Unity: Depth, Place and Space.

What do we mean when we speak of 'unity', particularly as applied to proto-Indo-European? I'm going to start with a discussion of what I accept to mean by 'unity' in this context, then will get to a discussion of the IE homeland.

For myself, I use 'unity' in the sense I apply to present-day English, and historically, to the 1000+ year period before today. We recognize substantial differences between the various Englishes you find in Britain, Ireland, Australia, New Zealand and North America; sometimes, these differences get in the way of mutual intelligibility, but when you analyze each of these 'dialects', they all reduce to near-100% grammatical and lexical unity, with only a partial, ultimately inconsequential phonological disunity. Under different historical circumstances, English could have, and probably would have broken into daughter languages, quite analogous to the case of Dutch and Afrikaans. Universal childhood education, and the status of English as the world-language, however, together with all the communication marvels of the present age are gently pushing all the various Englishes towards a vague convergent norm (Mid-Atlantic, as it's sometimes called, as with the English of Patrick Stewart and Alistair Cooke). English is forming its own sprachbund.

Present-day English is in historic unity with Middle English and Old English. Yes, we know these are three different languages, and we know that the Old English of King Alfred is not the dialect ancestral to Modern English, but they are all part of a single historic continuum. Old English leads directly to Modern English. We can speak of a 1300 year unity for English, from the time it diverged from its Low German ancestor.

We speak of Greek in the same manner. While we know there is some convergence of dialects involved (as with English), Modern Greek is essentially descended from Attic Greek. These are two different languages, but we can speak of a 2500 year unity, a 2500 year historic continuum.

In the historic sense, then, in the sense I have outlined here, a language can be in unity for a very long time indeed, and if one enlarges the definition to permit us to ignore extinct daughter languages, the time length can be even longer.

The thing that keeps a language in unity, it seems, is geographic extent. If a language is spread over too great an area, dialects develop at the edges, leading to daughter languages. The maximum geographic extent a language can occupy without daughtering is related to mobility. Despite being spread halfway across the globe, modern transport and communications keep English in unity, a unity that seems greater than it has enjoyed at any time in its past.

In earlier times, however, the area is necessarily considerably smaller. J.P Malory has suggested the area of modern Poland or Germany as the maximum territorial extent any language could occupy and remain in unity. When we look at the other languages of the world in this light, this size limit seems consistently vindicated.

[digression
'Imperial languages', such as Latin and Imperial Aramaic, and in the present day, English, Russian and Arabic, represent a special case. Once the center of Latin (Rome and her legions) collapsed, Vulgar Latin daughtered. With Arabic, daughtering is in progress, but as with modern English, modern technology is pushing the various Arabics into convergence.
end digression]

So, with all these understandings, how long was the language we call proto-Indo-European in unity? And, from when to when, and in what location?

If I read Piotr's views correctly, he locates the PIE homeland somewhere between the Elbe and Vistula river valleys, just north of former Czechoslovakia, and somewhat inland from the Baltic coast. This is a reasonably-sized potential homeland: this would be the TRB (=Trichterbecher, Funnel Beaker) Culture, 4500-2700, or the earlier, and as is usually said, its ancestor, the Linear Ware (LBK, Linearbandkeramik) Culture, 5500-4500. My main problem with this is the relatively huge area these two cultures occupy on my maps. TRB is from almost the mouth of the Rhine to well west of the Vistula, while the earlier LBK, while away from the coast, nonetheless ranges from the mouth of the Seine clear to Romania, and includes Austria and Hungary.

For these cultures to represent the IE homeland, it seems you have to posit a much earlier daughtering for the various branches. Presumably, Anatolian would have to be just north of the Danube, stretching northwesternly along the Dniester and perhaps to the headwaters of the Vistula, at 4500 BCE.

This would leave the Funnel Beakers in a reasonably sized region (though the Netherlands outlier and the Kiev outlier were probably mutually unintelligable). They could indeed represent residual IE, with Germanic breaking away early in the northwest.

Working this all out, with resort to the articles in the Encyclopedia of Indo-European Culture, I'm tempted to join Piotr's camp. I see that János Makkay is the principal exponent of this view. I'm still hesitant, however, about letting the various daughters emerge at such an early date. Yes, all the old questions about the Sherrats' views, as well as the words associated with wheeled vehicles.

With carts and wagons, it seems they were perfected sometime around 3500 and diffused quite nearly everywhere almost instantaneously (a more recent analogy to such rapid diffusion would be the near-instantaneous spread of firearms world-wide, to include equally rapid diffusion of later improvements; ditto for motor vehicles). Excluding Anatolian, we have an IE big bang running from 3500 to as late as 2500. This agrees with the usual dates for the standard kurgan model, as well as the linguistic evidence.

Can we keep an LBK-TRB PIE in unity over this wide area for 2000 years?

Mark.