Fwd: Deconstruction and modern linguistics

[Doug Millison]

Some of you will find this interesting, I think, from the PSYART list today.
-Doug



>Date:         Thu, 14 Feb 2002 12:47:57 -0500
>Reply-To: Institute for Psychological Study of the Arts
>              <PSYART at LISTS.UFL.EDU>
>Sender: Institute for Psychological Study of the Arts
>              <PSYART at LISTS.UFL.EDU>
>From: Norman Holland <nholland at ufl.edu>
>Subject:      Deconstruction and modern linguistics
>To: PSYART at LISTS.UFL.EDU
>
>Hi, gang!
>
>I'm including in this e-mail a quotation from Mark C. Baker, The Atoms of
>Language (Basic, 2001) criticizing the assumptions about language made by
>postmodernists (he takes Derrida as exemplary).  He argues that they are
>inconsistent with modern linguistic theory.  The book is very good, by the
>way,
>a study of the parameters of universal grammar that we know of today.
>
>                        --Best, Norm
>Norm Holland
>nholland at ufl.edu
>
>
>-------------
>       Here we can add a word about the underpinnings of postmodernism as an
>influential contemporary viewpoint. I hasten to note that I am no expert on
>postmodernism and am not qualified to analyze the movement as a whole. One
>does
>not have to be an expert, however, to know that it developed in part out
>of some
>of the midtwentieth-century trends in anthropology and linguistics that we
>have
>been discussing. The postmodern perspective is thus rooted in certain
>assumptions about language that I can comment on. Jacques Derrida in
>particular
>developed his very influential views in reaction /204/ to the Swiss
>structuralist linguist Ferdinand de Saussure. A contemporary of Boaz, Saussure
>is famous (among other things) for emphasizing the arbitrary relationship
>between the sound of a word and its meaning. In itself there is nothing so
>deep
>or remarkable about this observation: He was simply pointing out that
>there was
>nothing special about the sounds d, o, and g that makes their concatenation
>particularly suited to referring to domestic canines. On the contrary, the
>sounds chien (French) or perro (Spanish) or erhar (Mohawk) or ekita (Edo)
>denote
>man's best friend just as effectively. What is crucial is that there be a
>conventional association between sound and meaning that is shared by a
>group of
>speakers. Yet Derrida attaches great metaphoric significance to this
>arbitrariness and conventionality at the roots of language. He also points out
>(correctly) that the exact meaning associated with a sign like dog is hard to
>pin clown. This, too, is no great surprise to anyone who has been asked to
>define a word; it is a tricky task, and no one ever gets it completely right.
>There is a fluidity to how words are used that greatly complicates things.
>On a
>particular occasion, a speaker might refer to something as a dog in a loose or
>metaphorical sense that would not carry over to other times or uses. Now,
>if the
>most basic pieces of language are arbitrary, fluid, and not directly
>related to
>meaning, it is easy to conclude that all of language is like this. This leads
>naturally to the bewildering and ever-shifting postmodernist view of the
>world,
>in which nothing has a lasting or general meaning.
>
>Given the linguistic research presented here, this postmodern fine of
>reasoning
>does not really go through. The error is simply the common one of seeing words
>as the (only) atoms of language. Many of Derrida's observations about language
>are valid at the level of words. For this reason, the area of lexical
>semantics
>is perhaps the most problematic and least-developed area of contemporary
>linguistics. But words are not all there is to language; there is also
>grammar,
>the principles for combining words into sentences. One might think that if
>words
>are arbitrary and fluid in their meanings, larger constructions built out of
>them would be even more arbitrary and fluid because the indeterminacies of
>meaning would compound and magnify /205/ each other. Yet the opposite is true.
>Sentence structures are more rigidly and universally specified than word
>meanings are, as determined by the basic recipe for language together with its
>atomic variations, the parameters. Thus, whereas the lexicons of different
>languages vary widely, their grammars do not.
>
>The associations between form and meaning at the level of phrases and
>sentences
>also tend to be universal and not arbitrary. For example, we have seen that
>there is a broad tendency across languages for words that are related
>semantically to form phrases, so that they end up next to each other.
>Thus, even
>languages as different in their syntax as English and Japanese have
>essentially
>the same patterns of grouping into phrases. There are also universal rules
>that
>say that the agent of an action is expressed as the subject of the verb
>and the
>undergoer of the action is expressed as the object of the verb.
>Furthermore, the
>object of the verb consistently forms a phrase with the verb that does not
>include the subject (the verb-object constraint). Perhaps these relationships
>between form and meaning are ultimately arbitrary and conventional, but
>they are
>not variable. The entire human race seems to be fixed on the same conventions.
>Paradoxically, then, sentence-level form and meaning is much better understood
>than word-level meaning. Moreover, these constraints on sentence structure are
>powerful enough that they help narrow down the meanings of the words in the
>sentences. Although the meaning of a word in isolation is shifty and
>problematic, the same word in the context of a sentence is usually much less
>problematic. In context the word's meaning is constrained by both the sentence
>structure as a whole and the meaning of the other words around it. We can get
>away with flexibility in word meaning precisely because we know that the
>linguistic context will clarify what we intend (except, perhaps, in
>continental
>philosophy texts). The severe difficulties in dictionary writing are not an
>accurate measure of the difficulty of interpreting a sentence. Thus, the
>results
>of current linguistic research tend to refine the assumptions about language
>that lie at the foundations of the postmodernist perspective.
>