--- In cybalist@yahoogroups.com, "Brian M. Scott" <bm.brian@...> wrote:
>
> > The problem is that a the lexicon of a given language is
> > typically made up of several strata (multi-layer) due to
> > language replacement and contact processes, and it isn't
> > always easy to tell which is the "inherited" part.
>
> This is a commonplace. It's also of limited relevance to
> reconstruction of proto-languages. If F is a linguistic
> taxon, proto-F is simply the most recent common ancestor of
> F; its own history is largely irrelevant to its comparative
> reconstruction from F. For that history we must resort to
> internal reconstruction, and perhaps eventually to
> comparative reconstruction of a bigger taxon at a deeper
> historical level.
>
This is a description of the classical geneaological tree model, which IMHO is fairly inadequate to represent the IE family, which is a rather exceptional case, because it combines a very long time of evolution since the Upper Palaeolithic with a quick dialectal fragmentation into the historical attested languages in the Bronze Age.
Quoted from F. Villar et al.:
Lenguas, genes y culturas en la prehistoria de Europa y Asia suroccidental. Universidad de Salamanca, 2011., chapter 3 (my own translation):
The history related to us by the languages of Europe and SW Asia, both today's and the ones spoken in their territory of which there exists written attestations, doesn't reach to a deep chronology. In our continent languages have been repeatedly replaced in succesive events of élite dominance which in fact have happened in Europe since the Metal Age. [...]
The deepest hydro-toponymic strata which we've been able to identify in Europe are of paleo-IE character. There're at least two succesive ones, although not very distant in time. The first and more ancient one was rooted with the languages spoken by the European repopulators which set off from the various refuges after the Recent Dryas. The second one intruded in the Neolithic, mostly from Anatolia.
Upon these two paleo-IE strata are overimposed several toponimic superstrata, both historically IE (Celtic, Italic, Germanic, etc) and non-IE ones (Iberian). Nowhere with enough toponimic data available happens the other way around, that is, that the deepest stratum would be non-IE and the paleo-IE one lies upon it.
As a consequence, the common toponimic background of Europe make us suppose a paleo-IEity at least Mesolithic and Neolithic of the most part of the continent (excluding Finland and Estonia, which constitute the border with the Uralic world). Buit if the Mesolithic repopulators who set off from the refuges spoke paleo-IE languages, the consclusion that paleo-IE languages were spoken in these refuges is inavoidable. In turn, if the speakers of paleo-IE languages were confined and isolated in these refuges is because they were already in the continent in the
Last Glacial Maximum (LPG). This puts the paleo-IEity of the continent back to a Gravettian horizon. But at this point we must define what we understans as paleo-IEity.
Traditional IE-ist studies has made great and reiterated efforts in designing different versions of the genealogical tree of the IE languages. Below we include one inspired upon the ideas of E.H. Sturtervant (1964): [
A Comparative Grammar of the Hittite Language] [...]
All these genealogical trees have a common factor: the starting point is always
common IE, conceived as an unitary language spoken in a date and place which vary according to every author's position. And this common language is seen as a single language, more or less dialectalized, but with no contemporary relatives. [...]
In contrast with the interest in the genealogical trees intented for explaining the dialectal processes between common IE and the historical languages [see for example the one by
F. Rodríguez Adrados in
Nuevos estudios de linguística indoeuropea. CSIC, 1988, p. 38], the IE-ists never have been interested the ancestor genealogical tree of which this common language was only a branch coming from an older common phylum. Any ortodox IE-ist which make him/herself respect will consider any attempt to sail backwards from IE as leading to nothing more than speculations unworthy of being considered. And that leads them to behave in practice as if the PIE of their comparative reconstruction would have been at its time an isolated, famililess language, such as today's Basque (Iberian Peninsula) or Burushaski (Pakistan). One more exceptional point of IE, which accumulates then in amounts unparalleled in any real language.
The scholar who wished to find any approximation to other language groups is been forced to make a big jump across the comparative void. He/she has to pair off IE and macro-families such as Altaic of Afro-Asiatic and look for every all of them a very remote phylum supported by hardly tens of difficult correlations. From the genealogical tree quoted above, pretented to be spoken around 5,000-6,500 years ago, one has to jump to a very remote one, which our present knowledge of the prehistory of our species won't allow to put less than 20, 30 or 40 thousand of years. But this macro-comparative isn't of the linking of the ortodox who find, not without some reason, a clumsiness and inconsistency in the macro-comparative method which makes them to suspect of all this conglomerate of constructs.
We find there a temporal void which perhaps other families might fill with the complexity of numerous successive dialectal fragmentation processes, as in the case of Altaic, where one can find a linguistic aggregate so heterogenenous as the one made up from Turkic, Mongolian, Tungus-Manchu, Korean, Japanese or Ainu [sic]. In the case of IE there's no possible intermediate group between Nostratic and the IE of the 3rd or 4th millenium. The same traditional conception, fond of very short chronologies for IE dialectology, has to confront the inexplicable slowness by which the IE branch, which split from the Nostratic phylum between 15,000 and 45,000 years ago, has remained united, with none or very little dialectal fragmentation, up to 2,500 BC (4,500 in the Kurgan version), to then undergo a sudden dialectal fragmentation at explosive speeds.
The history related by the paleo-hydro-toponymy of Europe and SW Asia fits within this huge chronological void of several thousands of years between the real or supposed proto-languages from which these macro-families sprouted and the very recent (and no less supposed) proto-IE language of the Metal Age. In other words, [...] this indicates the reconstructed PIE wasn't an isolated language with no close relatives, [...] but one among the dialectal varieties of a macro-family which had been the main (although perhaps not the only one) character of the population events which happened in the European territory at least since the Gravettian. [...] All these dialects form a macro-family, of which the classical IE, if we continue to speak in the terms of the traditional view, would descend from a single dialectal branch. [...]
The evidences studied in this work reveal to us that IE wasn't a single species within its genus, but one of the many varieties of a very rich phylum located in Europe at least since the Mesolithic and seemingly since the Gravettian. And that not all the historical dialects come from a single paleo-variety of the Metal Age. We have to begin to think the genealogical tree of the IE macro-family not as a draw with straight lines successively biffurcate, but as an leafy arborescence whose multiple and tangled shoots have occupied a great part of the European prehistory. In fact, like it has already happened with the genealogical tree of Homo sapiens.