An informal dialogue between Michael Frank (M, italic) and Terry Winograd (T). This refers back to an earlier interview with Dretske. After conducting both of these interviews I wrote a paper about them.
M: Let's see. OK. Well, one of the things you stressed in the paper is that there's no - if you're going to have meanings, there's no objective way to define them for any set of meanings, in terms of objects and properties, that you have. It's just kind of a provisional realism, and you can always make further distincions that go above and beyond the set you have. And there's no `right' way of carving up the world into objects and properties, there're an infinity of different ways of doing it. And during discourse, we often move around from one perspective to another, carving up things in different ways. Now, I talked to Dretske, and he admits all that. He said in my interview that he doesn't believe in "perspectiveless positions", in that when you carve up the world a certain way, there's some absolute perspective that you can go to, that's the right way to do it. But he doesn't think his realist view requires that. There may be any number of ways of carving up the world, and none of them are objectively right, but, given any particular way of carving up the world into objects and properties and meanings, that are kind of propositions about them, the meaning is either supported in fact by the world, or it is not, that's what he thinks. And whether it is or not, is an important thing in determining whether a sentence is going to be true, and so forth. Like you mentioned in your paper that Perry had said that whether there really was a pitcher of water in the refrigerator or not seemed to make a difference no matter which way you defined water, I mean, some fact in the world is going to be important. So now, here's what I think you might say about that, and you can tell me if I'm right, and add to this, and correct it, and so forth. The problem is that, in practice, even philosohpers can never really specify any meaning precisely enough to fix upon something in the world that is objectively either supported or not. And even philosophers, when they're trying to describe some "meaning," they're still situated in this background, this context, the purposes of which are guiding their breaking-down, their indentifying of that meaning. So even if they think they've got something objective that's not contextualized, really it still is contextualized because it still contains terms from the language.
T: I accept the facticity of the world; I don't believe whether I would be alive or not if I walk out in front of a car is a matter of how I construct my this, that, and the other; if I walk in front of a car, I'm going to be dead, right, that's not something in my head, right? that's just going to happen to me. So in that's sense, I'm not saying that there isn't a reality to the world, but, exactly what you're saying, as soon as I make a proposition about the world, the very terms in that proposition are not objectively definable. If I say, "There is water in the refrigerator," and you say, "Okay, for your meaning of water, is it true or false?" there is no well-defined thing called "my meaning of water," that doesn't have boundaries that I could think of, I mean, I use it in various ways, and I have a, you know, all that kind of stuff. So that's, I think that's another way of saying the same thing.
M: Yeah. Of course, the realist might grant that whenever you're talking, you do have some background left. I don't know if all of them - I'm not sure if Dretske would admit that, but it seems like if he did, he might still say that in principle, there are still meanings, and any one proposition never fixes precisely upon one meaning, but there is, out there, kind of some context-free meaning objects, so to speak, and the world supports some of them and not others, and he has this intuition that that's the foundation of reality. But you might respond that, although it's a consistent view to postulate these objects out there, it's irrelevant and useless when talking about the mind, because we can't ever get to them really, and just the fact of their being out there doesn't seem to help any particular theory, like, you said... let's see... can't find that quote, but... so it doesn't help anyone to talk about this infinite set of overlapping fact in the world. You might think it's just an untestable ontological game that doesn't bear on anything. Do you think that's pretty accurate?
T: It's the argument I make, in various of these things, to Perry and Barwise. Saying that that is a logically consistent path, but it leads you to basically this infinity of slightly, vaguely different meanings, all of which are real, but then, you know, so what? There are basically as many different ones as there are people and their use(?) situations, and they're not very interesting, postulating their reality in some sense doesn't...
M: But that's just one kind of ontology. Do you believe that ontology in general is something that's useless? Postulating any sorts of existent objects, and so forth?
T: Well, ontology as an enterprise is very useful. Alright, now the question is how do you see - I mean, trying to think about what exists, right, and I think that's what I'm doing, right, the whole point I wrote the paper is that it's doing ontology, so if that's it, I'd hardly argue that ontology is wrong or useless. Right, if you just - I mean, you can imagine sort of mystical traditions in which ontology, trying to articulate the world, is the wrong thing to do, you just get enlightened, or something, right? That's obviously not my position. So now, the question is, what do you think you're doing when you do ontology. And if what you think you're doing is getting a handle on "the things that really exist," aside from the fact that you and your culture are engaged in the process of doing ontology, that's what I say is wrong. On the other hand, saying that you can elucidate what it is that you and your culture are doing in ontology, and that will - and it's this approximation business, right - to a large extent that will look like, well, there really are chairs and tables and atoms and electrons and all that kind of stuff, is fine. In the end, if you say, well, do Schroedinger probability waves really exist or not, right, then where I'm saying, well, that's, you've been misled by the nature of the question. They certainly exist in the discourse by which we understand our world, no question about that; and to sort of, you say, well, they don't exist, but photons do, right, is making this assumption that there's this sort of external valid - external notion of existence which is separate from that, which is not ultimately doable. On the other hand, if you say "do ghosts exist" versus "do people exist", I would say that's a different kind of question. You're taking for granted a certain notion of existence which is people observe them, so on and so on, and now you're simply saying, empirically, has that happened or not. But that's not ontology, that's really a different kind of thing.
M: Empirical studies, or something. Right, so you think that, let's see... Well, Dretske says that what you're doing is mostly epistemology, he says you're talking about what we can know, and how we can know it, and he's saying you're avoiding ontology altogether, or confusing it with epistemology, or some such. But you think you are stating an ontology, because you're saying that... you obviously know what [ontology is.]
T: Well, I think there's an element of truth in what he says, which is, I'm making the claim that ultimately we can't do ontology free of epistemology, that's it's not doable, to separate those. So to the extent that he says I'm therefore giving up doing ontology, I'm giving up a certain sense of what it would be to do ontology.
M: Even the philosophers cannot... everyday people don't do ontology, but even philosophers can't really.
T: Well, everyday people do do ontology in this other sense, that is, they don't do it in a formal way, but sort of wondering about, you know, does God exist, right, and that's ontology. Or yeah, somebody who just all of a sudden recognizes the solopsist paradox, right, which is I can't actually prove that this isn't all a dream. I'm sure most people at some point in their life, they sort of say, wait a minute, how do I know this isn't a dream.
M: Dreams seem real when you have them, and then they're not.
T: That's ontology, right? In some sense.
M: Okay. Here's one thing that realists might say, that makes them nervous about rejecting absolute objects and properties, they might say that the way the world is must obviously affect us somehow, and so that's why they think that there are there are these objective `things' in there that are doing the affecting. You might respond that yes, the way the world is does affect us, but it's sufficient to just say that, that there's a world and it affects us, and there's no real need to postulate this infinite breakdown of it into these multiple overlapping objects, which is Dretske's view of it apparently. He was talking to me about an example of, say, a chair that we think of as one object but someone else might habitually think of it as twenty objects, and then if Dretske says there's one object here in the room, he says, he'd be right, and if the other person says there's twenty objects in the room, he'd be right, and they're not disagreeing. So Dretske says there are these kind of overlapping facts, and so on. But, that's getting away from the point a little bit, which is that you don't really need to break down the world into these things in order to say that it affects us.
T: Right. ...stepping in front of a car. I really do believe that it's not my mental construct. But, you know, the notion that a "car" exists as a well-defined property in the world, and of course that one's a funny one, because it is a highly - it's an artifact, right, it was built specifically because somebody had a construct called "car" in their mind which they produced, so...
M: Another thing: Dretske seems worried that if you don't believe in these absolute objects and properties in some way, then the world is just going to look like a "homogeneous bowl of jelly," in his words. And I suppose you would say that you never said it was, and that you wouldn't mean to imply that the world was kind of a homogeneous thing, just by denying this certain way of breaking it down. I mean, you can characterize the world in any number of different ways, and each one's just a provisional reality for the purpose of the discourse at hand. And characterizing it in terms of an ontology of `things' that exist is just not informative or useful.
T: Well, ultimately. Right, it's certainly useful, if you care for all these pragmatic things. Obviously you think about this as a chair, or you'd have a lot of trouble ordering furniture.
M: In our normal provisional realities it is, but in...
T: And see, in some sense, I guess one of my fundamental intuitive differences here from I think many philosophers is, I believe there's more out there than is understandable by the human mind, I mean, it's sort of a humility of... There is truth out there, right, in some sense whatever is, is; it's not imagination. Now, the notion that the structure of that is something which we can capture by having our set of properties, objects, and so on, is making a certain assumption about the capacity of the kind of reasoning and the kind of language we use to sort of get at what's "really there." And that's where I'm just not convinced, right? I think we're getting a good approximation to certain stuff which is practical, and so on, and I think ultimately there may be things which are just, you know, it's a lot bigger than we are. See, there's this sort of feeling, which is an interesting one, that because we are able to be articulate and use logic and so on, we therefore have the mental tools, basically, to understand "the universe," fully. We may not have the details, you know, there's obviously all sorts of... we don't - but that's basically what it takes. And what I'm saying is I'm not sure I believe that. I'm perfectly willing to believe that our provisional reality is reasonably good, and we get along well, but in fact, what it really takes to understand it relates to how we think like how we think relates to an earthworm, right, it gets along too, right? But it doesn't have the particular tools we do. And there's no reason to think that ours, becuase somehow they're universal in a Turing sense, there's some sense in which we can claim they're universal, but I think that's a very weak sense, and therefore this notion that somehow because there is structure there, which I agree; it's not jello; that we should be able to carve up the jello with the kind of thinking we do, I think that's where I don't take the step.
M: So for instance, it could be that the laws of physics are really very complicated, and beyond that, they aren't even like laws as we know them.
T: It's not just complicated. If it's just complicated, then that's an empirical question, of how much time do we have, and how many scientists, and all that kind of stuff, but they're just different, in nature.
M: Or it could be, well, our laws for gravity and so forth have gotten more and more complicated, you know, Newton to Einstein, and quantum theories of gravity, and so forth, and maybe it's true that the actual physical law of gravity or some phenomenon would be infinitely long to describe precisely in our logical, discrete languages.
T: Or maybe there are laws of such totally different kinds that we can't imagine what they are, which govern relationships which we aren't even able to see, because we just haven't evolved to get to the point where we can see them. If you make the analogy, you see - there's an interesting jump, which is, if you look at the evolutionary chain, from bacteria up through dogs, or whatever it is, people have the sense, well, okay dogs clearly understand this world better than an earthworm, but it doesn't have ontology, but `we' have it. It's almost, it fits with the theological version, God created man in his image, it wasn't like we were just one more animal on the chain, the stuff which comes beyond us is as different from we as the things before us were, but rather somehow we've crossed the barrier from animal-like to god-like; we can really get the truth. When I think in sort of broad terms, it doesn't strike me as very plausible. I mean it strikes me as plausible that at every level, you would, whoever's there thinks they've got the truth. But it doesn't strike me as very plausible that somehow we've just crossed that boundary, and that's it.
M: So then, there could be, we could evolve later into having a different kind of mentality.
T: Or more likely, we'll all get wiped out and something else will evolve, and so on, but yeah. Or maybe they're already out there, in other galaxies, who knows. But the notion, sort of, our way of thinking is basically, it's got all the tools but just haven't put in the details yet. And I think that's where, in some sense, this notion that you can "really" seek the truth... this is why, emotionally, I'm not cut out to be a philosopher, because if you don't really believe that, then everything has this more relative view, this is how we think of it because of the kind of brains we have. That makes philosphers feel very nervous, they don't want it to depend on us; they want it to be the way... we're seeing the shadows in the cave, we want to know the real stuff there in the background.
M: And yet, they haven't done very well on actually agreeing on the truth...
T: Which, presumably, they see as simply not having thought hard enough yet. Everybody agrees that we haven't thought it all out right. But that you could, right, and in fact, maybe they're going to be the next great philosopher who does, right, that's the ambition.
M: But still, it seems like what Heidegger was doing was, he was being a philosopher, but he was denying the ability to do a lot of things that philosophers... So I think maybe there is some openness to this questioning of what we...
T: Yeah. Heidegger was questioning the very nature of questioning what is being.
M: It seems like there's a lot of philosophy that doesn't pay attention to Heidegger.
T: I think that's true, and it's because of the paradigm, the way he works. Basically, he is questioning something which is so fundamental to why they're even doing what they're doing, that it's not interesting, because if he's right, then why bother, and so let's assume he's not and continue.
M: Well, Heidegger apparently found a reason to continue doing philosophy after he had this insight...
T: Although, he did - I mean, it's interesting, because his later stuff is very different from analytic - I mean, not analytic because it's always different from analytical, but it's more almost religious, poetic, I mean, it generally has a very different... I mean, I haven't read a lot of it; I've been to various symposia at which Heidegger was discussed, so my view of Heidegger is the few things, the little window I have, plus secondary sources. But that in some sense, he really moved toward a more romantic, mystical, whatever the right word is...
M: It seems like you could believe what Heidegger does, and still do analytic kinds of thinking, but just think of it as a kind of game where you're setting up these systems, almost like matematics. If you take Heidegger's point of view then you could think of philosophy as a kind of matematical...
T: See, my view is, having said what I said about evoloution, what knowledge is, okay now, what is it we can make use of. What we can make use of, because we have the kinds of brains we do, is the kind of stuff we can figure out with our kinds of brains. And therefore it seems perfectly useful to worry about all these questions from that perspective, while recognizing that that's not getting necessarily the `truth' but the kind of provisional truth that's useful for the kind of animals we are.
M: But ultimately what we can get at with our mathematics and our philosophical systems is maybe not all that we could have gotten if we had different mentalities. Interesting. I think that's probably it. I might want to do some more philosophy papers sometime, because I think some of my best papers are philosophy papers.
T: Well, you clearly have a good sense of it. When you come in and say, well, here's what I think you would think, and I sit here nodding my head and say yeah, you're right, you're right, obviously you've understood it.
M: Unfortunately, I don't know quite as well what Dretske would think, because I'm more familiar with your point of view having taken your class, and so forth, than I am with the realist point of view. So Dretske might... if I were to tell him now what I think your response would be to what he says, then he might be able to come up with lots of things that I didn't anticipate, and kind of provide some arguments, and then I'd have to come back here and test my theories about how you might answer those.
T: I think most realists have some basic faith that the universe, or God, wouldn't have played a dirty trick like that on us. Right, that it feels like we know what we're doing, and it doesn't make sense to them - and of course, it probably has to do with how their parents raised them and how my parents raised me, right, I mean all these things end up - it doesn't make sense to them that it would seem to fit together so nicely logically to the degree it does, and that be an illusion. Right, I mean, `God doesn't play dice with the universe,' in that sense, a slightly different point, but... And I think it's just that basic faith that it couldn't be, it doesn't make sense to them to think that it could work that well without it being really the way it works, in some sense.
M: Yeah. I mean, there's still the question of how it all works, how we evolved these systems of reasoning that happened to do well in our world, and I think that's still an interesting question. How does the world come to be partitioned in our minds when it's kind of a continuous thing out there. It must be structural coupling.
T: Which, of course, is simply a license to look for things, not much of an - it's a type of explanation, not an explanation.