Cladistic language concepts

bwald bwald at HUMnet.UCLA.EDU
Sat Sep 5 19:08:42 UTC 1998


----------------------------Original message----------------------------
MG writes:
 
>               (I'm sorry I don't know your name, only your address!)
 
        Benji Wald.
 
>               Before reading your message I sent a long one to Steven
>          Schaufele about the relationship in question.  It may have
>          helped.  And if linguists are interested in that sort of
>          issue I am very pleased to be able to discuss such matters
>          with them.
 
I was relieved that Steven and some others gave you Saussure's position on
"langue" and "parole", as it relates to your question about the general
view of linguists on the issue of the relation between speaker and
"language".  I didn't want to go into the detail and various
interpretations.
 
>               You are quite right about the biologists' problem with
>          "life" and the linguists' problem with "language."  These
>          entities in the broadest sense originated once and only once
>          in the history of the world.  All organisms and lineages
>          thereof with which we are familiar have a common ancestor,
>          and the capacity for language is the property of a single
>          species.  Life in that sense and language in that sense are
>          individuals.
 
It remains moot whether all languages have a common ancestor.  However, if
we take the concept of "language" at its most fundamental, as the "innate"
(pre-programmed) ability to acquire any human language that has ever
existed as a first language, I guess there's no harm in making the
assumption that that ability derives from a crucial innovation in the
hominid line of evolution shared by all subsequent "humans" (whether or not
that innovation was limited to language, as innatists assume, or had a more
general nature).  We linguists have argued amongst ourselves whether that
necessarily means that *implementation* of that ability by creating
observable languages happened only once or several times (within a short
period of time, let's say, in different location) -- something like Newton
and Leibniz coming up with the calculus more or less independently (let's
give Leibniz some credit, although not as much as he felt forced to claim,
given that his knowledge of some of Newton's early work probably helped him
develop what Newton concealed -- and even in a form that has been accepted
as more convenient than Newton's).
 
NB. the difference between "ability" and "implementation" of that ability
seems clear, since, if I understand correctly, humans sociopathically
isolated at birth from normal social contact with other humans, have the
ability but don't implement it -- and, due to the relation of innate
language ability to various other maturational processes -- cannot
implement that ability to the extent that people normally socialised do, if
they are exposed to implementation (by others) after a certain (critical)
period of organic development.
 
A lot of their properties must be due to
>          historical accident, or contingency.
 
Because of what I said above, that is contentious.  However, most linguists
are aware of this issue and have to deal with it one way or other in
seeking linguistic universals.  Generally, they have no choice but to go on
the basis of empirical data.  If we can't agree on whether all observable
*particular* languages descend from one original *particular* language,
then we can't agree whether or not a common descent has limited the
considerable variety we do observe from a much greater *possible* variety.
Some linguists might even be inconsistent by supposing (even hoping -- for
whatever reason) that all languages are ultimately descended from some
*particular* original language, but by still insisting that the variety we
observe can safely delimit the possible variation in human language.  It is
not really all that harmful to the field, because the two considerations
never coexist in the same research project.  Anyway, strong hypotheses
about universals on the basis of observable language variation has proven
challenging and vulnerable enough, without having to worry about whether
possible but unobserved/observable languages would disprove them.
 
NB.  Virtually all linguists are excited when a "new" language is
"discovered".  However, that hardly ever happens anymore, as opposed to
dissemination of relevant information about a language which was already
known to some experts but had not yet become a matter of general knowledge
among linguists.  Linguists are much more concerned about the continuing
disappearance/cessation of languages that have not yet been (and never can
be) completely mined for their relevance to linguistic theory.
 
If we were able to
>          study similar entities that have evolved on different
>          planets, we might be able to come up with a definition of
>          some class of life-like entities.  Likewise with
>          language-like entities that have no common ancestry.  In
>          other words, defining life or language in that sense is
>          hindered by our familiarity with a class that has but a
>          single instance.
 
Linguists have (probably) always dreamed about this.  Few linguists are
naive enough to seriously expect that "human" language (or the capacity for
it) has evolved anywhere else in the universe -- but that does not hold for
the vaguer notion of "beings with superior intelligence", something that is
hard (actually *impossible*) for us to imagine without it having relevance
to human language.
 
NB.  Since we're dealing with science fiction, let's not worry about what
"superior intelligence" is supposed to mean, or how we could recognise it,
or whether it means "superior TO us" or "superior LIKE us", etc.
 
>               It is not quite that bad, because when we study
>          entities her on earth we find lots of species, and
>          species-like things, which allow us to come up with a
>          generalization that amounts to a definition of classes of
>          such things.
 
        Yeah, like I was saying above.
 
So we can define the class of species, and the
>          class of languages, even though Homo sapiens and French have
>          no definitions strictly speaking.  And we can define a class
>          that includes both the species and the language.  All
>          species and all languages are lineages etc.
 
Generally speaking, that's true for our assumptions about languages.  One
thing I'm not sure about is whether human beings raised in a social context
but without language, say, a set of congenitally deaf children, would
develop spontaneously a (sign) language -- in which case, the language
would have no lineage.  Innatists like Chomsky have suggested such a thing,
and I think some empirical studies of some deaf communities have proposed
such a development -- but I'm not well enough versed in such studies to
know whether or how intervention in such communities of speakers of
languages with lineages queers conclusions about spontaneous implementation
of language abilities under such circumstances.
 
NB.  It's not quite clear to me what the parallelism between prosody in
spoken language and facial gestures in sign language (vs. phonological
structure of words in spoken languages parallel to hand gestures in sign
languages) suggests about such issues.   (What does the difference between
word order in adjacent spoken languages and sign order in sign language
suggest?  I don't know what the conventional wisdom of sign language
experts is for this issue -- or if there is one.)
 
>               I hope this metaphysics is not too tedious, but it
>          shows I think the importance of studying such parallels in
>          clarifying our thinking about such matters.
 
I'm not really reacting to the issues on a metaphysical level, but in terms
of the empirical basis for such speculations.   --  Benji



More information about the Histling mailing list