the meaning of "genetic relationship"

[bwald]

----------------------------Original message----------------------------
Isidore Dyen invites comments on the following suggestion:
 
> There is little doubt that languages
>originated and the question is how. The problem  has a simple structure if
>it is set up properly. The first component is animal cries used as
>signals. The second component is the fact that all natural languages
>are characterized by a phonemic structure. What we need is a scenario (one
>type of hypopthesis) that gets us from a cry-structure to a phonemic
>structure.
 
I think this is a variant of the theory of language origin that Jespersen
called something like the "yowee" theory. (as opposed to the "ding-dong"
and "bow-wow" theories).  The only hypothesis that sticks in my mind with
regard to a transition is the quantum leap one about human neurological
evolution intervening between cries and words/roots/grammar/etc.  Focussing
on phonemic structure reveals the difference between cries as sounds and
language as sounds, but I also think of Martinet's "double articulation"
principle about language.  Cries are generally taken to be emotive, while
vocabulary is taken as symbolic (I hesitate to use the term
"representational").  Therefore, there is also a transition in creation of
meanings that must be taken into account.  It is unclear that cries
correspond in any way to most words/roots, apart from sharing vocalisation.
In view of the relation between gestures and meanings (e.g., beckoning and
dismissive gestures, perhaps referential pointing more generally ), the
same logic leads to consideration of the transition between "purposeful"
non-vocal animal gestures and human gestures that seem to have linguistic
equivalents, e.g., "come here", "go away/keep distant", "look at that",
etc.  To be sure, on the basis of the list of differences between human
language and what is known of animal communication (which I forget, Hockett
lists a bunch), of which referential displacement in human language is most
striking, innatists deny a direct connection.  They would say that
constructing a scenario between cries and roots/words is impossible (not to
deny a small overlap -- because human language can represent anything that
humans can perceive or imagine, including emotive cries).
 
ID goes on:
The scenario would be a lot simpler to construct if the most
>minimal element was meaningful.
 
I remember Swadesh in the 1960s suggesting that roots were initially
constructed as CVC syllables (and that animal cries can be analysed into
such units).  Abstracting from this notion, I suppose the notion of the
syllable would have to evolve prior to phonemic inventory or structure.  I
guess then the margins and nuclei started to be analysed as separate units.
That's as far as I would go with that idea.
 
ID continues:
Since all languqages use syntactic devices
>to reduce the ambiguity of utterances, one can take it for granted that
>syntactic devices developed for organizing different sequences
>of different meaningful (call them) cries.
 
As far as we know (I think), the isolation of "meaningful cries" (anything
like the roots of human languages) for displaced reference is already a
major step for the theory ID hints at.  Somehow, this must be accounted
for.  If not by neurological development which allows the evolution of
something qualitatively different from emotive (or "here-and-now") cries,
then what?  Similarly, syntactic sequencing remains mysterious in origin.
ID's suggestion reminds me of Bickerton's speculative distinction between
"pre-language" and "language", where, in "pre-language", thematic
roles/case relations are not specified, cf. early stages of children's
syntax, and there is a lot of "ambiguity".  Recall Lois Bloom's 1969? study
of "mommy sock" (18 month-old or thereabout), which, according to context
might mean "(look at) mommy's sock", "mommy, gimme a sock", "mommy has a
sock", etc.
 
ID continues:
Somehow phonemes were developed
>out of the mishmash that was going on at that stage. Perhaps you would
>like to take a hand in adding to this scenario, or to construct an absurd
>hypothesis of your own.
 
The efficiency element in constructing roots and words from phonemes is
striking, and parallels the efficiency of using syntax to construct
utterances out of words.  I don't know what to make of that for the origins
of language.  I can only appreciate that somehow the same kinds of logic
are involved on the syntactic and phonotactic levels, and that somehow this
has something to do with "reality", as humans perceive it, not just with
analytical tricks.  I'll leave it to others to propose theories of language
origins.  I'll just consider them to the extent that I can understand them
-- and criticise, if necessary.