Monday, July 31, 2006

Clarity in writing

Richard Rorty once criticized analytic philosophers for dismissing writing that was unclear as being a product of confusion or subterfuge. He went on to say that he had never seen an argument for the conclusion that clarity entails or promotes philosophical depth or insight. After giving it some thought, I don't think I've seen an argument for that either. There are a couple of reasons for this. It is a metaphilosophical point about which not that many people write, as far as I can tell. It is taken as a presupposition by a lot of people that clarity aids or entails insight. Why should this be though? If I have written a clear paper with a valid argument that is easy to follow, then it might be easier to read. This is not enough for insight, depth, or progress. The content of the argument needs to be interesting and lead to interesting places. Is there any reason to think that an argument laid out in a straightforward deductive style with all the premises explicit will facilitate interest? It is plausible that an argument laid out with some premises implicit, not necessarily with the conclusion at the end, could facilitate interest. If the argument reflects the thought processes of the author, it is quite possible that a lot of the argument had to be constructed after the fact to bridge dialectical gaps that emerged in the course of writing. This could all be done with a fair amount of clarity though. Why think that someone who is less than crystal clear, e.g. Heidegger, is not as deep as someone who is fairly clear, e.g. Quine (at times)? Suppose we take as our model of philosophical argumentation first-order logic, or some rigorous proof theoretic system. These are models of clarity; there is nothing hidden in their notation. If these are our models, then anything that requires close textual interpretation is already deviating from them. Those deviant texts have hidden premises and implicit assumptions. Switching gears some, we could take as our model scientific papers. I'm less familiar with scientific journals than with math and logic journals. Assuming they are fairly similar to computer science journals, then they will be fairly clear. The occassional metaphor might be used to illustrate a point, but not much else. There won't be any exigetical heavy lifting required. Deep results in science can be obtained without appealing to unclarity. However, some science writing is bad and is a bit hard to follow. This might undermine the point a little, but such papers still don't require the expertise of textual interpretation that Kant or Hegel do.

Here's another stab at it. If someone has written something that is difficult to follow and somewhat unclear, then we think that if they understand it well and there is some substance to it, they will be able to reformulate it in a form that we can better follow. If this is our implicit principle, then it says something about us too. The author should be able to reformulate her idea in such a way that the reader can more easily follow her thoughts. That's what it says about us. Following this principle as an author makes things easier on the reader, but again, it doesn't seem to have anything to do with depth of thought. Maybe it belies a bias towards extensionalism. If arguments/books/papers have a content which is extensional, then that content can be specified in different ways that are all extensionally equivalent. In the case of unclear writing, the author has selected one way whose presentation makes things difficult for us. She should be able to reformulate it in an extensionally equivalent, more accessible form. If this bias is dropped, arguments/books/papers could be said to intensional in that their content depends on its presentation in some important way. This is likely to be the case when an idea is put in an unusual way in order to draw attention to it and make the reader abandon some particular connection with that idea or presupposition. Possibly, once the reader has drawn out the moral she may kick away the intensional ladder and keep the extensional content, but that would only be once she's figured it out. The form of the presentation was crucial to reaching the insight. Some writers, like Wittgenstein, might use unusual forms or presentations to drive home this point.

I guess my tentative conclusions at this point are: (i) that the presupposition that clarity facilitates depth/insight is probably derived from taking science/math/logic as a model of inquiry, (ii) clarity is more an aid to the reader than the writer, and (iii) the presupposition might be false in some cases, e.g. Wittgenstein.

Why represent historical figures accurately?

Here are some thoughts I had about the history of philosophy. Why should one be concerned with accurately representing the views of historical figures to whom you are attributing? There are a few reasons that I can think of.

If you are criticizing their views, then it makes the most sense to present their views in the best light possible. The strongest formulation of their view will make the best target since it will allow you to attack the most important parts of their position rather than distractions that result from poor formulation. Dispensing with strawmen doesn't add anything to the conversation.

If you are presenting your view as an addition to an older view, then you want to remain true to the original so that you can piggyback on arguments and conclusions. If the historical figure presented an argument for some conclusion that you want to add on to, then changing the conclusion to fit your particular whim may invalidate the argument. This would defeat part of the purpose of attributing to the historical figure. If you want to use the sorts of arguments they used, in the rough forms they used them, then the substance of the attributed views needs to be close to the views they can reasonably be said to have held.

If you are presenting your views as being in the same tradition as another philosopher, then there seems to be less reason for making sure that you are attributing views to them that are exigetically accurate. You don't want to stray to far into left field with the attributions otherwise it will appear that you are (i) completely clueless or (ii) dishonest. Claiming a tradition like this requires a decent amount of honesty. At the very least, it should be clear how a reasonable person undertaking a reasonable reading could arrive at your interpretation. I'm assuming that part of tradition claiming is telling a cohesive historical narrative, which means that going through the steps of reading the historical authors could result in arriving at the end of the story you've told, with proper understanding and hindsight.

Friday, July 28, 2006

Intentions and meaning

One of the surprising things on the "In Conversation" tapes was Davidson's assessment of Grice's work. I think it was in discussion with McDowell or Dummett that Davidson said it. He thinks that Grice's project of analyzing implicatures in terms of intentions won't work for two reasons. One is that meaning-that is an intentional relation just like believing-that. The other reason is because he thinks that in any single utterance there are going to be many many intentions present and active and he does not think it is possible to single out any one of these as the source of the implicature or speaker meaning. He also said that he liked Grice's work in connecting meaning, action, and intention and showing how these notions are related. He didn't give any more argument for his position in the brief discussion than that though. I would have liked to hear him explain the multiple intention objection to Grice. Supposing that I do have many intentions involved with any particular utterance, why couldn't there be a single communicative intention that bears the burden of being recognized as an attempt to convey some content? I think this is probably connected to Davidson's holism and interpretation. In order to interpret someone and assign them certain intentions, I have to know a lot about them. In doing this I am already interpreting them as having many beliefs about different things and meaning certain things by their words. I guess the objection to the Gricean picture is that there are many intentions that will attributed to a speaker in interpreting them, and there are many intentions that could have been attributed to them just as well. Since there is no difference in the interpretation to recommend one over the other intention, there is nothing more to the story. This can't be right though. From the first person point of view (sounds Searlean), one knows more or less what one intends. There seems like there should be a fact of the matter about intentions. I have them or I don't, regardless of how you decipher them in interpretation. There has to be more to Davidson's objection than what I've given, although I think that would require digging into some of his papers to pull out.

Platonic question: appreciation (II)

Returning to my recent platonic question, I think I have an idea about what appreciation is. It is (i) being able to explore something to discover the subtleties and ways those details come together combined with (ii) actually using those abilities to explore something. To mangle a phrase from Kant, discernment without intake is empty, intake without discernment is blind. This captures the similarities between viewing art, listening to music, drinking a good cup of coffee, and watching an episode of the Simpsons. It leaves out the appreciation of air conditioning on a hot day, but I think there is a case to be made that that is a different, though related, concept. What is invovled in that is, maybe, more a recognition of the increase in comfort or pleasure. There doesn't seem to be any sort of subtleties or piecing together involved.

Thursday, July 27, 2006

Metaphorical commitments

I like the story that Brandom put forward in Making It Explicit and his Locke lectures. The focus on commitment and entitlement of assertions in a social practice seems like a good direction to go in. The constant focus on the truth of an assertion and cashing things out in terms of truth-values or truth conditions, while useful, seems to miss something, namely a lot of the interaction between people. That being said, the story isn't perfect.

One example is that it isn't really clear what I'm committing myself to when I make an assertion containing a metaphor. Suppose I assert, "You are my sunshine." I'm not committing myself to the claim that you are made of photons. I think all of the examples in the book and lectures were of literal, fairly straightfoward sentences. There is nothing wrong with startint there, but there are several things that still need accounting for. Metaphor is one of them. I think I am committing myself to the claim that you are important to me, but we need some story to get us from my usage of that to that particular claim. I'm not familiar with the metaphor literature, so I'm not sure what the canonical ways of dealing with metaphor are. The one I'm familiar with is a quasi-Gricean way that I don't think applies here since Brandom reverses the order of explanation that Grice requires. Brandom is starting with the pragmatic force and then gets to the semantic content while Grice went the other direction. A more interesting question to me is when am I entitled to make an utterance with a metaphor instead of a literal utterance that expresses roughly the same content? Are the entitlement conditions for "you are my sunshine" and "you are very important to me" different or is this just a matter of taste? There aren't going to be any extra epistemological conditions placed on the entitlement to one over the other and I don't think there are any particular collateral beliefs that would support using one rather than the other. An exception might be a belief about whether you think your addressee has a preference for poetic flattery, but could that be the only difference?

An overview of meaning and belief

Here's a nice way of describing the lay of the land of meaning and belief. Atomists like Fodor think that changes in beliefs do not affect concepts or the meaning of words. Holists like Quine and Davidson think that there is no way to tell the difference between a change in belief and a change in concepts or the meaning of words. Holists (extreme) like Brandom think that changes in beliefs are changes in meaning.

Philosophy and science: different in degree only?

Quine said that philosohy was on a continuum with science. He also said, "Philosophy of science is philosophy enough." I'm not sure if I've heard a good explanation of the way(s) in which philosophy is different only in degree and not in kind from scientific practice in general. Here are a couple of stabs at what that could mean. First, he could have meant that there is no first philosophy above science and all philosophy should come from science. This might not be too far off. Another way of taking it is that philosophy should be commentary on scientific results and the practice of scientists. This would explain the philosophy of science quip. A third way of taking it is that science is the regimenting of common sense practices of testing and forming beliefs to make sense of the evidence and philosophy is also a method of systematizing common sense and bringing it into reflective equilibrium. This would be a kind of David Lewis idea of philosophy. I don't think this is exactly what Quine had in mind, but it meshes well with his idea.

Wednesday, July 26, 2006

Davidson and rationality

I discovered that the library has a full set of the videos of "In Conversation: Donald Davidson" which consists of 19 tapes of one-on-one and panel discussions with Davidson and other philosophers. I've now watched almost all the ones that I want to watch. A few things struck me which I will comment on over the course of a few days. First was who Davidson talked about and who he didn't. There was a lot of talk of Quine, Wittgenstein and Sellars. The first two didn't surprise me but the last one did. I didn't get any Sellars as an undergrad but I'm coming to appreciate his work. The person that he didn't mention at all that surprised me was Kripke, apart from saying that he preferred Tarski's truth theory to Kripke's because it is more elegant by his lights and a brief mention of the rule-following puzzle. Maybe this just illustrates how overstated the importance of the development of rigid designators is.

The discussion with McDowell focused in large part on the importance of rationality, or the constitutive nature of rationality. McDowell asked whether we should retrospectively read this idea into the early pieces on truth and interpretation since it wasn't expressed until the coherence theory of knowledge paper. (I'll have to doublecheck that.) Davidson said it should be. He went on to say that the principle of charity was basically one way of stating the constitutive nature of rationality. Interpretation as a whole is trying to take what people say and organize them in a systematic fashion so that they ellicit a certain pattern. This pattern is the sort of inferntial links between various sentences that is distinctive of a rational mind. This, he goes on to say, is how we get to know other minds; we look for things that fall into this pattern in what other beings say and do. It also connects up with the paper on conceptual schemes. Davidson denies that there could be conceptual schemes that are uninterpretable by us yet still be thoughts becasue in order for something to count as a conceptual scheme, it has to have a certain structure. McDowell describes it as a way of telling what is a reason for what, which is a reformulation of the constitutive idea of rationality. If something has a way of telling what is a reason for what that we can understand as near enough to our own (providing some wiggle room) then we can attribute that thing rationality and thought. If we can't interpret it in that way, then it doesn't count as thinking. Why not? Because the concept of thought is our concept and so it is linked to our conceptual scheme. Something can't count as thinking unless we can count it as thinking because we have no way to get outside our conceptual scheme to check, so to speak, that it has some alien conception of thought. This seems like a kind of conceptual perspectivalism, that is, our concepts are what we are stuck with and it is not possible to grasp concepts which are in principle unintelligible to us. Maybe "perspectivialism" isn't quite the best word for it, but that was the thought that struck me when I heard it.

The discussion also cleared up the article on reference that I had trouble making sense of before. Davidson said that the truth theory sets up certain demands that require us to break sentences into words. We have to assign referents to these words, thereby setting up a relation of reference and satisfaction. These are governed by the demand that they ultimately allow for the attribution of rationality if interpretation is successful. We have to come up with a relation of reference, but it is not unique. There could be multiple such relations as long as each one preserves the structure of rationality and supports the truth theory. The relation of satisfaction involved then mixes objective parts (from the word-world connections) with normative parts (from the demands of rationality). That makes the article seem much more sensible.

Monday, July 24, 2006

Thoughts on language

There are at least two theories as to the main evolutionary benefit of language. (This is not the best way of putting it, but I think it captures how I remember each side.) One side thinks that the big advantage of language is communication. With language you can discuss things with others and exchange contentful ideas. The other side thinks that language has nothing to do essentially with communication and it is rather mental organization that is the main benefit. Language use adds a large amount of structure to the mind and it is what enables us to have the abstract thoughts that we do and make long-term plans involving various contingent and hypothetical possibilities. The communication stuff is a happy accident of this. The latter position is defended by Chomsky and his followers. The prior position is defended by pretty much everyone else I think. There is a prima facie harmony between the views (as indicated by my comment following the second description). One view says language is for exchanging contentful messages and the other view says it is for having contentful thought. This meshes with the idea that language is used for putting into a public medium the content of one's private thoughts. (That probably sounds quite naive and would not make partisans of either side happy.) I started thinking about whether there are any other possible pressures for the development of language or whether these two options exhaust the range of viable options. My initial instinct is that there should be more possibilities although I haven't come up with them.

I had more thoughts about how to approach language use that may seem naive. One thread in the philosophy of language sees language, particularly semantics, as a logical phenomenon. Logic is the primary tool for investigating and understanding language. As Richard Montague put it, syntax, semantics and pragmatics (!!) are all mathematical phenomena. Another way of thinking about it is as a completely naturalistic phenomenon that arose in a species that developed under evolutionary pressure. Pressures that resulted in language development should have resulted in various processing limitations as well in order to be adaptationally useful. Continuing this idea: tools that are useful for understanding evolutionary development should be useful for understanding restrictions on languages, in particular, (getting into uncertain territory here) information theory, noisy-channel models, and models of redundancy. Thinking about it, it is kind of amazing that evolution ended up giving us logic, to put it one way.

One other thing that I feel gets left out of the logical models of language is the fact that while languages might have a logical basis of some kind, they also have a historical basis. They are entities (for the moment begging the question about the existence of public languages) that have developed in haphazard ways in response to various contingent, historical developments that could have easily gone other ways. There are words and phrases whose meanings have shifted over time and parts of languages whose syntax has changed dramatically. These elements are all thrown together, at least in languages that have live modern versions. These considerations make the idea of a single, synchronically pure language an idealization. That is to say, I'm not particularly surprised that we don't have grammars for any natural language that generate all and only the sentences of that language.

Platonic question: appreciation

Here's a platonic question: what is appreciation? Is it recognition of the difficulty it took to produce something, as in a musical piece? This doesn't seem right because you can appreciate the flavor of food. Is it being able to recognize the subtleties of something? This leaves out appreciating something like air conditioning on a hot day. Could that last one be a different concept? Is appreciation a family resemblance concept? The dictionary glosses it as: "the recognition and enjoyment of the good qualities of something." Another gloss in the same dictionary is: "a sensitive understanding of the aesthetic value of something." Another gloss is: "gratitude for something." This lends some support to the family resemblance idea. I'm somewhat hesitant to say that my appreciation of a good view is the same sort of thing as my appreciation of a good cup of coffee, although that might be similiar to the apprecation of a good sumo match in that both took some work to acquire.

Saturday, July 22, 2006

Worries about relevance theory

The relevance theorists say that meaning is what is disclosed (transmitted?) via one's communicative intention (or was it informative intention?) in communication. They sound like they are saying that recognition of the communicative intention is necessary and sufficient for communication, i.e. getting meaning across. They go on to say that language is just one effective means of aiding the recognition of the communicative intention. I have a couple of thoughts about this.

First, if meaning is linked only to the communicative intention, then I am guessing that they deny that there is such a thing as sentence meaning. As Stephen Neale put it (I'm not completely sure it is fair to put him in the same camp, but the sentiment is the same), "Sentences don't mean; people mean things." That isn't a direct quote. It is a paraphrase from his paper "On Location" coming out in the volume Situated Semantics. If sentences, don't mean things, and sentences are compositionally built up from the meanings of words, then words don't mean things either. This has the odd result that lexical semantics should not be possible since that field studies the meanings of individual words. I'm not sure what the relevance theorist's response to this would be. Maybe she would appeal to the conventionalized use of words to signal certain sorts of intentions.

Here is another worry about tying meaning to intentions in this way, i.e. seeming to bypass words altogether. Suppose you have an unusual or underdeveloped theory of mind, e.g. you are autistic (fairly sure this is accurate). If understanding and recognizing intentions is linked to your theory of mind, then we would expect people with an underdeveloped or unusual theory of mind to have a great deal of trouble catching implicatures. This means that we should expect, e.g., autistic people to have a great deal of trouble with implicatures. I don't know if they do have trouble with that. It would be worth checking out. This might not even present any problems for the relevance theorist if they don't have trouble with implicatures. I'd expect a response along the lines of: their theory of mind may be deficient, but it is not deficient in the way that impairs the recognition of intentions and intentionality. Getting more into this will require reading up on some psychology I expect.

Friday, July 21, 2006

The meaning of "oops" and "ouch"

In Kaplan's (currently unpublished) "The Meaning of 'Oops' and 'Ouch'," he discusses the meaning (surprise!) of expressive terms like "goodbye" and "ouch." One of the conclusions he comes to is that the metalanguage for a language L containing expressive terms does not need to be translational. Like the metalanguage for a language containing indexicals, the metalanguage M for L does not need to contain expressive terms to translate the expressives of L. He thinks we can give the semantic content of expressives in L by using purely descriptive terms in M. He brings this out by saying that M doesn't need to contain honorificsc to explain the meaning of honorifics. We can't use expressives in M, but we can describe what they do. There is a surprising amount of resonance here with what Brandom says in his Locke lectures. Brandom says that you can have a vocabulary V that allows you to say what you are doing in a certain practice P, even though the practice of deploying V cannot do what is done in P. To rephrase it, you can have a metalanguage that can describe the use in the language without containing some of the described phenomena, i.e. expressives. Brandom points out the same point in the case of indexicals in the appendix to his second lecture. He brings it out by contrasting what he calls the Kaplan-Stalnaker semantics with the Perry-Anscombe pragmatics. I'll have to go back through the two papers to see what other points of contact there are.

Thursday, July 20, 2006

Thoughts on Quine and belief revision

One way of reading Quine is to say that we each have a set of beliefs, connected in an inferential network. This set forms something like a web (hence the metaphor) with a group of core beliefs and a beliefs lying along the edges of the web. Experience impinges on the web only at its edges and modificaitons are made primarily along the edges. Changes can make their way to the interior of the web only if the recalcitrant experience is strong enough to require modifying more central beliefs to preserve as much of the overall structure as possible. This is somewhat appealling as far as it goes, but I realized I don't understand the mechanism behind the change. What I mean is that it sounds like we have beliefs and we get sensory experience of the outside world which constitutes experience; this experience either supports or conflicts with our existing beliefs. If it supports it, so much the better we think. If it conflicts, then we must modify our web, making as few changes as possible. One of the problems is that experiences don't come labelled as recalcitrant or not. They first have to be recognized as such. Somone can have experiences that would seem to undermine other beliefs she holds without seeing those experiences as the basis of conflicting beliefs. She could refuse to draw the inferences necessary to make the conflict apparent. Often the links between what we experience and how that experience impinges on our already held beliefs is separated by a large gulf, so problems are not immediately recognizable in most cases. That is one problem.

Another problem is how the web is revised. Suppose I have some experience E that I view (or come to view) as recalcitrant with regards to my web of beliefs. I could make the minimal changes in my web by changing the truth values I ascribe to various sentences along the periphery of the web. Or, I could change the strength of the inferential connections in different places. I could use E as the basis of a new theory that I will tack onto my web. Or, I could reinterpret E such that it actually supports rather than undermines my beliefs. Or, I could suspend any revisions while I try to figure out if my beliefs are actually compatible with E. Or, I could suspend my revisions because I think there is some crucial information that I need to make sense of the combination of E and my beliefs, e.g. I need to learn more about some topic before I can say for sure what status E has in regards to my beliefs. Or, I could reject the recalcitrant experience because I think that any conflict with my core beliefs is either merely apparent -and not real- or based on reasoning whose fallacies I don't currently detect but am sure of due to the strength of my core beliefs. That makes seven different ways of responding to recalcitrant beliefs, some of which do not involve any modification to my web, even along the edges. I think that Quine had the fairly straight-forward minimal revision (of the sort found in some of the current belief revision literature in logic) in mind. No doubt some of these options can be removed by stipulation or idealization, but there are a lot of options. I guess my points are that belief revision shouldn't be taken as a functional process, experience needs to be interpreted (or recognized) as recalcitrant, and recalcitrant experience need not entail even minimal revision.

Wednesday, July 19, 2006

Propositional content in relevance theory

On my vacation I read a bunch of stuff since travelling affords one so much time for reading while also removing so many distractions.

I finally read Relevance by Wilson and Sperber. I got much of the same feeling from it that I did from Recanati's Literal Meaning. At times, both books seemed to be talking more about cognitive science or psychology than pragmatics per se. To put it a different way, at times it seemed more like cognitive science than philosophy of action. This is not a criticism but an observation of a difference in approach. One nice thing about the book was that its bibliography featured several articles on disambiguation and pragmatics, a topic that I could not find much philosophical literature on previously.

The final chapter of Relevance is about speech acts and relevance theory. Wilson and Sperber adopt the Frege-Searle propositional content hypothesis I've talked about before, but they put a twist on it. The representations they talk about have logical parts and conceptual parts (they might use the term "propositional," I don't remember and I don't have the book handy to check). Entailment is a relation based on the logical content of an utterance. It is not defined for the propositional part, but they use examples like: "A dog ran" entails "Something ran." This makes it kind of hard to see the demarcation between the propositional and the logical. I believe when they introduce these terms they say that it isn't a rigorously defined distinction. The propositional part is the fully semantic part of the utterance. They then go on to endorse the idea that "The door is shut," "Shut the door," and "Is the door shut?" all have a common propositional component while differing in force. Now, they seem to be in the position to accept states of affairs rather than propositions as the common content. They don't however. They insist on the common content being truth-apt. I'm not clear on why. They are fairly clear on this when they (briefly) talk about polar questions. They say that the questions have a complete logical form and a complete propositional form, where propositions are described as truth-apt when complete. They (roughly) echo Frege on the reason for this: an answer would be an assertion or a denial of the proposition. They also don't characterize the relationship between, e.g., the above assertion, imperative, and question. I think the line they give is roughly Searle's, i.e. obviousness. Is the propositional content hypothesis a dogma of pragmatics?


Offhand, I don't remember if they discuss non-'yes-no' answers to polar questions; if not, I will return to that topic later.