Re: IE common traits

From: markodegard@...
Message: 204
Date: 1999-11-09

junk "ivanovas/milatos" writes:
This question may sound absurd, but I've
always wanted to ask it in a learned forum (not meaning there's some kind of outcome I would prefer!):

Has anybody ever tried to prove that Proto-IE was the first (group of) language(s) present in all of Europe (may be identical with 'Old Europe', a place livable after the last ice-age), later developing singularly one by one, and those others, non IE-languages as Basque or Etruscan (which I believe to be IE) or 'Minoan' (the same), to be the newcomers? Mind you, I'm talking about that potential recorded languages - that might not exist at all, that unknown, hypothetical construct usually called Proto-IE for convenience (whose? If you don't have a picture, what are you constructing models for?), supposed to be a model of convergence (when looking back) of vaguely similar modern or written (how were they really pronounced, 'Tsaesar' or 'Kaesar'?) ancient languages.

In broad terms, all human languages are most likely decended from a single language; in English, this language is usually called proto-World. The time scale is probably identical to the time scale for our species (we are not so much Homo sapiens as we are H. loquens, 'loquacious man)'.

Greenberg has suggested that all known human languages, extinct and living, can be grouped into three huge super-families. The super-family to which Indo-European, Afro-Asiatic (Semitic, Berber, ancient Egyptian, others), Altaic (Turkic, Mongolic, others), Uralic (Finnish, Hungarian, others), and some additional smaller groups, is termed Nostratic. No one really disputes the idea that the world's languages can be placed into such supergroups. The methodology used to arrive at such groupings, however, is extremely controversial, and any conclusions made are at best, extremely tentative.

Historical linguists describe languages from the evidence presented by the language in question itself. It's like an archaeologist examining potsherds or inhumation practices and making the appropriate conclusions. Looking at Polish, Czech and Slovak, a linguist can instantly demonstrate, by internal evidence, that these three languages are not merely related, but that they are very closely related. This is done by comparing such things as pronouns, the way words are formed, the way verbs and nouns behave, etc. It's a mixture of grammar and vocabulary combined with an acute understanding of phonology and the laws of phonology -- particulary phonology, in that it is sound changes that propel just about every other important change in a language.

The exact date for the falling-together of proto-Indo-European is as controversial as is the date for its falling-apart. The standard theory puts the breakup ca. 3500 BCE. The date for coalescence is variously mentioned as somewhere around 5500 BCE or 6000 BCE.

What makes Indo-European a separate language family, and not just a branch of another language family are its distinct peculiarities. It is first, an inflecting language-family (all those noun cases!), which I've been told is something found only in a minority of the world's languages. Second, it's a nominative language-family, i.e., we analyze our sentences by searching for the subject and insisting that the verb agree with the subject -- which again (so I've read), is in the minority. These, and other grammatical and structural differences, combined with inherited vocabulary and phonology make the IE family of languages distinct from all other language families. It's the difference between apples and oranges.

In prehistory, we only know of a few other languages that were near to where the earliest IE speakers are usually posited to have lived (modern Ukraine somewhat north of the Black and Caspian Seas). Uralic is the one instantly mentioned in this context, and there seems to be a feeling that Uralic, of all other language families, is closest to IE, and was most likely in unity with IE sometime in the past (Indo-Uralic): this would have been at least 7500 years ago, and probably further back. The Caucasian languages, particular the Kartvelian group (Georgian, etc) are also mentioned, but this is highly controversial: there is simply not enough evidence present upon which one can make a definite statement.

The next group mentioned is the so-called Tyrrhenian group, which includes Etruscan and Lemnian (the language of the Phaistos Disk of Lemnos). Suggestions are heard that Pelasgian (the language spoken in Greece before Greek made itself known) should be placed here, but almost nothing survives of Pelasgian (there's even less than we have for Etruscan). It's been suggested that Tyrrhenian, Uralic and IE share common descent, again at a time depth of something like 7500 years ago. With Etruscan, I've read a tantalizing suggestion that it's something to do with the verbs.

Any relationship to Altaic is much more ancient, and ever harder to demonstrate. IE is clearly not related to Semitic, which was south of it, or Mitanni (spoken in what is now roughly Armenia). In the far west, we of course have everyone favorite isolate, Basque. Basque is known from Roman times (in its ancestor language, Aquitanian); Caesar seems to speak of it when he says Gaul is in three parts. Basque is as unlike IE as any language can be, notwithstanding more than 2000 years of interaction with Indo-European neighbors.

You read about something I remember being spoken of as the Atlantic Block. These would be the pre-IE peoples of western and central Europe, which may have all spoken languages within a single family, or just as easily, could have spoken utterly unrelated languages. Pitctish (the ancient non-Celtic language of Scotland) is usually mentioned in this context, as is the ancient substratum found in the Germanic group.  It is not unreasonable to suggest at least some of them spoke an Afro-Asiatic language (perhaps related to the Berber group). Someone had to build Stonehenge. Beyond this, very little else can be said; these languages are otherwise completely unknown.

The model as it is to my logic cries for this question. And if not, I'd like to ask another heretical question:

Whoever makes up a hypothetical language also needs to make up hypothetical people/places/communities

Proto-Indo-European is not hypothetical. It's like the text of Homer -- we know the text is real, even though there are thorny textual problems. PIE was a real language. The miracle is that the historical linguists have reconstructed it in astonishing detail. But then, the archaeologists have also performed seeming miracles too.

- a problem Dumezil didn't think of when putting up his beautiful, seemingly fitting three-fold myth-model (although this silently presupposed an advanced society that would be necessary for the model he probably based on later mythologies. But how about the archaeological facts for
the attributed time - seemingly simple, rural, agricultural societies without hierarchies -?really?. Are/were they as simple as excavations show? Or is Dumezil wrong? - a question I wouldn't want to solve. Who knows - that seems to be a question of 'academic'? attitude-).

Dumezil has been consistently and savagely criticized, yet, his tripartite divisions still cast their power over the standard model.
Still Dumezil's supposed similarities for IE myths look amazingly consistent. Are we turning in a vicious circle? (archeologists trying to find what linguists need and vice versa?) Gimbutas and Renfrew are somehow riding the same tide - going where?

The archaeologists are actually driven to distraction by the linguists, and vice versa. The linguists can give the archaeologist an extraordinarily detailed list of a peoples religious ideas, their social structure, even their name for 'mother-in-law'. An archaeologist can only describe what he sees, and relate this to other sites. Renfrew views have been given due consideration, but have been rejected; Renfrew carries archaeological technique to the ultimate logical extreme, and simply ignores the findings of historical linguistics Gimbutas is probably the greatest Indo-European archaeologist of all time. Her views are not to be discounted, but she, singularly, has been responsible for the situation where we end up presupposing an undifferentiated PIE being spoken from the headwaters of the Yenisey River (Tuva, just north of the westernmost point of Mongolia), all the way to the mouth of the Seine River: this is not possible.

If I'm looking for a king, I'll find one - even if he was just Asterix' 'Majestix'. And linguists will somehow prove his possibility - if my supposed word - well, I heard it from linguists, anyway, - somehow follows Grimm's or others' laws (cf. the example of 'wanax' lately discussed on 'Aegeanet'). But which modern word doesn't? What if my concept of king didn't exist yet? I'd only be proving the circle of my/our own thoughts.

I'm working on a posting for Grimm's Law and Verner's Law. You almost need Unicode to demonstrate it though. The Grimm-Verner Law is so predictively consistent that it really is a law in the sense that the 1st law of thermodynamics is a law.  It explains why Sanksrit pitar, Latin pater and English father are all descended from the same word. It also explains why latin centum (100) and English hundred are also cognate.

If *w(n)natks is PIE, my sources suggest is is rather late PIE, and perhaps a borrowing, but it's presence in Tocharian A and Mycenaean Greek (wa-na-ka) suggests PIE status. I forget where this point arises, but as I recall it also occurs in one of the supposedly non-IE Anatolian languages. It could be a borrowing from IE. There is, for example, the suggestion that the wine-word is PIE, and that the cognates in Semitic are borrowings! PIE donated its share loan words early.

As for Tsar, Czar, Kaisar, Caesar, well, they are all the same word. Gaius Julius Caesar probably pronounced it something like the German version. The English pronunciation is via Church Latin which reflects a shift from K to S/Sh. In Slavic, the sounds changed to ts, tz.

The rex-rix-raj parallel seem to speak of a religious king ('first function'), while wannax is closer to 'leader, lord' (the 'second function'). It's 'war-chief' vs. 'medicine man'.

I feel overpowered by different kinds of proof - they all sound right (and wrong) somehow. Is there a way out? Objectivity doesn't exist, we have just personal interpretations of - archaeologically speaking - random facts! (meaning: how about all those not found, not excavated or even not found worthy to excavate because 'just rural' etc.)
We're all overpowered. There is no doubt about the status of PIE, however. There is huge controversy over the details, however -- just as you find in archaeology.

I think the necessity in a subject like this is a 'just' personal opinion - albeit well founded by facts - because otherwise the usually trod lines of thought will never be looked out of. I don't want to open ways for stupid /or beautiful, as for that/ phantasies without foundation, but I'd like to hear some good ones, including possible not excavated, non-proof ways.

And if there are thousands: What are we talking about? With all my respect as a non-Linguist.

Sabine

--Mark Odegard