I've been meaning to comment on two posts from over at the Splintered Mind for a while now. They are about Donald Davidson and his treatment of reasons as causes. The first post asks if there is a problem with reasons. The second asks about the force of "because" in Davidson. I want to address the problem of reasons in Davidson but will have to hold off a little longer to get to the depth I wish. In the meantime I wish to raise a question I asked in the comments.
I'd asked Joshua if we could ultimately successfully disentangle talk about reasons as causes in Davidson from his anomalous monism. Davidson's monism, as you might recall, basically argues that there's just one kind of stuff out there but that we can't translate mental talk (including reasons) to physical talk. The reason I think this important is that if we talk about reasons as causes we're asking fundamentally about this issue of translation.
The way I take Davidson is in terms of holism. That is we might be able to translate mental talk to physical but only by telling everything about the ultimate "stuff." Now of course in practice that isn't necessary. (I suspect the contribution of Alpha Centauri to my reasons to type this are negligible) However what this means is that as a cause we have to take all that is referred to by our mental talk. That as a whole then acts as the cause.
Given that way, I think the issue of reasons as causes seems much more palatable. (Well, except for the problems some might have with anomalous monism. The other way to consider this is in terms of Peirce's consideration of mind and matter. I'd discussed that here. Roughly we can speak of mental causing mental and physical causing physical, but via slight changes we can relate the two. (How Peirce does this is caught up in his considerations of continuity and infinities. Something not everyone will accept)
There is an other issue to raise. That is, what are causes for Davidson? Davidson argues for causes relating elements of descriptions. That is causation is ultimately about narration rather than physicality. I'd discussed that in the past as well. Thus for any event there are many (presumably infinite or close to it) descriptions we can provide and causation relates statements of those descriptions to each other. Given that relating reasons as causes is much more acceptable.
A few points:
First: If by "telling everything about this ultimate stuff" you mean knowing all physical facts: No, Davidson denies that this would tell us what mental events obtained or failed to obtain. If by "everything about this ultimate stuff" you mean knowing all true sentences simpliciter, then it is trivially true that knowing everything means you know all facts about mental events, too. The reason for the anomalism of the mental isn't because we are ignorant of the physical facts which would give us what we want here, but because of what Davidson calls the holistic character of the propositional attitudes -- mental events, as propositional attitudes, are essentially normative, and so cannot have their relations assimilated to law-like relations. (Davidson expands on, and recants from the weak bit about the indeterminacy of translation supporting anomalism, in two essays in his "Truth, Language, and History" collection. See also John McDowell's "Anomalous Monism and Functionalism" in "Mind, Value, and Reality". Might be worth looking at; "Truth, Language and History" is a great collection anyways.)
Second: For Davidson, causal relations hold between events, not descriptions of events. This is what the token-token identity bit of anomalous monism is about. You can describe an event however you please, and it won't change what causal relations involving it obtain (though some may be opaque, for some descriptions; that's part of the reason we have multiple descriptions of an event).
Third: Again, for Davidson causation is an extensive relation between two events: event B caused event C. One doesn't have to have an exhaustive account of what all caused C to judge whether or not B caused C. Indeed, one can't have all descriptions of B (for we can talk about B however we please, getting into as strange an idiolect as it pleases us), but this doesn't matter for being able to judge that B, the event, caused C, the event, however one wishes to describe B and C. There's no need to say that a "whole" composed of a great mess of events (presumably everything prior in time to C) is the "real" cause of C; that's not how causal judgments work -- when I say "this caused that" I don't mean to exclude anything else having caused "that".
(The ball arcs through the air because the bat hit it. Nothing missing in this judgment. The ball arced through the air because I swung at it. Also fine. The ball arced through the air because I managed to be lucky, despite being terrible at sports in general. Also true, also causal. My being lucky really did make the ball move. We don't need to exhaustively catalog the casual nexus to see that something caused something else. Nor do we need to be able to bring all of our descriptions into a single vocabulary, so as to discern the "real" causes of an event. The "real" cause of an event is whatever other event(s) caused it, and any description of those events will pick them out. This is why "The orange juice was moved because I was thirsty" isn't in need of any fixing up, but stands as a legitimate causal judgment, despite it being quite hopeless to think that, someday, we might have general laws relating orange juice movements and my thirst.)
(I'm not going to argue for the bit where Davidson claims that all causal relations must be instances of deterministic physical laws, "The Principle of the Nomological Character of Causality," since I don't think it it's either necessary for anomalous monism or supported in its own right -- again, McDowell's "Anomalous Monism and Functionalism" is excellent on this point. If causation is an extensive relation between events, and not between descriptions of events, then it hardly seems bothersome if we can't describe some event or other in properly nomological terms.)
Daniel thanks for the comments. I'll be the first to admit that while I love reading Davidson he's a difficult and subtle philosopher. My limits are compounded by only reading his earlier stuff (thus far) However I picked up a couple of weeks ago Truth, Language, and History and hope to go through it this summer.
I have a few questions though. You say that the issue isn't our "not knowing all facts" but "holistic character of the propositional attitudes" I confess I don't see the distinction. I took Davidson's holism to entail the problem of knowledge since holism entails that meaning is always in terms of the whole. Now the normative issue is, of course, important. But I confess to not seeing how the normative aspects leads to the problem you point out. Does Davidson address this in his early works? I'll check if I have time tonight.
I may well be completely wrong on this. But something doesn't quite seem right.
What you seem to be saying is that "normative" entails "arbitrary" and thus no law is possible given the conflict between arbitrariness and law. (A point I think Davidson has in common with Derrida, if you are right) However the reason I don't think this is ultimately a problem is that laws are essentially a-temporal (at least as Davidson is using it I think) whereas it seems to me we can simply talk about temporally indexed laws due to context. That is accept that normative issues are themselves dependent upon context. But a context that is therefore wrapped up in knowledge of all facts.
For Davidson to do what you suggest he is doing (if you have him right) seems to beg the question. Don't get me wrong, I think the holism bit is important. But there's a big difference between translating the mental at a given time and translating the mental given a-temporal laws. The latter is obviously impossible. The former less so, although one can then bring up the issue of temporal externalism as tied to meanings. If meanings are always entailed by the future (as the Heideggarians appear to suggest and I think some take Davidson to assert) then Davidson's point follows in a somewhat limited fashion. (i.e. there's no way for a temporal individual to translate)
The bit on events needs unpacked somewhat. I'll comment on that a bit latter. Either I'm not communicating well (or perhaps I'm actually misunderstanding here) but it seems things are a tad more complex than you make them out.
The "holistic character of the propositional attitudes" isn't really tied to the (more general) holism that Quine leaves us with at the end of "Two Dogmas." Of course Davidson accepts Quine's holism; he adds a third dogma to deny, that of conceptual scheme and empirical content, which makes the picture of holism even more complete -- there's no longer reality "at the edges" impinging on it, since everything that can count as a reason is already incorporated within a web of belief. "On The Very Idea of a Conceptual Scheme" and "A Coherence Theory of Truth and Knowledge" are the main papers here, even though the second paper embarrasses Davidson by the time he writes its "Afterthoughts".
The special character of our mental vocabularies (Davidson also uses talk of "the constitutive ideal of rationality" to characterize them) is briefly dealt with in "Mental Events" and at some greater length in later essays such as "The Myth of the Subjective" and "Three Varieties of Knowledge." The latter paper is particularly relevant to the issue of the reducibility of mental vocabularies; p.215f in the "Subjective, Intersubjective, Objective" volume has Davidson explicitly answering the question "How does the normative element in mental concepts prevent their reduction to physical concepts?" One answer (he gives several) is that it is our membership in a community of minds, the possibility of our being corrected, that makes the objectivity of thought possible -- the possibility that I can have a thought that p, which can be true or false (depending on whether or not p). But since the objectivity of our thoughts depends on this community of minds, there's no way to strike it out of the picture -- the main thesis of "Three Varieties" is that knowledge of one's own mind, knowledge of the external world, and knowledge of the contents of others' minds all stand or fall together. So mental concepts can't be reduced to physical concepts, as they are not the same sorts of knowledge.
I don't see how "normative" entails "arbitrary" in the sense that laws and arbitrariness are opposed -- moving the orange juice because I am thirsty strikes me as a perfectly reasonable, non-arbitrary thing to do, but the fact that my thirst is a reason for the orange juice moving is a normative matter -- I desired a drink, so I poured myself a cup. (It wouldn't make sense to attribute, say, my typing this comment to my desire for a drink. That's the normative element to it -- some things are reasons for some thought or action, and others aren't.) I don't think that thought being "arbitrary" is something Davidson would countenance; for a mind to be a mind, it must be largely rational. Irrationality is only understandable as such against a background of largely rational behavior. (cf. "Paradoxes of Irrationality," which was originally delivered at a Freud conference.) The only sense I can think of in which thought is "arbitrary" is that there's no hope for forming strict laws along the lines of "anyone who has a reason R to do X, does X." Reasons can be causes, but they don't have to be; one can pass up a perfectly good opportunity, for whatever reason. (Perhaps one is just being contrary.) Reasons can be causes; causes can be reasons; reasons can fail to be causes; causes can fail to be reasons; mental events can fail to be reasons, even if they are causes; mental events can fail to be causes, even if they are reasons.
I don't see how speaking of temporally indexed laws in atemporal terms is a problem -- "if B obtains and the earth rotates about its axis once after B has obtained, then C obtains" strikes me as a way to say "B makes C happen, a day later" without having the law give a particular period for its validity. But I'm not sure what this has to do with Davidson at all; knowing all the physical facts about an event might still not tell you what mental descriptions hold true of it, since there are no strict psychophysical laws. I have to wonder if Pierce isn't causing trouble here; Davidson is never happy with the "truth is what would be reached at an end of inquiry" stuff, and always holds the good Convention T line when it comes to truth.
I have no idea what "meanings are always entailed by the future" could mean with reference to Davidson. Davidson doesn't posit any entities called "meanings" which link up with sentences, and I don't know where he speaks of theories of meaning and the future at all. (Except if you take talk of confirming an empirical theory of meaning, such as his radical interpretor has to work with, as dealing with the future. That seems like a stretch; I'm probably just missing the point of what you're saying here.)
I take the Heideggerian point of "meanings are entailed by the future" to be that the futural is essentially affordances, various opportunities for action which we may or may not take up. (This obviously disappears when one views things atemporally, inauthentically, since the affordances are only affordances for Dasein, and not just for "The One" in general.) I think this is a nice point, but can't see what it has to do with Davidson. Reasons can be viewed atemporally -- the attack on Pearl Harbor was a reason to declare war on Japan, now and forever. (One could disagree that it was the reason for our declaring war, or disagree that it was a good reason. But this has nothing to do with its temporality; if it wasn't a good reason, or wasn't the reason we acted, then this is likewise true now and forever. Though of course it's only comprehensible for someone who knows a little about 20th-century history.) The futural openings (as I recall, Heidegger calls them "destinies") are only there here-and-now, but whether or not they're rational things to act on will be something that could be discussed whether or not the opening is still open for me.
I'll admit, I haven't read most of his early stuff -- I got interested in Davidson based on the "Rorty and His Critics" volume Dave Maier was hyping on The Valve, and I've been slowly reading the Davidsonian corpus backwards, largely to help puzzle out some of the issues McDowell & Rorty are dealing with. "On The Very Idea Of A Conceptual Scheme" got me interested in analytic philosophy again; fantastic work, that. Can't say I see the connections make between Davidson and Derrida, largely because of that third dogma -- I noticed you posted in Holbo's Derrida thread on The Valve, and I look forward to someone bringing this up (they have to eventually, if linguistic idealism/"critical realism" is in the mix -- I don't know Derrida anywhere near well enough to contribute to the thread, but linguistic idealism I do know).
Sorry, this is fairly off-topic: You reference to Donaldson's 'caustion as narration' caught my interest (cf. my lds-phil post on hermeneutics and common law, and similarities between between Dworkin and Gadamer). Davidson's work is mostly about philosophy of mind, and philosophy of language (as a derivative of the former), no? But it seems this 'causation as narration' idea is a very interesting way to think about the philosophy of science, epistemology, and even ethics. Are these relationships taken up in analytic philosophy, or would such connections be viewed as Continental/hermeneutical gibberish?
Thanks for helping me see a good reason to study Davidson (and the necessary background). Do you want to offer any thoughts on similarities and/or difference between Gadamer and Donaldson (or Gadamer and Dworkin, or Donaldson and Dworkin--perhaps even via Rorty?)? This bit about 'narrative interpretation' seems (at least superficially) to be similar in many respects.... (So much to read, so little time!)
I've never been a big fan of the way topics are divided up in a lot of analytic philosophy departments -- as far as what he taught, Davidson taught pretty well all courses you could think of; he even had a fairly long-running series of seminars on Classical Literature that he taught as an English course (his autobiography in the Open Court "Library of Living Philosophers" series is a fascinating read; Davidson was a flat-out cool guy). As far as how to classify what he wrote about, I think he's similar to Quine in that epistemology and the philosophy of science are never off the radar, but they're never handled in quite the typical way, either. One of Davidson's big ideas was that what had been taken to be a fairly formal instrument in logic (Tarski's T-sentences and undefinability of truth) can do the work we'd wanted from a theory of meaning, and that this can clear up a great deal of philosophical muddles across the board.
As far as Davidson and Gadamer: "Dialectic and Dialogue" and "Gadamer and Plato's Philosopher" are explicitly dealing with Gadamer (Davidson says he ended up at many of the same conclusions as Gadamer, despite arriving at them by a very different route: toward the end of "Gadamer and Plato's Philosopher" Davidson just quotes a big chunk of "Truth and Method" and says "I am in agreement with almost all of this."). The main difference between the two is that Davidson denies that a speaker and a hearer must share a language beforehand to understand one another, at least in the sense in which many philosophers think that this is necessary. This is a pretty technical point when it comes to Gadamer, though; Davidson says that what's needed isn't a shared language (a shared set of conventions which give meanings to our sentences), but the ability to converge on "passing theories" in the actual give-and-take of a conversation. I take this to be very much in the spirit of Gadamer's "fusion of horizons," where understanding comes about not because speaker and hearer share a worldview beforehand, nor because one member picks up the worldview of the other, but because the two come to see the world together as their horizons shift. (I'm probably putting the Gadamer bits badly, but I hope it gives some impression of what Davidson is up to.) The key essay for seeing what Davidson is focusing on in his odd quibble about a "shared language" is "A Nice Derangement of Epitaphs". I'd probably start with that essay actually, if you're interested in where Davidson-the-analytic-philosopher-of-language starts to look an awful lot like a hermeneuticist.
As far as the relationship of Davidson to Gadamer (and hermeneutics generally) in analytic philosophy goes, John McDowell is certainly a big fan of both Gadamer and Davidson, and he's still fairly widely-respected (though seldom entirely agreed with) in analytic departments. Davidson himself is not as popular as he ought to be, but it's certainly not at all odd for a philosopher interested in Davidson to end up looking at some of the hermeneutic writers in the continental tradition. Gadamer, early bits of Heidegger, and Riceour aren't as liable to be mocked as later Heidegger, Derrida, Foucault, Lacan etc. are.
Nice to know I've helped you see a good reason to study Davidson. :-)
Sorry for the delay. Lots of sick kids and deadlines at work. But thanks for the comments. I do have some questions though and perhaps one or two objections. Hopefully I'll be able to address those tonight or tomorrow. (Yeah, my blog's been last on the "to do" list for a while now)
You did say one thing I wanted to address...
The main difference between the two is that Davidson denies that a speaker and a hearer must share a language beforehand to understand one another, at least in the sense in which many philosophers think that this is necessary. This is a pretty technical point when it comes to Gadamer, though; Davidson says that what's needed isn't a shared language (a shared set of conventions which give meanings to our sentences), but the ability to converge on "passing theories" in the actual give-and-take of a conversation.
I think this is important and gets Davidson a tad closer to Derrida. (Although, primarily due to Davidson still having a charity of interpretation, not all the way there) Davidson sees convergence whereas Derrida sees points of stability that may be temporary. Where I think Derrida would still take issue with Davidson as being too close to Gadamer is in the idea we can know when we've correctly interpreted someone. But once again the details get a bit more technical.
I completely agree that Davidson moves towards hermeneutics. I think the reason the latter Heidegger, Derrida etc. are mocked is more because of their style. Foucault's mocked simply because he makes quasi-empirical arguments that are clearly wrong. (On both madness and sexuality) I don't know enough Lacan to speak there. I read a bit and mentally put him on the side of continental thought I put Baudrillard on. i.e. uninteresting.
Oh, by arbitrary I don't mean irrational. Merely that it isn't demanded by nature. i.e. descriptions as expressing natural kinds. I'm here thinking more of semiotics and the arbitrariness of the sign. But I'll hold off saying more until I have time to write something cogent.
Daniel, thank you very much for these recommendations and thoughts on Davidson and hermeneutics, I'm excited to read this stuff.
I don't know very much about Lacan either, but I think the way he thinks about psychoanalysis sounds very interesting--in the same ways a text (or data or an other) can serve as a check on my beliefs, views, actions and self-deceptions/bad faith (esp.), I think a psychoanalyst is an interesting persona to consider, as sort of a live and interactive text. And, in a certain sense, I think any human-to-human interaction might be considered a mild type of psychoanalysis, and that this adds an extra dimension to how we might think about different levels of intimacy in interpersonal relationships.
It seems to me (though, again, I'm quite ignorant, so don't take my word...) that there's a danger in Levinas in conflating all Others as being the indecipherable from each other (the other might be my neighbor, or God, or some insane psycho-killer, and I have the same infinite obligation toward each...?). So I'm left wondering how the incomprehensible Other ever becomes comprehensible.
In analytic philosophy, it seems the tendency is to consider only interactions that are, in a significant sense, already comprehensible. (I know this is a gross oversimplification, which is why I'm anxious to get into post-positivist thought!)
So I see Lacan as striking at the middle of this (admittedly unfairly caricaturized...) problem: The incomprehensible Other begins to become comprehensible in the psychoanalytic process where the Other (psychoanalyst) calls the subject into question, and the subject is willing to take such calling-into-question seriously enough to actually change/transcend the state that the subject has been stuck in. Though I don't really know any details, I think this sounds like an interesting project....
OK, finally a few minutes...
First, the caveats. I fully confess I've only read the old Davidson stuff. I have two of his newer volumes here beside me but haven't read them yet and probably won't for a while. (Coincidentally I'd ordered them quite some time ago but Amazon just shipped my order last week) So I'm just going via my best attempts to understand Davidson from reading (fairly regularly) his early works. But I find Davidson difficult, so I'd be the first to admit I might get him wrong. I don't think I do, but then I'd rather be wrong than right since then I'd at least be learning something. (grin)
Now to be clear Davidson explicitly distinguishes causes from descriptions of causes.
. . we must distinguish firmly between causes and the features we hit on for describing them, and hence between the question whether a statement says truly that one event caused another and the further question whether the events are characterized in such a way that we can deduce or otherwise infere from laws or other causal lore that the relation was causal. ("Causal Relations", Essays on Actions and Events155, emphasis mine)
What I take Davidson to be arguing is both that Hume's critique of causality is true (160) in that "a caused b" entails a law but we don't know the law. Further that we have to distinguish the logical analysis of causality from the analysis of causality. (161) (A point Peirce makes as well) This is repeated by Davidson at the end of "Mental Events" (224) Davidson's conclusion is that we can know a mental event is identical to some physical event without knowing the law. (In the sense of giving a unique physical description) This is why one could know the physical history of the world without being able to predict or explain the mental. It's because the law-like relations aren't entailed by the totality of physical knowledge.
Now I recognize you'd agree with all that. I just want to point out that this is how I've read Davidson. Just to stay on some common ground.
Regarding descriptions, I may be reading too much Peirce into Davidson. But I see one of his complaints about descriptions as being roughly Peirce's point that signs are always general. That is, to follow Davidson ("The Logical Form of Action Sentences", 117) a particular event may make a general statement true but the general statement doesn't pick out the event. This is true (as Davidson shows by example) even of sentences that don't appear to be general.
This then entails a problem for any causal descriptions as well as physical or mental descriptions. And that's how I tend to read his work on causal relations and descriptions and also how I read the problem of holism in his Anomalous Monism. (In part) Certainly I agree Davidson isn't just adopting the holism of Quine. (Although that he adopts at least part of it is clearly relevant) Rather he is adopting a holism roughly akin to Heidegger's. That is we have a projective framework of actions, uses, aims, and so forth about objects. For the mental we have our projective framework of beliefs, intentions, desires, and so forth. It's only in that context of a framework that we can make descriptions. But that framework is, in Davidson as in Heidegger, what will always undermine the idea that a totality of physical facts can tell us about the mental.
Now, as you say, if we take the trivial sense that all facts entails all descriptions then of course we can do more. (And that's what's Peirce's limiting notion of truth as what the ideal community asserts in the long run achieves) But that's not useful for any finite inquiry. We're always stuck with a limited and fallible projective framework.
This too, getting to the Gadamer, Derrida, Davidson issue, is where I find Davidson's notion of charity problematic. Given that we always need these holistic backgrounds for sense to be possible, on what basic can we justify charity. Any charity is always in terms of this projection from our understanding which may well be doing violence to the person we're communicating with.
Anyway, I'd be interested in your thoughts.
Robert, (#8) the incomprehensible Other as such never becomes comprehensible. However there is always an "eruption" into our understanding that is comprehensible. While we can say that, in part, this comes from the Other we can't ever say it is the Other. If you can see the distinction.
(I find the issue far easier to explain via Peirce, but since not everyone knows Peirce I'll not do that)
The reason I think Levinas treats all Others as indistinguishable is both logical and also a matter of analysis. The logical reason is simple. We can't apply logic to the other so we can't make differences. Rather the Other is what makes differences possible and thus is logically prior to differences. (Derrida plays with this as lot in his notion of differance) The analytic reason is that the Other is the transcendent and thus is "depth." As depth rather than any "thing" it's pointless to make a distinction since you're just talking about a phenomenology of Otherness. To try and do otherwise is simply to stop talking about Otherness and start talking about things, which kind of defeats the point. This is the problem of Husserl and the Other. It's always tied up with thing talk as either things present or things expected. Otherness as a true Otherness really isn't allowed.
Oh, regarding arbitrariness. The problem ends up being that of signs always being general. However an other way to think of it is that descriptions are always made against a backdrop of projective theories. (I keep wanting to call it a projective metaphysics ala Heidegger's philosophy of science, but I'll not do that so as to avoid any misunderstanding) Anyways, given this projective stance, the theories are arbitrary in that the terms we use needn't be in a 1-1 relationship with natural kinds. Put simply we're always describing in terms of a natural language and not some ideal Aristotilean language. Given this, our terms needn't map onto the events in a 1-1 fashion.
This entails the point I made before. Particular events may make descriptions true but descriptions can't pick out individual events. The reason is different - it's because the terms we use have an arbitrariness to them even if, as with science, this arbitrariness may be limited it is always there. (An other way of saying artificial languages are always parasitic on natural languages)
This ends up being pretty important in both Peirce and Derrida. (Indeed I'd say it constitutes the main argument in part 1 of On Grammatology)
I have to wonder if Pierce isn't causing trouble here; Davidson is never happy with the "truth is what would be reached at an end of inquiry" stuff, and always holds the good Convention T line when it comes to truth.
Oh, that's always a worry I have. I have to be careful not to read into folks too much. (Although as Heidegger shows that can be interesting at times as well)
However I don't think I'm doing that here. I think I get Davidson on this point. I just think Davidson wrong. Although I confess his treatment of truth is still fascinating. But it has lots of problems.
One should note though that Peirce's notion of "in the long run" is bound up in his conception of infinities and isn't a normal convergence to truth the way one finds in most scientific realism. Peirce is much closer in practice to Derrida where we have points of stability in inquiry which may be permanent but we can't know they are permanent. So we can talk about the stability of belief. But truth proper is always this ideal community that is never an actual community.
Davidson's argument for charity is put most fully in "A Coherence Theory of Truth and Knowledge". The shorthand version he mentions often is that unless we need a certain amount of common ground to make disagreement intelligible. It's like the bit in "Alice in Wonderland" where the caterpillar asks Alice to recite a poem:
"Alice thought she might as well wait, as she had nothing else to do, and perhaps after all it might tell her something worth hearing. For some minutes it puffed away without speaking, but at last it unfolded its arms, took the hookah out of its mouth again, and said, `So you think you're changed, do you?'
`I'm afraid I am, sir,' said Alice; `I can't remember things as I used--and I don't keep the same size for ten minutes together!'
`Can't remember what things?' said the Caterpillar.
`Well, I've tried to say "How Doth the Little Busy Bee," but it all came different!' Alice replied in a very melancholy voice.
`Repeat, "You are Old, Father William,"' said the Caterpillar.
Alice folded her hands, and began:--
(poem cut for length; Carroll's parody of the normal "You are Old, Father William" poem goes here)
`That is not said right,' said the Caterpillar.
`Not quite right, I'm afraid,' said Alice, timidly; `some of the words have got altered.'
`It is wrong from beginning to end,' said the Caterpillar decidedly, and there was silence for some minutes."
The joke is that Alice, in attempting to repeat a poem she's memorized, ends up speaking a parodic version instead. (Part of the whole "Wonderland" thing.) She realizes that she's not said it quite right, because "some of the words have got altered." It's intelligible that she's said the poem wrong, because the boring old moralizing "You are Old, Father William" poem has become a silly thing about standing on one's head. The caterpillar, on the other hand, claims that Alice has said absolutely everything wrong -- but then in what sense has she repeated "You are Old, Father William" at all? If *all of it* was wrong, then it would appear that Alice has just said an entirely different poem. In which case she hasn't gotten it wrong after all.
In other words, it's only because I can see that you're trying to do this that I can see where you're doing it badly or well. Agreement and disagreement demands a background of massive agreement. (There doesn't need to be agreement between all speakers on any one point, but between any two speakers there will be mostly agreement. There aren't foundational-level sentences in a discourse, rather they all hang together in a web of belief. This is a Quinean point, of course.) Without a principle of charity, conversation is impossible, and thus the possibility of correction disappears. But if my thoughts cannot be corrected, if they can't turn out to be false or not, then it can't be said that I'm getting things right, either. But then it's not intelligible that my thoughts are objective at all -- that my thoughts are about anything. I can have the world in view only because it is a world I share with other thoughtful speakers with whom I interact. The principle of charity is needed not merely for interpreting another speaker, but for having any thoughts of one's own.
(Charity doesn't mean we agree on everything, of course. But where we disagree, the disagreement must be capable of being made intelligible. Certainly understanding between speakers is often a tricky affair, and progress often moves in fits and bursts, occasionally with things getting worse than they had been, but it is only because the speakers agree on enough that they can intelligibly said to be agreeing or disagreeing about this topic rather than that topic that we can puzzle out whether understanding is going on at all, never mind how well.)
I think the idea of a "projective ontology" as a sort of backbone for some sort of discourse (which thus renders that discourse limited to not getting at the thing-in-itself, so to speak) can't stand up to Quine's criticisms in "Two Dogmas of Empiricism." We can't sort out which of our sentences are true because of the "projective ontology" and which are true because of "the way things are," since all of them are revisable just as the rest are (though revising some of our sentences, such as the logical truths, will demand such massive shifts in our web of belief that it's almost inconceivable that the pragmatic move would be to revise those beliefs). This leads Quine to his famous statement at the beginning of "Empiricism Without The Dogmas": "The totality of our so-called knowledge or beliefs, from the most casual matters of geography and history to the profoundest laws of atomic physics or even of pure mathematics and logic, is a man-made fabric which impinges on experience only along the edges." Now, of course Quine doesn't mean by this any sort of skeptical point, or that our language falls short of the fact, fails to say how things really are. Quine's point is that it is not possible to take apart our sentences into what is "true due to meaning" and what is "true due to the facts" -- and so "the facts" don't play a strict role in correcting our beliefs (the way some theories of verification/falsification assume they do), but merely are occasions for revising our beliefs. And there are always many ways in which we can revise our beliefs to "save the data." So Quine says that reality only impinges "at the edges," since some given natural happening could be taken in a variety of ways, and each of these will demand that we revise our beliefs differently. It's only once a happening is internal to a web of beliefs that it plays a clear role in limiting what beliefs we can or can't rationally continue to hold.
Davidson takes up this end-point from "Two Dogmas" in his "On The Very Idea Of A Conceptual Scheme" and forces the point -- "reality" as something external to our conceptual schemes (or organizational systems, our theories of the world, our languages) doesn't impinge "at the edges," since only things internal to our web of belief (our "conceptual scheme") can play a rational role at all -- only a belief can be a reason for another belief. But this breaks down the dichotomy of conceptual scheme and empirical content: in forming our beliefs our schemes aren't "applied" to anything, since the "content" doesn't play any epistemic role, and so we can't say that some disagreement is due to applying different conceptual schemes to the same content. Nothing matters for the truth of "Snow is white" except whether or not snow is white. The connection between our beliefs and the world is thus direct -- if someone knows that "p" is true, then this is because p, and not because of something "below" the conceptual level which grounds it (states of affairs that sentences refer to; the reality which impinges on our web of belief); if he is wrong, it is because ~p.
The distinction between mental vocabularies and physical vocabularies is that we use one to do one thing, and another to do another -- one is used to keep track of propositional attitudes, and the other is used to track how things are in our shared world. It is not that each vocabulary is applied to some neutral "stuff" which is what makes both our mentalistic and physicalistic sentences true alike, and that it is just the variance of perspective which makes the two discourses untranslatable one to another.
Lot of topics flying around, so I'm inevitably going to be giving short shrift to some of your remarks, but that's rather the nature of the beast. (Not like this'll be the last post on Davidson/Heidegger/Gadamer that you make, I should think.)
Let's try to fix those tags. This is why I wish your blog had a "Preview" button. (I had an i before the slash, instead of after. Feel free to delete the previous comment; it's an ugly beast. I'm rather surprised that line returns apparently do not work when italicized, though. Odd, that.)
Daniel, first off thanks for the comments. Maybe if I have time I'll go through why I don't find the arguments in "A Coherence Theory of Truth and Knowledge" compelling. However here's the short version. The demand that we have something in common to make something intelligible tells us little about what it actually is we do or don't have in common.
Thus the Derridean tract (as we find in his Gadamer debate) would be to say, of course we have something in common (the Das Man if nothing else, whatever that is). However just because we have that in common doesn't mean there isn't stuff we don't have in common. By focusing in on this shared present and neglecting the absence philosophers end up hiding the problem. The absence always entails to real possibility of radical failure. (Peircean fallibilism as it were) That's because we may think we are speaking from common ground whereas actually we're not. But because the way we check if we have common ground which presupposes it, we can never be sure.
Of course this phenomena happens all the time. I've frequently been in discussions where we're using the same terms and having a fairly good discussion only to realize after quite some time that we meant completely different things by the words we used. We thought we were communicating whereas actually we were in the midsts of horrible miscommunication.
My opinion (and it's been a while, so I'd have to check the text to see if this is a fair comment) is that Davidson presupposes communication and asks what is necessary for communication. But this avoids the central issue of when we actually are communicating.
A couple more comments
I think the idea of a "projective ontology" as a sort of backbone for some sort of discourse (which thus renders that discourse limited to not getting at the thing-in-itself, so to speak) can't stand up to Quine's criticisms in "Two Dogmas of Empiricism." We can't sort out which of our sentences are true because of the "projective ontology" and which are true because of "the way things are," since all of them are revisable just as the rest are
Could you be more specific about what you see Quine saying that makes this problematic? (I think Quine goes too far ending up far more of a relativist than I think Derrida is, even though Derrida gets the rap whereas Quine doesn't) It does, however, seem to me that the undecidability of what is true because of "the way things are" (i.e. reality) and our "projective ontology" (the text; the phenomenological context; language) is an important point. But as you can see from my prior comment, I don't think that a problem, rather a fact of life. So I'm with Derrida there.
Your points about "On the Very Idea Of A Conceptual Scheme" are interesting. While I don't fully agree with Davidson (at least to the degree I understand him from his early writings) I think that the move he makes is almost the Peircean move. That is our assertions aren't grounded the way they are in the correspondence theory of truth. However reality acts leading inquiry to truth if conducted far enough. In the finite time though talking of truth and knowledge is problematic, it's better to talk about the fixity of belief. In the long run we can talk about truth and it is because the ideal community would assert p. This is nearly an idealist point and I think Davidson heads in that direction as well.
Of course once one rejects Frege, as Davidson famously does, then I think one is naturally lead to either relativism or something akin to what the pragmatists do. (Dewey is different than Peirce, but similar in important ways in this context - I think Derrida heads more in the Peircean direction)
The issue of the mental and the physical being tied to different "aims" is a good way to deal with Anomalous Monism. I think Anomalous Monism can easily be defended by an appeal to "worlds" in Heidegger and his grounding of understanding in ready-at-handedness and for-the-sake-of-which rather than, as you say, "neutral stuff." (Ontic entities, sometimes called de-worlded) Science aims as this ontic realism but I don't think we can move the mental to this deworlded since as a world, in the Heideggerian sense, is part and parcel of what mental descriptions involve. So there's a logical contradiction. You both have a world and don't have a world.
One last note. At the end of his essay on Plato and Gadamer Davidson does undermine the presentation I gave above and appears to differ from Gadamer. What counts isn't a shared language or understanding. He's explicit we can always misunderstand. (Making Derrida's critique) What counts is shared objects. Roughly that we are always in a world and it is our comportment with that world that makes all the rest possible.
I'm still thinking about that. I'm just not familiar enough with his later writings to say much about this. It does seem (to me) more Heideggarian than Gadamer even.
I see Quine making the anti-framework point in most of "Two Dogmas". The idea of a framework (of analytic sentences) which enabled us to have knowledge of the world (via synthetic sentences) was pretty much Carnap's idea. This might come out more clearly in the papers leading up to "Two Dogmas", where Quine is more explicitly in dialogue with Carnap. (Can't find the names of the papers I have in mind at the moment, since I didn't purchase the essay collections after I'd read what I wanted -- I am pretty sure Carnap was in the titles of the essays.) Quine counters that any sentence can be held true "come what may" and so there's no way to divide up between what is true due solely to how things are, and what is true because of the way we use language. This eventually leads to Quine's anti-mentalist rejection of epistemology (in favor of "naturalized epistemology" which takes it for a given that we do hold various sentences true, and looks at the conditions in which this occurs, rather than playing around with the old ideas of belief, opinion, knowledge etc. as classically understood).
I don't think Quine ends up as any sort of a relativist, because Quine's "ontological relativity" isn't the sort of thing Derrida and other po-mo types are accused of being. And I don't think that ontological relativity is worth holding onto anyway, since we can't make sense of truth, reference, ontology etc. being relative to a language. (See Davidson's "The Inscrutability of Reference" for a nice discussion of this -- he ties it nicely into the demand that the semantic features of language must be public, because learnable, and thus "What no one can, in the nature of the case, figure out from the totality of the relevant evidence cannot be part of meaning.") It is worth remembering that neither Quine nor Davidson posits entities called "meanings" which are correlated with sentences, and so it can't be the case that they posit meanings which are relative to a speaker, or private, or relative to a framework, or only come into play in momentary instances of understanding.
You're right, of course, that the principle of charity doesn't tell us what we hold in common, and that we can very well be wrong about this. Why would we need, or even want to, be able to determine this a priori, before we actually get to talking? It'd rather remove the point of discussion if we always already knew exactly what we agreed with one another on and what we disagreed over. The principle of charity does assure us that if communication is to go on, then it goes on against a background of massive agreement, and disagreement is a type of communication. If I cannot make intelligible that I might agree or disagree with you, then I can't make it intelligible that you have beliefs at all. But then a fortiori I can't interpret any of your bodily actions as assertions, or any other sort of speech behavior. Nor can I understand any of your bodily motions as actions proper, as a sign of a capacity for practical reasoning. Hence if I can understand any of your actions, then I can also know that most of our beliefs are both shared and true.
It is of course possible for debate on some issue to go on forever, to the satisfaction of neither party. But this is hardly compulsory. If neither party wishes to espouse a skeptical position, then neither does either party need to satisfy the skeptic. If we agree, then the time for giving justifications for our views has ended. (If one is not sure if there is agreement, or merely homophones being bandied about -- look and see! From what each of you does, do you agree or disagree on the matter? There cannot be more to what each of you means that what is discernible from your actions, for one can only intend to say what one thinks might be understood by one's hearer.)
(I am skeptical that your debates with evangelicals were quite so bad as you're claiming, even if it does come out in the end that your disagreements were due throughout to commitments which were not explicit until the end of the discussion. At the least, there was a shared understanding that what was under discussion was, say, one's religion, and you both understood in a broad sense the manner in which each other's sentences were structured, etc. etc.)
Davidson explicitly rejects the notion of truth as "ideal assertability" whenever it comes up. (Which is every time he mentions Rorty and pragmatism.) It's possible that we could be wrong about something forever, after all. (Perhaps some minor Assyrian figure has their name written down only once, and there it is transcribed wrongly, owing to a bad ear on the scribe. Future historians might unanimously get the fellow's name wrong. Just as an example of where truth and assertability would come apart, even if one is given as much time as one pleases.)
I don't think we can make sense of talking about "fixity of belief" as opposed to truth. To believe that p is just to believe that "p" is true. If we fix our beliefs regarding p, then there's no further issue for us about the truth of p. If the truth of "p" can be an issue for us at all, then this just shows that our belief in re "p" is not fixed. Certainly all of our beliefs are valid candidates for revision, that is the burden and duty of self-consciousness, but not all of them can be revised at once.
I'm not sure why you say that Davidson rejects Frege. Davidson holds to classical logic, meaning as truth conditions, affirms the Slingshot Argument etc. A Davidsonian theory of meaning has no use for Fregean Sinnen, but those mysterious entities hardly seem as important as the idea of meaning as truth-conditions. (It is perhaps worth remarking that a Tarski-style theory of truth for some language L functions by means of axioms which assign the reference/extension of the various terms used in a language. So "is a tree" is used only of things which are trees, "Bob Dole" refers to Bob Dole, etc. The point about the inscrutability of reference is that, insofar as these assignments can be varied without affecting the truth-conditions of the sentences of which the terms used are a part, reference cannot be fixed and does not need to be. Truth-conditions do the work we had wanted "Sinn" and "Bedeutung" to do.) I would say that someone like, say, Jaako Hintikka is a better bearer the "rejects Frege" title.
The point about Heidegger, anomalism, and worlds is worth thinking on. I'm not sure that it's right to say that science aims at being "deworlded" since the inauthentic view of things is still part of Dasein -- the view of "The One" is still the view of my "The One". I'm also not sure it wouldn't fall back into the view Davidson was initially concerned to combat, that the causal and the mental are unrelated -- that to give the reasons for something and to give the causes of something were unrelated activities, hence leading to the causal inertness of the mental. (I suppose one could try to finesse an understanding as causality, such that it isn't "the sort of thing that goes on when things happen in accord with scientific laws." But that strikes me as rather an unattractive way to handle the problem of mental causation.)
I'm not sure why Gadamer wouldn't be able to say that we can always misunderstand one another. It's not like our hermeneutical responsibilities ever end; eternal vigilance is the price of communication, so to speak. I can see why he wouldn't say that we can't *never* understand one another, but that is quite another issue. ("I might be hallucinating right now" versus "Maybe my thoughts are never about anything", basically. The first strikes me as always a possibility, while the second is a groundless skeptical confusion.)
What matters for communication to be successful is a shared understanding, but this doesn't need to be an understanding which is shared beforehand, or extends afterwards. It's the goal of communication, not a prerequisite for further goals. You're right that the emphasis on the importance of shared objects in a common world is something Gadamer doesn't stress as much as Davidson does, but I think this is partly because of the Heideggerian pressures that would be put on any talk of "the" world as opposed to "a" world. The objects in the world only matter for Davidson's account of interpretation because they (partly) determine the content of one's thoughts, after all. From what I know of Gadamer, he and Davidson are singing from the same hymnal.
With regards to frameworks don't we have to distinguish between frameworks as necessary for intelligibility and frameworks as essential to truth? The latter being a more relativistic position. Certainly Quine (and Davidson) don't think truth is relative to a framework of the sort Carnap invoked. Quine ends up being much closer to a traditional empiricist, from what I can see. (Although I'd be the first to admit that I get Quine wrong often when I pontificate on him, so I've learned to be cautious - it's simply been too long since I read Quine faithfully)
The reason I asked the question is that I think we have to be careful about what we mean by a projective ontology. I think, for instance, that's there's a big difference between say Heidegger's notion and Carnap's. As I understand it Quine wants to make our claims and understanding intelligible against a background of sense data whereas Davidson wants to do it against the events and objects the sentence is about. The problem I have with Quine is that, to me, empiricism simply isn't up to the job, as we can see with the mathematicization of physics. (Intelligibility comes about because the math makes it possible - we don't get the math from empirical experiences) Davidson I'm much more sympathetic to, except that we have to keep in mind that our encountering of objects is never in their fulness. (The old notion of presence in Derrida or the withdrawal in Heidegger) However for Heidegger the very move of authenticity towards objects in an individual moment of transfiguration is key to his philosophy of science. So it is surprisingly close to Davidson. What Heidegger gives us with the authentic/inauthentic distinction is two ways to consider sentences. One ends up being the social externalism the Davidson doesn't like whereas the authentic ends up being much closer to Davidson. One might even say that for Heidegger this authentic mode of Dasein makes truth possible. (Primordial truth allowing for later correspondence)
I'll be the first to confess that working out exactly what Davidson's particular form of coherency is difficult for me. (Hopefully reading these later essays will help) My sense is that I still have problem with him, but I may just be misreading him. It doesn't seem like normal coherency as I understand it. Indeed it often sounds almost like projective ontologies as coherency of beliefs. That is the projective ontology is simply the collection of coherent beliefs Davidson talks about. But I may well be wrong in that.
To the principle of charity. I obviously don't think we need keep straight when we think we understand each other. I just think there is a big difference between our "first person" state of relative ignorance and a more third person analysis where one can speak about being in agreement. To make an analogy, while I'm an externalist epistemologically it doesn't mean I don't find a focus on justification ("epistemology naturalized") important. There is an element of traditional epistemology that is a question and responsibility we can't reject. Likewise the problem of knowing when we are understood is important in language. Certainly we can, as a practical matter, just acknowledge a thorough-going fallibilism. However the question then becomes, how often do we take such fallibilism seriously philosophically and how often do we merely give it lip service?
Thus the Gadamer/Derrida difference. Both, as a practical matter, agree of course. Derrida says about Gadamer's principle of a "good will to understanding", "how could anyone not be tempted to acknowledge how extremely evident this axiom is?" However Derrida's point is simply to look at the power issues if it is not always true. (That is the point when charity breaks down)
That Gadamer and Davidson capture the majority of our cases is, of course, never in doubt. What Derrida's interested in is the philosophical import of when it does break down and "the other" erupts into our understanding.
Gadamer can, of course, acknowledge misunderstanding. Indeed he explicitly does. Once again the issue is the philosophical import of such matters. I think this gets into the issue of continuity - especially hermeneutic continuity. (Interestingly this may also be where Derrida and Peirce part company, although Peirce is very subtle in his notion of infinities especially relative to geometry. There's a whole subtle exegesis that I'll not bore you with since it gets into heavy mathematics)
So I don't dispute what you say. Indeed I take it as quite apparent. The question is whether we draw too much philosophically from the standard case rather than the aberrant case. Both need kept in check, IMO.
Note that Rorty's "ideal assertability" and Putnam's sometimes similar notion bear little resemblance to Peirce. I'll not bore things by going through the details. And, as Davidson got Rorty to admit, Rorty doesn't accept Peirce's notion of truth anyway.
Regarding truth and belief, of course what you say is true. (And indeed is key to Peirce's thought as I'm sure you're aware) I don't think I'm opposing truth and belief. (I agree that'd make no sense) But we have to keep in mind that they aren't the same thing in terms of finitude.
By saying Davidson rejects Frege I meant simply what you agree with earlier. That for Davidson there are no entities called meanings out there. Davidson has a bunch of anti-Frege arguments that I spent a lot of time last summer studying. So that's what brings it to mind.
Heidegger and the issue of deworlded entities probably deserves a separate thread. I admit that I follow, in a general sense, Taylor Carman, Mark Wrathall and others who see Heidegger as an ontic realist. So those comments of mine are wrapped up in that.
I don't see how ontic realism or de-worlding (in the Heideggarian sense or the somewhat similar Peircean sense) end up falling back into the problem of causality and the mental though. They seem rather unrelated. I suspect by "deworlded" you are taking me to be asserting something I'm not. But I'm not quite sure what that is.
The final point you note I'd mentioned in my earlier comment. For Davidson the issue is the entities themselves. In a certain sense that's true for Heidegger, Gadamer and even Derrida. "To the things themselves." But that takes us to the issue of how things are manifest phenomenologically. I'm not sure Davidson takes sufficient thought of that. (Although I may well be wrong - it's an area I'm just ignorant of relative to Davidson)
I think you're misreading Quine if you see him as a sense-data theorist; the world simply doesn't play that sort of role in his epistemology. It's not that we have a certain set of sentences (the ones about sense-data, which are a given for us) which we then have to revise the rest of our sentences to fit. We have to revise all of our sentences to fit with one another, with none having a particular priority, epistemologically speaking. Sentences about how the world appears to us (the sort of thing we might take reports of sense data to be) are revisable along with all the others; perhaps we decide that we saw things wrongly, or were hallucinating at the time, in which case the sense-data report gets worked over to cohere with the rest of our web of beliefs. The world which "impinges at the edges" of a web of belief doesn't do so by sending sense-data postcards, but by playing a causal role in determining some of our beliefs (primarily perceptual ones). (This is where Davidson attacks Quine in "On The Very Idea of a Conceptual Scheme".)
I'm not sure how you're drawing a contrast between Quine and Davidson, here. Quine agrees with Davidson that our ontology needs to include events, or else it's mysterious how we are supposed to handle adverbs. On the issue of sense data and objects, are you thinking of the distinction between the proximal and the distal causes of a belief? The proximal cause of a belief (Quine's "stimulus-meaning") isn't a sense datum, but a neural firing (or something similar -- the physical cause of the happening in the nervous system which is identical with the belief). Sense-data are private, and were supposed to allow for a meaningful private language, but language for Quine is always a social affair. Quine eventually came around on the distal vs. proximal point too, incidentally (sometime around '90 if I recall correctly; it was before his "Where Do We Disagree?" essay in the Library of Living Philosophers volume on Davidson, at any rate).
I think the point that we can't reduce science to sense-data talk is one of the points Quine is making in "Two Dogmas." We can't sort out which of the objects we speak of are "given" and which are theoretical posits, which is why Quine speaks of the "myth of physical objects." He doesn't mean to oppose physical objects to sense data, because he doesn't countenance sense-data at all. Certainly it wouldn't have come to a surprise to Carnap, even, that mere sense-data talk wasn't what we used in science; mathematics is always available for the manipulation of sense-data (as analytic knowledge, for Carnap).
I don't think Davidson has a notion of "projective ontology" at all. There's nothing for it to be projected on in his account, since (unlike Heidegger) he never endorses a non-linguistic relation to the world. Tarski-style theories of truth don't require some non-linguistic fact to obtain for a sentence to be true, but give truth-conditions for sentences which are always other sentences. A true sentence says how the world is, but the world doesn't play an epistemological role separable from the role played by the various sentences we hold true. The coherence of a belief with other beliefs gives us a ground for holding it to be true, given that we know from the principle of charity that most of our beliefs are true if they are beliefs at all. But what actually determines where "p" is true is nothing but p. "Snow is white" is true if and only if snow is white. Without already having mastery of a language, it's not intelligible that one has any propositional attitudes at all, so it's not a problem if I can't discern what makes a sentence true without using a language.
I can't see how a third party is able to say whether or not we agree or disagree any better than we (first-personally) are able to. For C to say if A means the same as B, C has to judge whether what C takes A to mean is what A takes A to mean, and if what C judges B to mean is what B take B to mean, and then whether what A mean and B mean are the same; for A and B to decide if what they mean is the same, A has to judge whether what A takes B to mean is what B takes B to mean, and B has to judge whether what B takes A to mean is what A takes A to mean. Adding extra parties just extends the hermeneutical circle further, but doesn't change the basic dynamics from what it is between the original two speakers.
I don't think a thorough-going fallibilism is something we should endorse, even in lip-service. I may be wrong about A, or B, or C, or... but I can't be wrong about everything. And the mere possibility that I may be wrong about p gives me not the slightest reason to doubt that p; I need some actual grounds to think that, perhaps, ~p. I have no doubt whatsoever that my mother loves me, though this belief is corrigible -- I might be convinced that I am wrong on this point, if someone or other presented me with evidence for thinking that she doesn't love me. The mere fact that I might be convinced that I have been in error does not make me doubt in the slightest that my mother loves me, even as lip-service. I can say the same for logical laws -- I have no doubt that the law of non-contradiction really is a law of logic, but this belief too is corrigible; perhaps Graham Priest will be enormously convincing when I get around to reading him. The mere fact that I know Graham Priest has rejected the law of non-contradiction as a universal logic law doesn't give me any reason to actually suspect that there might be some true sentences of the form (p & ~p). I think the corrigibility of all of my beliefs does the work that fallibilism is wanted to do (I am irrational if I refuse to give due regard evidence that I am wrong, no matter what the evidence is pointing to) without opening skeptical worries -- if it's possible that my thoughts are not about anything at all (or at least that none of them are about what I take them to be about), then it's opaque to me how I might claim to be justified in rejecting this worry (rather than just willfully rejecting it, because it smells false, or whatever).
I am not terribly clear on Gadamer's "good will to understanding" -- is it simply identical to Davidson (and Quine)'s mature principle of charity, viz. that most of anyone's beliefs will be true? Or is it something closer to Davidson's early (and embarrassing) formulation of the principle of charity as "Maximize Agreement" -- in Davidson's later work, a recurring theme is to apologize for the places where he used that slogan rather than "Maximize Understanding"? I couldn't even guess at what geometry had to do with any of this. (Also, I confess that from your Derrida/Gadamer post, it sounded like Derrida was relying on a pun to make his view compelling -- "good will to power" -- rather than actually doing anything to convince me that Gadamer didn't give enough consideration to misunderstanding. What's the problem with the will to truth being will to power, anyway, if it's still a will to truth? (to indulge some wordplay of my own.))
I don't see how truth and belief can be any different for an infinite community of investigators than they are for we finite creatures. If "p" is true, and I believe that p, then the infinite community can't do any better, because there is nothing more to do here. If "p" is true, but we can't know that p (we can come up with any story we like for why the fact that p is forever lost to us -- perhaps "p" says some fact about the number of scales on some particular dinosaur at some particular moment, and the poor dear was at ground zero when the asteroid hit earth and killed him & his fellows), then "p" doesn't stop being true just because the ideal community of investigators are ignorant of it, too -- unless Pierce's ideal community is just supposed to be simply omniscient, in which case I've not understood Pierce's theory of truth at all.
I think my concern with de-worlding and the causal inertness of the mental is that if the causal relations which hold between physical events only show up when we're concerned with viewing things with an eye for their physical status, then those relations aren't in view when we're concerned with understanding each others' actions as rational (reason-governed). Basically, that we'd end up back at Anscombe's view that talk of reasons and talk of causes are two different sorts of talk, and that we are confused when we try to mix them -- to take a reason for a cause, or a cause for a reason. It's certainly possible that the mix of Piercian, Heideggerian, and Davidsonian jargon has overcome me, though.
I don't have time to respond. I'd just say that by "sense-data" I don't mean it in the foundationalist sense. More just a generic empiricism. Probably not the best word choice on my part. My apology for the confusion. The rest I'll answer tomorrow if I have time.
OK, just a few comments.
We have to revise all of our sentences to fit with one another, with none having a particular priority, epistemologically speaking.
I'd actually talked about this a few months back. As I said, I've read a fair bit of Quine but still get him wrong frequently so perhaps I still am in this regard. But to me this aspect of Quine is what makes his "relativism" so problematic. That is all sentences appear to be on par, epistemologically speaking. I should note that I was thinking of Davidson's critique of Quine (although not that particular paper - he'd mentioned it in one I'd read two days ago) I just made a poor word choice when I brought in "sense-data." As you note I should have said, "proximal cause." Although isn't Russell's notion of "sense data" roughly the same? I could have sworn that many empiricists when you push for the physical meaning of "sense-data" end up with neural activity. Few adopt a true idealist position since there are few Cartesian dualists left in the world.
Regarding Davidson and frameworks. I don't think I asserted Davidson had a framework basis beyond a basis of beliefs which are "mostly correct." My question is at what point does a collection of coherent beliefs differ from a framework? One way is that one need not believe a framework I suppose. However typically everyone using a variant of a projective framework assumes that the person holding them believes them. The only exceptions might be the hard core Instrumentalists.
Your point about Davidson and non-linguistic relations to the world is an apt one, of course. The question then becomes what are the boundaries of the linguistical. But that doesn't really change the point. I should note that this has long been a complaint of mine against Davidson since I think a general semiotic rather than language is what is needed. I've had a few discussions on that point and can't convince any of the Davidson proponents I've encountered though. (grin)
Your point about propositions is, of course, correct. The problem is thinking propositions are the be all and end all of understanding. But of course if we're talking about truth we're talking about propositions. And I don't disagree with the Davidson points you make in that regard up to a point. (I'll hold off saying what that point is)
Regarding adding parties to understanding. It lessens the opportunity for error. This is an other place I tend to fault Davidson, although he does bring up Bayesianism in a few places. I'm not a Bayesian in the least but given science it seems obvious that more observers helps reduce error.
Certainly fallibilism need not imply doubt. Indeed that's a key point for Peirce. We have no freedom over our doubts and when we can't doubt that's as close to knowledge as really matters. What counts is continuing inquiry which may raise doubts. So with regards to belief you are correct. What is left out of the discussion is continuing inquiry which is where I think Peirce gets at these issues better than Davidson. (IMO)
With regards to Gadamer. I know him like Quine. That is I've read a lot of him but not close enough or consistently enough that I feel qualified to act as exegete for him. The issue of whether he's closer to the early or latter forms of Davidson's charity I'll leave alone. I have a few thoughts on the matter, but I think in general it ends up being taking the communications of your interlocutor and interpreting them in the fashion that makes the most sense in terms of your understanding. The danger in this, and why Derrida raises his pun of "good will to power", is that this ends up being a kind of violence. (In the Levinasian sense) That is we're taking our perspective as king and assuming that charity is making their words fit us. Whereas charity, to Derrida and Levinas, would much more be trying to free the person you're communicating and not leaping to such conclusions. Of course the practical differences are an other thing...
The example I give of this is typically from philosophy. When you read a philosopher you probably disagree with (say Plato, Hegel, Nietzsche, Russell), how do you read them? Do you try to read them in a fashion such that you do everything you can to assume they have a good point, are heading to the truth, and are coherent? Or do you read them looking for mistakes and so forth. Arguably to understand you have to try and read them assuming they aren't idiots and have a way of understanding the world. I remember when I first read Plato like that and suddenly "got" Socrates epistemology from the early dialogs. I honestly felt this period of disorientation when it all "clicked." Very weird experience. But I felt like I understood him in a fashion I hadn't before even though both times I was trying to read him carefully.
Of course the question then becomes was this "originary" reading of Socrates the Gadamer reading or the Derridean reading? That's the problem, as I said, in practice.
Regarding why the infinite community of inquirers is different from the finite one. If the world acts on us then over time this pure action will tend to correct our beliefs. For a finite set of inquirers there simply isn't enough time for the world to act to lead us to true beliefs in all our beliefs whereas (for Peirce) there is for the infinite community. One should note that infinity for Peirce included transfinite numbers much larger than the set of Integers or the set of Real Numbers.
Regarding deworlding. I think the issue of causality and deworlding is a good one. I don't know if anyone's actually written on it. I'd never considered the question before myself. My inclination is thus primarily a Humean one. I tend to think causality is always found as part of a world whereas deworlding can bring us the entities but not the causal relationships.
Everyone may have moved on, but just in case Daniel is still checking in, I have a question about Davidson's relationship to Quine, and more generally, about the legacy of Quine's "Two Dogmas" and "On What There Is."
Do aspects of Davidson's philosophy stand or fall with Quine's holism or his view about ontological commitment? Does he Quinize names and Russell away descriptions? Etc.
Are these staples of Quinean philosophy interesting historical facts about mid 20th century philosophy, or are they common tools being used by later and present thinkers. What I would really need is an article or book titled something like "The influence of Quine's philosophy." Is there such an animal?
N.N.:
I honestly can't tell you off the top of my head what Davidson does with "On What There Is." I don't remember seeing him cite that article, in anything I've read of his. I know he doesn't maintain Quine's physicalism, at least. Roughly put, Davidson's ontology includes whatever we talk about -- there's no suggestion that, though we can talk about chairs and snow and colors and numbers and Mormons, physical particles are all that's "really real" or that what exists is relative to a conceptual network. Davidson doesn't share Quine's "taste for desert landscapes." I'll have to reread "On What There Is" and see if anything leaps out at me; it's been a while since I've read that essay.
One thing that does come to mind -- Davidson nicely tears down Quine's ontological relativism in "The Inscrutability of Reference." If worries about Quine are making you shy away from Davidson, that's probably a fine place to see them disagreeing. (It's in "Essays on Truth and Interpretation.")
"Two Dogmas" is very important for Davidson (he read it in manuscript actually), as is Quine's holism, but the latter is importantly modified by Davidson's rejection of the "third dogma of empiricism," that of (empirical) content and (conceptual) scheme. "On The Very Idea of a Conceptual Scheme" is the key text for both Davidson's love of "Two Dogmas" and his rejection of the scheme/content dichotomy. I'm not quite sure what you mean by Quinizing names and Russelling away descriptions, so I'm not sure whether or not Davidson does either.
On a historical note, Quine was the professor that got Davidson to turn from his historical studies (he was a Classicist, thus the dissertation on "Philebus") to philosophy "in which it mattered that one try to get things right." So Davidson's enthusiasm for Quine doesn't stem entirely from his philosophical output; Quine the man was an inspiration for Davidson. The autobiography in the "Library of Living Philosophers" volume on Davidson is a pretty interesting read (there's a great bit about Quine on vacation in Mexico) -- and hey, there's a short article by Quine in that volume called "Where Do We Disagree?", which seems relevant to the question of where Davidson and Quine disagree. (Short answer in that article is, by the '90s they disagree about the scheme-content dichotomy and precious little besides. Of course, that's a pretty big thing to disagree on. And Quine goes through a laundry-list of topics he's come around to Davidson's view on.)
I would also be interested in a book or longish article along the lines of "Quine And What Happened After." Scott Soames's two-volume history of analytic philosophy suggested to me that Quine is widely disagreed with, but still a figure that one simply must engage with. I've seen it claimed that "On What There Is" was the start of "analytic metaphysics," but I've tried to avoid looking into that subfield any more than I had to; I know Brian Weatherson once made a joke about his metaphysical views sounding "crazy" to a layperson, and his example was "To be is to be the value of a bound variable."
Sorry I can't be more of a help here; hopefully I've pointed in some directions that might lead to what you're looking for. I am still checking in here, and on other related blogs.
Daniel,
Thanks, I'll look into those references.
"Quinize the name and Russell away the description" is a pithy way to put Quine's move of converting names to descriptions, and then moving descriptions into the predicate position. I can't remember where I read it.
I vaguely recall Davidson making a point about converting singular reference (as with names) to "there exists some X such that X is N and X=A" but I think it was in one of his earlier essays, which I just don't know as well as the later stuff. The task at hand was to clear up what appeared to be a difficulty for his Tarskian program by showing how it really didn't pose a problem for Convention T after all. (I can't actually recall what the problem being addressed was, sadly.) Of course, in the later essays there's hardly any formal logic employed at all; Convention T is straightforwardly about getting the truth-conditions of a sentence by translating it into a language I understand (or disquoting it if I don't need to translate it to understand it). If Quine & Russell are still playing a nefarious role in my ability to understand a sentence of natural language, then I'm not seeing where they're causing any trouble.
So: Yes, I think Davidson does "Quinize the name and Russell away the description" when he wants to put a sentence in formal logic, but I don't see why this is a problematic move. It's not a move he makes very often; the point of radical interpretation is to render matters in sentences I can understand, so formal logic isn't any more important than any other rendering I can read.
Might I ask why you find Quinizing/Russelling away in this manner worrisome?
I'm working up a criticism of Quine's position for publication. If successful, it would reach beyond Quine to anyone who makes use of his argument. Apologies, but I'm wary about putting the details online before I've had a chance to get it in print. If it does get accepted (I should have it off in a week or so), I'll be sure and post it on my blog.
Fair enough. I'll keep an eye out for it.
I've closed comments in order to avoid spam since I don't check this older blog as much anymore.
and Theology
Number of unique visitors:
Blogged by Clark Goble