Explain how an account of meaning in terms of truth-conditions differs from one in terms of verification-conditions. Which is correct?
'In this essay I shall attempt to outline first Davidson's account of meaning (that a sentence 'P' is true if and only if some fact Y) and second the verificationist's perspective, that the meaning of a sentence arises from the way that the sentence might be verified, and examine whether they really are conflicting theories, or simply examining different aspects of the same concept. I shall argue that for its simplicity Davidson's account may well fair better that the verificationists, though it still needs more work to answer some of the essential questions that he leaves open with his minimalist conception of meaning, especially in light of the large numbers of objections to the verificationalist's account, which seem to have convinced the majority of current logicians of its falsehood.
Davidson argues that the meaning of a sentence of a language is given by describing what would need to be the case 'in the world' for the truth-value of that sentence to be 'true'. His most famous example is that of the claim that ''Snow is white' is true if and only if snow is white'. The debate about Davidson's ideas seem to swing back and forth between his suggestion that this is a useful theory of meaning and his critics arguing that it is simply obvious, and cannot get us anywhere. Since the intuitive view (at least among non-philosophers) seems to be with the critics - they view the claim that ''Snow is white' is true if and only if snow is white' as an utterly apparent fact, and hardly a useful theory, even if it should be taken to be correct - the burden certainly appears to be on Davidson and his supporters to justify why this might be saying anything useful (or, indeed, anything at all).
I would agree, however, with Davidson's position that this is a useful idea, and suggest that his critics are making a mistake by not examining the punctuation of his example sentences closely enough; this is fairly strange, since his theory is evidently going to rest on how the punctuation marks work; I would happily agree with the suggestion that 'Snow is white if and only if snow is white' is obviously and fairly uninteresting, but the theory is about the relationship of the truth of sentences to facts in the world, not simply of facts to themselves. ' 'Snow is white' ' denotes a sentence (which happens to be a sentence of the English language), and is closely tied to rules about pronunciation and letter-forming, and is an example of a well-formed, syntactically valid sentence used by the English, Americans, Australians, etc. for purposes of communication. 'Snow is white', however (with one fewer set of inverted commas), denotes a fact in the world, the fact that snow is white. The difference between these two things is as marked as that Tony Blair is currently Prime Minister whereas 'Tony Blair' is a proper name pronounced in the same way as one would pronounce 'Toe-knee bl'air' and perhaps (depending on our inverted comma conventions) ''Tony Blair'' is a set of symbols of Times New Roman font, intended to represent the name 'Tony Blair' (which in turn represent Tony Blair). The sentence 'Snow is white' is therefore very different to the fact that snow is white, and the question that Davidson is trying to address is just that of how the sentence 'Snow is white' comes to express the fact that snow is white (and indirectly, how the symbols ''Snow is white'' connect to the sentence 'Snow is white').
Consider, for example, the (critics) example of the sentence 'Horses are called 'horses''. This is presented as an obvious fact that everyone will know if they understand the usage of 'are called', but what must be realised is that the first occurrence of the word 'horses' in the sentence is simply pointing at the class of things themselves, in a similar way (or at least the linguist's equivalent way) that one might point one's hand and say 'Those animals over there are called 'horses''. We could just have easily made the claim that 'Blargars are called 'horses'' where the term 'blargars' referred to a particular set (that happens to be the same as horses), or even 'Hesperus is called 'Phosphorus''. This is how it becomes possible to say things like 'Cars were once called 'horseless carriages'' - since the term 'cars' refers to the class of objects, rather than following contemporary usage of the term, which would mean that 'cars' would always simply point at the class of things that people meant when the uttered the word 'cars' (as an invented language - and a highly fiddly language at that - might have us use the term). Davidson, therefore, argues that we should, when formulating a theory of meaning, consider the relationship between a sentence, such as 'Snow is white' and a fact, such as that snow is white. He suggests that we consider what predicate might fit into the formula ''P' is T if and only if Y'. His suggestion is that 'is true' should fit the bill - that ''P' is true if and only if Y'; hence ''Snow is white' is true if and only if snow is white'.
Verificationism, on the other hand, argues that meaning is either analytic (that is, simply the way that the words are defined, in terms of other words we may already understand) or verifiable - prone to some sort of empirical examination, whereby certain instructions could be followed and the results observed to determine whether a sentence is true or not. According to the verificationist, therefore 'Snow is white' either expresses the fact that 'That white stuff that falls from the sky is white', taking it as an analytic fact about what snow is or verifiable (A more likely claim, in my opinion, since we probably mean 'frozen crystallised precipitated water' or something more scientific and specific by the term 'snow', and probably something that would allow us to call a similar black frozen, crystallised precipitated water-based substance snow, in spite of its unusual colour - meaning that it is not simply an analytic fact that snow is white) - that is to say, if we were to look at snow we would have the same colour experience as when we look at other things we call white.
As a result of this definition verificationism appears to deny that certain sentences that might commonly be considered to have meaning do in fact mean anything. Ayer's example is that of the Drogulus; he suggests that it would make no sense to say to someone that there was an invisible, intangible, impotent (and in fact in every way undetectable) monster called a 'Drogulus', and that one was standing right behind them. According to the verificationist this is meaningless, since the supposed existence of this creature has no effect of the world, and in fact makes no difference to anyone whatsoever.
It might seem, therefore, that the verificationist view is not entirely at odds with Davidson's theory, since Davidson argues that meaning is given by specifying the truth-conditions of a sentence, and the verificationist might be said to do the same, simply adding the additional requirement that the conditions must be in terms of empirically verifiable facts (if it is not analytic). This would, in my opinion, be a misreading of what the verificationists are saying. It is not simply that the means of verification enable you to ascertain the truth-conditions for a particular sentence, and thus the meaning, but according to the verificationist the means of verification is the meaning. So 'Snow is white' is meaningful because we have a means by which to verify it - going out and looking to see if it white - but this does not need to match up truth-conditions to the possible discoveries if we were to actually go out and see. Verificationism, therefore, says that a means of verifying a sentence is all that is needed for that sentence to be meaningful, and is entirely agnostic to what the truth-conditions are. It is exactly the sort of sentences that Ayer is objecting to that cause a rift between these two different theories of meaning. According to Davidson the sentence 'There is a Drogulus behind you' is true if and only if there is a Drogulus behind you (and thus is meaningful), whereas to the verificationist this sentence is meaningless, because we cannot specify a means to verify this claim - since we've stated that there would be no means for anyone to detect such a creature.
It seems to be the case, then, that you have to do a lot more work to convince a verificationist that a sentence is meaningful than you would Davidson. Davidson believes that if you have a sentence that is syntactically valid and a dictionary outlining the meaning of each of the words - i.e. you are able to apply semantics to the sentence - the sentence will be meaningful. A verificationist, however, would argue that even once you have a sentence that is syntactically valid, and you have a dictionary of the words, you still may not be able to say whether the sentence is meaningful or not. For Davidson, if word A holds the same position in the language as word B does, then in any sentence in which A appears, B can appear in just the same place and the sentence should still be meaningful (though clearly may well have different truth-values), whilst for verificationism this is certainly not the case. The sentence 'There is a Drogulus behind you' is meaningful for Davidson in just the same way that the sentence 'There is a cat behind you' is, yet the latter sentence is no proof at all of this for the verificationist. Since Davidson's account is very much intended to simplify language, by arguing that we are able to build up sentences from known component parts using the principle of compositionality, this seems to be very much in his favour. A verificationist definitely has a problem trying to explain why a sentence appears to be meaningful, when if fact it is not, but the problem this poses for communication seems just as great. If the verificationist is correct and every time I hear a sentence I have to evaluate, through an understanding of how I might go about checking if it were true, whether the sentence is meaningful I would be unable to compare it to previous sentences and thus understand a new sentence that I have never heard before. The only escape from this problem is for the verificationists to retreat to the position that meaningless sentences can have a recognisably similar structure to meaningful ones; luckily - with the verificationist theory of meaning, which divorces semantics from pure word definitions - this seems to be a tenable position to hold.
Overall, therefore, I suspect that Davidson's theory is satisfactory as a theory of meaning, if only for its simplicity, yet it will force many questions that people might like answered, such as whether it will ever be possible to say if a sentence is true or false, one stage further back. Since his theory is entirely agnostic as to questions such as these, it might well be possible to build them onto his foundations. Verificationists, however, with their stricter requirements with respect to meaningfulness, will always have a harder time convincing people that they are correct. I am not convinced that the verificationist stance is satisfactory, when we take into account the need for communicability in a language - to require that we analyse the sentence in terms of verifiability before we can tell whether the sentence is even meaningful seems a little strange, and Ayer simply arguing 'There's meaning and there's meaning' does not seem entirely satisfactory.'
is true if and only if
In this essay I shall attempt to outline first Davidson's account of meaning (that a sentence 'P' is true if and only if some fact Y) and second the verificationist's perspective, that the meaning of a sentence arises from the way that the sentence might be verified, and examine whether they really are conflicting theories, or simply examining different aspects of the same concept. I shall argue that for its simplicity Davidson's account may well fair better that the verificationists, though it still needs more work to answer some of the essential questions that he leaves open with his minimalist conception of meaning, especially in light of the large numbers of objections to the verificationalist's account, which seem to have convinced the majority of current logicians of its falsehood.
Davidson argues that the meaning of a sentence of a language is given by describing what would need to be the case 'in the world' for the truth-value of that sentence to be 'true'. His most famous example is that of the claim that ''Snow is white' is true if and only if snow is white'. The debate about Davidson's ideas seem to swing back and forth between his suggestion that this is a useful theory of meaning and his critics arguing that it is simply obvious, and cannot get us anywhere. Since the intuitive view (at least among non-philosophers) seems to be with the critics - they view the claim that ''Snow is white' is true if and only if snow is white' as an utterly apparent fact, and hardly a useful theory, even if it should be taken to be correct - the burden certainly appears to be on Davidson and his supporters to justify why this might be saying anything useful (or, indeed, anything at all).
I would agree, however, with Davidson's position that this is a useful idea, and suggest that his critics are making a mistake by not examining the punctuation of his example sentences closely enough; this is fairly strange, since his theory is evidently going to rest on how the punctuation marks work; I would happily agree with the suggestion that 'Snow is white if and only if snow is white' is obviously and fairly uninteresting, but the theory is about the relationship of the truth of sentences to facts in the world, not simply of facts to themselves. ' 'Snow is white' ' denotes a sentence (which happens to be a sentence of the English language), and is closely tied to rules about pronunciation and letter-forming, and is an example of a well-formed, syntactically valid sentence used by the English, Americans, Australians, etc. for purposes of communication. 'Snow is white', however (with one fewer set of inverted commas), denotes a fact in the world, the fact that snow is white. The difference between these two things is as marked as that Tony Blair is currently Prime Minister whereas 'Tony Blair' is a proper name pronounced in the same way as one would pronounce 'Toe-knee bl'air' and perhaps (depending on our inverted comma conventions) ''Tony Blair'' is a set of symbols of Times New Roman font, intended to represent the name 'Tony Blair' (which in turn represent Tony Blair). The sentence 'Snow is white' is therefore very different to the fact that snow is white, and the question that Davidson is trying to address is just that of how the sentence 'Snow is white' comes to express the fact that snow is white (and indirectly, how the symbols ''Snow is white'' connect to the sentence 'Snow is white').
Consider, for example, the (critics) example of the sentence 'Horses are called 'horses''. This is presented as an obvious fact that everyone will know if they understand the usage of 'are called', but what must be realised is that the first occurrence of the word 'horses' in the sentence is simply pointing at the class of things themselves, in a similar way (or at least the linguist's equivalent way) that one might point one's hand and say 'Those animals over there are called 'horses''. We could just have easily made the claim that 'Blargars are called 'horses'' where the term 'blargars' referred to a particular set (that happens to be the same as horses), or even 'Hesperus is called 'Phosphorus''. This is how it becomes possible to say things like 'Cars were once called 'horseless carriages'' - since the term 'cars' refers to the class of objects, rather than following contemporary usage of the term, which would mean that 'cars' would always simply point at the class of things that people meant when the uttered the word 'cars' (as an invented language - and a highly fiddly language at that - might have us use the term). Davidson, therefore, argues that we should, when formulating a theory of meaning, consider the relationship between a sentence, such as 'Snow is white' and a fact, such as that snow is white. He suggests that we consider what predicate might fit into the formula ''P' is T if and only if Y'. His suggestion is that 'is true' should fit the bill - that ''P' is true if and only if Y'; hence ''Snow is white' is true if and only if snow is white'.
Verificationism, on the other hand, argues that meaning is either analytic (that is, simply the way that the words are defined, in terms of other words we may already understand) or verifiable - prone to some sort of empirical examination, whereby certain instructions could be followed and the results observed to determine whether a sentence is true or not. According to the verificationist, therefore 'Snow is white' either expresses the fact that 'That white stuff that falls from the sky is white', taking it as an analytic fact about what snow is or verifiable (A more likely claim, in my opinion, since we probably mean 'frozen crystallised precipitated water' or something more scientific and specific by the term 'snow', and probably something that would allow us to call a similar black frozen, crystallised precipitated water-based substance snow, in spite of its unusual colour - meaning that it is not simply an analytic fact that snow is white.) - that is to say, if we were to look at snow we would have the same colour experience as when we look at other things we call white.
As a result of this definition verificationism appears to deny that certain sentences that might commonly be considered to have meaning do in fact mean anything. Ayer's example is that of the Drogulus; he suggests that it would make no sense to say to someone that there was an invisible, intangible, impotent (and in fact in every way undetectable) monster called a 'Drogulus', and that one was standing right behind them. According to the verificationist this is meaningless, since the supposed existence of this creature has no effect of the world, and in fact makes no difference to anyone whatsoever.
It might seem, therefore, that the verificationist view is not entirely at odds with Davidson's theory, since Davidson argues that meaning is given by specifying the truth-conditions of a sentence, and the verificationist might be said to do the same, simply adding the additional requirement that the conditions must be in terms of empirically verifiable facts (if it is not analytic). This would, in my opinion, be a misreading of what the verificationists are saying. It is not simply that the means of verification enable you to ascertain the truth-conditions for a particular sentence, and thus the meaning, but according to the verificationist the means of verification is the meaning. So 'Snow is white' is meaningful because we have a means by which to verify it - going out and looking to see if it white - but this does not need to match up truth-conditions to the possible discoveries if we were to actually go out and see. Verificationism, therefore, says that a means of verifying a sentence is all that is needed for that sentence to be meaningful, and is entirely agnostic to what the truth-conditions are. It is exactly the sort of sentences that Ayer is objecting to that cause a rift between these two different theories of meaning. According to Davidson the sentence 'There is a Drogulus behind you' is true if and only if there is a Drogulus behind you (and thus is meaningful), whereas to the verificationist this sentence is meaningless, because we cannot specify a means to verify this claim - since we've stated that there would be no means for anyone to detect such a creature.
It seems to be the case, then, that you have to do a lot more work to convince a verificationist that a sentence is meaningful than you would Davidson. Davidson believes that if you have a sentence that is syntactically valid and a dictionary outlining the meaning of each of the words - i.e. you are able to apply semantics to the sentence - the sentence will be meaningful. A verificationist, however, would argue that even once you have a sentence that is syntactically valid, and you have a dictionary of the words, you still may not be able to say whether the sentence is meaningful or not. For Davidson, if word A holds the same position in the language as word B does, then in any sentence in which A appears, B can appear in just the same place and the sentence should still be meaningful (though clearly may well have different truth-values), whilst for verificationism this is certainly not the case. The sentence 'There is a Drogulus behind you' is meaningful for Davidson in just the same way that the sentence 'There is a cat behind you' is, yet the latter sentence is no proof at all of this for the verificationist. Since Davidson's account is very much intended to simplify language, by arguing that we are able to build up sentences from known component parts using the principle of compositionality, this seems to be very much in his favour. A verificationist definitely has a problem trying to explain why a sentence appears to be meaningful, when if fact it is not, but the problem this poses for communication seems just as great. If the verificationist is correct and every time I hear a sentence I have to evaluate, through an understanding of how I might go about checking if it were true, whether the sentence is meaningful I would be unable to compare it to previous sentences and thus understand a new sentence that I have never heard before. The only escape from this problem is for the verificationists to retreat to the position that meaningless sentences can have a recognisably similar structure to meaningful ones; luckily - with the verificationist theory of meaning, which divorces semantics from pure word definitions - this seems to be a tenable position to hold.
Overall, therefore, I suspect that Davidson's theory is satisfactory as a theory of meaning, if only for its simplicity, yet it will force many questions that people might like answered, such as whether it will ever be possible to say if a sentence is true or false, one stage further back. Since his theory is entirely agnostic as to questions such as these, it might well be possible to build them onto his foundations. Verificationists, however, with their stricter requirements with respect to meaningfulness, will always have a harder time convincing people that they are correct. I am not convinced that the verificationist stance is satisfactory, when we take into account the need for communicability in a language - to require that we analyse the sentence in terms of verifiability before we can tell whether the sentence is even meaningful seems a little strange, and Ayer simply arguing 'There's meaning and there's meaning' does not seem entirely satisfactory.