caro: Bryan Register is a doctoral student at the University of Texas. He joins us now to hold forth on Truth and Facts--what more could one ask of a philosopher? Fire when ready, Bryan.
bryan: Phil Coates recommended a quick summary, so here is one. The paper aims to present a correspondence theory of truth. In order to do that, we have to get clear on what correspondence is, and we have to get clear on what sorts of things truths correspond to so that they get to be true.
Section one of the paper tries to lay some groundwork. I reject some criticisms of correspondence theories. Most importantly, getting clear on why we have a right to do correspondence theory tells us a lot about the structure of truths. This matters because of an argument I develop in section two. But the general structure of a simple proposition of the 'some thing has some property' variety, I argue, is always a rather complicated multiply-quantified first-order logical statement, to the effect that the there is a bearer of certain properties regarded as essential to some entity (which entity is the referent of the subject term of the sentence), and that that bearer is also the bearer of some other property (or properties) not so regarded. So I accept Kripke's views about names, but also a rather extreme Russellianism which does away with names. I try to show why this isn't contradictory, as it might appear.
In the second section, I argue that correspondence theories of truth which say that propositions (beliefs, assertions, etc.) don't share structure with the things to which they correspond are not adequate, and that there's some intuitive appeal to the early-Wittgenstein notion that propositions do in fact 'mirror' the structure of the bits of the world to which they correspond if they are true. (The argument I give for this position seems to me rather weak. But it is only meant to apply to intentional states which are propositional in structure; so beliefs 'mirror', but concepts and perceptual experiences and so forth don't. So we're not stuck with a general representationalism.)
I then move on to present a theory of facts which is right in structure but wrong in content. This is a Russellian-Wittgensteinian hybrid which appeals to universals and bare particulars. Then I present a view, Rand's I think, which allows us to do without such entities. We can have no universals because we have particular tropes ('abstract particulars') instead, and we can do without bare particulars because there's no problem of individuation because we don't have universals to give us that problem.
After some criticism of Rand's and Peikoff's remarks on facts and truth - which are almost universally completely wrong - I present my own view. On that view, a fact is always that some bundle of tropes, which contain certain tropes essential to the existence of some entity (which entity just is that bundle, and is the referent of the subject term of the sentence) also has some other tropes which are not so regarded.
I finish off by situating the theory in relation to some other theories. I think that the present theory dodges subjectivism of a sort and intrinsicism of a sort. Since I've argued that tropes are similar to one another, and entities token-distinct from one another, only in the context of a knowing subject, I've implied that facts (which rely for their obtaining on type-identity/similarity of tropes and token-dinstinctness of entities) also exist only in the context of a knowing subject. So facts exist only in relation to a knowing mind. However, we don't get to just up and decide what the facts will be; we're committed to a certain regime of acts by the categorizations we bring to the world.
bryan: Actually, I don't recall having read Peikoff discuss the matter. But the point you give is an old one from Strawson that Russell successfully dealt with. But part of the point of Russell's theory is that a sentence with a name in it doesn't have to have successful reference by that name in order to be true or false.
bryan: for Frank. 1. not particularly, because they're not informed by any epistemology, because I'm not an epistemologist. 2. I accept the Duhem thesis. 3. yes? what other alternatives might there be?
bryan: So, to finish with Robert, the Peikoff objection begs the question.
bryan: Hey, Carolyn, am I confusing matters by answering questions that are still only in the other window?
caro: If you don't mind, Bryan, I'm going to start sending in questions, so that the discussion makes some sense. You should probably hide your other window now.
caro: From Robert Campbell: Quaestio prima: Bryan, are you familiar with Peikoff's stated
objections to Russell's theory of definite descriptions? One that I have
always found plausible goes as follows. For Russell, "The present king
of France is bald" is false (since such a sentence asserts that there is
a present king of France, and there is none). For Peikoff the same
sentence is neither true nor false, because it presupposes the
existence of a present king of France, and there is none. I'm
sympathetic to your wanting to avoid naming relationships in your
account of linguistic meaning, but is Russell's theory of definite
descriptions going to do the job?
bryan: Done; it's shrunk to a singularity. If any problem should arise, I can just open it back up again.
caro: bryan: Actually, I don't recall having read Peikoff discuss the matter. But the point you
give is an old one from Strawson that Russell successfully dealt with. But part of the
point of Russell's theory is that a sentence with a name in it doesn't have to have
successful reference by that name in order to be true or false. (Resent to preserve caro's sanity.)
bryan: Due to the miracle of time travel, I just answered that. Was the answer okay?
caro: Questions for Bryan: 1. To what extent are your views informed
by evolutionary epistemology? 2. Can you comment on the Duhem
thesis, namely that sometimes we revise facts and sometimes our
theories? 3. Is truth first of all a semantic attribute of propositions? I
commend to all Paul A. Boghossian, "What Is Social Construction?" in
the February 23 _Times Literary Supplement_.
caro: Does that satisfy you, Robert?
caro: bryan: for Frank. 1. not particularly, because they're not informed by any epistemology,
because I'm not an epistemologist. 2. I accept the Duhem thesis. 3. yes? what other
alternatives might there be?
caro: Enlightenment: We Aim To Amuse.
bryan: No, Carolyn. 'Confuse'. 'We Aim to Confuse'. And follow up, if need be.
caro: Excuse, please. My comments seem to be randomly redistributed according to unknown principles. Hang on.
caro: Testing caro2
caro: OK, something miraculous _is_ happening now.
caro: I'm going to have to log out and return.
bryan: I am a Zen master. I patiently await the return of your absence. 8^)
caro: OK, better.
caro: I hope no new-agers are watching this. I'm going to get swamped with phone calls and emails requesting nirvana.
caro: Peikoff likes to leave a lot of his stuff in the oral tradition--so his reference to "The
present king of France is bald" comes in one of his lectures on modern philosophy, if
memory serves. The problem with Russell's theory that this example points to has not yet
been addressed--Russell's theory requires a sentence like "The present king of France is
bald" to be true or false. It doesn't just permit such a sentence to be true or false.
bryan: Well, *you're* the one who decided to call yourself 'Enlightenment' (!) after all.
caro: Yes, I sent it. He appears to be brooding now.
bryan: I don't know what the problem is. Russell says that these sentences are true or false, and Peikoff says that because they're not, Russell's theory is wrong. But what's Peikoff's reason for saying that 'The present King of France is bald' isn't false?
Better, let's define falsity. A sentence is false iff its meaning is a fact which fails to obtain. The fact asserted by the sentence in question fails to obtain: it's not the case that the present King of France is bald. There's all sorts of ways this might happen, and only one of them is that the present King of France isn't bald. So it seems that the sentence is false. Now, we have this intuition that something uncommon has gone wrong with this sentence, that it's sort of super-false; much more false, say, than 'The present Moderator of this discussion is bald'. But this super-falsity, on a Russellian analysis, can be preserved: a sentence is 'super-false' iff if the reason it is false is that the subject term fails to refer. Thus are our intuitions captured, and also the intuitions to which Russell appeals in putting forward his theory (such as that George IV isn't concerned with the law of identity and so forth). I'm not brooding, it's just early and I can't type fast yet.
caro: Sorry about this. My private comments keep getting publicized. BRYAN: Did you see Robert's further question?
bryan: Yep. I don't know whether I will satisfy him any better this time.
caro: Robert, are you still with us? If so, do you have a follow up?
caro: Quaestio secunda: Bryan, at the end of Section III, you say "I shall not attempt to say
how, in general, concepts are representations of the world. But they do represent sets of
objects or properties." Isn't it important to know how concepts represent, if you are going to
give satisfactory answers to many of the other questions you raise in your paper? For
instance, whether something like Wittgenstein's picture theory could be true depends on
how representing is done. The claim that if you believe that subject and object are internally
related, you must be an "exteranlist" also depends on assumptions about how representing
is done--specifically assumptions about mental states and their content.
bryan: Well, I won't attempt to say here and now. In fact, it seems to me that something like this is the case. In my JARS paper I proposed a capacity model of concepts, in which a concept is a capacity to recognize or perform something. In general, our recognitions of things as belonging to a certain type are beliefs, and the mental states we have in virtue of the having of which we perform actions are intentions. Both intentions and beliefs, it seems, are propositional in structure. So you actually have to get clear on how propositions work before getting clear on how concepts work, because what concepts do is make certain propositionally structured intentional states happen, and we have to know about those states before we can figure out what concepts must be so that they can make it so that we can have these states.
bryan: This amounts, BTW, to instrumentalizing the concept *concept*.
caro: Is it possible that you have to sort of work back and forth between understanding how propositions work, and how concepts work, to get an adequate account?
bryan: A word about all of my intuition-mongering. One human capacity which cannot be called into question is our ability to use our own language. So when O'ists complain about analytic philosophers constantly appealing to intuitions, they have a point only depending on what the analytic philosopher is studying. For instance, our moral intuitions could be wrong, because it's in question whether we are actually being moral. But it can't be in question whether we're actually using our language correctly (as a rule; of course particular misuses occur). So our intuitions about our language can't be wrong very much of the time. It's literally incoherent to say that someone is fundamentally wrong about how (their) language works, unless you're saying that the someone actually doesn't speak the language. So there's a quick defense of something which might be controversial.
bryan: to Caro: yes, I think so. The present theory has a bunch of tacit stuff about concepts in the background, like that as a rule individual terms (especially predicate terms) tend to 'go with' individual concepts. But I think that we can understand what concepts are, if they're capacities, only once we understand the things that they're the capacities to have. Concepts become a kind of theoretical entity designed to explain certain appearances, like that we speak in sentences and so forth.
caro: From Frank: What are facts as far as animals are concerned? And what things are facts for people
but not for animals? What does the literature you have surveyed say here? I think such
evolutionary issues make for good premise checking if for nothing else!
bryan: The only thing I know about animal beliefs is that Donald Davidson doesn't think that there are any. However, I'll hazard this. It doesn't seem to me that we are all that radically different from a rather large chunk of the animal population (say, a whole lot of mammals at least) with respect to having concepts (as opposed to some other kind of quasi-mental state which explains lower animal behavior). So my cats, it seems to me, have concepts, though they categorize things rather differently than I and will thence live in a world which has rather different facts in it. (This is not to say that their facts and mine contradict; check the end of the paper for why that can't happen.) The world of animal facts will no doubt be substantially impoverished with respect to ours. -- Oh, *that's* the kind of evolution you had in mind. I thought you were asking about Popper earlier. Same answer, though, because I'm still not doing epistemology. This is ontology: philosophy of mind and action.
bryan: Tell me, did anyone other than Robert, Frank, and Phil read the entire paper?
caro: Thanks, Bryan.
bryan: For? Sorry?
caro: From Robert: Yes, but is our overall organization of mental processes divisible into propositionally
structured mental states in the first place? Propositions are a logical notion; sentences are
a linguistic notion. Each describes some things we can do--but does either describe how we
do them? Another way of putting this would be to ask you to get more specific about your
conception of pragmatism. You mainly tell us that this is not Rorty's pragmatism--but I don't
see much about how you think mental representation is done under an objective type of
pragmatism, as opposed to a subjective one. I am sympathetic to many of the things you say
in your JARS paper, but I see what looks like a confusion between actually carrying out some
method (for instance, reasoning in a way that conforms to rules of logic) and having a
concept of that method. I've always understood Rand's notion of the "concept of logic" in
terms of having a concept about using the method, not in terms of the method or skill itself.
There is no account of skill in Rand's epistemology.
bryan: You're right, there is no such account. But let's try this. Does someone know how to play chess in virtue of being able to rattle off the rules of the game? No, we know how to play chess *even if we can't say what the rules are* if we can move the pieces right. So Rand's account seems over-linguisticized. - About pragmatism: I'm not a pragmatist, but it seems that this version of O'ist theory skirts rather close. I guess here's the difference. Pragmatism says that something is true iff it's handy for us to believe it. That's wrong; something could be true and useless. (The Battle of Hastings happened in 1066. Whee, let's go straight to praxis.) But O'ism seems to suggest that a concept is a good one (I don't know what the word replacing 'good' should be, but it's not 'true' or 'veridical' and I refuse to use 'valid') just in case it's handy for us to categorize in terms of it. So O'ism is pragmatism about concepts, but not about truth.
bryan: More: well, our 'folk psychology' is divisible into propositionally structured mental entities, as the man in the street will tell you. Neither propositions nor sentences explain how we have propositions or sentences, but we have to know the traits of these in order to know how we have them. Is that clear? Did I not understand?
bryan: (Carolyn: Oh, I see. I took for granted that *you* had read it. I meant among those who didn't *have* to read it.)
caro: Bryan, I take your comment about how we know to play chess as a call for an
epistemology of skill--not (necessarily) for analyzing skills in terms of concepts.
There are forms of pragmatism that do not revolve around a pragmatic theory of truth
(non-James and Dewey forms, I guess we could say). How about Peircean pragmatism, for
instance? Or James Gibson's notion that perception frequently detects affordances
(action-relevant properties of things in the environment, like squeezeableness or
caro: (Last question from Robert)
bryan: Well, actually I did have concepts in mind. Here's the reason. We have various propositional states. Some of these are beliefs, and we can express these beliefs in the form of sentences. Now, it seems that our notion of *concept* is going to play a role in our understanding of sentences, because our understanding of words is tightly connected with our notion of *concept* (though not so much, as I've argued, the other way round), and sentences are made of words. Now, it looks like the relationship between a sentence and a belief (the expression relation) is strikingly like the relation between many words and certain concepts; we like to say that, in the context of a sentence, we use 'red' to express the concept *red*. Now, assume that a compositionality thesis holds for both the sentence (sentences are composed of words) and likewise for beliefs: beliefs are composed of concepts. In that case, coming up with an explanation of our assertions will involve the notion of a proposition, which in turn involves the notion of a concept. A lot of that is just reasoning by analogy, but it strikes a chord. - Second step is where we suggest that concepts will play a similar role for any propositional attitude state, not just beliefs. So intentions will be understood in terms of concepts, and intentions are the folk psychological entity which explain actions. So it seems that the ontology of skill will require analysis in terms of concepts (The epistemology would then help explain, I guess how we acquire those concepts or something.) more...
bryan: I couldn't say anything about these forms of pragmatism: my own history of pragmatism consists exclusively of James, Quine, and Rorty. So take nothing I say to have any relevance to the other guys you're talking about, pro or con.
caro: From Jamie: I am a little confused about the ontology of tropes and bundles and I am hoping that you
might clear something up for me. In Rand’s direct realism, _what_ are we directly aware of?
Is it a property in the bundle or the bundle itself? I sometimes read her to be saying that
there is an intrinsic “this” that we grasp in order to be certain that it is a (unified) object. Is
that right? Or is ‘being a this’ a property?
caro: More from Jamie: In other words, I am having trouble seeing how we can have direct realism without
some sort of bare particular (or this).
bryan: for jamie: I'm glad you asked. It seems to me that in Rand's (and, I think more importantly, Kelley's) direct realism, we're directly aware of entities. That is to say, we're aware of bundles of tropes. An entity is identical to a certain bundle of tropes, in the same sense that Carolyn = Dr. Ray. If Rand thinks that there's a thisness, then she's wrong by her own lights, because thisness would be a bare particular and she tacitly rejects those when she lets Prof. ? heap scorn on prime matter. Now, if the thisness is a property of things, then it's the property of being token-distinct from other things, and token-distinctness exists only in relation to a knowing consciousness which perceives (at least roughly) the way we do. (And I mean *perceive*.) So there could be such a property; it would be the property of having been distinguished by a perceiver.
bryan: Direct realism doesn't, I hope, commit us to *bare* particulars. *Bare* particulars would be something in the entity which is distinct from any property the thing has, and which has all of them. I've certainly never directly perceived any such thing, and it would be hard to do, say, a scientific study of our perceptual interaction with propertyless inert ontological posits. Now surely there are concerte particulars: they're bundles of properties. And direct realism of Kelley's sort does commit us to *their* existence. more...
bryan: But it's not clear that it makes any commitment about the ontological status of the distinctness of such entities. Does Kelley's direct realism commit us to the idea that entities are *intrinsically* distinct? I hope not. It seems to me that our perceptual mechanism operates to distinguish entities from their background, but while the distinguishing is going to have a law-like connection to certain tropes which entities exemplify, that doesn't mean that the distinguishing is ontologically passive. It might be that we cause things to be distinct entities, and that we do this because they have certain tropes and we need to distinguish things with those tropes. So it's a queer sort of realism I've got really: take us away, and you change nothing, except that you get rid of all the entities and facts! Only the idea is that being-an-entity and facts are objective, not intrinsic: they exist in the world only in virtue of an interaction between us and the world. (And if you like: this is watered-down genericized Kant.)
bryan: BTW, these are good questions, everybody, thanks. I appreciate the attention having been paid to my unambitious, brief little paper. 8^) - I noticed yesterday about ten grammatical errors which obscured meanings... gross. Sorry about those.
caro: This sounds like a Ray-Radcliffe thesis, too. And in fact, I think Kelley's direct realism does commit him to the idea that entities are *intrinsically* distinct. It sounds like your realism is just like mine. That's a surprise!
caro: So bundles do not mind-independently exist. _We_ bundle properties, properties are a
causal-interaction between us and the outside world, and we label properties by tropes
terms. Is that right?
caro: (that last was from Jamie)
bryan: Well, not exactly a surprise. I was initially enthusiastic about your paper because it seemed to say what I'm saying here. That's why I cite your paper in mine, as saying maybe the same thing. I suspect that if we were to sit down and talk a long time, we'd end up having only semantic differences of a particular gross sort, but the differences really make your paper hard for me. I think also that you and Tom might not have recognized how weird this view really is, and so a lot of your language doesn't really suggest the exotic, whereas I hope that my own discussion does admit that the view is weird. I have in mind the paragraph where I say that it looks like I got rid of the tropes by making their differences from one another mind-dependent. I have a strong intuition that you can't really do that, and I just suppress it. more...
bryan: Jamie: Bundles mind-independently exist, but not qua bundles. All of the tropes constituting the bundle exist independently, but they're not a bundle without us. So yes, I guess we bundle properties. But *properties* aren't a causal interaction between us and the outside world, unless you mean something specific. All but a few relational tropes are mind-independent: they're the intrinsic. If by properties, you mean categories of bundles (what the realist calls a property), then yes, properties are mind-dependent. If you mean individual tropes, no, I don't think so. We label tropes by property terms; or better, the referent of a property term is all of the tropes which bear a similarity relation to certain tropes which are paradigms of that property to the speaker. Better?
bryan: That was to, not from, Jamie.
bryan: Here's the broad point. The view is Kantian. Facts, entities (qua entities), and similarity and distinctness relations are phenomenal; they exist only in virtue of an interaction between the mind and the noumenal world. The rest of the tropes are noumenal. Noumena and phenomena are *token-identical*: we look *directly* at the noumenal world, and by interacting with it, give rise to the phenomenal world, which **is** the noumenal world (as apprehended by us). (And the stresses in that sentence are key! It's not the the phenomenal world *isn't* the noumenal world, because it's not real because we made it. It *is* the noumenal [though it's no longer noumenal - you see the difference?], and it *is* real, and this despite the fact that it is there only because we interact with the noumenal world.)
caro: From Jamie: I am not familiar with Trope theory, so I am sure there is a subtlety that I am not
getting. When you say that “All of the tropes constituting the bundle exist independently”, I
am not sure how this is different from the realist approach.
bryan: About what do you think I'm being a realist? Universals or direct perception or something else?
caro: I am not sure what properties are (in your terms). I said "causal-interaction" because I
am think of colours: The property of 'red' is a causual-interaction with the outside world
(photons and electron shells) and our perceptions (which gives us the phenomenal
experience of RED).
caro: (that would be from Jamie, of course)
bryan: Ignore that there's a difference between the scientifically understood features of a thing and the phenomenal experience of it. You can ignore that, because the content of the phenomenal experience is the thing. So the causal interaction between us and the thing which gives rise to phenomenal experience isn't really relevant. Things really are red, independently of us, even though of course they don't sit out there looking red on their own. Being red implies: that you'll look red to a person whose brain is constituted in a certain way. But being red is 'really' having a certain feature known to scientists and not to me. Does that help clarify? That's the sense in which tropes exist independently of us. more...
bryan: I don't think I use the word 'property' a lot. I'm concerned with tropes. Take an intuitive use of the word 'property'. My white cup has a certain property, whiteness. And my white screen seems to have the same property. The realist about universals says that this is literally correct: there's a property, and it's possessed by two things. Nominalistic trope theory says no, it's sort of a figure of speech. There's not literally some feature of my cup which is also a feature of my computer. There's a feature of my cup which is strikingly similar to a feature of my computer. Each of these features is a trope. Better?
caro: From Phil: We seem to have a lot of terminological questions in this discussion--what is meant by
tropes, properties, etc. Some terms might be non-ambiguously defined by the original
philosophers. But this is a paper which could benefit from defining a lot more terms up front,
rather than assuming we all agree on usage.
bryan: I agree with Phil's complaint, though I think that trope theory is discussed in the paper. Uh, two hours are up. I don't have to rush away, but I do need to get lunch before the next session. Can I recommend that those who have specifically terminological questions send them to me by email (bregister_OF_mail.utexas.edu) and I'll compose answers during the next session and put them in the other room after that session has wound down? Maybe that can clarify enough to get to the heart of things in a follow-up or later on today.
bryan: BTW, Caro, I'm not running off. Just let me know when the threads naturally stop.
caro: It looks like there are no new questions on the table, so I'll end the session here. Eat, Bryan, eat! Of course, I think most papers would benefit from the inclusion of a glossary, so we'll be glad to get your terms in email, but I'll be especially enthused if you include them in your next draft of the paper.
caro: One more from Jamie, if you're still there: I am a little wary of ignoring the difference between the scientifically understood
features and the phenomenal experience... as I thought we were trying to figure out what is
mind-dependent and what is mind-independent. Should we not do that for the perceptual
level as well as the conceptual level?
bryan: Well, yeah. There's that. However, we can distinguish between the way something looks to me, which is surely an internal subjective state if there are any such things, and the content of the feel, which is, I think, an external objective thing. But we say that things, not experiences, are, say, red, so it looks like we should identify redness with the property of the things which physicists know about. Does that satisfy?
bryan: I'm not going away until Carolyn tells me definitely to do so, but since she's probably about to, I'll see y'all again in an hour and promise to post answers after the next session when some proper authority figure tells me that the other one has stopped. 8^)
caro: OK, then, Bryan. Close your session and log back in when we start with Ted, so he can be the speaker. Thank you very much for coming, and for inspiring the conference! It was a great idea and I think it's gone really well!