go home History

The Elusive Origins of the Indo-Europeans
John V. Day

The Proto-Indo-Europeans, they say, were the herdsmen who changed the world. But these days even the majority of well- educated people in the West have never even heard of them. They might tell you that the Aryans, who were Proto-Indo-Europeans under another name, had some connection with Adolf Hitler, but this information stretches their knowledge to the limit. This widespread ignorance among Westerners is cause for great shame, but it should be expected. For decades, educators in schools and universities have neglected Proto-Indo-Europeans. And although several scholars in recent years have written general books about them, readers seldom come across these works in bookshops.1 Non-readers never have the chance to learn about Proto-Indo- Europeans, either. It appears that neither the TV companies nor Hollywood have made a single documentary or movie on the subject. And yet, as the history of the world turned out, these Proto-Indo-Europeans may have been the most important people who ever lived.

Now, this is not Erich von DΣniken's "Chariots of the Gods" or some other fanciful idea dreamed up by the unhinged or those wanting to sell mountains of books for a quick buck, although it must be admitted that over the years one or two misguided souls have tried to locate Proto-Indo-Europeans in such unlikely places as Tibet, the Sahara, Antarctica, and outer space. The real story of the Proto-Indo-Europeans has been pieced together from meticulous work by brilliant linguists, mythologists, archaeologists, and anthropologists over the last two hundred years.

Scholarship understands a lot about Proto-Indo-Europeans, and yet they are still the most elusive of peoples. For one thing, nobody can pin down precisely where they livedùor even precisely when they lived, although it must have been at least four or five thousand years ago. Nobody knows what they called themselves or what their neighbors called them.

"Proto-Indo-Europeans" is our modern term. None of the Proto-Indo-Europeans' literate neighbors recorded what they looked like or which customs they practiced. And we have no documents, not even a single word, written by the Proto-Indo-Europeans themselves. In all probability, they had no writing.

Language of the Proto-Indo-Europeans

Yet scholars have identified the Proto-Indo-Europeans mainly by their spoken language. This language may not have been written down, but as groups of Proto-Indo-Europeans spread further afield in antiquity and lost contact with each other, so their original
language diversified into daughter languages, and linguists can reconstruct a good deal of Proto-Indo-European from these daughter languages that "were" preserved in texts.

Consider, for example, some words in ancient languages that mean mother.2 The word mother in ancient Greek was meter, in Latin it was mater, and in Sanskrit, a language spoken in northern India over 3,000 years ago, it was matar. All these words correspond so well that linguists can reconstruct from them the original Proto-Indo-European form for mother as mater. (The modern English word mother, incidentally, derives from Proto-Indo-European via another route altogether, from its Germanic branch in ancient northern Europe.) Similarly, linguists can compare Greek nephos, Latin nebula and Sanskrit nabhasùall words meaning mist, fog or cloudùto obtain the Proto-Indo-European form for cloud. These words indicate only that Proto-Indo- European people recognized their mothers and experienced cloudy days. But linguists can go much further. Among the hundreds of Proto-Indo-European words that have been reconstructed are the numbers one to ten; the other family members of father, brother and sister; the body parts of eye, ear, nose and mouth; such trees as ash, birch, pine and willow; and such domestic animals as cow, sheep, goat and pig. Proto-Indo-European vocabulary was so precise, linguists tell us, that it even distinguished between words for breaking wind audibly and inaudibly.3

Furthermore, the parts of grammar that survive in Proto-Indo-European's daughter languages closely resemble one another. Pupils who study Latin often begin by learning amo, amas, amat ─ I love, you love, he loves. These verb endings of -o, -as, and -at find parallels in other languages, such as the comparable verb endings in modern German of -e, -st, and -t.

Linguists use a similar comparative method to determine that Proto-Indo-Europeans sorted nouns by gender (masculine, feminine, or neuter) and number (singular, plural, or dual [for two of a kind]). Each noun, moreover, had eight cases, depending on its purpose in a sentence, and each one had a different ending. Thus every Proto-Indo-European who opened his mouth to speak a few words realized that a noun like mother or cloud had 72 possible endings to choose from. Proto-Indo-Europeans may not have used writing, which was being invented by their contemporaries in the highly centralized economies of Egypt and Mesopotamia to count goods and register taxes, but they evidently did not suffer from low IQs.

The daughter languages of Proto-Indo-European can be grouped into such branches as Celtic, Greek, and Germanic, so that in the modern world English, Dutch, and German languages, for example, all belong in the Germanic branch. We know from ancient written texts that Indo-European languagesùthe languages that the original Proto-Indo-European developed intoùhave for thousands of years covered much of Europe and Asia.

During this period, Celtic languages were spoken across vast regions from central Europe to Iberia. Consider the linguistic map of Europe and Asia during the 1st millennium B.C., the period in which some of the earliest evidence for the location of early Indo-
European languages appears.4 Across northern Europe, running from west to east, were Germanic, Baltic, and Slavic branches, while the so-called "Iranian" branch was spoken on the steppe before moving southward into Iran itself. In Italy existed the Italic branch, its best-known member being Latin, and further east in ancient Europe there were Thracian, Illyrian, Greek, and Albanian branches. During early historical times, the Armenian branch was sited in Asia's far southwest and the Indic branch in south central Asia. Languages descended from all these Indo-European branches of Europe and Asia survive today. But some other branches have died out, such as the Anatolian and Phrygian in Anatolia (which is what prehistorians call Turkey) and the Tocharian in northwest China.

As noted above, this particular survey of Indo-European languages dates to roughly the first millennium B.C. Any such map can have only a rough date, because, for a variety of reasons, the extent of languages will change over time. For example, Celtic used to be spoken over much of western Europe but is nowadays confined to Brittany and the fringes of Britain and Ireland. This doesn't necessarily mean that Celts themselves were driven to Europe's western rim by Romans invading continental Europe and Anglo-Saxons invading England. More probably, ancient Celtic-speakers and their descendants stayed put on the land, and, over time, simply changed their speech. When natives have new rulers who speak an alien language, it must be in the natives' interest to start learning it.

Race and Indo-European Languages

Incidentally, ideas about mass migrations being common during prehistoric times arose in the Victorian age, when Europeans really were migrating en masse to the Americas and the colonial empires. But prehistoric people had no guns, railroads, or steamships, and would have found it much harder than nineteenth-century European colonists to migrate and to conquer natives. Anthropologists rarely find skeletal evidence of mass migrations in prehistory, because the skeletal record largely speaks of biological continuity. So too does Europe's genetic record, for the most part, even going as far back as the Ice Ages.

The discovery that ancient and modern Indo-European languages were spoken over a vast area came as a big jolt to educated people in the nineteenth century. They were staggered that all these languages were descended from a single ancestor. Indeed, the great French linguist Antoine Meillet likened the impact of the discovery of the Indo-Europeans to Columbus's discovery of the New World.

Meillet was right. For one thing, because scholars can reconstruct a good deal of the Proto-Indo-Europeans' language--and, by similar comparative methods, their customs and mythologyùwe moderns can glimpse a prehistoric mentality. No longer restricted to such humdrum archaeological finds as stone tools and charred seeds, we can get inside the minds of the distant Proto-Indo-Europeans and understand their outlook on life.

Many people also find something intriguing in the idea that one fairly small prehistoric population and its descendants somehow
managed to expand across most of Europe and much of Asia, disseminating their language and culture on the way. After all, the Proto-Indo-Europeans' descendants provided much of the language and culture for the civilizations of ancient India, Iran, Greece, Rome, and Celtic and Germanic Europe.

Not surprisingly, Proto-Indo-Europeans were greatly admired by such earlier racial historians as the Count de Gobineau and Madison Grant and, of course, the Aryans were also the favorite people of Adolf Hitler. This enthusiasm for Proto-Indo-Europeans as the ancestors of the white race and European culture has contributed to the contemporary taboo against Westerners identifying too closely with their racial origins.

The racial origins of the Proto-Indo-Europeans are, like race and IQ or race and crime, a red-hot subject. Take the case of Professor Wolfram Nagel of Berlin University, who in 1987 argued in the journal of the German Oriental Society that Proto-Indo-Europeans must have been racially northern European.5 He didn't say they were a master race or destined to conquer the world, just that they were northern European. Although Professor Nagel had reached the top of his profession, his reasoned arguments based on ancient texts and artworks so appalled the learned society that they fired the journal's editors and debated whether to expel him (although in the event they allowed him to stay). This incident offers an insight into the totalitarian climate that intellectuals work under in "democratic" Germany.

Similarly in France, two intellectuals whose books and articles describe Proto-Indo-Europeans as racially northern European--Alain de Benoist,6 the leading figure of the French New Right, and Professor Jean Haudryùare routinely vilified as Nazis. Westerners are living in a strange world, when discussing the origins of their people and culture can land them in so much trouble.

The Search for the Proto-Indo-Europeans' Homeland

As noted above, the location of the Proto-Indo-European homeland has long been the subject of speculation. One might begin the search for it by deciding if the reconstructed Proto-Indo-European language offers any clues about where or when its speakers may have lived. Proto-Indo-European had words for houses, for taming animals, for wagons and for pottery, implying that its people must have lived during the Neolithic or even later, which gives us a general time-frame for the period of archaeological cultures and skeletal material that prehistorians should be examining.7 In addition, the earliest words from one of Proto-Indo-European's daughter languages, Hittite in Anatolia, appeared around 1900 B.C., and so Proto-Indo-European itself must have existed at least a few centuries earlier, before developing into Hittite, and so perhaps before about 2500 B.C.

Proto-Indo-Europeans can therefore be placed vaguely in time. But prehistorians struggle to pin them down geographically. Over the years, scholars and cranks alike have offered dozens of apparent solutions to the problem of the Proto-Indo-European homeland. Many seemingly ingenious proposals have seized on just one reconstructed Proto-Indo-European word, such as beech or salmon, to determine where these occurred in prehistoric times and delimit the homeland, but so far no proposal has worked. All these proposals turn out to be too vague. (One Icelandic linguist offered an especially bizarre idea, arguing that the harsh sound of some Proto-Indo-European words imitates seabirds living around the Baltic.)

Turning to more serious matters, once ancient people had given up hunting and gathering, which necessitated roaming across wide
territories, and had taken up the Neolithic, including farming and settling down into hamlets and villages, becoming more or less
rooted to the soil, their populations became relatively isolated from one another, and over time their languages also became isolated, accumulating more and more differences from one another. Judging by parts of the world that even now have a Neolithic way of life, the original homeland of the Proto-Indo-Europeans would have been more or less the size of, say, Poland.8

In tracing Proto-Indo-European origins, anthropology offers three main kinds of evidence in Europe and Asia. First, the genetic data, though so far almost all our data comes from modern populations. Second, the masses of information from ancient times about physical types, and most important of all about hair and eye pigmentation--information that comes from texts, artworks and mummified corpses. Finally, the ancient skeletal remains. Now, anthropologists cannot immediately deduce from any archaeological culture's skeletal remains that, in life, its people spoke Proto-Indo-European. All we can do with ancient skeletal material is determine cases of population movements, and then decide if any such movements match the relevant period of Indo-European expansions and the relevant lands penetrated by Indo-Europeans. Likewise with modern genetic material, we can use it only to locate ancient population movements that might correspond with Indo-European expansions.

The ancient texts and artworks recording human pigmentation offer a different kind of evidence. After all, these texts and artworks come from, or are about, historical societies that were certainly Indo-European-speaking, and so some, if not all or even many, of the people in these societies were descended from Proto-Indo-Europeans, as I hope to show later.

Anyway, let's begin with the genetic evidence. Any similar article written in 2020 will discuss at length the evidence of ancient DNA. Ancient DNA taken from human teeth and bone will revolutionize the study of prehistory. It will tell us about the sex of individual ancient humans, their familial relationships and their biological affinities and ancestries. Geneticists might one day draw up a family tree for all the populations of ancient Europe and Asia. And once geneticists have located the genes controlling hair and eye colors, we can speculate about the likely pigmentation of ancient human populations. We shall also use DNA from ancient domesticated crops and animals to explain how early farming expanded.

At present, though, ancient DNA has revealed only that modern humans are not, as Carleton Coon once believed, descended from
Neanderthals.9 But as for Indo-Europeans, current studies of ancient DNA tell us next to nothing.

Many prehistorians have used modern genetic data to work out where Proto-Indo-Europeans came from and how they expanded, but most of their ideas are chasing down blind alleys.10 For example, many analyses try to match modern genetic boundaries with modern or ancient linguistic boundaries, arguing that neighbors who speak different languages rarely marry each other, and so over time their populations have diverged genetically. But populations divided genetically and linguistically are also often separated by such physical boundaries as mountains and seas, and this factor complicates matters inextricably.11

This article touches very briefly on a few of the more important findings from genetic studies. First of all, it turns out that, in genetic terms, modern Europe is very homogeneous, and northern Europe even more so. Genetic distances between northern European populations are usually lowùbetween English and Germans, for example, English and French, and English and Irish. In contrast, many genetic distances in southern and eastern Europe are a good deal higher, such as those between Greeks and Hungarians, and Greeks and Yugoslavs.12

Genetically, Greek and Yugoslav populations are among the least typically "European." And the significance of this impinges on
Colin Renfrew's hypothesis that around 7000 B.C. Proto-Indo-Europeans were farmers in Anatolia, and indeed farming so well that their big population increases enabled them and their descendants to spread across most of Europe in the course of thousands of years, mixing with indigenous Europeans on the way.13 Yet it seems odd that Greeks should be divided by fairly large genetic distances from Hungarians and Yugoslavs if Anatolian farmers really had expanded via southeast Europe en masse. One might expect such a large-scale population movement to have homogenized gene pools in southeast Europe.

A particular kind of DNA is mitochondrial DNA (mtDNA), which has nothing to do with shaping physical or personality traits. Both males and females carry mtDNA, although only mothers pass it on, and when it is inherited there are occasional mutations. In consequence, geneticists can examine mtDNA lineages to determine how they evolved into new types. And comparing lineages from different populations allows us to work out where various lineages arose and, if we estimate mutation rates, when they arose.

Bryan Sykes and others classify modern European mtDNA in nine major lineage groups. Sykes finds that eight of these nine groups arose in Europe as long ago as the Upper Palaeolithic, during the time of the Ice Ages.14 But one lineage group which originated in southwest Asia entered Europe during the last 10,000 years and currently occurs across much of Europe, perhaps comprising 17% of modern European lineages, although another study puts it at more like 10%.15 This lineage group, Sykes argues, ran in two streamsùone common along the Mediterranean coast to Spain, Portugal and from there along the Atlantic coast to Cornwall, Wales and western Scotland, the other common in the river valleys of central Europe. And these two streams, he suggests, reflect ancient Anatolian farmers spreading northward and westward across Europe.

As for the problem of Indo-European expansions, Sykes's deduction makes a neat solution. It explains how Indo-Europeans managed during prehistoric times to advance across most of Europe and part of Asia. However, his theory doesn't makes any senseùat least, not as far as Proto-Indo-Europeans are concerned. Proto-Indo-Europeans appeared later on. For one thing, the age and distribution of the mtDNA stream along the Mediterranean and Atlantic coasts might be an echo of another migration altogetherùthe migration that thousands of years later took megalithic tombs around the coasts of western Europe.16

Sykes's hypothesis also runs into difficulties with the skeletal remains.17 Several studies of early Neolithic skeletal material find that, contrary to his hypothesis, remains from the Balkans don't really resemble remains from southwest Asia. So were these two populations related? In addition, we cannot be certain that early Neolithic remains from central Europe closely match remains from the Balkans. So this apparently unstoppable advance from Anatolia via the Balkans to northern Europe is, judging by the skeletal record, by no means proven. But even if it took place, such a population movement might still have no connection with the expansions of Indo-Europeans.

Indeed, linguists can apparently reconstruct Proto-Indo-European words for items of material culture that first appear in the archaeological record, as far as we know, only from the fourth millennium B.C. onwards. Some of these words are for wagon, axle, wheel, and reins.18 But if Proto-Indo-Europeans still existed as a unified population at this late date, then they cannot have begun separating as long ago as 7000 B.C., when wheeled vehicles were still unknown.

The pattern of languages tells a similar story. Archaic languages that preserve Proto-Indo-European forms are often found on the edge of the Indo-European world. Many correspondences link, for example, Germanic in northern Europe and Tocharian in central Asia. Indeed, the Indo-European branch retaining the highest percentage of reconstructed Proto-Indo-European wordsùabout 67% of themùis Germanic, followed by Greek with 60% and Baltic with 54%.19 Again, if farmers had taken thousands of years to migrate across Europe from an Anatolian homeland, one might expect that Germanic and Baltic would have the fewest original words, because migrants traveling ever further into new country for thousands of years, and marrying with natives on the way, would find their original vocabulary becoming more and more diluted.

Consider also the similarity between Indo-European mythologies. Scholars of religion consider the three great reservoirs of Proto-
Indo-European mythology to be Rome (think of Mars and Jupiter, Romulus and Remus), Scandinavia (think of Thor and Odin), and India (think of Indra the warrior-god and Agni the fire-god).20 Yet, just like the most archaic languages, these three regions sit right on the edge of the Indo-European world, thousands of miles apart. But if Anatolian farmers and their descendants had trekked across Europe and Asia, think how much Proto-Indo-European mythology would have been lost by the time, thousands of years after setting out, that they eventually settled in such distant lands as Rome, Scandinavia and India. So perhaps Indo-European settlers made fairly swift expansions to their new lands, where they established themselves.

Ancestral Clues From Antiquity

Information about how pigmentation was distributed in antiquity provides crucial evidence in understanding where Proto-Indo-Europeans originated and how they expanded, and is far too useful to be disregarded. Many students of Proto-Indo-European origins examine genetic data, and some even consider skeletal remains. But very few in recent years have said anything about ancient texts and artworks recording pigmentation. Linking Proto-Indo-Europeans with a specific pigmentation became a huge taboo once the National Socialists began promoting their doctrine of "blond Aryans," even though similar ideas go back as far as the 1820s.21 But we should ignore taboos of political correctness.

Yet ancient sources about pigmentation are often scarce. A further problem is that the earliest useful texts from each Indo-European branch vary so much in period. The earliest useful texts about pigmentation from India come from the mid-second millennium B.C., whereas those from Ireland and Scandinavia were composed two thousand years later. Ideally, ancient peoples would have compiled anthropological surveys, but these simply don't exist. Descriptions of historical figures provide a rough population sample, although even the Greeks of the classical period virtually never reported the coloring of their greatest men. And so anthropologists must also examine the pigmentation of mythical figures and deities, working on the assumption that their physical appearance mirrors the real-life people who admired or worshiped them.

The Indo-European world covers so many lands and eras that this article will consider just six of the many regions where Indo-European-speaking peoples lived in antiquity: Ireland, Rome, Greece, Iran, India, and Xinjiang (which used to be known as Chinese Turkestan) in northwest China.

For evidence of how the Celts described themselves, we might turn to the highly traditional society of early medieval Ireland. One
excellent source from Ireland is the epic Tßin B≤ Cuailnge, otherwise known as The Cattle Raid of Cooley, probably composed as late as the seventh or eigth century A.D. and lying at the heart of early Irish literature. The Tßin and other Irish works contain some valuable descriptions of mythological heroes.22 In the world they depict, beautiful women generally have fair hair and blue eyes, and the great warrior-heroes, although varying more than the women, also tend to have fair hair and, when bearded, always fair beards. Moreover, early Irish tales often regard men who have dark hair as somewhat alien, because some ugly giants and male slaves are dark-haired, and even a few important dark-haired warrior-heroes are regarded as marginal figures.

In ancient Rome, some valuable descriptions of physical traits are embedded in the biographies of early emperors. The earliest nineteen Roman emperors, from Caesar up to Commodus at the end of the Age of the Antonines in A.D. 192, offer a small but exceptionally useful population sample.23 Of these nineteen emperors, four have no descriptions and two are described only as gray-haired. But whereas one or perhaps two of the remainder have dark hair, five are described as having fair or fairish hair. And whereas three have dark eyes, nine have blue or grayish eyes, and indeed five of the first seven have blue or grayish eyes. For example, Augustus and Nero had fairish hair and blue eyes, Caesar had dark eyes, and Hadrian had dark hair and blue eyes. Although upper-class Romans tended to have a light pigmentation, they were greatly outnumbered by the Roman masses, who overwhelmingly had dark hair and eyes.

In Greece, Homer's two epics from the eighth century B.C., the Iliad and the Odyssey, are among the earliest texts in Europe containing useful information about pigmentation.24 When characterizing his Greek warrior-heroes, Homer says nothing about the coloring of Agamemnon, but he does picture Achilles, Meleager, Menelaus, and Odysseus as fair-hairedùa coloring that coincides with their youthfulness. Certainly by classical times, however, the great majority of Greeks had dark hair.

Useful information about real rather than fictional Greeks comes from Polemon, the second most important Greek writer on physiognomy, who wrote as late as the second century A.D..25 Polemon explains that "the pure Greek" of his time has fair skin and red hair, and resembles the man inclined to literature and philosophy, who has fair skin and fairish hair. Polemon may have drawn these ideas from Pseudo-Aristotle, the most important Greek physiognomist, who in his third century B.C. Physiognomica declares that the most perfect male type is the lion with its fair mane.26

Fair Indo-Europeans from the Caspian to TurkestanTurning to Iranians, I remarked earlier that speakers of Indo-European's so-called "Iranian" branch must have lived on the steppe before infiltrating southward to Iran, where non-Indo-European Elamites already had a civilization. Now, Greek and Roman writers in the centuries before and after Christ stated that Iranian-speaking peoples north of the Black Sea and Caspian had fair or reddish hair and blue eyes.27 One especially trustworthy source is Ammianus Marcellinus, because he had visited the Black Sea region, unlike the writers who simply relied on others' reports, and he portrays the Alans with fairish hair.

From Iran itself, although nowadays housed in the Louvre in Paris, comes the Archer Frieze of glazed bricks, which was created about 500 B.C. to represent the bodyguards of Darius I.28 Most of the eighteen or so archers on the frieze have dark skin, hair, and eyes, but a few have blue eyes. This frieze originally stood outdoors, causing the pigment for archers' skins to darken over time. But we do have some brick fragments showing paler skin, and Annie Caubet, the director of the Louvre's Department of Oriental Antiquities, told me in a letter that pinkish skin probably came from the frieze's portrayal of Darius himself.

Many similarities in language, as well as in mythology and culture, are shared by the Indo-European's Iranian and Indian branches, which implies that, wherever the Proto-Indo-Europeans had their homeland, the Proto-Iranians and Proto-Indians must even then have been neighbors who resembled each other physically. One similarity is that the Indo-European peoples in Iran and India both referred to themselves as Aryans. The Iranian king Darius I, in a famous rock inscription, calls himself "an Aryan of the Aryans," and the very word Iran developed from Aryan. These days, more and more linguists are returning to the older opinion that the term Aryan occurred throughout the Indo-European world, cropping up, for instance, in the Old Irish word aire meaning "noble, free," and hence Ireland's name of Eire. To the Proto-Indo-Europeans, it seems, Aryan meant peer or comrade or perhaps an ethnic term.29

In India, the earliest known Indo-European text, coming from the later second millennium B.C., is the religious work, the Rig-Veda.30 Only one god in the Rig-Veda has anything like a human pigmentation, and he is the great warrior-god Indra. In personality and attributes, Indra resembles the Germanic god Thor, and even his fair hair and beard resemble Thor's red beard.31 Throughout the Rig-Veda, Indra often helps the warlike Aryansùthe Indo-European invaders of Indiaùto battle against the native Dasas and Dasyus, who are portrayed as dark-skinned. In contrast, the Rig-Veda refers to Aryans as white and having an "Aryan color."

Later works from ancient India also offer useful sources, and one of them is the very long Mahabhasya, composed in northern India by the grammarian Pata±jali in the second century B.C. 32 In this work, Pata±jali, making a philosophical point about objects having and lacking attributes, casts around for an illustration that makes obvious sense to his readers. Nobody, Pata±jali says, would look at a dark-skinned man and imagine that he was a brahmin, from India's highest caste. Instead, he goes on, everyone knows that brahmins have fair skins and kapila-kesa hair, which translates as "brown" or "reddish-brown."

Finally, abundant evidence comes from Xinjiang in northwest China, the home of people speaking Indo-European's Tocharian branch. Unfortunately, ancient Chinese sources rarely comment on the physical appearance of foreigners. But they do record that the Yuezhi, who may have been Tocharians under another name, had fair skins, and that the Wusun's descendants, again possibly Tocharian, had green (or blue) eyes and red beards.33

Some of the best evidence for Tocharians is artistic, because they were painted on murals in Xinjiang during the later first millennium A.D. One example is the so-called Cave of the Sixteen Sword-Bearers at Kizil.34 Of these sixteen knights, five have white hair and eleven have light red hair. Marianne Yaldiz, the director of Berlin's Museum for Indian Art, where the murals are now housed, told me in a letter that the eyes are a sort of gray-green-blue. Although the men wear Iranian-style dress and stand in an Iranian-style pose, historians generally regard them as Tocharians.

Still, all of these sixteen figures are stereotyped. To find individual portrayals, we must turn to other murals in Xinjiang that are three hundred miles further east at Bezeklik.35 Murals at Bezeklik chiefly portray typically Chinese faces, although all these Mongoloids are stereotyped. In contrast, the minority of Caucasoids on the murals are rendered as individual portraits, as in one cave at Bezeklik which portrays about six or seven Buddhist monks who have Caucasoid features. These Caucasoid monks are apparently a distinct ethnic groupùunlike the Mongoloids, they all have heads shaven on top, and all wear similar gownsùand are almost certainly Tocharians. One or two of these Caucasoid monks have dark hair and brown eyes, but most have reddish-brown hair and blue or green eyes.

In recent years, archaeologists working in Xinjiang's Tarim Basin have excavated more than one hundred Caucasoid mummies, thanks to a desert climate and salty soil's having preserved corpses. Even some mummies 3,000 years old look as though they were buried days ago. These Caucasoid mummies have typically northern European faces, with prominent noses, unslanted eyes, and hair that is usually fair or light brown.36 Although the mummies' eyes have long since perished, we know that two infants were buried with stones placed over their eyes, one with green stones and one with blue, colors perhaps representing their irises.37 Judging by the mummies' location, historians conclude that at least the great majority of these Caucasoids were ancestral Tocharians.

Origins on the Steppe?

Does this brief survey of pigmentation in ancient Ireland, Rome, Greece, Iran, India, and Xinjiang tell us anything? I think it clearly does. Light-haired and light-eyed types were found all over the ancient Indo-European world, even in lands which at present are overwhelmingly dark in pigmentation, such as Rome and India. And traces of these northern European types occurred especially among the warriors who comprised each society's ruling class.

Ireland had an abundance of fair-haired warrior-heroes. Most of the early Greek warrior-heroes had fair hair, too. In fact, most of the early Roman emperors had fair or fairish hair and blue eyes. The Iranians who lived on the steppe north of the Black and Caspian Sea were also described by foreign observers as having fair or fairish hair. Indian Brahmins have been characterized as having fair skins and brown or reddish-brown hair, and the Indian warrior-god Indra apparently had fair hair as well. Finally, the mummies and murals of Xinjiang reveal that most Tocharians had fair or brownish hair and blue or green eyes.

Moreover, Indo-Europeans often seem to have been small minorities in the countries they penetrated: the Celtic warrior-class in Ireland; the Roman patricians; the few Homeric heroes and the so-called "pure Greeks" of later years; and the Aryans battling against the many natives in India. Then again, the majority of Tocharians in Xinjiang apparently had light pigmentation, as did most Iranians living on the steppe.

Indeed, many prehistorians believe that the Proto-Indo-European homeland lay on the steppe, which, if true, might explain why steppe Iranians retained the Proto-Indo-Europeans' northern European physical type. A steppe homeland, moreover, would have been a good basis for Indo-European expansions. Steppe groups during the third millennium B.C. and earlier lived mainly by cattle and sheep herding, and by at least the third millennium B.C. they had also domesticated the horse. Down to historical times, such other steppe pastoralists as the Huns and Mongols have been highly mobile horseriders and warlike, too, living in the midst of poor farmland and consequently raiding neighbors for food supplies. So if Proto-Indo-Europeans did originate on the steppe, they may also have been highly expansionist.

To confirm that Proto-Indo-Europeans did originate on the steppe, we must find traces in the skeletal record of prehistoric steppe groups expanding across the known Indo-European worldùto Xinjiang, Iran, India, Greece, Rome, and northern Europe. The evidence of pigmentation surveyed above implies that Indo-Europeans were usually minorities in the lands they entered, and must have expanded from their homeland in smallish groups. This finding tallies broadly with the skeletal record, which in general points to continuity in prehistoric Europe and Asia, where population movements on a large scale were the rare exception.

But prehistoric steppe groups did extend as far east as southern Siberia and Xinjiang, as demonstrated by both archaeological evidence and the remains of robust skeletal types.38 At present, though, traces of these steppe groups have not been found entering Iran or India, and neither can they be located as far west as Italy. In the northern Balkans, prehistoric steppe groups are
certainly represented by skeletal and archaeological remains, but did they penetrate as far south as Greece? Archaeological traces of steppe groups largely peter out before Greece, but the renowned Grave Circle B at Mycenae resembles steppe tombs, and the very rugged nobles buried here also resemble steppe groups.39 Steppe groups definitely expanded as far westward as central Europe as well, judging by the three thousand steppe graves known in eastern Hungary, and, although the females buried here seem lightly built, the males are similar to robust steppe types.40

But did steppe groups reach northern Europe? It is there that several important Indo-European peoples first emerge into history: Slavs, Balts, Germans, and Celts. The archaeological record is ambiguous: there are many disputed parallels between the Late Neolithic culture of northern Europe, known as Corded Ware, and Neolithic steppe culture, although vague cultural parallels can't
automatically be attributed to migrating groups. The skeletal remains are less ambiguous, however, because they show no traces of steppe groups reaching northern Europe. The typically Corded Ware skulls from Germany, Czechoslovakia , and Poland are high and have narrow faces, whereas steppe crania are low and have broad faces.41

The archaeological and skeletal evidence seems to leave us with three possibilities. First, that steppe groups did reach northern Europe, but in such small numbers it makes it nearly impossible to detect them. Second, that steppe groups didn't reach northern
Europe, which proves that, at least in this region, steppe groups were not transmitting Indo-European speech. Third, that steppe groups didn't need to reach northern Europe, because Proto-Indo-Europeans lived in a vast homeland that encompassed the steppe and northern Europe.

The puzzle remains. But steppe groups must somehow be implicated in Proto-Indo-European origins. They, and they alone among prehistoric groups, expanded to much of Europe and Asia where Indo-European languages were known to have been spoken. Perhaps one day archaeologists and anthropologists will determine exactly the prehistoric links, if any, between the steppe and northern Europe.

Scholars might also have a clearer picture about Indo-European influences in eastern Asiaùon the civilizations of China, where Indo-Europeans may have introduced bronze-working and the chariot, and Japan, whose mythology bears unmistakable affinities with Proto-Indo-European mythology.42

What we can declare is that Indo-Europeans tended to expand in small groups, and that in the great civilizations of Ireland, Rome,
Greece, and India they and their descendants were heavily outnumbered minorities who were concentrated in the ruling classes. I take it that Indo-Europeans were ruling these lands because they had somehow dominated the natives by force of arms, although the archaeological evidence for this assertion scarcely exists. So far, prehistorians have found it perplexing to explain from the
archaeological record how Indo-Europeans arrived in any land and established themselves as the commanding power.

And what happened to these Indo-Europeans? It appears that at least Europe and southern Asia were so densely populated by Neolithic times that small groups of newcomers would have made little biological impact on the natives. Since Proto-Indo-Europeans began expanding about five thousand years ago, two hundred generations have passed, and the few drops of their original blood have been lost in an ocean of non-Indo-Europeans. Traces of light hair and eyes crop up now and again in modern Iran and northern India, and even in Xinjiang, where Dolkun Kamberi, a local expert on the Caucasoid mummies, has green eyes and light brown hair. Light hair and eyes are more common in modern Greece and Rome and especially Ireland, although in northern Europe most traces probably predate any incoming Indo-Europeans.

In a journal about the West and its future, it is fitting to end this article by briefly recounting the fate of the Roman upper class. Among Indo-European peoples, the Romans offer an especially useful example because they left masses of records, enabling later historians to determine what became of them. The evidence found in ancient texts implies that this class descended largely from Indo-Europeans who had a decidedly northern European physical type, although that isn't something one reads in modern books about Roman history. In Rome, though, the upper class was always a tiny minority. Instead of protecting its interests, it allowed itself to wither away. Consider a bleak statistic. We know of about fifty patrician clans in the fifth century B.C., but by the time of Caesar, in the later first century B.C., only fourteen of these had survived.43 The decay continued in imperial times. We know of the families of nearly four hundred Roman senators in A.D. sixty five, but, just one generation later, all trace of half of these families had vanished.44

If we in the West want to avoid a similar fate, we must learn from Indo-European history.

John V. Day, Ph.D., is the author of Indo-European Origins: The Anthropological Evidence (The Institute for the Study of Man, 2001) and is the editor of the forthcoming anthology, The Lost Philosopher: The Best of Anthony M. Ludovici.

End Notes

1. For an excellent introduction to the subject, see Shan M.M. Winn, Heaven, Heroes, and Happiness: The Indo-European Roots of Western Ideology, Lanham, MD: University Press of America, 1995. See also John V. Day, Indo-European Origins: The Anthropological Evidence, Washington, DC: Institute for the Study of Man, 2001.
2. In the interests of clarity, linguists have transliterated the Greek, Sanskrit, and Proto-Indo-European words in the following examples into the Roman alphabet, removed any accents, and used only the root of each word without any suffixes.
3. For a survey of modern English words that ultimately derive from Proto-Indo-European, see Calvert Watkins, The American Heritage Dictionary of Indo-European Roots, Boston: Houghton Mifflin, 1985.
4. It is not only Indo-Europeans that had a wide language family. Uralic, for example, comprises Finnish, Lapp, Estonian and Hungarian and various languages of northwest Asia. But in modern Europe the only significant peoples who do not speak Indo-European languages are Basques, Hungarians, Finns, Estonians, and Georgians. Together, these non-Indo-European-speakers number only about 25 million out of over 500 million Europeans.
5. Wolfram Nagel, "Indogermanen und Alter Orient -- Rⁿckblick und Ausblick auf den Stand des Indogermanenproblems," Mitteilungen der Deutschen Orient-Gesellschaft 119, 1987, pp. 157-213.
6. Alain de Benoist, "Indo-EuropΘens: α la recherche du foyer d'origine; quatre remarques finales," Nouvelle ╔cole 49, 1997, pp.
7. Day, pp. 4-6.
8. Day, p. 8.
9. Bryan Sykes, The Seven Daughters of Eve, New York: W.W. Norton, 2001, p. 126.
10. Patrick Sims-Williams, "Genetics, Linguistics, and Prehistory: Thinking Big and Thinking Straight," Antiquity 72, 1998, pp. 505-27.
11. Day, pp. 253-66.
12. L. Luca Cavalli-Sforza, Paolo Menozzi, and Alberto Piazza, The History and Geography of Human Genes, Princeton: Princeton University Press, 1994, p. 270.
13. Colin Renfrew, Archaeology and Language: The Puzzle of Indo-European Origins, London: Jonathan Cape, 1987.
14. Day, p. 277; Sykes, pp. 184, 196.
15. Day, pp. 279-80.
16. Day, pp. 278-9.
17. Day, pp. 221-33.
18. J.P. Mallory and D.Q. Adams, eds., Encyclopedia of Indo-European Culture, Chicago: Fitzroy Dearborn, 1997, pp. 481, 625-8.
19. Day, p. 24.
20. Jaan Puhvel, Comparative Mythology, Baltimore: Johns Hopkins University Press, paperback edition, 1989, p. 39.
21. Day, p. 125.
22. Day, pp. 107-13.
23. Day, pp. 102-6
.24. Day, pp. 87-92.
25. Day, p. 94.
26. Day, pp. 94-5.
27. Day, p. 57.
28. Day, pp. 134-6.
29. Puhvel, p. 45; Mallory and Adams, p. 213.
30. Day, pp. 74-9.
31. Day, pp. 79, 115-16.
32. Day, pp. 80-2.
33. Day, pp. 58-9.
34. Day, pp. 137-8; J.P. Mallory and Victor H. Mair, The Tarim Mummies: Ancient China and the Mystery of the Earliest Peoples from the West, London: Thames & Hudson, 2000, plate XII.
35. Day, pp. 138-9; Mallory and Mair, Plate XIII.
36. Day, pp. 351-5; Mallory and Mair, pp. 181-200.
37. Day, p. 352 n. 1.
38. Day, pp.184-7, 190-4.
39. Day, pp. 199-200.
40. Day, p. 209.
41. Day, p. 206.
42. Mallory and Mair, pp. 324-6, 327-8; Michael Puett, "China in Early Eurasian History: A Brief Review of Recent Scholarship on the Issue," in Victor H. Mair, ed., The Bronze Age and Early Iron Age Peoples of Eastern Central Asia, Washington, DC: Institute for the Study of Man; Philadelphia: University of Pennsylvania Museum Publications, 2 volumes, 1998, pp. 699-715; C. Scott Littleton, "The Indo-European Strain in Japanese Mythology: A Review of Some Recent Research," Mankind Quarterly 26, 1985-6, pp. 152-74.
43. Arnaldo Momigliano and Tim J. Cornell, "Patricians," in Simon Hornblower and Anthony Spawforth, eds., The Oxford Classical Dictionary, New York: Oxford University Press, 3rd edition, 1996, p. 1123.
44. James Paton Isaac, Factors in the Ruin of Antiquity: A Criticism of Ancient Civilization, Canada: no publisher, 1971, p. 421. See also Nathaniel Weyl and Stefan T. Possony, The Geography of Intellect, Chicago: Henry Regnery, 1963, pp. 106-9; Ilse
Schwidetzky, Das Problem des V÷lkertodes: Eine Studie zur historischen Bev÷lkerungsbiologie, Stuttgart: Ferdinand Enke, 1954,
pp. 43-53.