4.9  The Evidence from physical anthropology

4.9.1. Continuity between castes

Half a century ago, Dr. Ambedkar surveyed the existing data on the physical anthropology of the different castes in his book The Untouchables.  He found that the received wisdom of a racial basis of caste was not supported by the data, e.g.: �The table for Bengal shows that the Chandal who stands sixth in the scheme of social precedence and whose touch pollutes, is not much differentiated from the Brahmin (�) In Bombay the Deshastha Brahmin bears a closer affinity to the Son-Koli, a fisherman caste, than to his own compeer, the Chitpavan Brahmin.  The Mahar, the Untouchable of the Maratha region, comes next together with the Kunbi, the peasant.  They follow in order the Shenvi Brahmin, the Nagar Brahmin and the high-caste Maratha.  These results (�) mean that there is no correspondence between social gradation and physical differentiation in Bombay.�

A remarkable case of differentiation in skull and nose indexes, noted by Dr. Ambedkar, was found to exist between the Brahmin and the (untouchable) Chamar of Uttar Pradesh. But this does not prove that Brahmins are foreigners, because the data for the U.P. Brahmin were found to be very close to those for the Khattri and the untouchable Chuhra of Panjab.  If the U.P. Brahmin is indeed �foreign� to U.P., he is by no mean . s foreign to India, at least not more than the Panjab untouchables.  This confirms the scenario which we can derive from the Vedic and ItihAsa-PurANa literature: the Vedic tradition was brought east from the Vedic heartland by Brahmins who were physically indistinguishable from the lower castes there, when the heartland in Panjab-Haryana at its apogee exported its culture to the whole Aryavarta (comparable to the planned importation of Brahmins into Bengal and the South around the turn of the Christian era).  These were just two of the numerous intra-Indian migrations of caste groups.

Recent research has not refuted Ambedkar�s views.  A press report on a recent anthropological survey led by Kumar Suresh Singh explains: �English anthropologists contended that the upper castes of India belonged to the Caucasian race and the rest drew their origin from Australoid types.  The survey has revealed this to be a myth.  �Biologically and linguistically, we are very mixed�, says Suresh Singh (�) The report says that the people of India have more genes in common, and also share a large number of morphological traits.  �There is much greater homogenization in terms of morphological and genetic traits at the regional level�, says the report.  For example, the Brahmins of Tamil Nadu (esp. Iyengars) share more traits with non-Brahmins in the state than with fellow Brahmins in western or northern India. (�) The sons-of-the-soil theory also stands demolished.  The Anthropological Survey of India has found no community in India that can�t remember having migrated from some other part of the country.� Internal migration accounts for much of India�s complex ethnic landscape, while there is no evidence of a separate or foreign origin for the upper castes.

Among other scientists who reject the identification of caste (varNa) with race on physical-anthropological grounds, we may cite Kailash C. Malhotra:

�Detailed anthropometric surveys carried out among the people of Uttar Pradesh, Gujarat, Maharashtra, Bengal and Tamil Nadu revealed significant regional differences within a caste and a closer resemblance between castes of different varnas within a region than between sub-populations of the caste from different regions.  On the basis of analysis of stature, cephalic and nasal index, H.K. Rakshit (1966) concludes that �the Brahmins of India are heterogeneous and suggest incorporation of more than one physical type involving more than one migration of people�.

�A more detailed study among eight Brahmin castes in Maharashtra on whom 18 metric, 16 scopic and 8 genetic markers were studied, revealed not only a great heterogeneity in both morphological and genetic characteristics but also showed that 3 Brahmin castes were closer to non-Brahmin castes than [to the] other Brahmin castes.  P.P. Majumdar and K.C. Malhotra (1974) observed a great deal of heterogeneity with respect to OAB blood group system among 50 Brahmin samples spread over 11 Indian states.  The evidence thus suggests that varna is a sociological and not a homogeneous biological entity.�

4.9.2 Family traits

This general rejection of the racial basis of caste does not exclude that specific castes stand out in their environment by their phenotypical or genotypical characteristics.  Firstly, any group that goes on breeding endogamously for generations will have �family traits� recognizable to the regular and sharp observer, at least to a statistically significant extent.  This does not mean that these family traits (rarely distinctive enough to be called �racial� traits) are in any way the reason why one caste refuses to intermarry with another caste, as you would have in the case of racial discrimination.

Secondly, intra-Indian migrations have taken place so that certain caste groups stand out by retaining the physical characteristics of their source region�s population for quite a few generations.  Thus, the Muslim invasions chased some Rajput castes from western India to the Nepalese borderland, and some Saraswat Brahmins from Kashmir to the Konkan region; geneticists ought to be able to find traces of that history.

It is well-known that the Brahmin communities of Bengal and South India originated in the physical importation of Brahmin families by kings who sought accession to the prestigious Vedic civilization and wanted to give extra religious legitimacy to their thrones.  These Brahmin families were brought in from northwestern India where, for obvious geographical reason, people are whiter and closer to the European physical type than in Bengal or the South. (Even so, due to intermarriage and the incorporation of local priesthoods, numerous Brahmins in South India are simply black.) Apart from Brahmins, numerous other caste groups throughout India have histories of immigration, putting them in environments where they differed in genetic profile from their neighbours, e.g. the Dravidian-speaking Oraon tribals of Chotanagpur recall having migrated from Maharashtra along the Narmada river.

The Chitpavan Brahmins of Maharashtra are often mentioned as a caste that stands out by its physical type.  Their slightly more �Nordic� build and the occurrence of blue eyes among them look like the perfect evidence for the theory that the Brahmins are the descendents of the Nordic Aryans who invaded India in 1500 BC.  In fact, it is only during the initial Islamic onslaught that the Chitpavans migrated from the Afghan borderland to their present habitat.

Nevertheless, the Chitpavan case shows that sometimes, such distinctive family traits do coincide with the difference between the higher or lower incidence of the distinctive traits of the white race, esp. the low pigmentation of the skin or, in this case, the eyes.  The difference between castes can in some cases be expressed in terms of the respective distances between their average characteristics and those of the European type.  And this is only to be expected given the basic fact that India is a large country with great variation in physical type and lying in the border zone between the major races.  The rich biological variety in the Indian chapter of the human species is due to many factors, but so far the Aryan Invasion has not been shown to be one of them.

4.9.3. Mixing of castes

The genetic differential between castes has recently been confirmed in a survey in the southern state of Andhra Pradesh. The main finding of the survey, conducted by human-geneticists Lynn B. Jorde (University of Utah) and Bhaskara B. Rao and J.M. Naidu (both with Andhra University), concerned the role of inter-caste marriages: men stay in their castes, while women sometimes go and live with a man from another, mostly higher caste.  In spite of the definition of caste as an �endogamous group�, the fact is that there has always been a marginal mixing of castes as well.  Likewise, even outside the marital framework, upper-class employers (in any society) have made passes at their maid-servants, while prostitutes got impregnated by their higher-class clients, all producing mixed offspring.

Factoring all these marginal mixed-caste births in, the cumulative effect over centuries is that the castes have mixed much more than the theory of caste would lead you to expect.  Over many generations, this mixing had to lead to a thorough genetic kinship even between castes of very divergent origins.  Given these known sociological facts, the scientists naturally found that genetic traits in the male line (Y chromosome) are stable, those in the female line (mitochondrial DNA) considerably less so.  Because inter-caste marriages are mostly between �neighbouring� castes in the hierarchy, the genetic distance between highest and lowest is about one and a half times greater than that between high and middle or between middle and low.

However, none of this requires a policy of racial discrimination nor an Aryan invasion into India: the known history of internal migrations and the general facts about relations between higher and lower classes in all societies can easily account for it. Moreover, the observed differences between Indian communities are much smaller than those between Indians collectively and Europeans (or Africans etc.) collectively.  A provisional table of the genetic distance between populations shows that North-Indians and South-Indians are indeed very close, much closer than �Aryan� North-Indians and �Aryan� Iranians are to each other.

Both sides in the debate should realize that this evidence can cut both ways. If an Aryan or other invasion is assumed, this evidence shows that all castes are biologically the progeny of both invaders and natives, though perhaps in different proportions.  Conversely, if the genetic distance between two castes is small, this still leaves open the possibility that the castes or their communal identities can nonetheless have divergent origins, even foreign versus native, although these are obscured to the geneticist by centuries of caste mixing.

4.9.4. Tribals and �Caucasians�

The one important general difference between two parts of the population is that between a number of tribes on the one hand, and some other tribes plus the non-tribals on the other.  V. Bhalla�s mapping of genetic traits shows that the latter category roughly belongs to the Mediterranean subgroup of the Caucasian race (though by the superficial criterion of skin colour, it can differ widely from the type found in Italy or Greece). incidentally, the term Caucasian as meaning the white race was coined in 1795 by the German scientist Johann Friedrich Blumenbach, who believed that the Caucasus region, particularly Georgia, �produces the most beautiful human race�, and that it was the most likely habitat of �the autochthonous, most original forms of mankind�. Thus, the typically Caucasian Rhesus-negative factor is �conspicuous by its absence� in the Mongoloid populations of India�s northeast, but the non-tribal populations �show a moderately high frequency of 15% to 20% but not as high as in Europe� of this genetic trait.

Bhalla lists a number of specific genes which are characteristically strong or weak in given racial types, and finds that they do define certain ethnic sub-groups of India, esp. the Mongoloid tribals of the northeast, the Negritos of the Andaman Islands, and the Australoids in the remaining tribal pockets of the south.  Everywhere else, including in many tribal areas, the Mediterranean type is predominant, but the present battery of genetic markers was not able to distinguish between subtypes within this population, much less to indicate different waves of entry.

In fact, no �entry� of these Mediterranean Caucasians can be derived from the data, certainly not for the post-Harappan period.  According to an older study, they were present even in South India in 2,000 BC at the latest: �The evidence of two racial types, the Mediterranean and the Autochthonous proto-Australoid, recognized in the study of the skeletal remains from the neolithic levels at Brahmagiri, Piklihal, Tekkalakota, Nevasa etc., seems to suggest that there was a thick population consisting mainly of these two races in South India around 2000 BC.�

The Caucasian race was present in India (like in Europe and the Kurgan area) since hoary antiquity.  Kailash Malhotra reports, starting with their geographical spread today: �The Caucasoids are found practically all over the country, though the preferred habitats have been river valleys and plains.�

In the past, the Caucasian presence was also in evidence: �Although a large number of prehistoric sites have been excavated in India, only a few of them have yielded human osseous remains (�) None of the pre-Mesolithic sites have yielded skeletal material; the earliest remains are around 8,000 years old.  An examination of the morphological features of skeletons from sites of the Mesolithic, Neolithic, Chalcolithic and iron age periods reveals the presence of Australoids and Caucasoids in all the periods, the absence of Mongoloids, and the existence of at least two types of Caucasoids, the dolichocephals and the brachycephals (�) The skeletal evidence thus clearly establishes the presence of Australoids and Caucasoids in India for at least 8,000 years.�

All that can be said, is that the population of India�s northeast is akin to that of areas to India�s north and east, that of the southeast to that of countries further southeast, and the bulk of the Indian population to that of areas to India�s west.  Probably a large demographic expansion from India�s northwest to the east and south took place during and at the end of the Harappan period (2,000 BC).  It is logical to infer that the populations of the Mediterranean type were more concentrated in the northwest prior to that time; but it does not follow that they came from the outside.  India�s northwest simply happened to be the easternmost area of Caucasian habitation, just like India�s northeast happens to be the frontier of the Mongoloid type�s habitat.

For politically correct support in denying the racial divide between tribals and non-tribals, we may cite the Marxist scholar S.K. Chatterjee, who dismissed the notion of distinct races in India, be they Aryan, Dravidian, Mongoloid or Austro-Asiatic.  He called the Indian people a �mixed people, in blood, in speech and in culture�.

Though the Christian missionaries have been the champions of tribal distinctness, Christian author P.A. Augustine writes about the Bhil tribals: �The Bhils have long ceased to be a homogeneous people. In the course of millennia, various elements have fused to shape the community.  During their long and tortuous history, other aboriginal groups which came under their sway have probably merged with them, losing their identity.  One can see a wide range of physical types and complexion.  The variation in complexion is very striking indeed, ranging between fair to quite dark-skinned (�) There is no consensus among scholars on the exact ethnic character of the Bhils, They have been alternatively described as proto-Australoid, Dravidian or Veddoid.� The same racial �impurity� counts for most Indians, tribal as well as non-tribal.  While not by itself disproving the Aryan invasion, it should prove even to invasionists that all Indians are descendents of both indigenous and so-called invader populations.

4.9.5. Language and genetics

While it is wrong to identify a speech community with a physical type, it is also wrong to discard physical anthropology completely as a source of information on human migrations in pre-literate times.  Lately, findings have been published which suggest that, for all the racial mingling that has taken place, there is still a broad statistical correlation between certain physical characteristics and nations, even language groups.

Thus, the percentage of individuals with the Rhesus-negative factor is the highest (over 25%) among the Basques, a nation in the French-Spanish borderland which has preserved a pre-IE language.  Other pockets of high incidence of Rh-neg. (which is nearly non-existent among the Bantus, Austroloids and Mongoloids) are in the same part of the world: western Morocco, Scotland and, strangely, the Baltic area, or apparently those backwater regions least affected by immigrations of the first Neolithic farmers (from the Balkans and Anatolia), the Indo-Europeans, and in Morocco also the Arabs.

Another European nation which stands out, at least to the discerning eye of the population geneticist, is the Sami (Lapp) population of northern Scandinavia: when contrasted genetically with the surrounding populations, the Sami genetic make-up �points to kinship with the peoples of North Siberia� eventhough they now resemble the Europeans more than the native Siberians. This confirms the suspicion of an Asian origin for the Uralic-speaking peoples of which the Sami people is one.

Where a small group of people have spread out over a vast area and lived in isolation ever since, as has happened in large parts of America in the past 20,000 years, genetic differentiation and linguistic differentiation have gone hand in hand, and the borderline between genetic types usually coincides with a linguistic borderline: �Joseph Greenberg distinguishes three language families among the Native Americans: Amerind, Na-Dene and Eskimo-Aleut. (�) According to Christy Turner of Arizona University, Native American dental morphology indicates three groups, which coincide with Greenberg�s.  Luigi Cavalli-Sforza from Stanford investigated a variegated set of human genes. His results equally point in the direction of Greenberg�s classification.�

Linguistic difference between populations may coincide with genetic differences; and likewise, linguistic mixing may coincide with genetic mixing.  A perfect illustration is provided by Nelson Mandela, leader of the anti-Apartheid struggle and belonging to the Xhosa nation.  His facial features are more Khoi (Hottentot) than Bantu, and his language, Xhosa, happens to be a Bantu language strongly influenced by the Khoi-San (Hottentot-Bushman) languages, most strikingly by adopting the click sounds.  In this case, genetic mixing and linguistic mixing have gone hand in hand.

However, in and around the area of IE expansion, a notorious crossroads of migrating peoples, the remaining statistical correlation between genetic traits and language groups is less important than the evidence for the opposite phenomenon: languages spreading across genetic frontiers.  In India, the only neat racial division which coincides with a linguistic borderline is between the mainland and the Andamans: though so-called Negrito features are dimly visible in the population of Orissa and surrounding areas, the pure Negrito type is confined to the Andamans, along with the Andamanese language group.  For the rest, in India, like in Central Asia or Europe, i.e. in areas with lots of migration and interaction between diverse peoples, genetic and linguistic divisions only coincide by exception.

Thus, the Altaic languages are spoken by the Mongolians, eponymous members of the Mongoloid race, and by the Turks, who have mixed so thoroughly with their Persian, Armenian, Greek and Slavic neighbours that they now belong to the Caucasian race.  The Hungarians are genetically closer to their Slavic and German neighbours than to their linguistic cousins in the Urals.  India being the meeting-place (or rather, mixing-place) of Mongoloid, Caucasian and Austroloid racial strands, it is naturally impossible to identify the speakers of the different Indian language-groups with different races.

Asked whether there are �concordances between genetic data and languages�, L.L. Cavalli-Sforza, the world�s leading population geneticist, explains: �Yes, very much so.  Our genealogical tree [of genetic traits] corresponds remarkably well with the table of linguistic families.  There are a few exceptions e.g. the Lapps, genetically rather European, have preserved the language they spoke in their Siberian-Uralic homeland.  The Hungarians, similarly, speak an Uralic language while being predominantly European.  In the late 9th century AD, the Magyar invaders in Hungary, then called Pannonia, imposed their language on the natives. (�) What counts from a genetic viewpoint, is the number of invaders relative to the natives.  As the Hungarians were not very numerous, they left only a feeble genetic imprint on the population.� So, the replacement of native languages by those of less civilized but stronger invaders is a real possibility (it is also what the Greeks did to the Old Europeans), though it becomes less probable in proportion to the size and the cultural superiority of the native population.

The reason why the replacement of native languages by the languages of genetically distinguishable invaders remains relatively exceptional, is this: �In a traditional culture, language is transmitted vertically from parents to children, just like the genes.  But in some conquests or in civilizations with schools, there is also horizontal transmission and substitution of languages.  The Romans organized schools in their part of Europe and thereby managed to replace the native languages by their own.  But this type of phenomenon is relatively recent.  In 90% of its history, mankind consisted of hunter-gatherers speaking tribal languages.  That is why the genetic tree has preserved a strong concordance with the linguistic tree.�

A typical example are the Basques: �The Basque language is the direct descendent of a language which must have arrived along with modem mankind, say 30,000 years ago.  It is [in Europe] the only pre-Indo-European language which has been preserved.  Why? 

So, the Basques are both biologically and linguistically the straight descendants of Old Europeans.  Most other Europeans are biologically the progeny of the non-IE-speaking Old Europeans, with some admixture of the Asian tribes who originally brought the IE languages into Europe.  These immigrants may have differed somewhat from the average European type, into which their smaller number got genetically drowned over the centuries.  Linguistically, most non-Basque (and non-Uralic) Europeans are the progeny, through adoption, of the IE-speaking invaders.

4.9.6. The original �Aryan race�

Is there anything we can say about the ethnic identity of the nomads or migrants who spread the early IE languages, if only to help physical-anthropologists to recognize them when found at archaeological sites?  Competent authorities have warned against the �semi-conscious prejudices on original genetic characteristics of the Indo-Europeans: they are supposed to be blond and blue-eyed�. This prejudice has even been reinforced recently by the discovery of blond-haired mummies of presumably IE-speaking people in the Xinjiang province of China.

The fact that the IE speech community includes people of diverse race, from the dark-skinned Sinhalese to the white-skinned Scandinavians, definitely implies that the spread of the language cannot be equated with the spread of a racial type.  Languages can and do migrate across racial boundaries.  That the IE languages crossed racial frontiers during their expansion accords well with established perspectives on the spread of IE, e.g. by I.M. Diakonov:

�These expanding tribes met local, poor and hungry sparser populations, often consisting of hunters and cattle-breeders.  The migrants started to merge with the local population, giving them their language and cultural achievements.  But in some cases, the local population may have been larger in numbers than the migrants.  In some historical situations the language of the minority, if it was widely used and understandable on a vast territory, could be accepted as lingua franca, and later as the common language, particularly if it was a language of cattle-breeders (cf. the examples of the Semites and the Turks).  The area of the newly created population became itself a centre of population spread, and so on.  Bloody conquests could take place in some instances; in others it was not the case, but the important thing to realize is that what migrated were languages, not peoples, although there had to be at least a handful of users of the languages, though not necessarily native speakers.�

On the other hand, the fact that the PIE-speaking community must have been a fairly small ethnic group, living together and marrying mostly within the community, implies that they must have belonged collectively to a fairly precisely circumscribed physical type.  Even if you throw together people from all races, after a few generations of interbreeding they will develop a common and distinctive physical type, with atavistic births of people resembling the pure type of one of the ancestral races becoming rarer and rarer.  Therefore, in the days before intercontinental travel and migrations, a speech community was normally also a. kinship group (or, in strict caste societies, a conglomerate of kinship groups) presenting a fairly homogeneous physical type.

During the heyday of the racial theories, a handful of words in Greek sources were taken to mean that the ancient Indo-Europeans were fair-haired and had a tall Nordic-looking build.  In Homer�s description, the Greek heroes besieging Troy were fair-haired.  The Egyptians described the �Sea Peoples� from the Aegean region (and even their Libyan co-invaders, presumably Berber-speaking) as fair-haired.  The Chinese described the Western (Tokharic) barbarians likewise.

However, the incidence of Nordic looks was not necessarily overwhelming. In classical Greek writings, the Thracians and Macedonians (most notably Alexander the Great), whose language belonged to an extinct Balkanic branch of the IE family, are mentioned as being fair-haired; apparently most Greeks were by then dark enough to notice this fair colour as a trait typical of their �barbaric� northern neighbours.  The Armenians have a legend of their own king Ara the Blond and his eventful personal relationship with the Assyrian queen Sammuramat/Semiramis (about 810 BC), who is known to have fought Urartu (the pre-IE name of Armenia, preserved in the Biblical mountain name Ararat). The use of �the blond� as a distinctive epithet confirms the existence of fair-haired people in Armenia, but also their conspicuousness and relative rarity.

All this testimony, along with the Xinjiang mummies and the presence of Nordic looks in the IE-speaking (Dardic/Kafiri) tribes in the Subcontinent�s northwestern valleys, does suggest a long-standing association between some branches of the IE family and the genes which program their carriers to have fair hair and blue eyes.  These traits give a comparative advantage for survival in cold latitudes: just as melanine protects against the excessive intake of ultraviolet rays in sunny latitudes, lack of melanine favours the intake of ultraviolet.  This segment of the sunrays is needed in the production of vitamin D, which in turn is needed in shaping the bones; its deficiency causes rachitis and makes it difficult for women to birth - a decisive handicap in the struggle for life.  The link between northern latitudes and the light colour of skin, hair and eyes in many IE-speaking communities only proves what we already knew: IE is spoken in fairly northern latitudes including Europe and Central Asia.  Yet, none of this proves the fair-haired and blue-eyed point about the speakers of the original proto-language PIE.

Suppose, with the non-invasion theorists, that the original speakers of IE had been Indians with dark eyes and dark hair; then, according to I.M. Diakonov: �if this population had migrated together with the languages, blue-eyed Balts could not have originated from it.  Blue eyes, as a recessive characteristic, are met everywhere from Europe to the Hindu Kush.  But nobody can be blue-eyed if neither of his/her parents had blue-eyed ancestors, and a predominantly blue-eyed population cannot originate from ancestors with predominantly black eyes.�

This allows for two possible scenarios.  Either the PIE speakers were indeed blue-eyed and fair-haired: that is the old explanation, preferred by the Nazis. Or the blue-eyed people of Europe have not inherited their IE languages from their biological ancestors, but changed language at some point along the genealogical line, abandoning the pre-IE Old European language of their fair ancestors in favour of Proto-Germanic, Proto-Baltic, Proto-Slavic etc., based on the language of the invaders from Asia.  The latter scenario would agree with I.M. Diakonov�s observation: �The biological situation among the speakers of modern Indo-European languages can only be explained through a transfer of languages like a baton, as it were, in a relay race, but not by several thousand miles� migration of the tribes themselves.�

That this is far from impossible is demonstrated by the Turks who, after centuries of mixing with subdued natives of West Asia and the Balkans, have effectively crossed the racial borderline from yellow to white.  But against using this Turkish scenario as a simile for the story of IE dispersal, one could point out that some eastern Turkic people, such as the Kirghiz and the Yakut, are still very much Mongoloids.  However, far from forming a contrast with the IE state of affairs, this makes the simile more splendid: if IE spread from a non-white to a white population, it also remained the language of numerous non-whites (though technically �Caucasians�), viz. the Indians.  On the Eurasian continent, South-Asians still constitute more than half of the wider IE speech community; the Indian Republic alone has more IE speakers than the whole of Europe.

It is perfectly possible that the PIE language and culture were developed after a non-white group of colonists from elsewhere settled among and got racially immersed in a larger whitish population.  As we saw in our speculations about IE-Austronesian kinship and about Puranic history, it is at least conceivable that Aryan culture in India started after �Manu� and his dark-skinned cohorts fled the rising sea level by moving up the Ganga and settling high and dry in the upper Ganga basin, whence their progeny conquered areas to the northwest with ever whiter-skinned and lighter-haired populations: the Saraswati basin, the upper Indus basin, the Oxus riverside, the peri-Caspian region.  By the time these Indian colonists settled in eastern Europe with their Kurgans, their blackness had been washed off by generations of intermarriage with white people of the type attested by the Xinjiang mummies. (Likewise, their material culture had been thoroughly adapted to their new habitat, hence de-indianized.)

So, it is perfectly possible that the Aryan heartland lay farther to the southeast, and that, like eastern Europe in the later 5th millennium BC, the Panjab area a few centuries earlier was already a first area of colonization, bringing people of a new and whiter physical type into the expanding Aryan speech community which was originally darker.  While the Panjabi is physically very similar to the European, the Bihari, Oriya or Nepali is markedly less so, and yet it is possible that he represents more closely the ultimate Proto-Indo-European.

4.9.7. The race of the Vedic Aryans

As for the Vedas, the only ones whom they describe as �golden-haired� are the resplendent lightning gods Indra and Rudra and the sun-god Savitar; not the Aryans or Brahmins.  At the same time, several passages explicitly mention black hair when referring to Brahmins. These texts are considerably earlier than the enigmatic passage in Patanjali describing Brahmins as golden- or tawny-haired (piNgala and kapisha). Already one of Patanjali�s early commentators dismissed this line as absurd.  To the passage from the grammarian Panini which describes Brahmins as �brown-haired�, A.A. Macdonnell notes (apparently against contemporary claims to the contrary): �All we can say is that the above-mentioned expressions do not give evidence of blonde characteristics of the ancient Brahmans.� Considering that Patanjali was elaborating upon the work of Panini, could it have anything to do with Panini�s location in the far northwest, where lighter hair must have been fairly common?

On the other hand, demons or Rakshasas, so often equated with the �dark-skinned aboriginals�, have on occasion been described as red- or tawny-haired (also piNgala or kapisha, the same as Patanjali�s Brahmins). Deviating from the usual Indian line that all these demon creatures are but supernatural entities, let us for once assume that they do represent hostile tribals racially distinct from the Vedic Aryans.  In that case, reference can only be to certain northwestern tribals, among whom fair and red hair are found till today, indicating that they at least partly descended from a fair-haired population.  If the Vedic Aryans were dark-haired and migrated from inside India to the northwest, these odd coloured hairs may have struck them as distinctive.

In modern Anglo-Hindu publications, such as the Amar Chitra KathA religious comics, Rakshasas are always depicted as dark-skinned, a faithful application of the AIT.  Yet, there are instances in Vedic literature where �blackness� is imputed to people whom we know to have had the same (if not a lighter) skin colour than the Vedic Aryans: the Dasas and Dasyus, as Asko Parpola has shown, were the Iranian cousins and neighbours of the Vedic Aryans.  Physical (as opposed to metaphorical) blackness or more generally skin colour was never a criterion by which the Vedic Aryans classified their neighbours and enemies; that precisely is why we have no direct testimony on the Vedic Aryans� own skin or hair colour except through a few ambiguous, indirect and passing references.

4.9.8. Evidence of immigration?

A very recent study, not on crude skull types but on the far more precise genetic traits, confirms the absence of an immigration from Central Asia in the second millennium BC.  Brian E. Hemphill and Alexander F. Christensen report on their study of the migration of genetic traits (with reference to AIT advocate Asko Parpola): �Parpola�s suggestion of movement of Proto-Rg-Vedic Aryan speakers into the Indus Valley by 1800 BC is not supported by our data.  Gene flow from Bactria occurs much later, and does not impact Indus Valley gene pools until the dawn of the Christian era.� The inflow which they do find, around the turn of the Christian era, is apparently that of the well-known Shaka and Kushana invasions.

Kenneth A.R. Kennedy reaches similar conclusions from his physical-anthropological data: �Evidence of demographic discontinuities is present in our study, but the first occurs between 6000 and 4500 BC (a separation of the Neolithic and Chalcolithic populations of Mehrgarh) and the second is after 800 BC, the discontinuity being between the peoples of Harappa, Chalcolithic Mehrgarh and post-Harappan Timargarha on the one hand and the late Bronze Age and early Iron Age inhabitants of Sarai Khola on the other.  In short, there is no evidence of demographic disruptions in the northwestern sector of the subcontinent during and immediately after the decline of the Harappan culture.  If Vedic Aryans were a biological entity represented by the skeletons from Timargarha, then their biological features of cranial and dental anatomy were not distinct to a marked degree from what we encountered in the ancient Harappans.�

Kennedy also notes the anthropological continuity between the Harappan population and that of the contemporaneous Gandhara (eastern Afghanistan) culture, which in an Aryan invasion scenario should be the Indo-Aryan settlement just prior to the Aryan invasion of India: �Our multivariate approach does not define the biological identity of an ancient Aryan population, but it does indicate that the Indus Valley and Gandhara peoples shared a number of craniometric, odontometric and discrete traits that point to a high degree of biological affinity.�

And so, Sir Mortimer Wheeler, one of the great pioneers of the AIT, may be right after all.  Indeed, even he had remarked that �the anthropologists who have recently described the skeletons from Harappa remark that there, as at Lothal, the population would appear, on the available evidence, to have remained more or less stable to the present day.� If anything Aryan really invaded, it was at any rate not an Aryan race.

There are no indications that the racial composition and distribution of the Indian population has substantially changed since the start of the IE dispersal, which cannot reasonably be placed much earlier than 6,000 BC.  This means that even if the IE language is imported, as claimed by the AIT, the IE-speaking people in India are nevertheless biologically native to India.  Or in practice: the use of the terms �aboriginal� and �indigenous� (AdivAsI) as designating India�s tribals, with the implication that the non-tribals are the non-indigenous progeny of invaders, has to be rejected and terminated, even if the Urheimat of the IE languages is found to lie outside India.

One of the ironies of Indian identity politics is that those most vocal in claiming an �aboriginal� identity may well be the only ones whose foreign origin has been securely established.  The Adivasi movement is strongest in the areas where Christian missionaries were numerously present since the mid-19th century to nourish it, viz. in Chotanagpur and the North-East.  Most tribals there speak languages belonging to the Austro-Asiatic and Sino-Tibetan families.  Their geographical origin, unlike that of IE which is still being debated, is definitely outside India, viz. in Southeast Asia c.q. in northern China.

The Tibeto-Burmese tribals of Nagaland and other northeastern statelets are among India�s most recent immigrants.  Many of those tribes have entered during the last millennium, which is very late by Indian standards.  As for the Munda tribes in Chotanagpur, it is not even certain that the ancestors of the present tribes are the authors of the attested Neolithic cultures in their present habitat.  In H.D. Sankalia�s words: �It is an unanswered but interesting question whether any of the Aboriginal tribes of these regions were the authors of the Neolithic culture.� Those who want to give the Austro-Asiatic peoples of India a proud heritage, will find more of it in China and Indochina than in India, e.g. in the Bronze age culture of 2300 BC in Thailand.

On the other hand, biologically the Indian Austro-Asiatics (unlike the Nagas) are much closer to the other Indians than to their linguistic cousins in the east.  Exactly like the Indo-Aryans in the Aryan invasion hypothesis, they are predominantly Indian people speaking a foreign-originated language: �Whereas the now Dravidian-speaking tribals of Central and South India can be considered to be descendents of the original inhabitants of India, who gave up their original languages in favour of Dravidian, Tibeto-Chinese speaking tribals (Northeast India) and Austro-Asiatic speaking ones (East India) immigrated into India since ancient historical times.  Most likely they came in several waves from Southern China (Tibeto-Chinese speakers) and from Southeast Asia (Austro-Asiatic speakers) respectively.  Without doubt these immigrating groups met with ancient Indian populations, which were living already on their migration routes, and thus one cannot exclude some cultural and also genetic contacts between immigrants and original inhabitants of India, at least at some places.�

In the case of Indo-Aryan, by contrast, its speakers have obviously also mixed with other communities, but its foreign origin has not been firmly established.

4.9.9. Conclusion

We may conclude with a recent status quaestionis by archaeologist Jonathan Mark Kenoyer of Wisconsin University at Madison: �Although the overall socioeconomic organization changed, continuities in technology, subsistence practices, settlement organization, and some regional symbols show that the indigenous population was not displaced by invading hordes of Indo-Aryan speaking people.  For many years, the �invasions� or �migrations� of these Indo-Aryan-speaking Vedic/Aryan tribes explained the decline of the Indus civilization and the sudden rise of urbanization in the Ganga-Yamuna valley.  This was based on simplistic models of culture change and an uncritical reading of Vedic texts.  Current evidence does not support a pre- or proto-historic Indo-Aryan invasion of southern Asia.  Instead, there was an overlap between Late Harappan and post-Harappan communities, with no biological evidence for major new populations.�  

We repeat that physical anthropology is going through rapid developments due to the availability of new techniques, and we don't want to jump to conclusions in this moving field. But we notice that whatever new technique is applied and from whichever new angle the question is approached, it has so far consistently failed to yield evidence of the fabled Aryan Invasion..

Footnotes: 
 

Dr. Ambedkar: Writings and Speeches, vol.7, p.301.

Dr.  Ambedkar: Writings and Speeches, vol.7, p.301, with reference to G.S. Ghurye: Caste and Race on India, London 1932.

N.V. Subramaniam: �The way we are.  An ASI project shatters some entrenched myths�, Sunday, 10-4-1994.

K.C. Malhotra: �Biological Dimensions to Ethnicity and caste in India�, in K.S. Singh: Ethnicity, Caste and People, Manohar, Delhi 1992, p.65. Reference is to H.K. Rakshit: �An Anthropometric Study of the Brahmins of India�, in Man in India #46; and P.P. Majumdar & K.C. Malhotra: OAB Dynamics in India: A Statistical Study, Calcutta 1974.

Pallava Bagla: �Study shows caste system has changed genetic makeup of Hindus.  Studying 200 men in AP, Indo-US team finds that lower castes have over the years become �genetically different� from upper castes�, Indian Express, 18-10-1998.  See also the subsequent critical editorial: �Questionable enterprises.  DNA and caste can make a deadly combination�, Indian Express, 22-10-1998, which points out that the study merely confirm what observers of caste relations had known all along.

Thus, Kancha Ilaiah (Why I Am Not a Hindu, Samya/Bhatkal & Sen, Calcutta 1996) offers a description of the differences in life style between upper castes and Shudras, with the declared intention of getting the reader indignated at the injustice and absurdity of the typically Hindu castle system.  Yet, his testimony unwittingly shows just how similar Hindu caste inequality is to the social inequality in other societies, e.g. Ilaiah�s repeated observation that women are more controlled in upper castes and more assertive and free in lower castes is or was just as true for Confucian China or the feudal and bourgeois societies of Europe.

Luigi Luca Cavalli-Sforza: �Genes, Peoples and Languages�, Scientific American, November 1991.

Quoted in Simon Rozendaal: �Ras - wat is dat eigenlijk?�, Elsevier, 14-10-1995.

V. Bhalla: �Aspects of Gene Geography and Ethnic Diversity of the People of India�, in K.S. Singh: Ethnicity, Caste and People, P.51-60; specifically p.58.

B. Narasimhaiah: Neolithic and Megalithic Cultures in Tamil Nadu, Sundeep Prakashan, Delhi 1980, p.195:

Kailash C. Malhotra: �Biological Dimensions to Ethnicity and Caste in India�, in K.S. Singh: Ethnicity, Caste and People, p.63.

Kailash C. Malhotra: �Biological Dimensions to Ethnicity and Caste in India�, in K.S. Singh: Ethnicity, Caste and People, p.63.

S.K. Chatterjee: Indianism and Indian Synthesis, Calcutta 1962, p.125.

P.A. Augustine: The Bhils of Rajasthan, Indian Social Institute. Delhi 1986, p. 2-3.

Hilde Van den Eynde: �Genetische kaart van Europa tekent oorlogen en volksverhuizingen�, De Standaard (Brussels), 20-7-1993.

Hilde Van den Eynde: �Biologen en archaeologen moeten Amerikaanse taalknoop doorhakken�, De Standard (Brussels), 3-8-1990; see also Joseph H. Greenberg & Merritt Ruhlen: �Linguistic Origins of Native Americans�. Scientific American, November 1992.

Interview in Le Nouvel Observateur, 23-1-1992.

Interview in Le Nouvel Observateur, 23-1-1992.

Interview in Le Nouvel Observateur, 23-1-1992, emphasis added. 

T.V. Gamkrelidze and V.V. Ivanov, in Journal of Indo-European Studies, 1985/1-2, p. 182.

See e.g. the fall/winter 1995 issue of Journal of Indo-European Studies, almost entirely devoted to the Xinjiang mummies.

I.M. Diakonov: �On the Original Home of ther Speakers of Indo-European�, Journal of Indo-Europen Studies, 1-2/1985, p.92-174, specifically p. 152-153.

I.M. Diakonov: �On the Original Home of the Speakers of Indo-European�, Journal of Indo-Europen Studies, 1-2/1985, p. 153-154.

Related with details and undisguised favour by Alian de Benoist: Les Indo-Européens (Nouvelle Ecole no. 49, Paris 1997), p.47.

I.M. Diakonov: �On the Original Home of the Speakers of Indo-European�, Journal of Indo-Europen Studies, 1-2/1985, p.153-154.

Atharva-Veda 6:137.2-3 is a charm, for making �strong black hairlocks� grow, apparently on the heads of bald or albino or greyed people. Paramesh Choudhury (The Aryan Hoax, p. 13) also mentions Baudhayana�s Dharma-Sutra 1:2, �Let him kindle the sacrificial fire while his hair is still black�, also cited in Shabara�s Bhasya on Jaimini 1:33, as instances where Brahmins� hair is off-hand assumed to be black.

Patanjali: Mahabhashya (comment on Panini) 2:2:6.

Quoted from his A Practical Sanskrit Dictionary by Paramesh Choudhury: The Aryan Hoax, p. 13.

E.g. Mahabharata: Adiparva 223, describes a Rakshasa as red-haired, as pointed out by Paramesh Choudhury: The Aryan Hoax, p. 13. He also mentions that Ravana�s sister Surpanakha is described by Valmiki as having pingala eyes, but remember that Ravana�s family is described as a Brahmin family immigrated in Lanka from northern India.

Hemphill & Christensen: �The Oxus Civilization as a Link between East and West: A Non-Metric Analysis of Bronze Age Bactrain Biological Affinities�, paper read at the South Asia Conference, 3-5 November 1994, Madison, Wisconsin; p. 13.

K.A.R. Kennedy: �Have Aryans been identified in the prehistoric skeletal record from South Asia?�, in George Erdosy, ed.: The Indo-Aryans of Ancient South Asia, p.49. On p.42, Kennedy quotes the suggestion that �not only the end of the [Harappan] cities but even their initial impetus may have been due to Indo-European speaking peoples�, by B. and F.R. Allchin: The Birth of Indian Civilization, Penguin 1968, p. 144.

Note that many scholars assume an (albeit somewhat irregular) etymological kinship between GandhAra and the Greek word Kentauros, meaning a horse-man.  The rough terrain of Afghanistan was unfit for chariot-riding and required horseback-riding.  To people from countries unfamiliar with horses (as India must have been in some pre-Vedic