> You contradict yourself quite a lot.
Yes, sorry about that. It came from trying to reply to you point-by-
point when I should have stepped back and reorganized everything so
it wasn't so disjointed. I'll do that now. I got some help, too.
First, terminology: instead of Gamkrelidze and Ivanov (not Schleicher
as you thought) we'll use your groupings:
2) Proto-Italian, Proto-Celtic and Proto-Germanic;
3) Proto-Greek (centum);
4) Proto-Armenian and Proto-Albanian (satem);
5) Balto-Slavic undergoes RUKI (satem).
1) Anatolian is irrelevant to this since it is such an aberration. It
could have left India before Pre-PIE and travelled to Anatolia. Or
Beekes could be right and it's not nearly as archaic as it seems but
has simply mixed with many other languages; in which case it would be
part of PIE. Either way, it doesn't really matter much in this
discussion about a postulated OIT although it is important to the
larger IE question.
2)and 3) Proto-Italian, Proto-Celtic, Proto-Germanic and Proto-Greek
are the core of the undifferentiated PIE.
4) Armenian and Albanian are attested only very recently. Albanian
from the 15th century -- almost 3 millennia distant from Greek and
Indo-Iranian! -- and Armenian from the 5th century -- two millennia
distant! These languages are actually attested at dates closer to
modern times than PIE times -- and this is assuming conservative
dates for IIr. Albanian is also mixed with Latin, Greek, Slavic and
Turkic so it is hard to tell how it might be related to PIE times and
so it is not considered. Armenian shows heavy influence from both
Greek and Iranian.
It is possible that Armenian's satem terminology is due to Iranian
It's important to note that the only major attested really ancient IE
languages are Anatolian, Greek and Indo-Iranian.
So the proposed migration is as follows:
0) Anatolian does whatever (irrelevant since it is unclear and it
doesn't directly affect the theory since it is anomalous).
1) Pre-PIE consisting of Proto-Italian, Proto-Celtic, Proto-Germanic
and Proto-Greco-Armenian (along with Proto-Tocharian) head north out
of NW India or the BMAC into Central Asia. Tocharian splits off at
this time. During its sojourn, from Central Asia to Eastern Europe
near the Danube, PIE loses the words for South Asian flora and fauna
(this is just the reverse of what is usually proposed so it is
feasible.) Once they reach Central Asia or E. Europe (depending on
your theory of choice), they split according to standard IE theory.
Perhaps Greco-Armenian goes to the Balkans where they split and Greek
goes south and Proto-Armenian goes southwest and displaces the
Luwians and Urartians near modern Armenia.
2) Meanwhile back in South Asia, various Vedic dialects dominate NW
India while the so-called "Proto-Indo-Iranian" branch leaves via the
Helmand Valley (or the Khyber Pass) and reaches the Iranian Plateau.
Significantly, after the PIE split has taken place some of these
people continue to the NW in a movement similar to the later Scythian
expansion and they loan words into the Uralic languages. They also
satemize Armenian or maybe this happened later during the other
Iranian migrations. This expansion across Central Asia separates the
Tocharians from the other centum languages.
3) All of the languages in South Asia have now undergone RUKI. Slavic
is not attested until the 9th century and was unified before then.
Baltic, i.e. Lithuanian (plus Latvian) is only known since the 16th
century but is considered ancient. It is commonly thought to be
areally associated with Proto-Germanic but Beekes says: "There is no
pressing reason to assume that Germanic and Balto-Slavic underwent a
common development." p30 Therefore it is not true that Balto-
Slavic "leaves India in extreme haste: it's a long way to go and they
have to catch up with Proto-Germanic and become areally associated
with it." They have several millenia. And besides, when you say "they
have a long way to go", I must point out that it's not any farther
from India to Eastern Europe than it is from the home of PIE in
Eastern Europe to India. :-) Nevertheless, they probably do leave
4) Indo-Aryans expand from the late Harappan cities into BMAC
(reversing the direction of Parpola's and Sergent's IA establishment
of BMAC) and invade the Kassite and Mitanni areas. Their path is
erased by later Iranian expansions since it was probably only a small
number of warriors (elite dominance.)
OK, so I admit there's more than one migration. The IIr and
Kassite/Mitanni I didn't count since they were "in the family" but
the Balto-Slavic would have to count as a second one.
But it's not really the number that matters -- the important fact is
that this theory does not require them to maintain their association
across thousands of miles. That is the absurdity that Hock
highlights. Multiple migrations are not a problem and are well-
> > ... Some will say Anatolian is explained by mixing with other
languages. Obviously any Indian homeland will have to go with the
anti-Indo-Hittite school of thinking, cf. Beekes "its simplicity need
not suggest antiquity, but could rather be owing to loss."
> "Antiquity" is not the issue. Hittite and Luwian are approximately
as "ancient" (close to PIE) as Mycenaean Greek, Rigvedic or Old
Avestan. What's disputed is which particular differences between
Anatolian and the Rest result from common innovations in the latter
rather than loss in the former, but few linguists doubt the validity
of regarding Anatolian as a first-order branch in the IE family tree
(I certainly don't), even if they don't like the term "Indo-Hittite".
You'll have to complain to Beekes about his use of the
word "antiquity." :) Antiquity is an issue though-- it is the fact
that Anatolian split from IE *EARLY ON* that makes it important as
well as the nature of the differences. Minor point, I think.
> But the characteristicaly IIr features are innovations, not
retentions (unless you go with Misra all the way), so you'd really
need an independently developed "Doppelgänger" of Proto-Indo-Iranian
in eastern Europe -- the most remarkable case of parallel development
ever seen -- to account for what people regard as evidence of
I believe this is answered above.
> > There is very little consensus on which of these words are IE and
which are not. Kuiper will say they are not IE, Mayrhofer will say
they are, Masica will question both of them. The few that are clearly
not IE can be explained by adstrate and especially through trade.
> "The few"? I should think the majority of them are
uncontroversially non-IE. The debate concerns individual items, not
the bulk of them. I don't always agree with Witzel, but his
article "Substrate Languages in Old Indo-Aryan" is definitely worth
reading. It includes a discussion of what characterises a non-IE word.
I don't always agree with Witzel either -- especially with regard to
that article! One of the tenets of this theory is that the disconnect
between Vedic and IE means that pure Vedic is not a pure IE language.
It has many roots and morphologies that are unique to it and that it
does not share with IE. Many of these are shared with Munda and
Dravidian but they are mostly loans FROM Vedic. Vedic *is*
Masica's "Language X."
> > Nichols disagrees.
> I have to disagree with anyone who claims that homeland areas are
characterised by lower-than-normal heterogeneity. Such a claim is
falsified by real-life examples.
Duelling experts -- you pays your money and places your bets.
> This would mean that Balto-Slavic stayed in India long enough to
participate in the Satem and RUKI innovations and was the last branch
to leave before the separation of Iranian (and Nuristani). But then
how do you explain the fact that Indo-Iranian, Greek and Armenian
share a number of typological traits not found in Slavic?
That could be explained by postulating that the Greco-Armenian(-
Anatolian?) branch split off first from PIE while it was still closer
to Indo-Iranian. This also matches the attestation dates of documents.
> If the migrating group was differentiated from Indo-Iranian, this
also means that Indo-Iranian was differentiated from the migrating
group. At this point I have the right to ask: can you propose _any_
plausible common innovations defining non-Indo-Iranian IE in
opposition to Indo-Iranian? Without such a demonstration you can't
claim that the differentiation proposed above ever took place.<
Perhaps I don't understand the question but aren't the common
innovations among Indo-Iranian that aren't shared among the rest of
the IE languages exactly the reason that Indo-Iranian is considered a
> The degree of dialectal fragmentation in the US has little to do
with the national character of the Americans but all to do with the
history of their expansion in America.
Perhaps. But the extreme conservatism among Indo-Iranians has
everything to do with the lack of dialectical fragmentation among
their oldest texts so it's relevant to a comparison of the two
> >> The loans (very likely all of them) are distinctly Indo-Iranian
and many of them can't be dated to the post-split period, since the
most archaic of them show features like unpalatalised velars before
front *e (still retained!), and satem stops still reflected as
> > This would be the character of this so-called IIr but is really
the Iranian split.
> Untenable. The stage in question is ancestral both to Iranian and
to Indo-Aryan. By the time the two subbranches separated, some
characteristic changes (like the palatalisation of velars before
front vowels, and *e > *a) had already taken place.
The main point about this theory is that it accepts the existence of
several dialects. It does not try to shoehorn Sanskrit into being
Proto-IIr, PIE, etc. That is definitely untenable. So the so-
called "Proto-IIr" in this theory is a dialect that has these archaic
features. It seems to be the only possible explanation.
> Now you're beginning to see the light. If you're ready to admit
that there were some Proto-Indo-Iranian communities in the BMAC and
sufficiently farther north to interact with Proto-Finno-Ugric, why
not take the next step and hypothesise that there were no _other_
Proto-Indo-Iranians anywhere? :)
I suppose I should explain why he came up with this theory in the
first place. He doesn't actually believe it!!! ;-) Well, to be more
precise, he has been reading the doubts of the archaeologists and has
come to the conclusion that his earlier belief in the Aryan Invasion
may have been premature. There is not enough evidence to answer the
AMT question conclusively at this time. But he decided to try and see
if he could construct a feasible hypothetical Out of India theory.
Much to his surprise, the more he dug into it, the less solid the AMT
looked. He's still not convinced either way, but he feels it's
important to develop a testable alternative theory in order to
prevent preconceptions from prejudicing the experimental designs and
interpretations. Hopefully this will lead to increased interest in
doing the necessary research to finally answer the question. The
linguistic theory is just a small part of a larger holistic re-
evaluation of the evidence from all of the relevant disciplines,
which he hopes to publish in a few months.