the Wheel and Dating PIE

[X99Lynx at aol.com]

I wrote:
<<I'm beginning to suspect that the 4000BC "last date of PIE unity" is pretty
much a linguistic conclusion and - be it right or wrong - the material
evidence does not especially favor that date versus an earlier one.>>

Sean Crist replied (dated 10/28/99 2:09:32AM):
<<It's quite true that the reconstructed IE vocabulary puts certain
constraints on the possible dates for the final IE unity.>>

But that is not what I said and I don't believe it's true.  Reconstructions
are relative dates - there's no constraints imposed by the linguistic
evidence alone. WHAT I SAID was that the archaeology will not necessarily
support the dating - not that reconstructions put any constraint on the
objective findings.

Sean Crist continued:
<<For example, we reconstruct "wheel" for the PIE lexicon.  We conclude that
the speakers of PIE at its final unity belonged to a culture acquainted with
the wheel.>>

The dating of the wheel is NOT linguistic evidence.  This cannot be
understated.  If there's supposed to be some CONCLUSION ABOUT DATES drawn
from the commonness of words for the wheel, it is most definitely NOT
primarily based on linguistic evidence.  It MUST BE BASED on the earliest
dates given to the wheel by archaeology.  (Unless of course you are referring
to lexicostatistical analysis, which I'm pretty sure is not the case.)

<<Dating the final PIE unity to 7000 BCE is therefore quite inacceptible,
because the wheel is not attested in the material record until much
later....>>

"Quite incceptible" is quite unacceptible.  There's no reason to say that
this is  finally conclusive or not questionable or anything more than
tenative.  There's actually very little to be sure about here.

Starting with the linguistic side - very clearly, there are cases where
technological and cultural innovations carry their own terminology with them
and enter different languages with a common name long after those languages
have separated.  (In some cases, they have even been conformed after the fact
to the local sound rules.)  IE speakers could have become "acquainted" with
the wheel after they separated and adopted the traveling wagoneer's,
wheelwright's or merchant's word for the item.

And let me question whether the universal shared character of the attested
words for the wheel is even accurate.

What was the Hittite word for wheel?

Awhiles back, Miguel Carrasquer Vidal wrote (3/9/99 12:12:33 AM):
<<The other words related to horse technology yield no Hittite
cognates (Hittite "wheel" is not *kwekwlo- or *rotHo- but
<hurki>, related only to Tocharian <wa"rka"nt> "circle, wheel"),
except for two curious items: "shaft/pole", Hittite hissa ~ Skt.
i:s.a:, Grk. oie:ks, Slav. oje(s)- and "(to) harness", Hittite
turiia- ~ Skt. dhu:r-,...>>

And, BTW, how does Greek or Mycenaean jive with the statement that wheel has
a shared form in all IE languages?

Miguel wrote to me:
<<AFAIK, "horse" in Mycenaean is i-qo (Class. Greek hippos), not a regular
reflex of PIE *ek^wos.  "Wheel" is a-mo (Cl. Greek harma "chariot").>>

And a simple reading of the Illiad (as Chapman noted a long time ago and Buck
gave a nod to) makes it rather clear that Homer's specific word for a wheel,
a chariot wheel, a potter's wheel, wheel tracks and a spinning wheel is
<trochos>.  What is equally pretty clear is that <kuklos> refers not
specifically to a wheel, but to anything circular, including a circle of
counselors or the walls surrounding a town.

If <kuklos> did in fact originally refer to a circle (rather than
specifically to a wheel) then there is no surprise that the word would trace
back to PIE long before the wheel - and the fact of the wheel's introduction
would only reflect a later shift in semantic meaning of <kuklos>.  <*rotHo>,
the other supposed IE universal, does not even approach the meaning "wheel"
in Homer.

So, it appears that most of the words for the wheel that are attested or show
a clear path from the 2d millenium BC (Hittite, Tocharian, Mycenaean, Homeric
Greek) don't give much support at all to the universality of wheel
terminology.  Or therefore to the 4000BC dating.

So how is it that such a claim keeps being made?  Is it because all these
early languages "lost" the PIE form?  Or are we assuming that PIE MUST have
had a word for the wheel - based on what?

And how will this rigid theory - about the correlation between the wheel and
the latest dates of PIE unity - change as the evidence for the wheel keeps
moving backward in time?

Archaeologists have begun to adopt the understanding that the earliest
finding of a well-developed technological item cannot be accepted as the
"earliest date" of that technology, anymore than it is likely that the
earliest wheel found is actually the first wheel ever made.  Evidence for
wheel tracks contemporary with early European megalithic material culture may
have been found and is being actively pursued. (see, e.g., Maximilian O.
Baldia, Causewayed enclosures, the oldest roads, the first wagon tracks, and
the development of megalithic tombs in southern Scandinavia and Central
Europe, a copy of which is on the web at
http://209.217.18.237/SAA_1998_Roads.htm.)  The dating would mean the PIE
speakers would be acquiring their common word(s) for the wheel in Germany
about 3750BC about the same time they were picking up the same terminology in
the vicinity of Sumeria.  Not.

But here's the important question: how could a piece of archaeological
evidence that is so uncertain as to both ultimate dating and conclusiveness
become such a rigid date for some linguists?

<<If this is what you mean by "a linguistic conclusion", then yes, dating
the latest IE unity to 3500-4000 BCE is "a linguistic conclusion".  Are
you saying that this is a bad thing?>>

The way this has been done, yes.  There's nothing wrong with saying that the
archaeological evidence looks like right now the earliest wheel is at about
4000BC and the linguistic evidence MAY point to that as a possible time of IE
unity.  It's quite another to dismiss earlier dates as "quite inacceptible."
Or to build a whole house of cards on such a date and then act as if it were
unquestionable proof of the chronology of PIE or its descendents.

The nature of the evidence does not allow anything but a quite TENTATIVE
conclusion, at best.  Your conclusions here are based strictly on things
found in the ground and you can be no more certain about those things than
the people who find them and know what they mean.

There is no reason for any kind of a conflict between the archaeological
evidence - which is becoming harder science everyday - and the linguistic
evidence.  It just requires a realization that nothing can be particulary
written in stone on either side.  And there are many, many unknowns.

Regards,
Steve Long