In 4000 BC, Indo-European was spoken somewhere, but its location is very controversial.
Central Europe at the beginning of the 4th Millenium BC Broadly speaking, three major economic lifestyles were in competition in Central Europe during the 5th Millenium BC.
These were Cereal farmers (called Linear Ware or Long House after its pottery and architectural styles) who had inhabited the Danube and Rhine basins since the late 6th millenium, using slash/burn techniques to replace forest with farmland. (Linear Ware is considered sibling to the earlier Impresso culture in the Mediterranean area. Linear Ware farmers apparently practiced crop rotation and animal herding for manure fertilization, but use of the plow probably came 2500 years later.)
Stockbreeders of the east European steppes. These people are called Kurgans, after the Russian name for their burial mounds. In various theories this culture may be derived from pig farmers to the West (Bug-Dniester ca 6000 BC) sibling to the Danubian farmers, hunters from the Baltic area to the North, or ancient farmers and goat herders from across the Caucasus Mountains to the South. Pre-neolithic people who engaged in hunting, gathering and fishing. These were called the ``Megalithic'' (Big Stone) people because they were entering an advanced phase with spectacular artifacts leading eventually, for example, to Stonehenge. Though shown in the North on the map, these people were also present in forests throughout Europe (and obviously in hostile competition with the forest-clearing farmers).
This was the era of the ``Secondary Products Revolution'' with inventions like cheese, leather, beer, and, most notably, the wheeled wagon. (Cast copper axles have been discovered near the Cucuteni dominion dated to the 5th millenium BC; and Kurgans are known to have bridled horses for riding by 4000 BC.)
By 4000 BC, three mixed-farming (dairy) cultures were in competition in East Central Europe; these were Tripolye (and Cucuteni), a branch of the Danubian Linear-Ware farmers who, however, did not practice cereal farming, but rather had an economy based on orchards, cows, sheep, and pigs.
Sredny Stog (and Kemi-Oba), branches of the Kurgan breeders whose economy featured horses, cows, goats, barley, and animal byproducts like leather. TRB/Funnel Beaker, believed to be a branch of the Erteboelle Hunters, who began to build primitive villages and adopt some of the economic ideas of their neighbors, including barley and dairy farming.
There is no universal agreement on which of these three groups provided the proto-Indo-European language, and you can find sober scientists guessing that Indo-European was spoken by any combination of these groups, including none or all three!
Although all three of these groups -- Tripolye, Sredny Stog, and Funnel Beaker -- could be described as ``early dairy farmers;'' in fact the cultures were quite distinct: Tripolye was an organized village society with egalitarian matriarchal customs; Sredny Stog was a semi-nomadic patriarchal society which stressed individualistm; the haphazard lifestyle of Funnel Beaker villagers betrays their recent development from unsettled foragers.
(Cereal farming stresses patience, while stockbreeding requires physical strength -- this may explain why domesticating large animals changes a matriarchal society to patriarchal. Furthermore, the contrast between land-fixed self-growing crops and mobile animals needing to be tended, may help predict whether ancient economics will be based on communal or individual property rights.)
The geographical placements on the map are only approximate. Moreover there was overlap: Kurgan tombs from this era are found as far West as Czechoslovakia, while Tripolye had settlements in central Ukraine. Finally, the indicated cultures are only roughly contemporaneous: Lengyel was 5th millenium, and TRB/Funnel Beaker mainly 4th millenium.
With sophisticated mining, smelting and casting, this era might be called the ``Advanced Copper Age.'' There were metallurgical centers in the Karanovo area in the Balkans, as well as in West-Central Asian areas accessible to Kurgan traders, and the Carpathian Mountains, situated roughly at the central point between the three competing dairy cultural styles, was a rich source of copper ore.
But, although the Bronze Age would begin to emerge 1000 years later, archaeological evidence suggests that by 4000 BC the Balkan metal industry was entering a ``Dark Age'' lasting several centuries. Was this apparent conflict related to competition by the competing dairy cultures being played out in the fertile Danube Basin just to the North of the Balkans?
A thousand years later, new cultures have emerged, and the locations of Indo-European branches can be inferred. (It is good to remember that there was rapid change even in prehistoric times -- Europe's population may have tripled between 4000 and 3000 BC, although both dates fall in the ``Late Neolithic.'')
Central Europe at the beginning of the 3rd Millenium BC
Again dates and places are approximate: The Bell Beaker culture emerged about 900 years after Globular Amphora.
Although Indo-European languages do not enter the historical record until the 2nd millenium BC, there is wide agreement about Indo-European geography in 3000 BC. Most supporters of both the Gimbutas Kurgan Theory and Danubist or Anatolian hypotheses would agree that Usatovo culture can be tentatively identified with the first speakers of proto-Greek, and both theories usually identify Tocharian with the Afanasievo culture far to the East in Asia. Similarly the identifications of Indo-Iranian with Yamnaya, Balto-Slavic with Battle Axe, and Germanic with Corded Ware (see below) are not controversial. Most of the other identifications shown in the map might also be tentatively accepted by theorists on both sides of the Kurgan-Danubian debate.
In other words, many would agree that the Balkan-Pontic area of the 5th or 4th millenium BC was a locus for early Indo-European expansion; the debate is whether Tripolye ``converted'' the Kurgans to speak I-E, or vice versa! For most experts, the signs of Kurgan culture among the early Indo-European speakers are unmistakable. As just one example, the warrior heroes in Homer's Iliad are buried in Kurgans (though of course Homer doesn't use that Russian word).
In the map above, it should not be inferred that the Battle-Axe people spoke ``proto-Balto-Slavic,'' nor that the Yamnaya people spoke ``proto-Indo-Iranian.'' The languages had diverged too recently for that. Instead both groups spoke the same language, called ``Late Indo-European'' or ``proto-Satem,'' but were developing divergent dialects which after a few more centuries would become the familiar distinct proto-languages of the Indo-European group. Only proto-Anatolian (and probably a few other Centum branches) existed as distinct Indo-European languages in 3000 BC.
Did Indo-European Language Originate with the Kurgan People?
Proto-Indo-European has been reconstructed and shown to contain many words related to horses and stockbreeding. The word kwe-kwlo (cognate of wheel and cyclos) is reconstructed for the wheel, but in all theories besides Gimbutas', proto-Indo-European had already separated into its branches before the wheel was invented.
Opponents of Gimbutas' theory of Indo-European origin base their case on five assumptions:
The Kurgan culture is not the sort of giant advance needed to explain language replacement.
The language of the Danubian Linear-Ware culture and related cultures like Tripolye dominated Europe for about two millenia, and couldn't have disappeared without a trace.
Expansion circa 4000 BC is too late to explain the diversity among Indo-European branches. The similarities among Indo-European cultures are the result of coincidental, parallel developments. kwe-kwlo meant ``rotate'' and different peoples independently adapted their versions of the word to denote ``wheel.''
Let me answer these ``charges'' one by one:
The language of the Kurgan horse-riders did expand. All scholars (except the so-called Indocentric crackpots) admit that Indo-Iranian was a Kurgan language, and that the languages of northern India have been replaced by Indo-Aryan, even though there is no evidence of a major invasion. And it does seem suggestive that Middle-Easterners were using Indo-Aryan words to describe horse-riding, even before Indic makes any other historical appearance.
Language replacement is common. The pre-Roman languages of Spain and France quickly disappeared (except for Basque); the Negrito languages of the Philippines have disappeared; Pictish was dominant in northern Britain during Roman times but hardly a word of it is known today.
To the contrary, similarities among the earliest recorded Indo-European languages (Hittite, Homeric Greek, Sanskrit, Latin, Old Irish) are about what one would expect if they had diverged just a few millenia earlier. Today's Lithuanians can make some sense of Sanskrit, which would probably be impossible if these languages had diverged before 5000 BC. Anyway, don't overlook that liturgies and written records serve as a brake on language change, so languages evolved much more rapidly in pre-literate cultures.
Archaeologoists gasped in surprise when they unearthed Tocharian clothing that looked just like Old Irish clothing thousands of miles to the west. The ancient Indic horse-sacrificing ritual of asvamedha has detailed similarities to the horse-sacrificing rituals of the ancient Romans and Irish; even the compound word asvamedha appears to be directly inherited from a proto-Indo-European word meaning ``horse-drunk.''
kwe-kwlo survives in many branches of Indo-European so must have been a basic vocabulary item, a common word used everyday. Primitive peoples didn't speak of ``rotation'' much before they invented the wheel.
Finally, the genetic language tree of Indo-European would have a different structure if Celtic and Italic were spoken in Central Europe before the Kurgan intrusions. I try to explain that on a separate page, but briefly:
All sensible theories agree that Indo-Iranian was a Kurgan language, i.e. the Andronovo descendants of Yamnaya/Pit-Grave. If this were an adopted language, one would expect major changes; instead phonology and grammar of this branch closely follows proto-Indo-European.
West European languages like Italic and Celtic are non-Kurgan in any anti-Gimbutas theory. Thus the I-E tree would involve two major branches: Western and Kurgan. The Afanasievo culture (sibling to Yamnaya) is the only logical candidate for proto-Tocharian. A Balkan or Danubian (non-Kurgan) origin of Afanasievo is farfetched. Were the (controversially lumped) Greek-Phrygo-Armenian languages derived from Kurgan or not? In either case, these languages (and Tocharian) would be either in the Western branch or the Kurgan branch of I-E. Instead they form co-equal branches, with micro-tree structure best modeled (cf. Ringe) as westward Kurgan migrations (with the Centum-Satem shift in Yamnaya occurring after Greek separated).
Balto-Slavic/Germanic Equals Battle-Axe/Corded-Ware
Battle Axe and Corded Ware were sibling cultures, and some scientists do not distinguish the two. Yet, Baltic seems to be the Indo-European branch which most closely preserves the prehistoric proto-Indo-European language, while Germanic has undergone major changes in grammar, phonology, and lexicon. Do the prehistories of Battle-Axe/Corded-Ware culture and the early Germano-Balto-Slavic language shed light on each other?
The reason Baltic preserves ancient Indo-European most closely is that before the Battle-Axe culture emerged, the only non-Kurgan people in the Baltic area was a small population of hunter/fishers: there was no need for proto-Balto-Slavic to absorb an indigeneous language. The Corded-Ware invaders into present-day Germany, however, encountered the entrenched Funnel-Beaker Folk, who resisted Corded-Ware culture for several centuries.
Germanic evolved as a hybrid language, with elements of proto-Balto-Slavic, the language of the Baltic Battle-Axe invaders, an Indo-European dialect, probably sibling to proto-Italic, derived from an earlier Kurgan-derived intrusion (e.g. Globular Amphora or Baden culture), an adstratum of the non-Indo-European languages of the indigenous Funnel-Beaker (and Erteboelle?) culture. The term ``language hybrid'' may offend professional linguists, since independent languages don't ``interbreed.'' But mutually intelligible dialects of the same language do interbreed and during the Corded-Ware expansion Baltic and Italic speakers could probably make sense of each other, as the common ancestor was only a few centuries in the past. (Similarly, Saxon and Viking languages were probably mutually intelligible at the time of Alfred the Great: hence the many Viking words in English.)
The very name ``Corded-Ware'' provides a strong affirmation of the Gimbutas Theory. This refers to leather cords which were added to pottery as ornaments. (Anti-Gimbutists don't like to admit it but before the advent of Corded-Ware, the Kurgan people were applying cord ornaments to pottery.) The relationship between cord and ornament is preserved in Germanic languages! Consider two cognates in Dutch:
touw -- cord tooi -- ornament
