# Objection to Davidson's truth-theory of meaning

Hello,

Was just wondering of what some of you would make of the objection (“the extension problem”) to Davidsons truth-theory of meaning:

Davidsons argument (brief): We can construct for every utterance of a language a “T-sentence” which will allow us to deduce under what conditions that utterance would be true. For example:

(T) ‘Snow is white’(s) is T iff snow is white(p)

Thus Davidsons concludes: A truth-theory essentially is a meaning-theory.

The objection of :the problem of extension: Such a theory of meaning overgenerates T-sentences for a single given utterance. Such as:

(T1) ‘Snow is white’(s) is T iff snow is white(p)
(T2) ‘Snow is white’(s) is T iff grass is green(p)

And whereas (T1) is interpretive (one has understood the meaning of the term) (T2) is uninterpretive - one has not understood the meaning of the term. Now we cannot tell for certain what “snow is white” means by it’s truth conditions! We cannot tell them apart because both sides of the biconditional are true: ‘Snow is white’ is true and so is ‘grass is green’. (T1) and (T2) are equally true - but obviously (T2) is false. Thus Davidson is wrong.

In my oppinion this could prove fatal. Thoughts anyone?

Thanks

But I have no idea of what you are questioning or where you got your “T2” or what it has to do with anything.
It all seems to be much to do about nothing… ?

Hello James,

T1 and T2 are equivalent in truth-value. That is:

(T1) ‘Snow is white’(s) is T iff snow is white(p)

Is true - because the meaning of “snow is white”(s) is dependent on the truth-value of of it’s truth-conditions(p). So aslong as p makes ‘snow is white’ true, we can substitute it with another true truth-condition, such as:

(T2) ‘Snow is white’(s) is T iff grass is green(p)

This is relevant because obviously (T2) is false, but it is equivalent in truth-value to (T1): so we haven’t really learnt anything at all by the utterance. (Ergo - Davidson is wrong).

My question is: Views on my lines of reasoning? (Naturally I think I’m right)

The wonderful world of linguistic philosophy .

Thanks

There is a significant difference between T1 and T2 in that p in T1 directly associates to the stated p`. In T2, the p has no association to the p`, thus constitutes a nonsequiter.

Logic is the recognition of associated conceptual definitions.

Yes, and in Davidsons notion of Radical Interpretation highlights than if an agent alien to the language was to hear both - he would not be able to discern this.

We only understand the difference because we already know english. Non-english speakers woulden’t know this.

Yes, and in Davidsons notion of Radical Interpretation highlights than if an agent alien to the language was to hear both - he would not be able to discern this.

We only understand the difference because we already know english. Non-english speakers woulden’t know this.
[/quote]
I don’t think there’s a single theory of truth that doesn’t depend on knowing the meaning of the statement to be analysed. I don’t see why this is an argument particularly against Davidson, though.

Hello Only_Humean,

According to Davidson this is exactly the case (to a large extent i agree with you). Again, his T theory of meaning:

S is T iff P

“Schnee is weiss” is True if and only if schnee is weiss (“Snow is white”) - regardless of whether one know german or not.

This is because the meanings are derived from the truth value of state of affairs, not individual meanings of terms or prior linguistic knowledge.

A language could be constructed such that it was determinable that statements were true without knowing what they meant, but it would be a cumbersome language. One could even construct a language that got garbled and yet automatically corrected. But what’s the point?

If he was working on the whole “talk to aliens” thing, realize that aliens can’t get away from logic either.

“whats the point” is no argument against a theory of meaning. Neither is it being cumbersome. Some of us don’t choose to be nihilistic.

Well I think your undercutting your own argument. This is very similar to what Davidson is trying to say; but you’ve already considered this cumbersome and superfluous.

I was kind of hoping someone would have thoughts on my lines of reasoning and not why Davidson isn’t relevant lol. The latter isn’t a widely held view at all.

Well, “widely held” is a bit meaningless around here. I apologize for the distraction to insignificance (that is also a bit meaningless around here). Perhaps you can tell us why Davidson’s concerns are so relevant?

And by “cumbersome”, I mean “impractical”. Perhaps you can tell of why it is important to go to the trouble of enduring impracticality on this issue?

I don’t think the last sentence follows. The truth value is dependent on the state of affairs, more or less. I can say “Santa is a fat bastard” and “Santa is a skinny angel” and both are meaningful, so now what?

hello captaincrunk,

Yes, and they are both false. Whereas “Santa is an imaginary person” is true, and the meanings differ.

Again - whether or not we know the meaning of the sentence.

I think Davidson addresses this objection in one of his essays. I can’t remember where, but it was in The Essential Davidson. The one problem I remember having with Davidson is idioms, metaphors, questions. He addresses some of this in What Metaphors Mean, but I was unimpressed with it (can’t say why right now though). I mean, if I’m at the grocery store and I ask the cashier, “Hey, how you doin?” she understands what I’m asking, even though no truth conditions can be given for it. She knows I don’t care about how she’s doing. I’m just saying something like, “I acknowledge your existence.” There’s meaning there that’s grasped without truth conditions.

in other words, it’s a truth theory of truth, and it really has nothing at all to do with meaning, and this guy is an idiot for including the word meaning in his work?

either that or you’re failing to explain it

Davidson wasn’t writing a dictionary. He was saying ‘this is what the nature of meaning is’, not telling us how we could discover what specific words mean. As OH points out, he admitted that you would need to understand the sentence before you found the truth condition.

In philosophy of language, there is this ‘dilemma’: everybody knows and understands what words ‘mean’, but no one seems to know what ‘meaning’ is. This is his solution to that ‘dilemma’. He isn’t solving the first part: we all know what words mean already, and how we know what words mean is just not Davidson’s issue.

We can easily tell them apart because these two have the same truth value in our world but they have different truth conditions. Davidson said the meaning of a sentence was its truth condition not its truth value. There is a possible world in which ‘snow is white’ is true but grass is not green. However, there is no possible world in which ‘snow is white’ is true but snow is not white. T1 and T2 don’t have the same truth conditions, thus according to Davidson’s theory they are two different transcriptions of the one statements meaning. We can use our linguistic knowledge to say that the first is the correct transcription.

Further note that you do not need to know a statement’s truth value to know its truth condition. You know the truth condition of ‘brevel_monkey is sitting in bed’, but you do not know its truth value.

The real problem with Davidson’s theory is that much of language is not truth apt. ‘What’s that thing up there?’, for example, is not truth apt. ‘Red Riding Hood went to visit her grandma’ is also very difficult to create a truth condition for. As is ‘I would have done it if you hadn’t have changed your mind’. In fact, the amount of language which is not truth apt is pretty large. Yet none of these sentences are meaningless. After Davidson, philosophers of language spent a lot of time trying to figure out how they could translate all such statements into truth conditions. But they couldn’t make it. Because some sentences don’t have truth conditions. Because Davidson was wrong about the nature of meaning.