Sociological Linguistics

[Nicholas Widdows]

[ moderator re-formatted ]

JoatSimeon said:

>>>> Middle English is not one iota more "effective" at communication than
>>>> Old English.  It's just different.  Languages change because they do.

Patrick Ryan replied:

>>> Everything in life of which we have knowledge shows a development from
>>> the simple to the complex.

Nicholas rebutted this and ended with:

>> They're just different.  Languages change because they do.

Pat, aghast:

> I was with you all the way until this final sentence.
> Is this Zen?

In this famous koan of the Enlightened JoatSimeon-sensei, later commentators
believe he is stating that you don't need overarching principles guiding
language change into any particular direction (such as simple to complex).
Entropy, random drift, local conflicting processes like analogical levelling
and regrammaticalization, will make all languages change, but the change is
local, generation by generation. It's not part of the evolution of human
culture from the pre-human.

I don't think you are with me quite all the way, because our ideas on the
time scale are so different. Let me make it more explicit, with apologies
for doing so in the wrong forum. Then I'll hold my peace.

Like you, I find monogenesis the only reasonable opinion. And I would
certainly _like_ to believe that new statistical analyses can peer further
back, to Nostratic, to Nostratic-Amerind and Dene-Caucasian, and just maybe
the odd word of Proto-World. If monogenesis is true, some such ancestors
necessarily existed, though we might be deceived in which particular modern
families branch from which ancestor. But a common feature of all recorded
languages is that they're all about equally complex. I think there's a
strong consensus on that. [ignore sound of one hand raised in disagreement]
Therefore as a matter of logic it was likely to be a common feature of their
ancestors.

You think you can see much of modern language -- the salient difference
between vervets and Verlaine -- being invented somewhere back near Nostratic
times. If that was say 10 000 years ago, and Nostratic-Amerind must be
pre-Clovis so call it a round 15 000, then where do Aborigines and Andaman
Islanders fit in? Modern humans are known to have lived in Australia 50 000
years ago; there were more results just last week suggesting it could be 70
000. The precise age doesn't matter, because modern Pama-Nyungan languages
have a wide selection of ergatives and instrumentals and causals, past and
future and subjunctive, they don't have weird and wonderful phonology, their
semantic fields are as diverse and as similar as anything from Senegal to
Siberia. In short, there's nothing in Australian languages to suggest that
the speakers branched off three or four times longer ago than we can
possibly trace even the boldest Nostratic, yet the archaeology seems to say
they did.

What I, with many others, conclude from this is that all our present baggage
-- such as the idea of marking the verbs of dependent clauses whether of
French or Bidyara-Gungabula -- arose in its present form in a language
ancestral to the last time we were all one ethnic group, somewhere in Africa
before the division, 50 000 or 100 000 years ago. Complexity can't have
arisen from proto-human simplicity in anything reconstructible from Sumerian
or Egyptian, because they're a scratch on the surface compared to the actual
time depth in which this complexity has been with us all.

Nicholas Widdows