Quine “Mind and Verbal Dispositions”

From Moore’s Intro:   Quine’s argument against the existence of senses (and thus against the identification of meaning with sense):   Any two competing assignments of sense to a segment of linguistic behavior (like the uttering of a particular word in a particular situation) are undetermined by the physical facts. (That is, the physical evidence supports equally both of any two contradictory theories about what the meaning of an utterance is.) Thus meaning (sense) is indeterminate.  But to assume senses is to assume determinacy, thus, since there is nothing determinate to answer to the name “sense” there is no such thing as sense.

Why, according to Quine, does the assumption of sense require assuming that sense are determinate? Moore, in his intro, says there are two reasons, but what he goes on to say is quite useless.  

Turning to Quine’s article:   Quine gives one tentative argument against sense (note, in Quine’s article, sense = meaning): On p. 83.  Meanings are supposed to explain the understanding and equivalence of expressions, but they do no such thing. We could just as well replace talking of “knowing the meaning, and of giving the meaning, and of sameness of meaning” in favor of “talk of understanding an expression, or talk of the equivalence of expressions and the paraphrasing of expressions.   Continuing Quine’s project: to show how the understanding of declarative sentences can be done in terms of behavioral disposition.  

(Why not physiological? Probably because Quine thinks that internal physiological synchronicities between speakers is unnecessary for communication.  Only conformity of words, behavior, and observable (distal) stimuli are necessary. Compare Quine’s remarks on subjective colour inversions on  p. 81.)  

First step, p. 84: acknowledge that understanding a declarative sentence involves understaning it’s truth conditions.  Thus, one understands “Neige est blanc” only if one understands that it would be true if and only if snow is white. But how to account for this sort of thing, involving understanding and knowledge, in terms of behavioral dispositions?  

The answer is to think of the dispositions one has to assent or dissent when queried, for instance “Is this red?” in the presence or red objects.  So, suppose we met someone (Call him Bruce) who said “Flizle bizzle” (an utterance in a language unknown to us) and we were trying to figure out what the truth conditions of “Flizle bizzle” were.  We would point to various objects with various properties in the presence of Bruce, and ask “Flizzle bizzle” suppose that we found that Bruce smiled and behaved in an assenting way when and only when we were pointing to an orange golf ball while asking “Fizle bizzle?’, and shook his head and otherwise behaved dissentingly when queried in any other circumstances. Would we have thereby determined the meaning of “Fizle bizle?” . . . .????  

(Note that Quine never tells us how to be sure when an alien speaker is assenting or dissenting.  But what counts as assent is as much a question of interpretation as anything else. . .)  

So far, the strategy works ok for occasion sentences (a particular utterance of e.g. “this is red”) as opposed to standing sentences (utterance types of which an utterance “this is red” is a particular instance).  It won’t work for standing sentences, according to Quine. “ . . .we cannot hope to correlate standing sentences generally with observations, because the sentences one by one simply do not have their own separable empirical implications.  A multiplicity of standing sentences will interlock, rather, as a theory; and an observation in conflict with that theory may be accomidated by revoking one or other of the sentences—no one sentence in particular” pp 85-86.  

Here Quine is expressing his version of semantic holism (the opposite of semantic atomism): the view that items have their semantic properties only in virtue of the relations that they bear to other similar items. So, for instance, the meaning of a statement like “snow is white” depends on its relations to other statements assented to by users of the language in which it is an item.  Imagine another group of language speakers who speak something very similar sounding to English: call it Zinglish.  Suppose we show a Zinglish speaker and an English speaker some snow, and they both said “That is snow”.  Suppose also that, unlike the English speaker, the Zinglish speaker would deny the truth of “snow is frozen water” and assent to the statement “all things below 50 degrees Ferenheight are snow”.  Would you say, then, that Zinglish speakers mean something different from English speakers by “That is snow” in the presence of snow? If so, then you probably are a semantic holist.