. And I think, as Mark pointed out, this is definitely true, that we have to think of perception as having an element of inference in it.
So just phenomenology of it is we see, for instance, a surface and we infer the whole object. Right. We don't see the whole object and in fact the whole object. There's a sense in which it's not something that's visible because it's a structure. And you could say a lot more about that. In fact, Thomas Reid, who I want to. I want us to do at some point, big critic of Hume, says that what we see are actually signifiers. When we see objects, the raw phenomenon that we're seeing is basically a signifier. It's like concept use in a way. To see the object when we see a part of the object is to see a signifier.
And I don't know if infer is the right word, but to go beyond the signifier to the whole thing. Okay, so what is a Nyaya argument against that? I'm looking at page 25. It's. Or it's at least the argument, the, the Nyaya argument against the Buddhist view
14:27 Mark: just that yes, there are concepts involved, but the object perception that exists is ultimately real. It is endowed with universals or class properties. So they're realists about natural kinds or all kinds or natural kinds. They're just that. Well, I don't know if they're natural kinds, but they're not arbitrary constructions by the mind. Which is what the Buddhist wants to say, that if it's conceptual, then it's culturally relative, it is an invention of the mind, is not ultimately real. It's all bullshit. And no, it's actually right in the thing.
And again. But if you want to prove that, well, it seems like there has to be pragmatic grounds that how do we know that a snake is different than a rope, that is different than a mechanical snake that is made for cats to play with because of the obvious behavioral differences and your success in dealing with them based on one or another judgment.
15:24 Dylan: But part of it's also, this was mentioned earlier is this point of view of basically saying starting from, well, my perception is probably right and that you don't start from the standpoint of everything's wrong. You don't start from a position of complete doubt and then try to build up certainty. You start with, well, it's probably right. But then you're doing checking on that.
15:48 Mark: Right.
15:48 Dylan: And that's a pretty different point of view. And I think it's actually probably closer to the way in which we actually operate in our lives. Right. We don't operate in our lives at all by calling into question everything we perceive all the time. We don't do that. We actually make judgments immediately, very quickly. And then we have, depending upon the circumstances, ways of checking that or verifying it, often based upon how important it is that we're right or not there
16:21 Mark: being an actual legitimate challenge. So I was a little doubtful when you just said don't trust a single instance of anything because we usually trust a single instance of anything because there's no controversy about it. It's only when we want to establish it as part of a scientific system, for instance, that somebody might say, okay, well let's double check this, let's. But as a practical matter there.
16:41 Dylan: Yes.
16:42 Mark: Most of the judgments we make are in unique one off situations.
16:46 Dylan: This is true. Yeah.
16:47 Mark: That we can never exactly repeat that or have somebody. Even if somebody else comes in the room and checks what you just looked at. Well, they're at a different time. They're seeing it with their perceptual. There are issues that make the. Any analogy between two situations only approximate at best.
17:05 Seth: No, that's a really good point. Mark that. And again it comes back to this pragmatic coherence notion. So there's two ways. Just thinking of Seung, remember him? Oh, Error can be made in two ways.
17:21 Wes: And spitting, just like Phillips spitting on the audience as they talk.
17:27 Mark: That was part of the knowledge transmission, is that it's physical through the ether.
17:31 Wes: You got to have direct physical contact or else it doesn't.
17:34 Seth: That was. That's how perception works.
17:36 Mark: People are listening. Do you have the new iPhone that spits a little in your ear whenever I say anything?
17:40 Seth: That would be a really cool feature. Could give you a little tongue while you're.
17:45 Wes: Right.
17:46 Mark: I had that on when I was watching the debate last week. It was horrible. Sorry.
17:51 Seth: So no, you can be unsure or you could have an experience that. So one way is like I see in the distance. Is that a post or is that a human being? Right. I don't know. I gotta get closer or I gotta ask somebody to figure out which it is. But if you had the experience of something you expected to be a post and instead it was a human being or if you had an experience that just didn't accord with your past experience, it would cause you to do a double take. Or perhaps you would call in another testimony, another experience and testimony, because. So in other words, the standard by which you determine whether a one off situation is you can act on and is whether it accords with what a certain set of expectations that you have.
Or at least maybe expectation isn't the right word, but I think it actually is. Right, you get my point. It's perfectly acceptable to act pragmatically on a single instance of something, but only if you're sure about it and it's not out of the ordinary, whatever ordinary means.
19:00 Wes: The response, I guess Mark, you summarized it is they're realists about universals and class properties. Is there anything else to get from the section? It's a bit complicated. There's a reference to Bessie is a cow, which again is a. Is an example that produces suspicion. Unless they really were using naming cows Bessie back then. Well, although when they give the example of a person's name, it's Dita, so they didn't feel the need to modify the person's name. But the cow got a updated western
19:35 Dylan: name as I put in parentheses. Dita being a common name for a person like John.
19:42 Wes: It's really strange.
19:43 Dylan: Interesting editorial, juxtaposition choice.
19:45 Wes: Yeah, but. Okay, so. But at the beginning of the section, the. I'm not going to try and pronounce this commentator's name, but they claim that it's Vachaspati.
19:54 Mark: He's the latest one, number three. We could just say. Yep.
19:59 Wes: Yeah. So he's claiming that the charge is that there's an incompatibility between genuine perception and something that's expressible in words. That when we get to the point where we're using words, we've already gone beyond perception and it's no longer properly called perception. And the response is in part to say that yes, that would be true if the object were just a bare particular without all these universals and properties encrusted on it. My word, I don't know if that's the right word, but it's not. It has universality built into it. And then language hooks into universality. Universality doesn't need language, as we said before. But universality is the thing that can be hooked into with language. Okay. I mean, are we satisfied with that account?
20:59 Mark: Yeah, that's interesting. I'm just that saying that this was a response again, this is what, 900 CE, way after the other folks. And there's been a lot of back and forth with these Buddhists saying that, yeah, the true object of perception is the bear is the individual and I. It just. This didn't even seem to be something that was on the radar of these earlier commentators of. Is it when we perceive. Right. The individual cow, are we perceiving it qua cow or. We were just perceiving it and I thought that. Yeah, we had gotten this model.
21:34 Wes: This un-Hegelian.
21:36 Mark: Yes, this. That we gotten the model of. Well, the this is the first moment, but then we actually do perceive it. While the universal as well, quality individual, it cannot be conceptualized.
21:47 Wes: Well, the Hegelian point, of course, is that this is already universal even to say a. This. Right. And then beyond that. Yeah, we need universals even to. We don't start with the this. We got to come back to the. To the this via. Via universals.
22:01 Mark: Yeah. So I guess there is no concept of sense data, right?
22:04 Wes: Yeah. I mean, we do get an interesting claim here that the words. So if you say, okay, yeah, you can have a proper name. I thought it was saying that you could have a name that refers. So if you say. If you point and you say cow, it refers both to the universal and to the particular. Sure, I think that's right.
22:24 Mark: Yep.
22:25 Wes: Yeah.
22:25 Mark: And we are perceiving both of those things. We're perceiving the trope of his cowness and we are perceiving the universal cow through it.
22:33 Wes: The ideas we have about the thing are not generated by the object because the thing's unity is constructed imaginatively. We have already pointed out in the introductory comments about it not depending on language. Objects do not contain with themselves the words that. Okay, so I'm a little confused by this. I haven't asterisks, but the ideas we have about the thing are not generated by the object because the thing's unity is constructed imaginatively.
22:59 Mark: Right. That's why the commentators were saying that actually Vachaspati Mishra is making is sort of going halfway to the Buddhist position here that this is probably a significant. If we want to say, what's the Nyaya view? Well, what stage of Nyaya are you talking about? That this is actually a movement from the earlier folks. A way of interpreting does not depend on language.
23:22 Wes: I mean, he's saying this is wrong. Right? It's wrong to say that the ideas we have about a thing are generated by the object because of the objects, because we construct that unity imaginatively. It's. This is another way of saying that the unity, it's a real unity. It's out there in the object. Just. And the universals is.
23:41 Mark: There's a double negative here. That makes it. It is wrong to suppose that the thing. That the ideas are not generated by the object. Okay. Yes. So it is correct to suppose that they are generated by the object. Yeah.
23:54 Dylan: So this is the realist.
23:55 Wes: It's not just constructive. Yeah. Imaginatively.
23:57 Dylan: It's.
23:57 Mark: Yeah.
23:58 Wes: In the object. Okay.
23:59 Seth: I don't.
23:59 Wes: It's saying. I think this is. We're repeating something we've covered a few times now. Universals are immediately perceived, and if they weren't, children and the mute would not have sensory presentations.
24:11 Mark: Right.
24:11 Wes: Since they have no memories of words. We can't just require words. But. Yeah.
24:14 Seth: And is it. Can we in this case say that the substance is the unity and it is also immediately perceived?
24:21 Mark: I mean, I think we're gonna have to do this chapter on substance if we really want to understand that, but. Yes, that seems reasonable.
24:29 Wes: Yeah. What about inference? Yeah, yeah.
24:32 Mark: The end of the perception chapter is just about apperception, which it seems like if you. If everything has to come from the senses, then what about your knowledge of your. Of yourself, of your own mental states? That's not. There's no sense organ. And he wants to say, actually there is. It's the mind. The mind is the inner sense organ, which I think is. We've seen that in the Western tradition as well, somewhere. It doesn't seem bizarre to me. Anybody want to dispute that or.
25:03 Wes: No, I think. Yeah, we have seen it. We're running out of time, and we haven't even gotten to chapter two. But you should definitely talk about inference a little bit.
25:11 Mark: Yes. So the question, as when we were summing this up and talking about in the first half of the discussion, was, is inference just something like induction? No, there's all these other things that it could be. But the key here is you're looking for in the way they talk about is you're looking for an inferential mark that indicates some other thing. So you're smoke.
25:32 Wes: Yes, Fire.
25:33 Mark: Right. So smoke is the inferential mark on the situation. And you always see it when you've seen fire, there's a fire situation, then smoke. You're. In this case, you're saying, smoke is the mark. So wherever I see that mark in the future, then I'm gonna think fire. What did you make of these? Commentator 1 is, well, I guess it's right. It's right in the Sutra itself. Next is inference, which depends on previous perception and is threefold from something prior, from something later, and through experience of a common characteristic, which I didn't see this as very fundamentally Different. It was just a matter of what later I see the. I see smoke, or maybe I see smoke and fire together that I'm remembering.
26:16 Wes: It just seems to calm the something later, right? We see a river brimming with rushing water and we can infer that something happened before that to cause that rained upstream through a common characteristic. What is that example? The change location. So we infer the sun is moving because we see it in two different locations.
26:37 Mark: Which is a weird example because the change in location is not a common characteristic. It's a difference between the common characteristic is everything else about the sun. It's sun quality over in location A and the sun quality is over in location B. Therefore there's been movement like that. Doesn't know it just means it's the same sun as it was before. This is. It seems a terrible example.
26:58 Wes: So in. But the interesting part of this for me was these alternative. These different. He gives those basic explanations of something prior later common characteristic. But then he wants to give us more interesting versions of those. So alternately, something prior we could infer something that's unseen. So I'm not sure about that example, but from something later that could be argument by elimination. So this is, I think, getting us towards a more scientific approach where we can start doing inferences to the best explanation. For instance, I think falls in this category or giving a model.
27:36 Seth: Right?
27:36 Wes: We can infer an atomic model from the phenomena from the data that we have. But it doesn't mean that we need to see that we don't. It's not like we're seeing two conjoined things. Oh, here's the. The spectra. And then, oh, there's the model over there. And I see them conjoined. And then no, we just. We come up with the model ourselves based on that data. So it becomes. I found this interesting in the sense that the commentator extends. Right. It's a bit like again, it's saving the appearances. They're taking these very terse sutras and they're fleshing them out to make them as comprehensive and interesting as possible and to cover more sophisticated theories about what counts as knowledge or how knowledge works.
28:18 Mark: We have another example of the self here. Vatsyayana is saying from. From common experience. When the relationship between an inferential mark and what it marks is not perceived, one infers an unperceived object through a mark that has a similarity to something else. For example, one infers the existence of the self from desire and the like. Desire and the like are qualities and qualities inherent substances. That substance in which desire and the rest are situated is the self.
28:44 Wes: Yeah. We know that properties like red are attached to substances. And then we say, oh, desire is a property just like redness, therefore there must be a substance. We get, I think therefore I am the property of desire. So there must be the substance underlying it. But we didn't have to see those two things conjoined and then draw that inference. Based on the conjoined, we can invent the self. It's much like the atomic model.
29:08 Seth: Again, I actually like the next paragraph too, although I think it raises some issues. So perception targets existing things, while inference targets both things that exist and things that do not exist. How so? It's because inference ranges over the three past, present and future. For objects in all three times are grasped by inference. It will be, it is, it was. And indeed things past and future are non existent. But when you recall a memory, the self of the past and the self of the future are also non existent.
29:42 Wes: Yeah, I mean, in a way a memory itself is something like a sign. Right. So we can infer from the memory that the thing did exist. It's no longer here with us.
29:53 Mark: I think earlier somebody has referred to memories as perceptions. That how we establish that an inference is legitimate is because I have seen smoke and fire together in the past. Well, I'm not perceiving that right now, but yet I am. I can as part of this pattern, perceive the memory. So I'm perceiving current smoke. I'm then adding a second perception, which is my memory. I'm perceiving the memory. And then the third perception is, oh, now I see the smoke as indicating fire. So this was again like the three moments, like the Hegelian moments in a perception, they all really happen at once. But yeah, this one of the commentators here talks about it be a threefold.
30:37 Seth: Yeah, that's right, the three stages.
30:39 Mark: Yeah, I think that's Uddyotakara who says that stuff.
30:43 Wes: So now we get the editors make a move here to talking about the way in which composite substances inhere in their parts.
30:54 Dylan: Yeah, just as a.
30:55 Wes: All the way down to atoms. Yeah.
30:57 Dylan: Just to clarify my understanding, these sections, long sections with the editors, are them summarizing interspersing points in ways that they made the judgment that the text just isn't as worth reading at this point. And they're going to summarize the result.
31:13 Mark: Yep, yep. So they're not just commenting.
31:15 Wes: Yeah, sometimes the commenting. It's a bit confusing because of that. Sometimes they're commenting, sometimes they're Filling in the space.
31:21 Dylan: Yeah.
31:21 Wes: So summarizing and commenting, but how is inherence here? I'm trying to figure out the relevance to inference.
31:28 Dylan: So composite substances such as a pot or a wagon can adhere in things. A wagon inheres in its parts and the parts in further parts, down to non composite substances, atoms. But all substances either inhere in nothing such as atoms, or they inherent two or more things such as a dyad, which is something that inheres in two
31:48 Wes: atoms, sound inheres in ether.
31:50 Dylan: Yep. This is back to your comment about there's simultaneously a emergence claim about inherence. That the wagon inheres in the parts of the wagon as well as the parts are inhering in the wagon is more of a compositional claim. The way sound inheres in the mo. In the ether.
32:12 Mark: Yeah.
32:13 Dylan: In the ether.
32:14 Wes: Yeah. I don't know why this is in the inference section here.
32:17 Mark: I think it's just because it's a connection between two things that we can then infer one based on the other. So if you know that a blanket is made of threads, then I mean, what, you can infer it either way. Right. If you see threads, therefore there's a blanket. I think it's only the other way. If you see blanket, then you can infer, even if you're seeing it from far away, that the blanket is made of something. It's made of threads.
32:40 Dylan: Well, you could have the case where it would be an open possibility if you see some threads that, oh, maybe there's a blanket, maybe it's a blanket. This is the way you would say that when you see a wagon wheel, that's really. You're also seeing a wagon. That's why it goes that direction. It's just like, I see a nose, I see a human nose.
33:01 Mark: No, I think that works especially if like I was unconscious and I open my eyes and I see before me a wagon wheel or a nose or something, then like it is reasonable based on past experience to infer that's that's probably a wagon. It's probably not just a wagon wheel that somebody propped next to my head as I was sleeping.
33:18 Seth: Yeah.
33:18 Dylan: And that's also, I think that that's why it ends up being in the inference section, is the conclusions you're making based upon those perceptions.
33:26 Mark: So it seems like inference should be way more difficult to do correctly than the other kind. Right. I mean, of course you could have perceptions that are faulty in some way. You didn't get a good look at it. But there's Just many more ways for inference to go wrong. Right. Just because you've seen something happening in the past does not mean.
33:47 Wes: Right.
33:47 Mark: It's perfectly legitimate that there really could be a wagon wheel without a wagon. It fell off the wagon. The wagon hasn't been built yet. If you're at a wagon wheel store, there's so many ways that it's not actually going to be a wagon. And so you just at least have to supplement this with perception. Or I think it's another one of those things like, well, if it's uncontested, if your inference is uncontested, then we just go with it. And it's only when then you have two inferences both ways that he brings up the second. This concept of Tarka, of, well, okay, you say that there's a full wagon there and I say there's not.
Well, I show that your view has some unacceptable consequences. So we can wipe that off the map and then we're left with my view, which was reasonable in the first place, supported by whatever evidence it has. And there's no reason now to brush it aside. So I haven't proven it.
34:38 Wes: I don't, I don't know that I like this inference from parts. But. Right, the other way, where we do a model of the phenomenon and we say there's microstructure, there's atoms, that I think is, certainly is inferential. But we're talking about emergent phenomena which are important because they. So atoms. Right. Obviously we don't see atoms and make inferences to objects. That's not the, the point here. But there is an important relationship between atoms and emergent macro level phenomena. And again, those phenomena are important because they are a way in which the parts can be more than the whole.
Sorry, the whole is more than the parts. To put it in Western terms, when we see a macro level phenomenon that we want to call unity, and we try to explain it, so we can do that by appealing to an Aristotelian form, something like that. But that's really just a placeholder for more sophisticated explanation in terms of microstructure. At least that's the way we think about it today. So we infer that there is some thing that makes this a genuine, this a genuine unity. We can't just appeal to the macro level phenomena because those are fuzzy and there might be a leg missing on the dog, or what if the dog is an example?
Or a wheel missing on the wagon? So we automatically, even if we're not scientists and talking about atoms, we're automatically saying, well, there's something, there's some metaphysics, there's some substance that's unified at bottom that explains the unity of the phenomena as we perceive them. So that I can see experience as being inferential in that sense. So when we write. So when we talk about perception as theory laden, we're not just talking about the fact that we need that perceptions already come to us with universals. We're talking about the fact that we inferentially create a.
This. Anyway, that's probably not very clear, but I think that might be part of what's going on here.
36:44 Mark: Yeah, I don't know that we. Again, I was sort of maybe improperly looking for some sort of toolkit here for like, how can I judge whether this is a good inference? But instead we just get some of these categorizations like, oh, well, you could have a positive negative inference. The inferential mark is there when the thing is not, but it's not there when it's not. So if there's no smoke, there's no fire, or just a positive one. An inferential mark is sufficient for the property to be proven. There's no fire without smoke or negative. The conclusion is negative, motivated by a lack of some inferential mark.
If no fire, no smoke. So those were all very subtle. It's like we're listing some logical rules, just that if A is necessary for B, then B is sufficient for A. It's just these logical rules like this, like, okay, I was clocking. Those were interesting, but they don't actually. I don't know. They're true, but not as helpful as I would like for these actual examples that what they want to give, like, so a analogy might be a separate thing from inference, or maybe it's a type of inference that's sort of up in the air between the commentators. But the example they gave is a buffalo is like a cow.
Or there's a different name for buffalo, but whatever, you can see it and identify it because it looks like a cow. So if I see something later that looks like a cow, after you've told me, oh, there's this thing that's kind of like a cow, but it's not a cow, then I might say, oh, that's what you were talking about. But you could see how that could obviously be wrong. Like if there ends up being more than one thing out there that's, oh, that's not a buffalo flow. It's a different thing. I see a cat and I'm not familiar with cats that sort of looks like a cow or I see a, a bison, I don't know.
38:22 Seth: Yeah, that example seemed kind of weak to me because it's. In that case, it's talking about naming. And so I think the implication is not so much that it's purely linguistic. Like, oh, now I've never seen a buffalo. I come across this, but you tell me a buffalo looks like a cow. And then I come across this thing that kind of looks like a cow but isn't a cow. And I say, ah, that must be a buffalo. It's not simply that I now have a name for the object. It's not a linguistic thing. It's that this is in essence, you've given me testimony. I'm combining inference and testimony in order to be able to put myself in an epistemic situation where I can have direct perception of a buffalo and have it be veridical.
39:10 Wes: So I think this, it seems to me this is about how we can learn to connect a word to a kind of object. So the word gavaya is the name or buffalo is the name of that animal. So how do we do that? How do we reliably create a connection between a word and a kind of animal? Especially if we assume that this is a new phenomenon for us, right? This is a new type of object that we're encountering. The suggestion seems to be that we have to build off of some previous name object combination like cow. And this. I can on analogy connect buffalo to that. And it helps to have some common qualities in order to, to be able to, to draw that analogy in order to name something.
I haven't fully figured this out, but it seems to be something like that. So that actually is quite an interesting account. Do we have to build off of our previous linguistic associations in order to make a new linguistic connection? Because really this is about how do we perceive the gavaya or the buffalo as an actual unity, as a thing, as opposed to something accidental, something that's not, not worthy of a name, Right? Not, not everything. Right. A heap of stones doesn't. Oh, there, let's give that a natural kind name. We don't do that. But a child would know, right?
With only see, having seen one buffalo ever in its life, it would know that, okay, that is a real unity, a real mind independent unity. Of course, they wouldn't put it that way. They would get very excited. Look, and they point. They wouldn't necessarily have the name for it, but they would understand that it's a. This, it's worthy Of a name.
41:01 Seth: You say that a child would know that. I would say that the child would know that in the context. In its cultural context.
41:08 Wes: This is where I completely disagree. I don't think they've ever. It's not induction. It's not. There's a name already there and they've been taught it.
41:15 Seth: No, no, no, no, no.
41:16 Wes: Actually, in. It's our phylogenetic heritage, natural kinds of. Yeah. We are equipped to see a buffalo never having seen it before and to know that it is nameworthy. Let's just call it nameworthy. That it is a kind. Yeah, that's my view. I'm not going to defend it here, but.
41:35 Seth: Okay.
41:36 Mark: Should we talk about testimony for five minutes before we're done here?
41:39 Wes: Sure.
41:40 Mark: Luckily, it's. I pitched this to you. You were excited about it, Wes, because. Oh, we get to talk about testimony. I have a lot to say about testimony, but what they have about testimony is much more scanty in this text than about the previous ones. That's just testimony is instruction by a trustworthy authority.
41:57 Wes: And this is not right. I think. And I think this is often what people get wrong about testimony, because it's not a particular person that is authoritative. As much as we say we want to defer to experts and expertise and people with credentials, I don't think that's right. I think it's not that there's not something to that, but it's really the discursive process that is authoritative and the discursive chain of evidence. And we might even say, right. The scientific process. It's. For me, I have a process oriented explanation of how testimony is authoritative.
And ultimately it comes down to our. What we trust is our own ability to evaluate the psychology of other human beings. Right. And to do consistency checks and to use reason to cross check against different sources of testimony and all that stuff. But most importantly, we don't. We're not conspiracy theorists and we think that overall, Right. The existence of the moon landing, or even of Lincoln, let's say we never met Lincoln. The pictures could all be fake. The accounts could all be faked. We trust authority in that sense. Not because our teacher, the historian, ooh, they must know, but because the discursive chain of custody, we expect that to be intact because we think the psychology of it, that someone would fake that over hundreds of years is kind of crazy.
You see what I'm saying?
43:23 Dylan: And also not just someone.
43:24 Seth: Right.
43:25 Dylan: In some of those cases, it's another sort of multiple evidence. The amount of the scale at which there would have to be Some collusion. And in such a conspiracy theory becomes much more unlikely. Right? Yeah. All those different people generating all those different books about Lincoln over all those, all that, I mean it becomes a tenable.
43:48 Wes: Two different forms of psychological implausibility are at stake there. One is just that people would actually be motivated to do that and have the energy to do that. We expect human behavior to be motivated in some way. And yes, people lie, but we think we, we understand why it is they lie and what the way they lie. And we do that because we project our own motivations on others. We think we understand ourselves well psychologically and we can use that to understand others. And that helps us discard this idea that anyone would be motivated to have a vast conspiracy.
And about Lincoln. And even if they did, we know that people can't keep their mouths shut. There's just too many failure points in that because there are too many people who have to be involved over an enormous time span. The truth will out, in other words, and the truth will out. The more likely it is to out, the more members of the conspiracy that you have, the more failure points that you have. So those are two forms of psychological implausibility. And I think this, we should do a whole episode on this because I think it's very important because most of what we say we know has nothing to do with perception.
It has to do with perception at the periphery. Right. Someone has perceived it. Someone met Lincoln. But not our particular perception. It's almost entirely testimony based. But we think we know it's a knowledge is a social phenomenon.
45:04 Mark: Yeah. So there's acknowledgment in this section that we really need testimony that we don't just go on our own perception. But the definition that Vatsyayana commentator number one gives a trustworthy authority is someone who knows something directly. So that sounds like you need. It's only one stage removed from perception. Whereas what you're talking about is. No, it's actually probably many stages removed from many perception and modeling and discussion among people. And the way that we don't just establish there's a sense if and when we get to talk about Chapter 2 in a future discussion where these folks are looking to build systems.
So again, we're not doing a Cartesian doubt. We're not trying to get down to foundations. We like take some things as established and that could be on the basis of long standing testimony. We don't even remember where the original, where this originally came from. But unless there is like some serious reason to want to verify this Particular thing. We don't just need to periodically go through all the foundations of our knowledge and say, well, let's question this, and let's question this just for the sake of questioning. Like, if there's. It's just too.
It would be too hard to do that. We've never be able to grow the founding.
46:18 Seth: Yeah. Well, in a weird way, this. It works against and is rescued by the pragmatic piece and the coherence piece. It's like, okay, well, if most people take the testimony of this authority as sufficient for such and such a situation to motivate action, and I do too, I don't really need to verify it firsthand. On the one hand, there seems to be this sense in which perception is privileged as the knowledge source, and that ultimately, in a weird way, I don't want to say you don't ever know anything until you perceive it directly. But there's a sense in which perceiving it directly is superior as a form, as a knowledge source to other sources. But the idea is that somewhere down the chain, somebody did perceive it. But in reality, just as Wes just pointed out, almost everything is testimony from inference.
47:17 Wes: Yeah. So if we were to think about this as a system, right. You could recast this to say, the tent poles of the system. There have to be nodes, right, that are trusted authorities or that are perceptions and correct inferences and all of that. It's not that I perceived it, but the system has to have perceived it. Right? The Borg at some point has to have perceived it. And if I'm a conspiracy theorist, I don't trust that whole discursive system and that discursive process. I ironically, it's not that I'm not scientific enough. It's that I take science too literally and become hyper scientific.
So I'm sitting at home saying, well, oh well, to know how the towers fell on 9 11, I'm not going to trust the scientists. I have to see it for myself. I need empirical evidence. I need to do my own experiment. And here I am dropping a brick on like some plastic model and oh, it didn't fall, so it couldn't have. It had to have been blown up on 9 11. So really it leads to a kind of stupidity. But the interesting irony to that is it's hyper scientific in the sense that it demands. The conspiracy theorist demands their own perceptual evidence. They demand their own experimental evidence instead of taking that from the system.
And they can always say, yeah, well, look, people are politically motivated and they're conformist. That's a really good argument. So the system must be corrupt. But of course, there are reasons why. There are reasons to trust the system and the discursive chain. It's not infallible, but overall it's the best we have. Is that discursive process in that discursive system, even though individually human beings are imperfect and they're liars and they're this and they're that.
49:07 Seth: Right.
49:08 Wes: It's when you globalize that dissatisfaction with other individual human beings and their fallibility and you globalize it to the discursive process as a whole. You get the conspiracy theorist.
49:18 Mark: Yeah. You're referring a lot to some bad inferences that this experiment that I'm doing at the Brick has anything to do with. With the legitimate how a building would work. Whatever. I guess there's still a question whether testimony is reducible to perception and inference,
49:35 Wes: but also just the impulse to try to do it oneself. Is that lack of trust, that, hey, there are others who know about this. And I'm plugged into that in a reliable way is what I'm pointing to. Yeah. It's also bad experiment.
49:48 Dylan: Yeah.
49:48 Mark: I mean, I think sort of want to leave it as an open question because it's not addressed in the text, is to what degree is testimony reducible to. I perceive that somebody has told me something. There's certainly perception involved. Did you hear that person?
50:01 Wes: Right.
50:01 Mark: Did you hear the witness on the stand? Were they speaking clearly? And inference. How do we determine if somebody is trustworthy? Well, is this a person that has lied to me in the past? Is this the kind of person, if I'm in a doctor's office that my HMO will pay for or whatever. So it's probably a legit doctor, not just a doctor that I found on the street when I was drug seeking and is telling me something. To what degree do I. It seems like it is based on past experience. Even if it's the experience of your parents that told you about this and they learn. So it seems like it has to. Ultimately, we learn from testimony. We learn from testimony. But that somebody's making an inference based on perceptions.
50:41 Wes: Yeah, but I think it's A lot of it, though, is inference from psychological principle. And we get that. Yes, of course, from experience with a people, but we get that also from. I'm a kind of a Cartesian about this. We get it from our experience of our own motivations for action. So if someone is on the stand. Yeah, if someone is on the stand in a courtroom. Yes. We can evaluate the credibility we can evaluate it really well in a lot of different ways. The same thing with the newspaper article. People act like we're kind of just screwed. Oh, is this a bad source of news?
That's not a trusted authority, therefore, no. We have lots of sophisticated tools for evaluating testimony. Those include right logic, right internal consistency. If something is. If an article or a person on the stand is contradicting themselves, we're less likely to trust what they're saying. But also if they have bdis. Sorry, they're shifty. No, but also there are psychological cues, lots and lots of psychological cues to when some. And sometimes it's just even a vibe. So we have lots of psychological experience from which we can draw inferences about the credibility of someone or a newspaper account, any sort of account.
51:59 Mark: So we're going to republish this book that we have the Vatsyayana, we have a Uddyotakara and then the west commentary on that.
52:06 Wes: West. What's a good Indian name for me?
52:11 Mark: Well, this has been enlightening enough conversation that we want to spend a little more time with this book, actually get a little bit more into probably 3, 4 and 5 just to get with the metaphysics that they're arguing for. So that's what we're doing. Next time, if you'd like to suggest something different, something along these lines or along completely different lines, feel free to email us pel@partiallyexaminedlife.com or comment on our Facebook or Twitter or Instagram posts or many other ways. Check out the book partiallyexaminedlife.com book check out the class I'm going to teach in the fall partiallyexaminedlife.com class Good night, everybody.
52:49 Dylan: Good night.
52:50 Seth: Good night.
Leave a Reply