The Neolithic Hypothesis

[Miguel Carrasquer Vidal]

X99Lynx at aol.com wrote:
>I don't want to give the impression that much of what you've said about this
>isn't plausible or compelling.  The question I think it does not address is:
>What is the latest that PIE could have emerged from a local homeland and
>expanded to where we find it when the evidence gets direct?  (I.e., written
>evidence)

Given the enormous variations in rate of change, that's not easy
to answer.  We have Mycenaean Greek, Vedic Sanskrit and Hittite
attested c. 1500 BC, and comparing those three, one doesn't get
the impression that the split had been very recent.  Even Greek
and Sanskrit, which are reasonably close to each other within the
whole of IE, are more divergent than any Romance or Slavic
language, maybe roughly comparable to West Germanic and North
Germanic now, so somewhere between 2000 and 3000 years of
separation would be a fair guesstimate.  Give or take, your
mileage may vary, etc.  That would take us back to 3500 ~ 4500
BC, which is a good match for the Kurgan expansions into the
Balkans.  The distance between Hittite and the other two is
doubtlessly greater.  But it's even harder to put a date on that.
One, two millennia more?  My own theories call for a date of 5500
BC.

><<any initial dialect gradients that came into being have been
>destroyed by later language spreads.  Celtic has been largely
>swallowed up by Romance and Germanic, the Slavic and Hungarian
>spreads have replaced whatever gradients there were in Eastern
>Europe with new dialect gradients. >>

>I don't think that is an answer in itself.  Because the later languages you
>mention fall into Mallory's category of "state languages."

Slavic too?

>Whether the exact
>term is right or not, the meaning is that there is a standardizing agent or
>agents that prevents splintering.  Whether it's trade or the school marm or
>the Academie or the Latin grammarian or the mass media or even a dictionary,
>you have a strong force working against splintering and gradients.

Part of the "state" effect is optical.  Dialects still exist, but
people write in the official language.  Of course the Roman
Empire had a unifying effect for quite a long time.  Whether the
rate of change of Latin itself was affected (other than
optically) is a different matter.  I doubt it.  So the "state
effect" means we have less variation in space, but probably the
same variation in time.

>Without this standardizing force, epecially among sedentary populations, like
>sedentary farmers separated by thousands of years from one another, one should
>not even expect PIE to have been PIE a relatively short distance from the
>place of origin.  Especially, if one follows Renfrew's "a few square miles per
>year" rate of expansion

The "wave of expansion" model is just a model.  What I like about
it is that it explains how a language group might have spread
across a whole continent without anybody actually setting out to
do so (no Anatolian farmer said: "let's invade Europe").  But the
model is too imprecise to accurately reproduce what really
happened.  The process was not so gradual and uniform: farming
quickly spread from Anatolia to Greece and the Balkans
(7000-6000), but then the advance completely stopped for more
than a millennium, until a new wave (LBK) spread rapidly from
Hungary across most of temperate Europe (5500-5000).

>In the 5000 years that separates first agriculture from direct evidence of the
>languages of Europe (aside from Mycenaean), there should have been dozens of
>"swallow-uppers" that preceded Celtic or German or Romance.  And these might
>have moved back and forth all across the continent in a way that would put
>many ancestors between the first historical IE languages and PIE.  There might
>be many proto's between Proto-German and PIE.

Yes, I think so.

><<All we can say with confidence is that the number of
>languages has been going down on average since the Paleolithic.>>

>But you know, Mallory's expert on the North American tribes says the exact
>opposite.  That the number of languages and language families actually
>substantially increased over time and with the coming of agriculture.  The
>reason is obvious.  Farming causes stabilization of location and localization
>promotes local diversity in language.  Standardization is only something that
>happens with centralization - a very different event.

I don't know much about agriculture in North America, but I would
have thought that the linguistic map of Northeastern North
America, with wide-spread language families like Algonquian,
Iroquoian and Siouan, shows very little diversity compared with
places like California (hunter-gatherers) or NW Coast (sedentary
salmon fishers).  Isn't that the consequence of a relatively
recent spread of agriculture up the Mississippi, Missouri and
Ohio?  Of course, with time, these more or less sedentary farmers
would have developed as much diversity as the Pacific NW people.
Standardization (in the sense of little divergence) is something
that happens with centralization or with expansion.  Divergence
is the result of decentralization and just time.

>But the unique thing about IE languages is not their diversity but their
>commonality, something that makes the reconstruction of the proto language at
>all plausible.

The unique thing about IE is the amount of data that we have.
There's nothing unique about being able to reconstruct a
proto-language.  Given enough data, we can do that for any group
of languages that stem from a common ancestor.

>I think when you look for the latest possible date for a unity you get closer
>to the truth.  PIE gives evidence of having been a standardized language in
>some way early on.  The kurgans may explain it.  Agriculture doesn't - not by
>itself.  It creates the opposite effect without a standardizing agent.
>If Latin had been PIE, for example, it accomplished a lot of what it did in
>less than 600 years.  And it did it without mandating conversion - in
>comparison for example to the German laws against speaking Wendish in the
>middle ages.  I don't think elite dominance describes the Latin phenomena
>either.  Whatever the Romans did, it seems to be one of the best historical
>model we have for what happened in the days when PIE was just another local
>dialect BUT on its way to turning into "the first ancestor" of a whole new
>family of languages.

I disagree completely.  The mechamism(s) by which PIE and its
daughter languages (Latin excepted) expanded was nothing like the
Roman Empire, and given the time-frame, it couldn't have.  PIE
was not a "standardized language" in any way.  It was just a
language like any other, and it fell apart into different
dialects like any other.

=======================
Miguel Carrasquer Vidal
mcv at wxs.nl
Amsterdam