Monogenesis and polygenesis

[bwald]

----------------------------Original message----------------------------
I once wrote something similar to Histling before, but the use of
"polygenesis" reminded me of it.  In pidgin/creole studies, "polygenesis"
usually refers to the descent of a language from more than one natural
language, contrary to the tree concept of language diversification.  In the
current discussion, "polygenesis" is being used in a somewhat different,
but equally valid, way, to suggest that languages evolved "from scratch" in
different places (at different times is beside the point here).
I don't have anything much to add to that discussion.  Except to say that
it seems as plausible as monogenesis in the same context of discussion.
The point may have been made without me recognizing it, but it seems to me
that when whatever neural development evolved that allowed "humans" (for
the sake of argument) to acquire language in the natural way that they now
do, it does not automatically follow that that ability immediately
transformed itself into language as we know it today, particularly with
respect to a core lexicon in a unique homogeneous human community.  That
assumption is essential to our use of lexicon to reconstruct ancestral
languages according to the monogenetic model, and it has no direct bearing
on any "innate" language faculty, since no linguist is likely to claim
(again?) that such a reconstructed lexicon in its sound-meaning
relationships would be fundamentally non-arbitrary.  (If it had a
non-arbitrary lexicon, it seems to me it would be impossible to reconstruct
on the basis of attested languages, and it would certainly not be like
attested languages in that respect, nor can I imagine that its speakers
could be like current humans, pace Herodotus)  So, it seems to me as likely
as not that humans developed much, if not all, of even their "basic"
lexicon in various independent sites/groupings.  I'm not saying that it's
MORE likely than not, just that until we know more about early human social
organizational dynamics, it's JUST AS likely.
 
(Think about it.  If, say, one or a group of current chimpanzees evolved by
quantum leap the capacity for language, syntax or whatever, do you think
they would develop a basic lexicon to exploit that language potential
before the genes spread to other chimpanzee communities?)
 
Now to the point that I really wanted to make.  One might hope that the
above problem can be resolved simply by developing better methods of
lexical reconstruction, so that we can eventually say, aha! we have
succeeded in reducing the lexicons of all languages to enough of a single
reconstructed proto-lexicon to suggest a proto-lexicon which would have
been adequate for a society of language users as we have reason to imagine
it at that time depth.  And it is improbable that we could have done that
if language had a polygenetic origin, such that all other independent
stocks just happened to die out leaving no trace (and no accompanying
problem for monogenetic reconstruction).  (That probability argument could
be debated.)
 
But our ignorance of the social dynamics of early language users encounters
a SECOND stumbling block.  That is polygenesis in the pidgin/creole sense.
While the single ancestry/tree diversification assumption works quite well
for many, even most, languages UP TO A CERTAIN TIME DEPTH, there is still
the possibility that if we work back far enough with improved
reconstructive tools, we will still find that particular sets of languages
for which that assumption works eventually hit a point at which the
assumption of more than one ancestral language is necessary, still quite
short of the point at which "original" polygenesis or monogenesis can be
decided.  In other words, all attempts at reconstructing the ultimate
proto-language by reconstructing further into groups of families and
lumoping them together may eventually hit at one point or another,
"creole"-like languages (in origin) which frustrate further attempts to
continue the same method to reconstruct an original language.  Maybe that
presupposes that in some, or many, instances early independent groups of
humans merged with each other into new cultural formations, including
lexicon, instead of annihilating one or the other. What's wrong with that
presupposition on the basis of what we know?
I think we at least know enough to discuss it, if not to resolve it. It
might at least be therapeutic to anticipate what to do if such a problem
arises, instead of trying to squeeze more ink out of the dried up ballpoint
pen of the monogenetic/tree theory of language diversification (which has
its limits even in subgrouping within well established "monogenetic"
families).
 
That was something like what I first thought a message containing
"polygenesis" in the title might suggest -- but since it wasn't, I thought
I'd throw that consideration into the brew.