GENETICS ON MIGRATIONS IN HISTORY

George Santayana said: "History is always written wrong, it needs always to be rewritten." When it comes to Indian history, we need to identify the people who inhabit the country. Recent findings in population genetics can shed light on the problem.

N.S. Rajaram

Background: Mismatches

When we examine the accounts of ancient India as given in history books still in use against the background of empirical data and the primary (literary) sources, we find fundamental mismatches between data and historical theories. These mismatches are both qualitative and quantitative in nature. Qualitatively speaking, we may note the following:

1.  There is no archaeological record of any invasion and/or massive migration from Eurasia in the Vedic period. If anything we find traces of movement in the opposite direction- to West Asia and even Europe.

2.   The geography described in the Rigveda corresponds to North India in the

      fourth millennium BC and earlier and not Europe or Eurasia.

            3.   The flora and fauna described in the Vedic literature, especially those found in

                  the sacred symbols, are tropical and subtropical varieties and not from the

                  temperate climate or the steppes.

            4.   The climate corresponds to that found in North India.

In quantitative terms, there is huge time gap-exceeding a thousand years-between the dates assigned to significant features and what we actually find. These include:

1. Indian writing is supposed to be based on borrowings from the Phoenicians, but 

    the Indus (Harappan) writing is more than a thousand years older than the

    oldest Phoenician examples known.

2. Naturalistic art with realistic depictions is supposed to have been brought to

    India by the Greeks, but we find superb realistic depictions in Harappan

    remains. To summarize John Marshall: "The Indus artist anticipated the Greek   

    artist by more than 2000 years."

3. Indian astronomy is claimed to be a borrowing from the Greeks, but the

    Vedanga Jyotisha cannot be dated later than the 14^th century BC. The name  

    Vedanga indicates it is later than the Vedas, so the astronomical references in

    the Vedas must be older.

4. Migrations: The major migration or invasion is supposed to have taken place

    after 2000 BC, but the genetic evidence shows that the people of India have

    lived where they are for tens of thousands of years.

It is clear that we need a serious re-examination of history- both of the chronology and the descriptive accounts. Two fundamental tasks suggest themselves: (1) establishing independent chronological markers that connect literary accounts and datable physical features; (2) determining the identity of the people of India on scientific grounds, independent of historical and/or linguistic theories.

The Indian people

Indian literary records are the most extensive in the ancient world. Indian archaeologists have also found many important settlements going back to prehistoric times. They include not only the well-known Harappan and pre-Harappan sites in India and Pakistan but also Neolithic and Paleolithic sites as well as cave dwellings like those at Bhimbetka. These are now being supplemented by underwater discoveries like the ones at Poompahar in Tamil Nadu and off the coast of Gujarat. In spite of this profusion of records, a basic question remains: who are the people that created all this? This remains a closed book.

It is not as if Indian records are silent about them. We have references to various peoples like the Purus, the Turvashas, the Druhyus, the Yadus and others who remain unidentified in archaeological terms. In even broader terms, although there are many theories about their languages and culture, the Aryans and the inhabitants of the south have not been archaeologically delineated. Existing historical theories make arbitrary assignments and then use these to define important kingdoms, events, and even chronology. This is a highly unsatisfactory situation. What are needed are objective methods that lead us to chronological and historical identification of the different people of India and their contribution- in archaeology and in literature.

This identification has to address two basic issues- chronological and descriptive. We need to find objective ways of relating the archaeological dates with the dates assignable to literary passages. For example, if the Harappan sites in a particular region may be assigned to a particular period-say 3000 to 2000 BC-we need to find references in the literature that refer to this period. We may note that this corresponds to what is known as the Krittika period (equinox in Pleiades in Taurus). So references to this in the literature, found for example in the Yajurveda and the Shatapatha Brahmana, may be taken as the first step in establishing a chronological link between archaeology and literature.

When we come to the identity of the inhabitants, the problem becomes qualitatively different and at once more challenging. First, we need to ask a basic question: are the inhabitants of India primarily of indigenous origin or are they the result of successive migrations? The historical theories that have dominated textbooks for over a century lean towards the latter view, especially as regards the creators of the Vedic civilization. Of late, there is a tendency to attribute even the Harappan civilization to foreign inspiration, notably from West Asia and Elam. The languages of India, the two main classifications-Aryan and Dravidian-are said to be imports. The first from some "Aryan or Indo European homeland' (or Urheimat) and the latter from Scythia.

As a result, we may say that the first task in establishing the identity of the Indian people is to get an idea of the evolutionary history of its inhabitants going back some tens of thousands of years. This takes us to the last Ice Age. In pursuing such an inquiry, we may take a clue from paleontology where data are scarce, but intelligent use of molecular biology, especially genetics, has led to important breakthroughs in our understanding of human origins in different parts of the world. As Luigi Luca Cavalli-Sforza, one of the world's foremost human geneticists put it:

"Historians relying on written records can tell us nothing about 99.9 per cent of human evolution which preceded the invention of writing. It is the study of genetic variation, backed by language and archaeology which provides concrete evidence about the spread of cultural innovations, about the movements of peoples across the globe... and the sheer scientific absurdity of racism."

He further notes speaking of his most recent and most ambitious work: ¹ "It is a history of the last hundred thousand years, relying on archaeology, genetics, and linguistics. Happily, these three disciplines are now generating many new data and insights. All of them can be expected to converge to a common story; and behind them must lie a single history."

Needless to say such an approach, while promising, is fraught with pitfalls some of which the eminent author is not entirely successful in avoiding. Upon critical examination, one finds the work in the field to be marred by oversimplifications, uncritical acceptance of discredited historical theories, and unsupported flights of fancy all of which serve to highlight the promise as well as the pitfalls that challenge researchers in the field. This perhaps is to be expected in any work that seeks to break new ground. A critical examination therefore is the need of the hour. For this reason, various claims and conclusions need to be subjected to careful scrutiny. This is what is done here, beginning with the notion of race- a colonial legacy that has so dominated historical discourse in India.

Science on race

            Racism has no scientific basis; the genetic variation within any geographical or any other group is always greater than the variation between different groups.² In quantitative terms, the human species on the planet now has 99.9 per cent of its genetic material in common. As a result, samples taken from any two human groups will be overwhelmingly similar. Further, we share 98.4 per cent of our genes with the chimpanzees. According to the evolutionary tree used by some paleontologists, humans are a subspecies of chimpanzees. So, anyone using genetic similarity to argue that a particular people (like the Indo Europeans) invaded a particular geographical region had better be aware that an almost equally strong case can be made for an invasion of chimpanzees!

            (An even stronger case can be made for an invasion theory of Parsees, who happen to speak Gujarati. It may be argued that the invading Parsees drove away the Dravidians from western India and imposed the Gujarati language and culture. This can draw further support from the fact that numerous Kannada inscriptions have been found in the region- a legacy of the Chalukya and Rashtrakuta dynasties that ruled the region in early medieval times.)

            To return to race, many biologists once believed that different races carried dramatically different sets of genes. But studies in human genetics have shown that they were completely wrong. Geneticist Mark Feldman puts it this way: "The biological notion that we used to have of races is not compatible with the reality of the genetics we are finding today."

            So, when one finds claims that try to revive old race based theories and their conclusions in the name of genetics research, it is necessary to look for reasons that lie beyond the frontiers of science. They will usually be found to lurk behind social, political and religious beliefs. As a rule, one cannot apply biological or other scientific rules and causes to man-made classifications like race, caste or creed. Yet, theories that purport to do precisely this persist in the literature, especially in history and anthropology. Nature does not respect man-made laws. We need to understand nature on its own terms. This is what is examined next, in the context of human populations.

Genotypes and phenotypes: no "blending" of traits

            Many historical and anthropological theories, especially those created during the European colonial period, start with the assumption that civilizations in different parts of the world began with a massive migration from a central homeland. The most famous-or infamous, depending on one's viewpoint-is the Aryan invasion theory (AIT) of India, now being repackaged as AMT or the Aryan migration theory. It is too well known need elaboration. Though discredited, it persists in various forms because of entrenched academic and political interests.

This is not to suggest that migrations have not made an impact on history and civilization. They certainly have. But a migration thousands of years ago, is not easy to establish based on the genetic profile of a population today. To take an example: Parsees now living in Western India migrated from Persia more than a thousand years ago. We know this only because of historical records. Genetics alone may or may not be able to establish the fact, unless there is a specific genetic polymorphism like a particular blood group, which the immigrants carried- one that is found in Persia but rare in India. Physical appearance is the most misleading of indicators; what we see is only the phenotype or the reaction of the individual to the environment, and not any blending of imported traits (like skin color) into an existing population.

            The idea that phenotypes (observable traits) somehow ‘blend' from parents to offspring is probably the greatest fallacy that people generally have. It was this fallacy that gave rise to colonial anthropologists like Herbert Risley coming up with classifications like Nordic Indo-Aryans and others that are supposed to be the result of the blending of the invading Nordic races with the natives, thousands of years ago. Gregor Mendel demolished the whole notion in 1865 by showing that heritable traits are transmitted from generation to generation by discrete units that we now call genes. So, if we see a gradation in color in India as we move north, it is for the same reason that we see a similar gradient in Europe- adaptation to the environment through natural selection. ³

            How about race? Here genetics makes it clear that the determining factor in physical similarity is geographical proximity. In Cavalli-Sforza's words, "...the genetic distance between two populations generally increases with geographic distance separating them." This is still not the whole story, for the differences between individuals within a group are always greater than differences between different groups. That is to say, human beings now inhabiting the world are extraordinarily close, genetically speaking, though they exhibit great variability in observable traits (phenotypes) like physical appearance. This is a very complex issue: the phenotype is the result of interaction between the genotype (inherited factors) and the environment. And the same genotype can produce different phenotypes in different environments. The simplest example is language. Though all of us inherit the capacity to learn language, the language we actually use is strictly the product of the environment. The same applies to height, weight and other features that strike the eye. They are a complex mix of inheritance and environment.

            Genes determine the pattern of variation in response to variation in the environment.  This makes it virtually impossible to recover the past history of a people based on genetic analysis, because, the environmental conditions in which the population underwent the changes leading to its present genetic state are lost forever. A possible exception is a clearly defined polymorphism (different form) like blood group that is constant across environments. Even this can be lost due to natural selection, which may favor some groups but not others. We see this phenomenon both in the Basque region of Spain and among South American Indians. This means that the best we can do is speculate and offer plausibility arguments. This is the reason that human geneticists like Cavalli-Sforza (and his lesser imitators like Spencer Wells) frequently fall back on the phrase "it is likely" in presenting a particular scenario.

Norm of reaction

            The great difficulty in using genetics as a tool in history arises from the fact that we are all incredibly similar. Geneticists use a measure known as the genetic distance to measure differences between population groups. This is based on the frequency of some observable genetic trait like the blood group. As Cavalli-Sforza observes: "For most genes, the frequency differences between populations are nil to very slight and their contribution to the global genetic distance between populations is close to zero." In the face of this, it would take a very brave man indeed to try to reconstruct the history of some current practices (like caste) based on genetic data.

            This is further complicated by the fact that the same genotype can give rise to different phenotypes, and also, different genotypes can produce the same or similar phenotypes. In general, the relationship between the genotype and phenotype varies from one species to another, or even between phenotypic traits that are superficially similar within a species (like skin color). There is no simple way of characterizing the relationship between the genotype (even if it is known) and phenotype. For this reason, geneticists use a technical term they call the ‘norm of reaction.' It describes a range of responses of a genotype to different environments. But our norm of reaction, the manner in which human genes (or genotypes) react to different environments, is unknown.

As Harvard Geneticist Richard Lewontin observes: ⁴ "Except for such traits as the presence or absence of blood-group antigens, which are absolutely constant across environments, we do not have the norm of reaction for any human trait."

            This fact, there is no known norm of reaction for any human trait, should be firmly kept in mind in evaluating any historical claims made supposedly on the basis of genetics. Again as Lewontin observes (Op. cit. p 163): "Reconstructing the evolutionary past of the human species is almost as difficult as predicting the future, although both are common exercises that biologists engage in, especially when they address a nonscientific public." This is doubly so for history where the time scale is shorter than in evolution and the influence of culture is incomparably greater. Regrettably, even knowledgeable geneticists (including Cavalli-Sforza) tend to indulge in speculative reconstructing of human history, especially of migrations, based on genetic data. There is of course no shortage of such theories created by historians and anthropologists in the past. This leads to an illusion of scientific plausibility to theories that have already been discredited, giving them a lease of life which they don't deserve.

Most variation is within groups

            What then is the story of the spread of human populations that our genes do have to tell? This may be stated very simply: we humans are incredibly alike. Something like 85 percent of the genetic variation is within any group or tribe. And it does not matter what race you assign them to- Caucasian, Mongoloid, Negro or anything else. Africa is the most varied continent, but Yorkshire, for example, has 85 percent as much variation as all of Africa. (This means, among other things, that the genes responsible for the white skin of Europeans were already present in Africans before they made their way to Europe! The same holds for Aryan genes, Dravidian genes, Dalit genes and so forth even if they exist.) At the same time, at the individual level, there is enormous variability, which is what makes possible genetic matching (as in parentage) and identifying inherited disorders. By this combination of extreme similarity on a global scale and great variability among individuals, nature has conspired to keep our past locked from scrutiny- or so it seems.

            This means that any two population groups can be shown to be similar by looking at the genetic inheritance. The difficulty lies in finding the differences- contained in 0.1 percent of our genes spread over a population of over 6 billion. So, a classification based on language families (which is what Indo European studies claims to do) can be shown to have a genetic basis by comparing Indian and European populations. (More rigorous studies based on blood group frequencies have shown this to be unfounded.) Some genetic marker or other can always be found that shows up a statistical quirk to which historical significance can always be attached. This doesn't make it historical fact.

            The statistical method that many population geneticists use is the principal component, which realigns multi-dimensional data along directions that represent the smallest set of coordinates. (This is done with the help of eigenvalues and eigenvectors for those who might be mathematically minded.) This is based on the pioneering work of the late Sir Ronald Fisher, who carried on the work of Mendel and developed many statistical techniques that are in use today. ⁵

Garbage-in, garbage-out: language theories

It is important to note, however, that any mathematical entity (like a principal component) is meaningful only in a physical context. For example, Einstein's famous equation E = mc² is important only because of its physical content: it establishes a fundamental relationship between mass and energy. Without the physical meaning, it is a commonplace equation that describes a parabola. Assigning arbitrary meanings to some computable mathematical quantity does not make a theory valid. A dubious theory cannot be made rigorous by appealing to mathematics; it may only obfuscate the truth. But this is what some scholars have done in attempts to correlate statistical results with theories in linguistics and anthropology. Even competent scientists like Cavalli-Sforza can fall into this trap when they allow their enthusiasm to get the better of judgment. The following examples illustrate the dangers.

The first example relates to the work of Kruksal, Dyen and Black who applied statistical tests to the languages that make up the Indo European family. ⁶ The most important member is of course Sanskrit, but their analysis threw up a major contradiction: Indian and Iranian languages failed the grouping test! This is a bombshell, for according to Indo European linguistics, Indo-Iranian is the lynchpin of the whole discipline, but the one quantitative test that was applied to the hypothesis discredited it.

Cavalli-Sforza is too good a scientist to ignore the problem. So he contents himself by observing that his (and his group's) analysis gives a similar linguistic grouping, "but slightly contradicting the results of Kruksal and his colleagues." It is more than a slight contradiction. In fact, the author himself had noted in his earlier work that the Kruksal, Dyen and Black study "...found no similarity at all between Italic and Celtic languages, nor between Indian and Iranian ones... The non-identification of an Indo-Iranian group by Dyen, et al. is the major departure from the conclusions accepted by the majority of traditional linguists." (Great Human Diasporas, Addison-Wesley, 1995: page 190)

This is quite unequivocal, and not slightly contradicting, as he suggests in his more recent Genes, Peoples and Languages.  To his credit, in a more recent article, Cavalli-Sforza appears to have gone back to his original position- that there is a problem with the conclusion of linguists. He and his coauthors note that genetic data don't support the linguistic, or caste and/or race based division of the Indian population.⁷ The real meaning of all this: if the empirical base of a subject is weak, genetics and mathematics may be used to give an appearance of scientific soundness, but it is a bubble that waits to be punctured by critical scrutiny. "Garbage-in, garbage-out" works in genetics no less than it does in computing and statistics.

It is the same story with the Kurgan homeland theory of Indo European speakers. (This refers to prehistoric settlements in the region north of the Black Sea in Ukraine.) Since there are no genetic methods for detecting what language a people spoke thousands of years ago-or speak today-its proponents have to exercise considerable ingenuity in support of their theory. This they have done. The Kurgan people have left no written records of any kind. Noting this Cavalli-Sforza observes: "Without written documents, it is very difficult for archaeologists to say what language was spoken in this region at the time." Difficult? Impossible seems a better description. Apparently, the argument for the Kurgans as the original Indo European speakers is based on the fact that some horses have been found buried in the region. The logic of this lies beyond the grasp of this writer who happens to be a mathematical scientist.

Methodological confusion

Several observers have noted that the tree diagrams used to classify language families bear a striking similarity to tree diagrams used in biology, and later, in evolutionary genetics. From this some have concluded that there must be some mysterious connection. This is no accident. It is simply an artifact of analysis, which it would be naïve to see as the manifestation of a natural law. Here is the story in brief. August Schleicher (1821 - 1868), an amateur biologist, borrowed the idea of tree diagrams for classifying species from biology and applied it to languages. Later, scientists applied the same tool in evolutionary biology and genetics. So, the fact that language tree diagrams bear a resemblance to tree diagrams used by some geneticists is simply the result of this historical accident.

Indo European scholars are now hailing the resemblance between the tree diagrams used in genetics and linguistics as scientific validation of their conclusions. It is nothing of the sort: it is simply a restatement of their method and the fact that both- population genetics and linguistics have borrowed the same tool from the same source. This was noted as far back as 1935 by Leonard Bloomfield: ⁸ "The earlier studies of Indo-European did not realize that the family tree diagram was merely a statement of their method: they accepted the uniform parent languages and their sudden clear-cut splitting, as historical reality." But some linguists and anthropologists are now going further, attempting to revive discredited race theories in the name of genetics research. This is what we may examine next, looking especially at migrations and the important but neglected concept of genetic drift.

Evolutionary forces: Founders' Effect

The human species is young, no more than 10,000 generations or 200,000 years old. Modern humans evolved in Africa between 100,000 and 200,000 years ago. As things currently stand, the major geographical races-meaning the people that inhabit different parts of the world-are believed to have diverged from their common African ancestors between 40,000 and 50,000 years ago. This corresponds to some 1500 generations. This means that all the physical differences we see in non-African humans today are the result of changes in these 40 to 50 thousand years that took place after our ancestors left Africa. But the seeds of these or the potential for these changes - like the capacity to learn language - must already have been there in our African ancestors. Evolutionary forces acting in conjunction with the environment have brought about these changes. In the process they have produced the rich diversity of peoples and cultures that we see in the world today.

Four evolutionary forces have contributed to the evolution of modern humans: natural selection, mutation, genetic drift and migration (or gene exchange). Natural selection selects those members of the population that have the best chance of survival in a particular environment. It is in reality an elimination process, with those members with low probability of survival and reproduction being allowed to become extinct. Mutation is random change in the genetic structure that is passed on to successive generations. The great diversity of organisms and species that we see on the planet is the result of mutation and natural selection working together. It is the main creative force in evolution; evolution is the great unifying theme in biology.

Observable features in different geographical regions, like the high concentration of brown skin pigmentation in parts of India, Southeast Asia and Africa are adaptive characteristics. The same is true of the high frequency of hemoglobin S in parts of Africa, Yemen and South India. This confers an advantage to its possessors in the form of resistance to malaria caused by Plasmodium falciparum. This advantage was lost when some of these Africans ended up on the American continent as slaves. The same hemoglobin made them susceptible to the disease sickle cell anemia. This shows the importance of the environment in natural selection.

Drift (or genetic drift) works differently and accounts for fluctuations in gene frequencies in small populations. "The smaller the population, the greater the fluctuation in the gene frequency over generations," is how Cavalli-Sforza puts it. This allows small population groups to survive, provided the environment is benign. A good example is furnished by the Parsees of Western India. They began as a small group of immigrants arriving in India more than a thousand years ago. When successful, as the Parsees were, the members of such a group become ‘founders'. For this reason, the phenomenon is also called the Founders' Effect.

The key though is survival by adaptation in the new environment. Founders' Effect is important because, if the founders succeed, the population that it gives rise to will also succeed rapidly. This was the case with the Parsees in India a thousand years ago and may also be the case today with the Indian immigrants in the United States. On the other hand, the Vikings who landed in North America several centuries before Columbus's voyage could not survive in the hostile environment. They became extinct within a few generations. Unlike the Parsees, they did not become founders. The environment was against them.

Genetic drift is important in history and can work in different ways. But a genetic analysis cannot reveal if a particular feature (phenotype) in a population is the result of drift or natural selection. For example, if there is some observable feature in Parsees that distinguishes them from their surrounding populations (this is not known), we can only attribute it to their being of Persian origin; we cannot prove it by genetic analysis. We may assume that this is due to genetic drift, only because of our knowledge of history. At the same time, a proper understanding of drift can be a useful tool in reconstructing history, for the Founders' Effect may well prove to be a major historical force, as it has been the case in the United States and in Australia. This may also have been the case 50,000 years ago, when our modern ancestors moved out of Africa and settled in different parts of the world, possibly mixing with or even replacing earlier inhabitants. This is what seems to have happened when modern humans encountered the Neanderthals of Europe. (See Figure 1)

Drift and migration in India

            The Founders' Effect is usually the outcome of migration of a small population followed by drift. It need not be permanent. To begin with, it needs a benign environment in which the population can thrive in isolation. But drift can accentuate negative features that work against long term survival due to excessive in breeding in the small population. In addition, the genetic identity of the small population may be lost due to changed historical and social conditions, when the founders and their descendants mix freely with the local populations. This is what seems to have happened following the Mongol invasions in Russia and West Asia. This is true also of the Delhi Sultanate established by Central Asian Turks, and later, the Mogul Empire. There was no Founders' Effect. (We should not confuse cultural traits like language, art and dress with the Founders Effect, which is strictly a biological and demographic phenomenon.)

            Until recently, and even now to some extent, histories of India have emphasized the impact of migrations (and invasions) going back to ancient times. Karl Marx, who incidentally knew next to nothing about India, asserted that the history of India is nothing but the "record of successive intruders." We may now have the scientific tools to test these theories- like the Aryan invasion. All attempts so far to establish the presence of European (or Eurasian) genetic traits in the Indian population, especially the upper castes, have failed. Brahmins for example, share the physical traits of the rest of the people in the region where they come from. Kerala Brahmins for example, look like other Keralites and not like Kashmiri Pandits, much less European invaders who were supposed to have been their ancestors thousands of years ago. The same is true of their genetic traits to the extent they are known.

            A comparison with the Parsees is illuminating. The Parsees are a Founder Group. According to the various Aryan invasion-migration theories, the upper castes, including the Brahmins, should also be a Founder Group. But unlike the Parsees, who until recently were concentrated in small isolated communities in Western India, Brahmins are found all over India. They show no indication of any peculiar genetic traits that may be attributed to the Founders' Effect or even genetic drift. (Cultural influences like the study of Sanskrit and rituals are not genetic traits.) Also, literary and archaeological records show that India has been heavily populated for at least 5000 years. So, it would have been impossible for Founder Groups of Brahmins to be established in all parts of India through diffusion, while simultaneously genetic drift would have preserved their genetic traits. This is a clear contradiction. It is a serious indictment of scholarship that this simple analysis has not been applied to these theories.

Genetic studies of European populations based on the frequency of RH gene (blood group), suggests a gradual movement of humans from south and east. This is the reverse of the gradient that would have appeared if the migration were from Europe. (See Figure 2) All the different forms (known as alleles) of the RH gene are present in Africa and Asia also, though the frequency is less than in Europe. Combined with archaeology, some scholars attribute the increasing RH gene frequency in Europe to the diffusion of agriculture from Asia into Europe. But we do not know when the spread took place, if at all it did, so the picture is hopelessly confused.

A common error is to confuse linguistic and other cultural diffusion with genetic drift caused by migration. Migration doesn't always result in the Founders' Effect and drift, except under highly favorable conditions. But the confusion persists with its authors generally unaware of the phenomenon involved. In fact, it can be said that the whole discipline of Indo European studies rests on this confusion. This is the reason that biological (race) and cultural (language) features are readily interchangeable, and often lead to the same conclusions.

Genetics of caste and race ⁹

            In the past few years, there have been several articles purporting to show that the Indian caste system was imposed by invading (or migrating) Indo-Europeans-or ‘Aryans'-though the term Aryan has lost some of its appeal following Hitler. This is claimed to have scientific support in the form of genetic evidence of the present day Indian population. In an article titled "Genetic Evidence on the Origins of Indian Caste Populations," eighteen authors, mainly from Utah in the U.S. and Vishakapatnam in India, led by Michael Bamshad of the Department of Pediatrics from the University of Utah make the claim that there were several waves of immigration into India, the last of which (from Europe) was responsible for the caste system. In their words: ¹⁰

            "In the most recent of these waves [of immigration], Indo-European speaking people from West Eurasia entered India from the Northwest and diffused throughout the subcontinent. They purportedly admixed with or displaced Dravidic-speaking populations. Subsequently, they may have established the Hindu caste system and placed themselves in castes of higher rank."

            To begin with it is palpably absurd to mix genetics and language: how does one distinguish the genes of "Dravidic-speaking populations"? In his press statements, Bamshad has gone a good deal further, claiming, "we are able to demonstrate unequivocally that the upper castes are more similar to Europeans than lower castes..."

            The first point to note is that it is impossible to trace the origin of a small segment of a population like upper caste Indians through genetic analysis. We have already seen that the genetic variation within any group-like the upper castes-is always greater than the variation between two groups like the upper caste Indians and Europeans. Scientifically speaking it requires no great effort to demolish the claim. This can be further illustrated with the example of the Zoroastrian Parsees who are fairly recent immigrants (by Indian standards of time), having emigrated from Persia only some 1200 years ago.

            This of course, we know from historical records. But Bamshad and his colleagues would have us believe that they can detect this from genetic analysis, and even "unequivocally demonstrate" it. This means their genetic analysis should be able to tell us how the Parsees living in Western India happen to be speaking Gujarati. To go by the Aryan invasion analogy, the migrating (or invading) Parsees brought the ancestor of the Gujarati language and imposed it on the native population, which the Bamshad method can prove with the help of genetic analysis!

            A point to note is that despite its appeal to science, it is simply a reincarnation of nineteenth century colonial race prejudice. No less a ‘scholar' than Max Müller wrote: " ...how the [English] descendants of the same [Aryan] race, to which the first conquerors and masters of India belonged, return ...to accomplish the glorious work of civilization, which had been left unfinished by their Aryan brethren."

            There was more than scholarship involved in this, for Max Müller was as much a colonial-missionary politician as he was a scholar, for which he was richly rewarded. As he proudly declared in his Autobiography: "Lord Derby, then Secretary of State for India, declared that the scholars who had discovered and proved the close relationship between Sanskrit and English, had rendered more valuable service to the [British] Government of India than many a regiment." The most eloquent statement relating this ‘history' to the British rule came from Stanley Baldwin, in a speech he made in the House of Commons in 1929:

            "Ages and ages ago, there sat, side by side, the ancestors of the English, Rajputs and Brahmins [upper caste Indians]. Now, after ages, the two branches of the great Aryan ancestry have again been brought together by Providence.... By establishing British rule in India, God said to the British, "I have brought you and the Indians together after a long separation, not in order that you should lord over them, or that you should exploit them, but in order that should recognize your kinship with them.... It is your duty to raise them to their own level as quickly as possible, and work together; brothers as you are, for the evolution of humanity...." "

            From brotherhood, it took only a step to turn them into oppressors. This was done with a sleight of hand by missionary ‘scholars' and colonial officials. Anthropologists like Herbert Risley, a colonial official, equated caste with race with the help of what he and his colleagues claimed were measurements of facial features, of the nose in particular. Risley wrote in 1891: "The social position of a caste varies inversely as its nasal index.... Community of race... is the real meaning of the caste system."

            All this ‘science' was part of the agenda to divide the people of India into antagonistic groups under labels like Aryan and Dravidian to facilitate colonial rule and conversion to Christianity. The most influential figure in this was Robert Caldwell, Bishop of Tirunelveli, who wrote the highly influential Comparative Grammar of the Dravidian or South Indian Family of Languages. First published in 1856, it continues its influence today, especially in Tamil Nadu. This is due less to its scholarly content than the fact that it became the political manifesto of Dravidian politicians who not infrequently engaged in obscene conduct towards innocent people that were labeled as descendants of Aryan oppressors.

            This digression is necessary to understand the sources and motivations of the research of men like Michael Bamshad who would seek to prove all this with the help of genetics. This is worth a brief look.

Semantic confusion

            A study that appeared in February 2004 takes this exercise further.¹¹ Specifically it claims that tribal females and caste females are of indigenous origin while caste males are descended from Eurasian ancestors. Further, they claim that this is supported by linguistic and archaeological evidence. Upon closer examination, one finds that their ‘linguistic evidence' is nothing more than the conclusion of linguists derived under the assumption of the Aryan invasion, which, as previously noted, cannot withstand scrutiny. The archaeological evidence is only an interpretation (by Walter Fairservis) of archaeological remains, again assuming the Aryan invasion theory to be valid. So it is no evidence at all but a circular argument that proves itself. It is nothing but a combination of ignorance and shoddy science.

            It is worth noting that the authors of the study depend on the work of Michael Bamshad and associates that has been refuted, with one of his co-authors repudiating his claims. ¹⁰ For example, Bamshad's label of Europe was applied to "anything West of the Indus," and his sample was from the East Coast of India. Further, as previously noted, claims were made about Indo European language speakers, who, needless to say are not distinguished by genetic features. Further, there is the claim that these Eurasian (Aryans)-males only-arrived in India 3500 years ago (1500 BC). There is no genetic method of determining this date; Max Müller derived the 1500 BC date for the Aryan invasion based on the Biblical belief that the world was created was on 23 October 4004 BC!

            In evaluating all this, it helps to recall Risley's claim, "Community of race... is the real meaning of the caste system." We are asked to believe that theories about racial invasions and migrations, concocted by nineteenth century colonial officials like Risley and missionary propagandists like Bishop Caldwell-and modern Dravidian politicians-has all been proved by the magic of genetic research!

            A close look at these studies show them to be full of technical sounding terminology in an attempt to convince the uninformed reader that high science is being used to derive profound results of great importance. But an informed reader, familiar with quantitative methods, will have little difficulty in recognizing this as scientific name dropping- an example of what physicist Robert Park calls "Voodoo Science." ¹²

Recent revisions

            Happily, some recent studies in which Cavalli-Sforza has been a leader, have taken steps to correct this error. ⁷ As previously noted, an important article based on a study of Indian tribal and caste populations by Cavalli-Sforza and his colleagues observes: "Taken together, these results show that Indian tribal and caste populations derive largely from the same genetic heritage of Pleistocene southern and western Asians and have received limited gene flow from external regions since the Holocene. The phylogeography [neighboring branches] of the primal mtDNA and Y-chromosome founders suggests that these southern Asian Pleistocene coastal settlers from Africa would have provided the inocula for the subsequent differentiation of the distinctive eastern and western Eurasian gene pools." (Italics added)

            Put in non-technical language, it means that the Indian population-upper castes, tribals (or indigenous peoples), Dravidians and so forth-are mainly of indigenous origin, and the contribution of immigrants (gene flow) is negligible. This is a major blow to the many invasion-migration theories that continue to dominate academic (and political) discourse in India. These of course hold that only the adivasis (‘first settlers') as tribals are often called, are indigenous while caste Hindus are descended from later immigrants, with the upper castes made up of immigrants from Europe. What these careful scientific studies have shown is this is simply not true.

            A basic fallacy in many of these analyses is to compare paternal lineages using Y-chromosomes and maternal lineages using mitochondrial DNA (mtDNA). (Y-chromosomes are passed on along the male line while the mtDNA are transmitted along the female line.) The mtDNA mutate (change) much more rapidly than other DNA (or nuclear DNA) - by as much as 15 to 20 times. The Y-chromosomes on the other hand hardly change at all. As Walter Bodmer, one of the originators of the Human Genome Project observed: ¹³ "No matter where they look, scientists have found that the Y chromosome, regardless of race or creed, has a molecular configuration that hardly varies at all. It does not matter whether the sample is Japanese, or from a Bushman, or a South American Indian, its composition produces extraordinarily few distinctive characteristics and none that can distinguish human populations from each other..."

Origins of Indian populations

Where does all this leave one with regard to the origin of Indian populations? Put simply, unlike the Americans or the Australians, Indians are overwhelmingly of Indian origin. There must have been migrations in past several thousand years, but their influence is negligible. The social structure of Indian society, or the linguistic evolution for that matter, cannot be attributed to migrations. But there is more: the Y-chromosome patterns suggest that modern Indians are descended from migrants from Africa who settled along coastal regions, that is to say in peninsular India. (Note that this Y-chromosome pattern changes hardly at all.) In addition, some mutations in the blood groups found in South India, which evolved as resistance to malaria, are found in Africa and South India. This is further evidence of migration from Africa to Coastal India. This took place according to modern studies-by no means a consensus-some 40,000 to 50,000 years ago. ¹⁴

This means: all of us are immigrants, but hardly any the result of any recent ones - like the Aryan invasion 3500 years ago or an earlier Dravidian migration from Scythia as Bishop Caldwell would have it. To understand the evolution of the Indian people, and the civilization they created-Pre-Vedic, Pre-Harappan and so forth-we need look deeper into the early settlements in peninsular India, especially in the coastal regions. It is probably no accident that the oldest settlements in India are being found in the peninsular region, like the caves of Bhimbetka, and, more recently, underwater settlements off the Gujarat coast and also Tamil Nadu. This is where historians need to look. But a multi-disciplinary approach is essential.

Conclusion

            The picture one obtains from a study of our evolutionary past is one of great diversity- biological diversity as well as cultural diversity. They do not lend themselves to simple, not to say simplistic, explanations based invading people bringing forth their genes, cultures and languages supplanting local populations- all in a few thousand years. We now have the beginnings of the technical tools needed to unlock the secrets of human past, but it needs many more years of dedicated study and research in several disciplines. It is important, however, to familiarize oneself with the basics of human genetics to be able to make use of the wealth of ideas and data that are becoming available to historians. ¹⁵ This probably calls for a new generation of historians that is less intimidated by science and also less deferential towards anything coming from Western academia as valid. Indian scientists and technologists have achieved this progress, but humanities scholars in India are yet to.

The following words of Sri Aurobindo written seven decades ago remain substantially true, especially in history: "... [that] Indian scholars have not been able to form themselves into a great and independent school of learning is due to two causes: the miserable scantiness of the mastery in Sanskrit provided by our universities, crippling to all but born scholars, and our lack of sturdy independence which makes us over-ready to defer to European [and Western] authority."

NOTES AND REFERENCES

1. Genes, Peoples and Languages by Luigi Luca Cavalli-Sforza 2001 Penguin, U.K. Pages: 228 + xii. Price: £7.99, Preface. See also The Journey of Man: A Genetic Odyssey by Spencer Wells (2002). Penguin, India. It is based on a television series drawing upon Cavalli-Sforza's work, but often fails to distinguish between science and speculative theories, especially in linguistics and history. Quite different is Mapping Human History: Genes, Race and Our Common Origins by Steve Olson (2002), Mariner Books: New York. Olson makes a distinction between what is scientifically established and speculation about origins, especially when writing about linguistic groups.
2. For example, this means that the variation within one caste group (like upper caste Hindus) will be greater than between upper castes and so-called adivasis or aborginals.
3. The skin color tends to get darker as we move closer to the equator and lighter as we move towards the poles. This is due to the effect of natural selection. Human pigmentation has evolved to be dark enough to prevent sunlight from destroying the nutrient called folate but light enough to foster the production of vitamin D. The fact that we see wide variation in skin color in India and Europe is evidence that they have lived there long enough for natural selection to work and therefore not recent migrants. Even then, there is a tendency to argue from extreme cases. There are fair skinned Keralites and dark skinned Kashmiris, but we tend to ignore them!
4. Human Diversity by Richard Lewontin (2000), New York: Scientific American Library, p 22.
5. R.A. Fisher was the founder of two important closely related disciplines- population genetics and mathematical statistics. Cavalli-Sforza carried forward Fisher's work in population genetics while C.R. Rao, another of Fisher's students, came to be recognized as the world's foremost mathematical statistician. (Thirty years ago, I was fortunate enough to be Rao's student at Indiana University in Bloomington, where I learnt some of this at first hand.)
6. The Vocabulary and Method of Reconstructing Language Trees: Innovations and Large Scale Applications by J.B. Kruksal, I. Dyen and P. Black, in Mathematics in the Archaeological and Historical Sciences (1971) edited by F.R. Hodson, D.G. Kendall, and P. Tatu, Edinburgh University Press, Edinburgh.
7. The Genetic Heritage of the Earliest Settlers Persist Both in Indian Tribal and Caste Populations: by T. Kisilvid, S. Rootsi, M. Metspahi, S. Mastana, K. Kaldma, J. Parik, E. Metspalu, M. Adojan, H.-V. Tolk, V. Stepanov, M. Gölge, E. Usanga, S.S. Papiha, C. Cinniğolu, R. King, L. Cavalli-Sforza, P.A. Unterhill and R. Villems. 2003. American Journal of Human Genetics, 72: pp 313 - 332.
8. See Language by L. Bloomfield (1935), London: Allen and Unwin, p 310. See also Chapter 4 (Emperor's Clothes) in The Politics of History (1995) by N.S. Rajaram, New Delhi: Voice of India.
9. The material in this section (except for facts on genetics) is discussed in detail in The Invasion that Never Was by Michel Danino (2001), Mysore: Mira Aditi. See also, The Politics of History, Op. cit.
10. Genetic Evidence on the Origin of Indian Caste Populations by M. Bamshad, T. Kivisild, W.S. Watkins, M.E. Dixon, C.E. Ricker, B.B. Rao, J.M. Naidu, B.V.R. Prasad, P.G. Reddy, A. Rasanagam, et al. 2001, Genome Research 11, pp 994 - 1004. (This article has a checkered history. When I pointed out the fallacies contained in the article to the editor of Genome Research, he stated that he was not the one responsible for its publication!)
11. Independent Origins of Indian Caste and Tribal Parental Lineages by Richard Cordaux, Robert Aunger, Gilian Bentley, Ivan Nasidze, S.M. Sirajuddin and Mark Stoneking, Current Biology, Vol. 14, February 2004: 231 - 35. This division into male migrants and indigenous females became necessary because, with no European traces in the mitochondria, the females could not have come from Europe. Cavalli-Sforza on the other hand concluded that females moved around more than the males, who for the most part stayed put in one location!
12. Voodoo Science: The Road from Foolishness to Fraud by Robert Park (2000). New York: Oxford University Press. Park makes the telling observation: "This brings up another symptom of pathological science ...: there does not appear to be anything resembling progress. The evidence never gets any stronger. Decades pass, and there is never a clear photograph of a flying saucer or the Loch Ness monster. Ten years after the announcement of cold fusion, results are no more persuasive than those obtained in the first weeks." (P 199) In the case of these race theories, including their linguistic incarnations, not decades but centuries have passed. To Park's observation, I would add another: while no evidence may be forthcoming, terminology multiples without end. This is what we are witnessing in these publications- the same conclusions being recycled in ever more profuse technical sounding verbiage.
13. The Book of Man: The Quest to Discover Our Genetic Heritage by Walter Bodmer and Robin McKie (1994), London: Abacus, p 225.
14. This is not accepted by several scientists, especially those who rely more on fossil data than on molecular analysis alone. See for example, The Multiregional Evolution of Humans by Alan G. Thorne and Milford H. Wolpoff that appeared in the April 1992 issue of Scientific American. This was updated in the Special Edition of Scientific American, Volume 13, Number 2, as A New Look at Human Evolution, that appeared in the summer of 2003. Thorne and Wolpoff state (p 46): "Both fossil and genetic evidence argue that ancient ancestors of various human groups lived where they are found today." This pushes back the date our ancestors in India by hundreds of thousands of years. See also The First Chimpanzee: In Search of Human Origins by John Gribbin and Jeremy Cherfas (2003), New York: Barnes and Noble, for an interesting account of the various controversies surrounding human origins and world populations.
15. A useful place to start is Human Diversity by Richard Lewontin (1995), New York: Scientific American Library. It is a popular work by one of the world's foremost human geneticists. Those more technically inclined might consult, The Molecular Biology of the Gene by James Watson, Tania Baker, Stephen Bell, Alexander Gann, Michael Levine and Richard Losick, 5^th edition (2004), Delhi: Pearson Education. It does not, however, discuss population genetics. Genes and Evolution by A.P. Jha (1994), Delhi: MacMillan, India, is more comprehensive but can be hard going. Several others are cited in the article.