i have to admit sometimes i don't know why davidson thinks he can carry out a program similar to tarki's. tarski explicitly said his T-schema only applies to FORMAL languages with unambiguous predicates, no self-reference, and only a defined, pre-determined set of propositions and objects. davidson pretty much ignores this and goes ahead and tries to apply the T-schema to natural languages only to get meaning out of sentences. So we have something like:
(1) 'snow is white' is true iff snow is white.
which seems true, but doesn't really who much about meaning. but of course, because of the definition of an "if and only if" statement, it also makes sentences like
(2) 'snow is white' is true iff grass is green
true as well. of course this is absurd. yes, the biconditional is true, but it tells us nothing about the meaning because the name 'snow is white' doesn't designate whether grass is green. davidson kinda says that the precise theory of meaning wouldn't allow for cases like (2), or at least for not many of those cases, because each word in any statement will have a set of axioms that interpret the referent of words. for example, there will be axioms such as:
A1: 'dogs' refers to dogs
A2: 'snow' refers to snow
A3: 'is white' refers to the property of being white
...
An
such that cases similar to (1) will not occur (or at least not very often). but in effect all this does is equate meaning with (potentially mistaken) translations.
in effect, what i'm trying to say about davidson's adoption of tarki's t-schema adapted to get meaning is that it's both too strong and too weak. by this i mean that there are cases where the whole biconditional is met and we still don't want to say that we have meaning (such as in the case of (2)).
AND there are cases where the biconditional is not met yet we still want to say there is some meaning. this type of thing happens often in natural languages: self-reference, sentences without truth values, vague predicates, metaphors, similes, malapropisms, and other stuff.
this position i take against davidson is probably due to the fact that my idea of meaning is (not surprisingly) akin to peirce's, where the meaning of something is its effects... but that's for a post at another time...
Showing posts with label truth. Show all posts
Showing posts with label truth. Show all posts
07 November 2007
31 October 2007
truth theories
yeah... i've been hung up on beliefs lately. now comes some thoughts on truth. the way i'm gonna go about it is offering some theories of truth and me just basically running my yap about them. quickly, before i go on to do just that, i say i stand by what i roughly said last post, that if you believe homo sapiens evolved from australopithecus, then you'd say of the sentence "Homo sapiens evolved from australopithecus" that it is true. but what is being said to be true there? the sentence? the proposition expressed by the sentence? the belief? not sure yet. let's keep these questions in mind when we analyze these theories of truth
*truth as ideal (plato) -- what's true is what resembles some ideal of truth. well i think for something to be true it at the very least has to predicate on something in the (one) real world. so this gets thrown out almost immediately.
*truth as correspondence (russell, early wittgenstein, aristotle?) -- something is true iff it corresponds to the facts or the real world or something of the sort. i think there's something to this, although most of the time when expressed it makes you commit to some pretty insane metaphysics. logical atomism is the most extreme case here and it's just tough to swallow that there're pretty much two worlds matching in structure exactly, just one in the physical world and one in the logical world. again, seems a bit crazy if you adjust the theory with the metaphysics. the bad part is that if you don't commit to crazy metaphysics, the correspondence aspect of the theory seems kinda trivial, and doesn't make much of a difference. still, seems to be on to something.
*truth as semantic (tarski, davidson?) -- there are several ways to phrase this, but i'm gonna put it like this: "R" is true iff S, & R names S. in english: "snow is white" is true iff snow is white. this theory would run into a lot of problems except for the fact that he includes the addendum that R must be a name for S. anyway, this seems to agree a bit with what i was saying... in fact it's not far from it, really, although i was mentioning it in terms of beliefs being true. i'm not quite sure what "true" predicates over here, but i'll say this sounds about right.
*truth as disquotational/deflationary/redundancy (ramsey, quine?) -- seems very similar to the last one. ramsey said that for all p, it is true that p iff p. e.g. it is true that convertibles are a type of car iff convertibles are a type of car. this is what i believe to be the most harmless of the theories: it doesn't force you to adopt any outlandish metaphysics, it is still connected to the real world through real kinds, and it retains the concept of realness being an objective quality some things have while other things don't. possibly, the nonrealists and/or nominalists out there will still object due to the use of real kinds. [i might be running together theories that don't really belong... if you think i am then let me know]
*truth as final opinion (peirce) - quick overview of peirce: he used the pragmatic maxim, which stated that the meaning of x is the practical consequences x has. so to peirce, truth is that which will keep being affirmed by a hypothetically infinite (timewise) community of inquirers (coi). so if this coi were to say "dinosaurs became extinct due to a meteor" at some time far into the future, and there is little to no undermining evidence, then that is the truth. this kind of view compensates for the fallibility of the coi at any point before that which they gave their final opinion.
*truth as cash value (james) -- true is good in the way of belief. a true belief will benefit most. what's true is what's expedient in our thinking. these three formulations of vaguely the same idea seem to be interpretable in at least two ways: short-run and long-run. in the short term, it does seem to me that certain false beliefs could benefit more than true ones, making this theory contradictory. but in the long run the truth will be more useful than the false... at least to people seeking to understand reality. however, both in the short and the long run, it does seem to me that as a general rule truth is more valuable than falsity.
*truth as coherence (hempel, putnam, davidson?) - truth is what coheres to a set of logical statements. obvious objections: maybe p coheres with a set, and so does not p. this theory downplays the role the real world has when it comes to truth. it doesn't completely rule it out, but once you try to incorporate some connection between the real world and the coherent set, it starts to sound more like correspondence. eh... again, like the cash value, there does seem to be at least something here. at some ideal point in the future when we have most if not all true statements (statements about the world in some way), then they're bound to cohere, merely due to the fact that there couldn't both BE and NOT BE a real kind, or any certain thing. but this is kinda trivial.
*truth as conversational (rorty, late wittgenstein, foucault, french literary theorists) -- deny that truth is at all objective or even characterizable. all they say is that truth is what's "defensible from all comers" or "withstands all conversational objections" (quoting rorty here). some of these are more extreme than others, but all share a type of relativism that ultimately does away with truth and replaces it with whatever they prefer (tribes, games, power). although i'm convinced that this is wrong, there is a bit you can tease out which might be true. again, if seen from the long run, and if assumed that "truth" is a real predicate, then what happens to be true will be defensible from all comers, precisely because there will be irrefutable evidence (gathered from the real world). likewise, the truth will withstand all conversational objections but precisely because it isn't solely based on conversation, because it's anchored in the real world, somehow.
so.. i think most of these (all but one, actually) have something going for them, but some (probably redundancy and semantic) appeal to me more because they have some real-world element without making me commit to much. did i characterize any of these unfairly? which do you think is closer to the truth??
EDIT (11/1): i think i forgot to mention that i tried to organize the theories in order from making TRUTH a big deal to making it the least big deal.. well denying it. i kinda like ramsey's redundancy and the theories close to it most because it admits truth is out there, but it's not a big deal metaphysically. further, epistemologically speaking, ramsey's theory still allows for there to be an anchoring in the real world which requires inquiry to find out if something is true or not (or at least to gather evidence.)
*truth as ideal (plato) -- what's true is what resembles some ideal of truth. well i think for something to be true it at the very least has to predicate on something in the (one) real world. so this gets thrown out almost immediately.
*truth as correspondence (russell, early wittgenstein, aristotle?) -- something is true iff it corresponds to the facts or the real world or something of the sort. i think there's something to this, although most of the time when expressed it makes you commit to some pretty insane metaphysics. logical atomism is the most extreme case here and it's just tough to swallow that there're pretty much two worlds matching in structure exactly, just one in the physical world and one in the logical world. again, seems a bit crazy if you adjust the theory with the metaphysics. the bad part is that if you don't commit to crazy metaphysics, the correspondence aspect of the theory seems kinda trivial, and doesn't make much of a difference. still, seems to be on to something.
*truth as semantic (tarski, davidson?) -- there are several ways to phrase this, but i'm gonna put it like this: "R" is true iff S, & R names S. in english: "snow is white" is true iff snow is white. this theory would run into a lot of problems except for the fact that he includes the addendum that R must be a name for S. anyway, this seems to agree a bit with what i was saying... in fact it's not far from it, really, although i was mentioning it in terms of beliefs being true. i'm not quite sure what "true" predicates over here, but i'll say this sounds about right.
*truth as disquotational/deflationary/redundancy (ramsey, quine?) -- seems very similar to the last one. ramsey said that for all p, it is true that p iff p. e.g. it is true that convertibles are a type of car iff convertibles are a type of car. this is what i believe to be the most harmless of the theories: it doesn't force you to adopt any outlandish metaphysics, it is still connected to the real world through real kinds, and it retains the concept of realness being an objective quality some things have while other things don't. possibly, the nonrealists and/or nominalists out there will still object due to the use of real kinds. [i might be running together theories that don't really belong... if you think i am then let me know]
*truth as final opinion (peirce) - quick overview of peirce: he used the pragmatic maxim, which stated that the meaning of x is the practical consequences x has. so to peirce, truth is that which will keep being affirmed by a hypothetically infinite (timewise) community of inquirers (coi). so if this coi were to say "dinosaurs became extinct due to a meteor" at some time far into the future, and there is little to no undermining evidence, then that is the truth. this kind of view compensates for the fallibility of the coi at any point before that which they gave their final opinion.
*truth as cash value (james) -- true is good in the way of belief. a true belief will benefit most. what's true is what's expedient in our thinking. these three formulations of vaguely the same idea seem to be interpretable in at least two ways: short-run and long-run. in the short term, it does seem to me that certain false beliefs could benefit more than true ones, making this theory contradictory. but in the long run the truth will be more useful than the false... at least to people seeking to understand reality. however, both in the short and the long run, it does seem to me that as a general rule truth is more valuable than falsity.
*truth as coherence (hempel, putnam, davidson?) - truth is what coheres to a set of logical statements. obvious objections: maybe p coheres with a set, and so does not p. this theory downplays the role the real world has when it comes to truth. it doesn't completely rule it out, but once you try to incorporate some connection between the real world and the coherent set, it starts to sound more like correspondence. eh... again, like the cash value, there does seem to be at least something here. at some ideal point in the future when we have most if not all true statements (statements about the world in some way), then they're bound to cohere, merely due to the fact that there couldn't both BE and NOT BE a real kind, or any certain thing. but this is kinda trivial.
*truth as conversational (rorty, late wittgenstein, foucault, french literary theorists) -- deny that truth is at all objective or even characterizable. all they say is that truth is what's "defensible from all comers" or "withstands all conversational objections" (quoting rorty here). some of these are more extreme than others, but all share a type of relativism that ultimately does away with truth and replaces it with whatever they prefer (tribes, games, power). although i'm convinced that this is wrong, there is a bit you can tease out which might be true. again, if seen from the long run, and if assumed that "truth" is a real predicate, then what happens to be true will be defensible from all comers, precisely because there will be irrefutable evidence (gathered from the real world). likewise, the truth will withstand all conversational objections but precisely because it isn't solely based on conversation, because it's anchored in the real world, somehow.
so.. i think most of these (all but one, actually) have something going for them, but some (probably redundancy and semantic) appeal to me more because they have some real-world element without making me commit to much. did i characterize any of these unfairly? which do you think is closer to the truth??
EDIT (11/1): i think i forgot to mention that i tried to organize the theories in order from making TRUTH a big deal to making it the least big deal.. well denying it. i kinda like ramsey's redundancy and the theories close to it most because it admits truth is out there, but it's not a big deal metaphysically. further, epistemologically speaking, ramsey's theory still allows for there to be an anchoring in the real world which requires inquiry to find out if something is true or not (or at least to gather evidence.)
Subscribe to:
Comments (Atom)
