Total Posts:3|Showing Posts:1-3
Jump to topic:

Two Dogmas of Empiricism

dylancatlow
Posts: 12,242
Add as Friend
Challenge to a Debate
Send a Message
4/25/2016 9:59:10 PM
Posted: 7 months ago
Two Dogmas of Empiricism: A Critique

In one of his most influential essays, the philosopher W. V. O Quine set out to debunk what he regarded as the two most pernicious "dogmas" of empiricism. The first of these was the assumption that statements can be divided into two distinct classes: those that are analytic and those that are synthetic. The second was the assumption that all meaningful statements can be converted to logical form. I'll only be focusing on the first.

He starts out by defining the analytic-synthetic distinction. As he understands it, a statement is analytic "if it is true by virtue of meanings and independently of fact," whereas a statement is synthetic if it is true merely because of "accidental matters of fact." His strategy for undermining the validity of this distinction is simply to challenge the tacit assumption that "analyticity" is a well-defined concept to begin with. His main complaint is that all attempts at explaining the term are ultimately circular, which he takes to be a major problem. To be precise, he argues that the notion of "analyticity" crucially relies on the notion of "synonymy," which itself can only be understood in terms of "analyticity".

For Quine, one of the more promising efforts at explaining "synonymy" appeals to the notion of interchangeability, so that two statements are synonymous if and only if they are "interchangeable in all contexts without change of truth value." His first counterargument is that what are normally taken to be synonyms e.g., "bachelor" and "unmarried man" are not interchangeable in this sense. For instance, "bachelor" and "unmarried man" do not have the same number of letters, and one could construct sentences in which the two terms are not interchangeable for that reason. Quine does not spend much time on this argument and gives his opponents an easy way out: a distinction could perhaps be made between linguistic synonymy and cognitive synonymy, so that only the meanings of synonyms are interchangeable. However, Quine is unsure that we possess an understanding of "meaning" and "word" adequate enough for a clear distinction to be drawn between them. As we shall see, Quine's linguistic skepticism runs deep.

Quine then argues that, nevertheless, interchangeability is no assurance of cognitive synonymy, and uses a hypothetical language to demonstrate this point. He conceives of a language in which all of its words refer strictly to things out there in the real world, and calls it an extensional language. Within this language, "any two predicates which agree extensionally (i.e., are true of the same objects) are interchangeable." Now, imagine that there exists a single creature with both a heart and a kidney. Within this hypothetical language the sentences "Creature with a heart" and "Creature with a kidney" are thus interchangeable because they both refer to the same thing; anytime one is talking about "a creature with a heart" one is also talking about "a creature with a kidney." Quine argues that these sentences are not cognitively synonymous yet are interchangeable. Is he right? I would argue no, that in fact the sentences are cognitively synonymous. In a normal language he would be right, but not so in the "world-obsessed" language he constructed. Within his language, the sentence "Creature with a heart" can't be interpreted in the way we intuitively want to interpret it. "Heart" does not refer to some abstract set of criteria by which we could judge whether hypothetical configurations of atoms would qualify as "hearts". Same goes for creature. In fact, "Creature with a heart" is a redundant statement within his language because the meaning of "creature" is totally equivalent to something which has a heart. Thus, both statements simply reduce to "Creatures with the properties of creatures" which are the same.

In any case, Quine's demand that analyticity be definable in a non-recursive fashion lacks any basis, and can't reasonably be met. If all words in a language can only be properly explained by appeal to ever more basic terms, no word in the language can be comprehended by someone who doesn't already possess an understanding of at least some of the words in the language. In other words, if you take a term at random and ask when it means, you must know the meaning of the terms used to define it if the explanation is to make any sense. But if you don't know what those terms mean, then you better understand the terms used to explain those terms, or the terms used to explain those terms, etc. At some point you just need to know what words mean, and this means that from your perspective those words can only be explained in terms of themselves, because if your understanding of the words were dependent on other words, then we're back to the infinite explanatory regress. What's more, the meaning of analyticity must be grasped in terms of itself because of the very nature of the term. If analyticity is explained by other terms in the language, then that is an analytic statement to the effect that analyticity is defined in such-and-such way and that the words employed by the explanation accurately illustrate what this "way" is, which presupposes an understanding of analyticity. In this sense, analyticity is just as fundamental a concept as "meaning". In fact, Quine's arguments apply equally as well to the concept of "meaning" -- all attempts to explain it presuppose it in some sense or another, do they not? Are we therefore to conclude that meaning is also not well-defined? But that would just be to claim that the word "lacks meaning" -- a sentence which, according to Quine, we shouldn't be able to understand.

Why does Quine claim not to understand the term "analyticity"? He does give a reason, but it's pretty unconvincing. He says: "I do not know whether the statement 'Everything green is extended' is analytic. Now does my indecision over this example really betray an incomplete understanding, an incomplete grasp of the "meanings," of 'green' and 'extended'? I think not. The trouble is not with 'green' or 'extended,' but with 'analytic.'" The implication of this argument is that unless someone can decide whether "Everything green is extended" is an analytic statement or not, they have no right to say they understand the term. This is a bit like saying that unless someone can name every number, they don't know what a number is. There's a difference between not knowing what is subsumed within X according to the definitional criteria of X, and not understanding X.
dylancatlow
Posts: 12,242
Add as Friend
Challenge to a Debate
Send a Message
4/25/2016 10:09:55 PM
Posted: 7 months ago
To clarify, "creature with a heart" can't be separated from "creature with a kidney" because they are essentially the same sentence. They are both saying "Creature with a heart and a kidney" just by saying "creature".
sdavio
Posts: 1,798
Add as Friend
Challenge to a Debate
Send a Message
4/26/2016 1:46:45 AM
Posted: 7 months ago
At 4/25/2016 9:59:10 PM, dylancatlow wrote:
In a normal language he would be right, but not so in the "world-obsessed" language he constructed. Within his language, the sentence "Creature with a heart" can't be interpreted in the way we intuitively want to interpret it. "Heart" does not refer to some abstract set of criteria by which we could judge whether hypothetical configurations of atoms would qualify as "hearts". Same goes for creature.

Such an account of an "extensional" language seems to me to end up as a language consisting of either proper nouns, or terms losslessly exchangeable for proper nouns. However, even a proper noun could be seen as an unacceptable "categorization" rather than directly referring to a singular essence. For instance; I see John on Monday, and then see John on Friday, and may not know that they are the same person - basically this is the morning star / evening star issue. All terms must be directly exchangeable, then, to references to objects (or the meanings or essences of objects) which are absolutely present to our understanding in an indivisible way which precludes multiple interpretations. This is where I think Leibniz's identity of indiscernibles comes in, because any two things, by virtue of the fact that they are two and not one, are not absolutely losslessly exchangeable. And if the exchange were *not* lossless, this would simply be another way of saying it's meaningful, that is, could not be arrived at a priori. Therefore you would need to posit, not an exchange of the terms as whole entities, but an exchange of some meaningful essence which forms a substratum of the terms.

There is the written mark or spoken sound of the term, which clearly cannot be a candidate for this essence because it is infected by arbitrariness; obviously, for the fact that there are different languages. Also, two different physical marks are always meaningfully different - since they aren't really conceptual there doesn't seem to even be a grounds for exchange other than resemblance.

We might also say that there is a concept or meaning between the mark and the object itself, which forms the essence which secures their referential relationship. The problem there is that then, we are not accessing the object itself, but we are in some sense accessing our access to the object. This leads to a certain kind of infinite regress, the working out of which formed the whole thrust of modern philosophy up until Kant's total abstraction of the noumenal essence from our understanding.

So, the essence which regulates the exchange of meanings cannot be the physical symbol, nor can it be a mediating concept between the mark and the object. What if we just say, then, that the meaningful essence of a term is the object itself? Then, the sign would not contain its essence, but refer outward to its meaning. But, by accepting a view like this, we would already be on the path to a view like Quine's, because if the meaning of a term is never totally contained within it, then all statements of definition which are not purely repetitive could be informative.
"Logic is the money of the mind." - Karl Marx