Urheimat in Lithuania? (was Re: the Wheel and Dating PIE or NW-IE)

[Robert Whiting]

On Tue, 28 Mar, JoatSimeon at aol.com wrote:

>whiting at cc.helsinki.fi writes:

<snip of what whiting writes>

>Since, for example, some of the innovations that define
>Proto-Germanic seem to be quite recent -- Iron Age, judging by
>the development of the Celtic loanwords for things like iron
>technology and some social terminology like "king" or "servant"
>-- what would be the distinction between pre-proto-Germanic and
>pre-Balto-Slavic in, say, 1500 or 2000 BCE?

>Not a question that can be settled, of course, but interesting to
>contemplate.  (My own guess would be "not much".)

Solving the distinctions between Baltic, Slavic, and Germanic
seems to be one of the major problems of IE classification.  As I
remember the discussions of the UPenn tree, this is where it had
the most problems determining the branching.  Not to mention the
question of whether there actually was a Balto-Slavic unity or
not.

The reason for this, I think, lies in the fact that there has
been extensive reconvergence among the early forms of these
stocks so that the original branchings have been obscured.  There
is fairly conclusive evidence that at one point there was a
circum-Baltic sprachbund that included Germanic, Baltic, Slavic,
and Balto-Finnic.  Reconvergence under the effects of the
sprachbund is a great little blurrer of earlier distinctions.

>>but after this change this branch seems to have become very
>>conservative (Sanskrit, Slavic, Baltic).

>-- here I would disagree, to a certain extent.  Baltic and
>Slavic yes; but Indo-Iranian, no.  After all, Sanskrit appears
>conservative precisely because the version we have was fossilized
>as a "learned" and liturgical language, rather like Latin.  Its
>forms date from a very early period, analagous to that of our
>Mycenaean and Hittite records.
<snip>

Yes, I grant you this.  In fact (Sanskrit, Slavic, Baltic) almost
looks like a "which language doesn't belong here?" question.
Sanskrit looks archaic because, well, it's archaic -- as you say,
it is at least 3000 years closer to PIE than modern Lithuanian
is.

>>(actually, one of the other Baltic languages, like Old Prussian
>>or Curonian, may have been even more conservative, but since
>>little or nothing of them survive it is not possible to say).

>-- very true; entropy strikes again.  On the other hand, if you
>run Latvian backward, you get a proto-language very much like
>Lithuanian!   Plus, of course, we know that much of the
>territory now occupied by Latvian was originally Uralic.

Yes, and it would seem that Baltic (or one of its previous
manifestations) has pretty much always been in contact with
Uralic.  First we have the evidence of contact between PIE and
PU, then there are the Old Iranian loans at the PFU level, but
there are also Slavic loans in all the FU languages (but not,
apparently, inherited from the proto-language, and who can really
tell the difference between Slavic and Baltic at this stage).
Then there are the sprachbund effects leading right down to the
present day when Baltic and Balto-Finnic are still in contact.

>>But again (or still) this doesn't say anything specific about
>>where they started out geographically except that Baltic was
>>always in close proximity to Slavic.

>-- not definitively, no.

>However, at the earliest historic attestation, Baltic was
>directly north of Slavic and Slavic extended from east of the
>Vistula into the forest-steppe of the Ukraine.  The relationship
>of Baltic and Slavic and the lack of identifiable substrata other
>than some influence from Uralic _and_ the presence of Baltic
>river-names in the eastern and northeastern areas later colonized
>by Slavic (Russian and Beorussian particularly) would argue that
>they had occupied both this relative position _and_ their
>respective actual territories for a very, very long time.

I grant you all this, but note that this argument does not make
use of linguistic conservatism, which, I think, is as it should
be.  It is quite possible that Baltic is very conservative in
some features and that it may very well have been near the IE
homeland, but I don't think that this is a cause and effect
relationship.  I would be willing to give more weight to the
persistent connection with Uralic as an indication of not having
moved much than to linguistic conservatism.

>>But the presence of more archaic lexical items and fewer obvious
>>substrate borrowings could equally well be a result of the
>>refusal of the language to accept loans as of its still being in
>>its original home.

>-- quite true; we know, for example, that Anglo-Saxon ended up in
>England due to migration and supplanted a Brythonic-Celtic
>language (and Latin), but you couldn't prove it by linguistics
>alone.  11th-century Anglo-Saxon, the Wessex dialect
>specifically, was still an extremely ordinary West Germanic
>language and probably fully mutually comprehensible with its near
>kin in the Low Countries.  It was still marginally mutually
>comprehensible with Scandinavian, for that matter.   And there
>were very few Celtic loan-words -- about 12, if I remember
>correctly.

But this is more or less typical of substratum influence on
superstratum languages.  Using Amerind influence on English as an
example, words taken into the superstratum will typically be
toponyms (Mississippi, Mississinewa, Missouri) or words for local
flora (squash) and fauna (woodchuck, chipmunk) or local
implements (moccasin, tomahawk, wampum, wigwam) or institutions
(powwow) that the superstratum language lacks.  Other words will
tend to be used pejoratively or for comic effect (papoose,
squaw, mugwump).

If the superstratum language came from a similar area (as in the
case of Anglo-Saxon) so that the flora and fauna are familiar,
there is not likely to be new words coming into the language in
this area, and if the superstratum language preferred to rename
the topological features in its own language or calque or folk
etymologize these names from the old language, there would not be
many here either (but note the persistence of Latin-based place
names in England into modern English).  And so on through the
list.

>So the archaism and lack of non-IE loanwords in the Baltic
>languages _by itself_ would not be a firm indication of
>anything, as you say.  However, when taken in _combination_ with
>other factors, we're in somewhat different territory.

Agreed that the more evidence that points in the same direction,
the more likely that direction is to be the correct one.  But
likely still doesn't mean definitely so it is still a purely
hypothetical solution.  But it is still useful to be able to
disprove a theory piecemeal, because these potential disproofs
provide falsifiability for the parts of the theory.  And if the
parts of the theory can't be falsified when there is a test for
falsification available, then the entire theory gains in
strength.  In essence, it makes the theory scientific even if the
theory as a whole has no test for falsification.

So there is nothing wrong with saying that while the individual
pieces of a theory could all be explained differently, doing so
stretches the limits of both coincidence and credulity and
therefore the simpler explanation provided by the theory is more
likely to be correct.  This is enumerative induction as opposed
to eliminative induction.  A while back we had a lengthy
discussion on the validity of the laryngeal theory.  The opponent
of the theory claimed that all the effects of the laryngeal
theory could be accounted for by 16 or 17 different rules
operating in various places.  When it was pointed out to him that
the statistical probability of this happening was virtually nil,
he eventually accepted the laryngeal theory as correct in its
essentials.  (Of course, we're still arguing about the number of
laryngeals and their values, but that is a different matter.)

>Eg., Anglo-Saxon/Old English is geographically peripheral to the
>main mass of the Germanic languages, with salt water in between,
>and there -are- a number of Celtic place-names in its territory,
>increasing in number as you move west.  Even if one knew nothing
>about the history prior to 1000 CE, you'd still have enough for
>an informed guess that Anglo-Saxon was a fairly recent offshoot
>of the main Germanic zone. (And, taking in similar evidence from
>the Continent, that Germanic in general had been expanding at the
>expense of Celtic and Romance.)

And at the expense of Slavic too, but that also is a different
matter.

>>English, on the other hand, just gobbles up loanwords and
>>neologisms regardless of whether they violate English
>>phonotactics or not (e.g., aardvark, gnu, syzygy).  This is
>>simply not a function of how close these respective languages
>>are to their original homelands.

>-- true; although, of course, we know that German is much closer
>geographically to the proto-Germanic _urheimat_.

But not *in* the Germanic Urheimat.  There is still not a cause and
effect relationship.  If the Goths came from Gotland, then
Swedish is closest to the Germanic Urheimat.  And modern Swedish
has about as many loanwords as modern English does.  But because
most of these loans come from Low German they are not as
noticeable as the predominantly Latin/French loans in English so
the point is not often raised.

>Interestingly enough, English only became exceptionally open to
>loan-words after the Norman Conquest.  Prior to that, Old English
>was notably resistant to foreign lexical influence.

No, it is only after the Norman Conquest that the loans become
easily noticeable.  There are a lot of loans from Scandinavian in
English that predate the Norman Conquest that most native
speakers don't even recognize as loans.  This is the effect of
adstratum borrowing from a closely cognate language (in the case
of Scandinavian and Old English, practically a matter of dialect
borrowing) and is very similar to the effects of Swedish
borrowings from Low German.

In adstratum borrowing, very common, everyday words can be
borrowed for quite a number of reasons.  Among the most common of
these is greater differentiation or specialization in meaning.
There are thus quite a number of Scandinavian borrowings in which
both the inherited word and the borrowed word still exist but are
specialized in different uses.  For example:

    Native English         Scandinavian

        shirt                 skirt
        shuttle               skittle
        shoe                  sky (both originally meant 'covering')
        'em                   them
        yard                  garden
        rear                  raise
        whole                 hale

Unless one has been specifically taught that the Scandinavian
forms are borrowings, one is not likely to recognize them as such
(with the exception of those that begin with sk- since it is
fairly obvious that any word that begins thus in English is a
loan [or a dialectal form]), in contrast to French borrowings
like 'facade' 'chalet' or 'chandelier' which jump out as
loanwords from their spelling and pronunciation.

>>If we didn't have a historical record of the situation, the
>>argument of conservatism plus lack of substratum influence could,
>>other things being equal, be used to claim that Iceland is the
>>original Scandinavian homeland.

>-- good point.

>Although there, we know that there was no prior population -- and
>my original argument was that Baltic probably entered an area not
>far away from the _urheimat_ ... _and_ one which was very thinly
>populated.  (What's now the eastern part of the Baltic was late
>being neolithicized, if memory serves me correctly.)

Yes, knowing what happened makes it easier to do reconstructions.
But the fact that we know that Icelandic had no substratum
influence and we suspect that Baltic didn't have much, doesn't say
anything about how far away from their original homelands they
are.  If it did, your hypothesis would be in real trouble,
because if Icelandic has no substratum influence and is
conservative and is a long way from its homeland, then if Baltic
also has no substratum influence and is conservative, then it
should also be a long way from its homeland.  These are just not
one cause - one effect relationships.  If one effect can have
many causes, then lack of the effect does not say which cause
is responsible.  Lack of substratum influence can be the result
of a lack of a substratum or of a refusal to be influenced.

As an example consider the following:  There is no evidence that
Nixon was involved in the Watergate coverup.  This implies:  a)
Nixon was not involved in the Watergate coverup; b) all evidence
that Nixon was involved in the Watergate coverup has been
suppressed or destroyed.  Which is true?

>>Negative evidence cannot be used to construct specific
>>scenarios.  Negative evidence only means that there is no
>>evidence.

>-- it's not demonstrative, as positive evidence is.  However, I
>think it can be legitimately used _in conjunction_ with other
>supporting evidence to say that one of a number of alternative
>explanations is more likely than another.

Yes, negative evidence is not completely worthless, but it can
only lead to negative conclusions which may allow you to
eliminate some of the potential causes.  It is particularly
useful if there is a reciprocal one effect - one cause relationship.
If there is no smoke without fire AND no fire without smoke, then
no smoke means no fire.  But it still doesn't tell you why there
is no fire.  But if it is possible to have a fire without smoke,
then no smoke just means no smoke.  In this case, smoke means
fire, but no smoke doesn't necessarily mean no fire.

>As Holmes said, the crucial thing was what the dog did in the
>night.  When Watson pointed out that the dog had done nothing in
>the night, he replied: "Exactly."

Well, close enough.  But the point of this is not so much that
the dog didn't bark, but what it would have taken to make the dog
bark.  If the dog never barked at anything, then the fact that
the dog did not bark in the night would not have been interesting
at all.  Since the dog didn't bark, either there was no dog or
nothing happened to make the dog bark.  Now if the dog barks at
strangers or at rabbits but not at people that it knows, then if
some unaccountable act is performed in the night and the dog
doesn't bark (and we know that there was a dog) that eliminates
strangers (and rabbits) as having performed the act.  But it
still doesn't tell which non-stranger/non-rabbit did perform the
act.

This is why negative hypotheses are generally scientific
(falsifiable) while the opposing positive ones are not.  "There
is no Santa Claus" is a scientific theory because it can be
falsified by producing the one and only true Santa Claus.  "There
is a Santa Claus" is not a scientific theory because it cannot be
falsified.  It can only be disputed with negative evidence.

Of course, scientific doesn't mean necessarily true, it just
refers to the way in which it can be investigated.  So there is
nothing wrong with believing in Santa Claus.  It just means that
no one can prove that you are wrong if you do.  And if you are
sure that all the evidence has been destroyed, this can be a very
comfortable position to be in.

Bob Whiting
whiting at cc.helsinki.fi