The Homeland of Indo-European
and Culture: Some Thoughts
presented at a seminar organized by the Indian Council for
Historical Research on the same theme in Delhi on 7-9 January 2002
There is an academic tradition that while discussing the origin of the railway engine one has perforce to get back to the story of the water-filled kettle which, when heated, emitted hot vapours and the person watching it got the brilliant idea that from the steam thus produced one could invent a steam engine that could propel very heavy weights. Though this simile is not an exact one in the present context, one must nevertheless go back to 1786 when a Calcutta High Court judge, Sir William Jones (1788), sprang a surprise by declaring in his Presidential Address to the Bengal Branch of the Royal Asiatic Society that Sanskrit, Greek and Latin had so much in common that it was difficult to view these as mutually exclusive entities.
Little did the poor judge realize that his innocuous judgement will lead to an unending cut-throat debate for centuries to come.
The observation that Sanskrit, Greek and Latin were very close to one another led to a series of formulations. Thus, it was argued that if these languages were so very similar there must have been an earlier language from which these emerged. To this hypothetical language the name given was “Indo-European”, since the three languages just mentioned belonged, on the one hand, to India and, on the other, to Europe. It was further argued that there must have been some common people who spoke this ancestral language, and they were called “Indo-Europeans”. And the final corollary was that there must have been an “original home” of these people, and thus began the hunt for this “Urheimat” — an exercise that knows no ending in spite of sustained efforts of hundreds of scholars for over two hundred years.
A formulation of the branches and sub-branches of this hypothetical Indo-European language has been made out to be as follows:
Initially, some scholars opined that India, being the home of the earliest extant literature (viz. the Vedas) of the Indo-European group of languages, must have been the original homeland of these languages. However, soon the canvas got enlarged so as not to limit it to India but to include a large part of central Asia. Thereafter the scenario was taken to Europe and, not surprisingly, almost every part thereof was declared to be the homeland: Scandinavia, Finland, south-west Russia, the Baltic area, Germany, Lithuania, Hungary, the Danube valley and so on. In fact, the rat race was so much that no part of Europe was left out. In a very sarcastic comment on this race, Jean-Paul Demoule (1980:120) averred: “We have seen that one primarily places the IE’s [Indo-Europeans] in the north if one is German … in the east if one is Russian, and in the middle if, being Italian or Spanish, one has no chance of competing for the privilege.”
However, in the course of time this Eurocentric approach began its climb-down and many new regions were upgraded. As of now, the more important claims to the Indo-European homeland pertain, in their geographical locations from west to east, to: (i) the Anatolian region of western Asia; (ii) the Black Sea-Caspian belt; (iii) the steppes of southern Russia; and (iv) Sogdiana in south-central Asia. We shall deal with these claims first, after which the possibility of North-west South Asia having been the homeland will also be considered.
The first question that has to be addressed while looking for the original home of these Indo-Europeans is at what cultural level were they: hunter-gatherers, pastoral nomads or sedentary agriculturists? An attempt in this direction was made by trying to find out the common words pertaining to agriculture, animals, plants, climate, etc. in the various Indo-European languages and thereby to determine the techno-cultural level and the geophysical surroundings of these hypothetical Indo-Europeans. Since the reconstructed terms in many cases are faulty, no consensus on this issue could be arrived at. Thus, some scholars would like to visualize these Indo-Europeans as nomads roaming about on steppes in cold climate, whereas others would regard them as sedentary agriculturists, living in an area with a moderate climate and endowed with wild cereal plants and animals that could be easily domesticated.
The Anatolian homeland : The noted proponent of the Anatolian homeland is Collin Renfrew who believes (1987) that the Indo-Europeans are first identifiable in Anatolia where they practised agriculture around 7000 BCE. And it was from here that one of their groups moved westward to Europe, crossing the Bosporus and another group, moving eastward, via the region south of the Caucasus mountains and the Caspian Sea, into Iran from where it must have subsequently entered Afghanistan and India. In an alternative scenario, Renfrew thinks that the Indo-Europeans split up after entering Europe and then the eastern branch went to south-central Asia, via north of the Black and Caspian Seas, whence it moved on to north-eastern Iran, Afghanistan and India. The Anatolian hypothesis, however, falters on at least two major counts. In the first place, if the Europeans, on the one hand, and the Indo-Iranians, on the other, had once lived together as agriculturists in Anatolia, they ought to have a common vocabulary for agricultural items, which unfortunately is not the case. Secondly, the Hittite language of Anatolia, on which this commonness has been perceived, was a “minority” language, probably of the elites, whereas the basal language was non-Indo-European. This is hardly tenable with the concept of the Indo-Europeans having been the original inhabitants of this area. On the Anatolian agricultural setting vis-à-vis the Indo-Europeans there comes a bitter comment, from across the Atlantic, by Lamberg-Karlovsky (1988:2) who snaps: “… [The whole issue has been] simplified by Professor Renfrew to the ludicrous formula 7000 B.C.E. Anatolia = farming = Indo-Europeans.”
The Caucasus Region : Shifting the focus a little to the east, Gamkrelidze and Ivanov (1995) place the homeland in the region between the Black and Caspian Seas. Their thesis is based primarily on linguistic palaeontology, applying which they aver that the homeland was a mountainous terrain, replete with lakes and fast-flowing rivers, which their proposed region, they argue, has in ample measure. In an attempt to add further weight to their thesis, they argue that since the reconstructed Indo-European language has a number of Semite loan-words, the homeland could not have been far away from the Semitic world. However, the thesis of there being a large number of Semitic loan-words in the Indo-European language has been shown by many a scholar as a misplaced belief. Besides, there is another linguistic incongruity in the thesis, viz. that the Armenian language, which is spoken in the area between the Black Sea and the Caspian Sea, is replete with a strong element of non-Indo-European vocabulary, suggesting that there was a sizable population substratum and consequently the Indo-European could not have been the original language of the area.
The “Kurgan” Homeland : In yet another scenario, the venue is shifted to the steppes lying to the north of the Black and Caspian Seas. Over here archaeological remains of a culture typified by burial barrows (called “kurgan” in the Slavic language) have been met with. The chief proponent of this thesis is Maria Gimbutas (1966 & 1997). According to her, the reconstructed linguistic evidence suggests that the Indo-Europeans were horse-riding warriors who used thrusting weapons and could easily overrun other areas, and did do so in so far as central Europe is concerned, around the fourth-fifth millennia BCE. On the techno-cultural level, the Kurgan people were essentially at a pastoral stage. Discounting this equation, Renfrew (1999: 268) holds that on the European scene mounted warriors appear only as late as the turn of the second-first millennia BCE and these could in no case have been Gimbutas’s Kurgan warriors predating the facts by some 3,000 years. On the linguistic turf, there comes a severe attack by Kathrin Krell (1998) who finds a great incongruity between the terms found in the reconstructed Indo-European language and the cultural level met with in the kurgans. For example, Krell holds that the Indo-Europeans had reached an agricultural level whereas the Kurgan people were just at a pastoral stage. There are others, like Mallory and Schmitt, who are equally critical of Gimbutas’s hypothesis.
The Sogdiana Homeland : Descending to the south-east, another homeland has been suggested, namely that in Sogdiana, by Johanna Nichols (1997 a and b). From this homeland, Nichols holds that there was a spread of the Indo-European language to the area surrounding the Aral Sea. From there a two-fold spread has been envisaged: a major one to the areas lying to the north of the Caspian and Black Seas, and a comparatively minor one along the southern side of these seas. However, a more noteworthy point in Nichols’s schema is that there was only a language-spread and not a migration of people. While such a stand absolves the proponent from producing proof in terms of the material culture as well, one has yet to be fully convinced of the hypothesis that a language can keep on spreading from area to area, without involving the language-carriers, viz. the people themselves. Also, the Sogdiana model does not fall in line with the normative model wherein there is a centre-to-periphery spread, like the centre-to-periphery ripples created when an object is dropped on the surface of placid water and the ripples start moving in circles towards the periphery. In the Sogdiana thesis, one might well ask: “Why was there no movement towards the east?” The matter requires an in-depth study, but a provisional answer may well be that if there is an obstruction on one side, say the wall of the water pool close to the centre from where the ripples start, the obstruction will not let the ripples go in that direction. Maybe a very strong presence of an altogether different language, in terms of both its lexical content and structural behaviour, stood as a buffer against any penetration of the Indo-European language to the east. Nichols’s model, proposed only recently, has yet to be fully evaluated by linguists. However, as of now there are no noteworthy dissenting voices.
In the preceding pages we have seen the “kurganization” (i.e. burial under burrows) of three of the four theses, viz. those relating to Anatolia, Caucasus and the Russian steppes. Also, as already mentioned, the Sogdiana one has yet to stand some rigorous tests. We have also noted that in the case of all the three aforementioned theses, for each proponent there are at least half-a-dozen opponents. (For the sake of the brevity of this paper, which is to be presented within a maximum of thirty minutes, I have referred to only one or two instances in each case.) A very interesting point that emerges from these controversies is that the dispute is not merely between archaeologists on the one hand and linguists on the other, who flaunt their own discipline as being superior that of the other, but amongst archaeologists themselves and similarly amongst linguists, indicating that not even two practitioners of the same discipline see eye to eye. Completely disillusioned with such a scenario, Mallory (1989:143) rightly observed: “One does not ask ‘where is the Indo-European homeland?’ but rather ‘where do they put it now?’ ” (Emphasis mine.)
North-west South Asia : On my part, however, I would not like the search to be given up and would only indicate the region to go to now. Let us move full circle and try out the Indian homeland thesis which was proposed at the end of the eighteenth century, but could not hold the ground. I am aware that there would be an instantaneous uproar at this proposal, but why be allergic to the very idea itself? The reason for a fresh examination of this proposal is that in those early days there was a total absence of any archaeological data from north-west South Asia, which we now have in abundance. Let me also make it clear that I want the north-west South Asian region as a whole to be re-examined and not merely what is now left over as north-west India. Let it also be emphasized that the present-day political boundaries did not exist during those ancient days we are dealing with here.
The evidence from the excavations at Mehrgarh (Jarrige) has demonstrated that the north-western part of the Indian subcontinent had reached a neolithic, i.e. settled agricultural stage, by the seventh millennium BCE. Here it may also be emphasized that the Mehrgarh neolithic complex stands in marked contrast to that of Western Asia. For example, whereas in the West Asian neolithic there is the domination of sheep and goat amongst the domesticated animals and of wheat amongst the cultivated cereals, in the Mehrgarh context the cattle dominated over other animals and barley over other cereals. Thus, the Mehrgarh neolithic has its own identity, having no generic relationship with its West Asian counterpart. In other words, the Mehrgarh people were the “the sons of the soil”.
Further, there is a continuous story from the succeeding chalcolithic level onwards, taking us through various evolutionary stages to the Early Harappan from which there emerged the Harappan Civilization itself, around the middle of the third millennium BCE. Again, after a thorough study of the human skeletal remains, Hemphill and his colleagues (1991) have shown that there was a biological continuity right from 4500 BCE to 800 BCE. A question may now be posed: “What language did these chalcolithic people speak?” Though the Harappan script has not yet been deciphered, in spite of so many tall claims, we have yet another way of tackling the issue.
In the Rigveda, the Sarasvati has been stated to be a mighty river flowing from the mountains to the sea (RV 7.95.2). By the time of the Panchvimsa Brahmana (XXV.10.16) it dried up. When did this drying up of the Sarasvati take place? The answer is provided by the evidence from the excavations at Kalibangan which stood on the bank of the Sarasvati, now going by the name of the Ghaggar. Radiocarbon dates indicate that the Mature Harappan settlement at Kalibangan had to be abandoned around 2000-1900 BCE. And, as the hydrological evidence indicates, this abandonment took place on account of the drying up of the Sarasvati. This latter part is duly established by the work of Raikes, an Italian hydrologist, and of his Indian collaborators. Raikes (1968) has very significantly titled his paper, “Kalibangan: Death from Natural Causes”. Thus, an in-depth study of the literary-cum-archaeological-cum-hydrological-cum-radiocarbon evidence duly establishes that the Rigveda (which, to recall, speaks of the Sarasvati as a mighty river) must antedate ca 2000 BCE. By how many centuries, it can be anybody’s guess.
We may now take up the geographical evidence yielded by the Rigveda itself. The famous Nadi-stuti hymn (RV 10.75.5-6) refers to the then familiar rivers in a serial order from the east to the west, beginning with the Ganga and ending up with the Indus along with its western tributaries such the Kabul, Kurram, etc., encompassing eastern Afghanistan as well. One might now pose another question: “Which archaeological culture/civilization was there in this very area during the period preceding 2000 BCE?” The inescapable answer will have to be: “The Harappan Civilization”. In other words, the entire circumstantial evidence points to a correlation between the Vedas and the Harappan Civilization. A final seal on this, however, can be put only when the Harappan script is satisfactorily deciphered. Lack of time prevents me from dealing right here with the various objections raised against the Harappan-Vedic equation. These have been dealt with in great detail in my just-published book, The Sarasvati Flows On.
Putting together the various parts of this jigsaw puzzle, it would mean that if the Vedas reflect the literary counterpart of the Harappan archaeological complex, the Harappans spokes a language called Sanskrit. And since the Harappan Culture had its roots going deep at least into the fifth millennium BCE, it would imply that the Sanskrit-speakers were there in this area as early as that. Further, had the Sanskrit-speaking people not been the original inhabitants of this region, we would have got evidence thereof in terms of a substratum language, which we really do not have. The presence of a few Dravidian words in the Vedas can be explained by an adstratum and not necessarily by a substratum. As explained elsewhere by the present author (in press), the Harappans came in lateral contact with the Southern Neolithic people who, in all probability, were speakers the Dravidian language.
We now turn to yet another important piece of evidence. The Boghaz Kuei inscription, dating back to the fourteenth century BCE, refers to Indra, Mitra, Nasatya and Varuna as witnesses to a treaty between the Mitanni king Matiwaza and the Hittite king Suppiluliuma. There is also the evidence furnished by a text on the training of horses, which uses typical Sanskrit terms like ekavartana, trivartana, etc. Further, there are many Indian names in the region going back to circa seventeenth century BCE. After a thorough examination of the entire evidence, the renowned scholar T. Burrow (1955:29) came to the conclusion: “The Aryans appear in Mitanni from 1500 BC as the ruling dynasty, which means that they must have entered the country as conquerors.” If so, from where could have these conquerors come? Around 1500 BCE there was no other country in the entire world except India where these above-mentioned gods were worshipped. (As we have shown earlier, the Rigveda is datable to at least 2000 BCE, if not earlier.) Putting two and two together, it is clear that the immigrants to Anatolia were from no area other than India. This movement is likely to have taken place along the belt lying south of the Black and Caspian Seas.
The above model is not much different from that of Nichols. The shift of the “original homeland” from Sogdiana to a few hundred miles to the south — i.e. to the region now comprising eastern Afghanistan, Pakistan and north-west India should not upset anyone, since the archaeological-cum-literary evidence from this area is more positive than that from Sogdiana.
The author would gratefully welcome comments on his hypothesis.
© Indian Council for Historical
Research, New Delhi.
Reproduction or publication not permitted.
P.S. When this paper was being written in November 2001, the author received from Dr. Edwin Bryant a copy of his book, The Quest for the Origins of Vedic Culture (Oxford University Press, 2001). Dr. Bryant deserves to be congratulated on presenting a very good analysis of the entire issue, particularly without any pre-conceived notions.
Burrow, T. 1955 (Reprint). The Sanskrit Language. Faber and Faber.
Demoule, Jean-Paul. 1980. “Les Indo-européens: Ont-ils existé?” L’Histoire 28:108-120.
Gamkrelidge, Thomas V. and Vjacelav V. Ivanov. 1995. The Indo-European and the Indo-Europeans. Trends in Linguistic Studies and Monographs 80. Berlin: Mouton and Gruyter.
Gimbutas, M. 1966. “Proto-Indo-European Culture: The Kurgan Culture during the Fifth, Fourth and Third Millennia BC.” In George Cordona et al. (eds.) Indo-European and Indo-Europeans, pp. 155-197. Philadelphia: University of Pennsylvania Press.
---------- 1997. The Kurgan Culture and the Indo-Europeanization of Europe. Ed. Miriam Dexter and Karlene Jons-Bley. Washington, D. C.: Institute for the Study of Man.
Hemphill, B. E., J. R. Lukacs and K.A.R. Kennedy. 1991. “Biological Adaptations and Affinities of Bronze Age Harappans”. In R. H. Meadow (ed.) Harappa Excavations 1986-1990, pp. 137-182. Madison, Wisconsin: Prehistory Press.
Jarrige, Catherine, et al. 1995. Mehrgarh: Field Reports 1974-1985 (From Neolithic Times to Indus Civilization). Karachi: Department of Culture and Tourism, Government of Sindh, Pakistan, in collaboration with French Ministry of Foreign Affairs.
Jones, William. 1788. The third Anniversary Discourse. Asiatick Researches, Vol. I, pp. 415-431.
Krell, Kathrin S. 1998. “Gimbutas’ Kurgan-PIE Homeland Hypothesis: A Linguistic Critique”. In Roger Blench and Mathew Spriggs (eds.) Archaeology and Language, II:267-289. London: Routledge.
Lal, B. B. 2002. The Sarasvati Flows On. New Delhi: Aryan Books International.
-------- (In press). “Aryan Invasion of India: Perpetuation of a Myth.” In E. Bryant and L. Patton (eds.) Evidence and Evocation in Indian History. London: Curzon.
Lamberg-Karlovsky, C. 1988. “Indo-Europeans: A Near-Eastern Perspective”. Quarterly Review of Archaeology 9 (1): 1-10.
Mallory, J. P. 1989. In Search of the Indo-Europeans. London: Thames and Hudson.
Nichols, Johanna. 1997a. “The Epicentre of the Linguistic Spread.” In Roger Blench and Matthew Spriggs (eds.) Archaeology and Language I: 122-148. London: Routledge.
---------- 1997b. The Eurasian Spread Zone and the Indo-European Dispersal. In Roger Blench and Matthew Spriggs (eds.) Archaeology and Language, II. London: Routledge.
Raikes, Robert. 1968. Kalibangan: Death from Natural Causes. Antiquity, XLII: 286-291.
Renfrew, C. 1988. Archaeology and Language. New York: Cambridge University Press.
--------- 1999. Time Depth, Convergence Theory and Innovation in Proto-Indo-European: “Old Europe” as a PIE Linguistic Area, Journal of Indo-European Studies, 27 (3-4): 258-293.