J. D. Anderson

Curiously enough, the systematic enquiry into the physical race-characteristics of the Indian peoples was due to a daring assertion by Mr Nesfield, of the Indian Educational Service, to the effect that, so far as physical signs go, there is practically only one Indian race and one Indian caste. This was a hasty but quite natural generalisation from experience of a part of India, the United Provinces, which is in the heart of the Aryan settlement in the Gangetic do-ab (the area between "two rivers"). Here caste has long been a settled institution, and innumerable sub-castes, professional or the result of outcasting, have come into existence. Mr Nesfield was driven by his local observations to assert the unity of one great Indian race; he denied the truth of "the modern doctrine which divides the population of India into Aryan and aboriginal": he sturdily declared that it was impossible to distinguish a scavenger from a Brahman, save by costume and other artificial and accidental marks. Even in the United Provinces this uncompromising statement awoke dissent. In other parts of India, as, for instance, on the north-eastern frontier, the crowded home of many races and languages, dissent was eager and loud. It was evident, on the face of it, that Mr Nesfield's new dogma was based on too limited a study. Caste, for him, was a mere matter of hereditary function and profession; since most castes in the sacred "midland" of Hinduism have assumed that guise. There is no reason to suppose that castes have usually or even often been formed as professional guilds. They come into being for many reasons, some of which will be presently stated; and in civilised communities, where the division of labour and specialisation of professional skill are well established, a caste gradually assumes some distinctive means of livelihood. But on the borders of Hinduism, where the Hindu social system is still assimilating new races, instances abound of racial castes, tribal castes, perhaps even (though this is a more doubtful matter) totemistic castes.

Those who had the widest experience of the Peninsula were convinced that its races were at least as varied as those of Europe: those who, like Mr Nesfield, had made a close study of one limited tract, might have continued to believe that under the superficial distinctions of caste and class lay a real unity of race. But Mr (afterwards Sir H. H.) Risley had spent the early years of his Indian service among the Dravidian tribes of Chota Nagpore, and was aware that they differ more widely from the people Mr Nesfield had studied than an Englishman differs from a Turk. The difference, indeed, was almost as great as that between a European and a Chinaman. Could such differences be registered and described in such a way as to convince minds accustomed to scientific accuracy in statement? Mr Risley thought he saw his way to an ethnological classification of Indian races and castes by means of the then comparatively new methods of anthropometry. In 1891, he published in the Journal of the Anthropological Institute a paper which marked the beginning of systematic ethnological studies in India. It contained a summary of the measurements of eighty-nine castes and tribes of Bengal, the United Provinces, and Bihar. It dealt, therefore, with the great alluvial plain, created by the Ganges and Indus, which lies between the Himalayas and the massif central of the Deccan. Here is the home of the Aryan immigrants, where the great Indo-European languages are spoken by communities as numerous as the larger European nations. Anthropometry showed in the plainest, the most incontrovertible way, that the caste system of marriages had sorted out men into classes possessing definite and recognisable physical characteristics. There were local differences, and caste differences. It only remained to extend anthropometrical measurements to other parts of India to prove that the many languages and religious beliefs of India are associated with an even greater variety of physical qualities. Such enquiries are still in progress, but many notable results have already been obtained, especially by Mr Edgar Thurston, in his now famous investigations into Dravidian ethnography.

The most important and significant measurement is that of the shape of the head. It is, of course, impossible to take a man at random and to say with certainty that the excessive length or breadth of his skull proves him to belong to a given race. But the average skull-measurements of a race are distinctive, and confirm, on the whole, the impressions created by general aspect, colour, language and other vaguer indications. The general result is as follows. At either end of the Himalayan range, in Baluchistan on the west, and in Assam and Burma on the east, broad heads prevail. Broad too are the heads of the mostly Mongolian races inhabiting the valleys of the southern slopes of the Himalayas, and in a belt of country running down the western coast at least as far south as Coorg. In the Panjab, Rajputana, and the United Provinces, tracts where the climate is dry and healthy, where great summer heat is compensated for by a bracing winter, where wheat is for the most part the staple food, long heads predominate. In Bihar, travelling eastwards, medium heads are most common. In the damp and steamy delta of Bengal, inhabited by over forty millions of rather dusky rice-eating people, there is a marked tendency towards the Mongolian brachy-cephaly of Tibeto-Burman races. It is visible among the Muhammadans and Chandals of Eastern Bengal, people who are probably indigenous in this tract, it is more marked among the Kayasthas, the writer-caste of Bengal, which claims a western and Aryan origin. It reaches its maximum development among the Bengali Brahmans. South of the Vindhya mountains, where the population is chiefly Dravidian, with a comparatively small and ancient mixture of northern blood, the prevalent type is mainly long-headed or medium-headed. The coast-population has been much affected by foreign influences. On the east coast Malayan, Indo-Chinese and even Portuguese settlers have altered the local type. On the west coast, Arab, Persian, African, European, and Jewish immigrants have mingled with local races, and have changed their physiognomy, stature, and character of mind and body. It is still a moot point, which the Mendelists may some day settle for us, whether head-form is a true hereditary race-characteristic, whether the osseous structure of the body generally is not a result of climate, food and other such circumstances of environment. Yet the shape of the head as shown by average measurements does mark off races of men which are separated by other differences than those of habitat. They do correspond to those vaguer yet unmistakeable characteristics which enable us to tell one race from another. The Mongolian, even when he settles in the plains of Assam, Bengal, or Burma and takes to a diet of rice and fish, keeps his round head and his smooth hairless face. The Aryan of the north-west has a markedly long head, which, in his case, goes with a fair complexion and luxuriant beard. The Dravidian, darkest of Indian races, with a tendency to crinkly or curly hair, has also a long or medium head. The mixed races of Bengal have, it is not surprising to find, medium heads, which tend in the upper castes to become broad.

Another significant index to race is the measurement of the nose. The results of nose-measurements roughly divide the peoples of India into three classes—those having narrow or fine noses (leptorrhine), in which the width is less than 70 per cent. of the height; those having medium noses (mesorrhine), with an average index of from 70 to 85; and broad-nosed (platyrrhine) people, the width of whose noses exceed 85 per cent. Here we get a physical means of distinguishing between the long-headed people of north-western India, fair and stalwart, and the almost equally long-headed dusky folk of the south. For the average nose of southern India, in Madras, the Central Provinces, and Chota Nagpore, is broad. In the Panjab and Baluchistan we get fine noses of what, to us Europeans, seems an aristocratic type. In Afghanistan, noses are so long and hooked as to give the tall and vigorous Afghan a Jewish aspect. In the rest of India, and especially down the west coast, noses are of medium type. A still more interesting discovery is the fact that anywhere outside the Aryan tracts of the north-west, the broad nose is a distinct sign of aboriginal blood. In Bengal, for instance, the lower castes have broad noses. The priestly and writer castes, for all their broad heads, have fine noses, which support their claim to a western origin. Roughly speaking, the broad nose goes with primitive forms of social organisation, with totemistic exogamous clans. Finer noses are usually associated with communities of a more modern type; and above these again come social units, castes and tribes, which claim descent from eponymous saints and heroes.

A third physical measurement enables us to effect a further sorting out of Indian races. What is called the "flatness" of the Mongolian face is plain to the most careless observer. This is due chiefly to the formation of the cheekbone, and its relation to the socket of the eye and the root of the nose. This can be measured and expressed in figures, with the result that the Mongoloid people of the north-east and the Himalayan region can be definitely distinguished from the broad-headed races of Baluchistan, Bombay, and Coorg.

Finally, it is possible to arrive at the average stature of various Indian races and communities. The tallest races are found in the north-west, in Baluchistan, the Panjab and Rajputana. A progressive diminution is seen as we go down the valley of the Ganges, until we find very short folk among the Assam hill tribes. The Dravidians of the south are shorter than the Aryans of the north. The smallest Indian tribe is that of the Negritos of the Andaman Islands, whose average height is only 4 feet 10½ inches.

From a careful comparison of these measurements, Sir Herbert Risley arrived at the classification of Indian humanity, which, for the moment, is the accepted division, into seven main physical types. Beginning with the north-western frontier, these are as follows:—

(1) The Turko-Iranian type, which comprises the Baloches, Brahuis and Afghans of Baluchistan and the north-west Frontier Province. These are probably the result of a fusion of Turki and Persian blood, and are all Muhammadans. The general aspect is wholly different from that of other Indian races, and no one who has ever seen an Afghan or Baloch, with his long Jewish nose and plentiful hair and beard, can ever confuse this type with any other. In temperament also these men of the border differ from other Indians. They are a fierce and warlike race, engaged in constant blood-feuds with one another.

(2) The Indo-Aryan type, with its home in the Panjab, Rajputana and Kashmir, has as its most conspicuous members the Rajputs, Khattris and Jats. These, in all but colour (and even in colour they are hardly more dusky than the races round the Mediterranean) closely resemble the well-bred European in type. In stature they are tall, their complexion is fair; "eyes dark; hair on face plentiful; head long; nose narrow and prominent, but not specially long." One significant peculiarity of this group is that there is little difference in physical character between the upper and lower classes. This, as we shall presently see, is what we should expect from what is known of the history of these peoples. The upper social ranks probably represent the blood, but little diluted with indigenous mixture, of the Aryan immigrants. Even in the lower classes, the typical Aryan characteristics are now so prominent that any indigenous strain that exists is no longer noticeable in average measurements. Only in height, a quality especially sensitive to differences of food and sanitation, are the lower castes inferior. Here we get a remarkable modern instance of transformation of type. The preaching of the Sikh reformers, involving a change of food and the inculcation of martial discipline and fervour, has converted the despised scavenging Chuhra into the soldierly Mazhabi, once a redoubtable foe of the English, and now one of the finest soldiers in the British army.

(3) The Scytho-Dravidian type, including the Maratha Brahmans, the Kunbis, and the Coorgs of western India. These peoples differ from the Turko-Iranian races in being shorter, in having longer heads, higher noses, and flatter faces.

(4) The Aryo-Dravidian or Hindostani type, which exists in the United Provinces, in parts of Rajputana, and in Bihar. This type appears to be due to a mixture of Indo-Aryan and Dravidian strains. The higher classes resemble Indo-Aryans, the lower have a distinctly Dravidian aspect. Yet, even to the eye, they form a type apart and are easily recognised. In this type, the average nose-index corresponds exactly to social status. The noses grow broader as we go downwards in the social scale.

(5) The very interesting Mongolo-Dravidian or Bengali type which is found in Bengal and Orissa. Here Aryan influences may still be detected in the upper classes, but there has been extensive mingling with Tibeto-Burman and Dravidian peoples, and other aboriginal inhabitants. The main distinguishing feature is the broad head, which is most conspicuous in the upper classes. It is shared equally by the Bengali Brahman, who claims a western origin, and the Chittagong Mag, whose Tibeto-Burman origin is not denied. The Brahman, on the other hand, inherits a fine and narrow nose, which may very well be due to Indo-Aryan ancestry. Recent investigations tend to show that Buddhism survived till a comparatively recent date in Bengal. Hence, no doubt, a temporary disregard of caste restrictions and a freer mixture with local strains.

(6) The Mongoloid type of the Himalayas, Nepal, Assam, and Burma. "The head is broad: complexion dark, with a yellowish tinge; hair on face scanty; stature short or below average; nose fine to broad; face characteristically flat; eyelids often oblique." Here we have races which, if somewhat dark, correspond to the ideas most of us entertain about the external aspect and temperament of the Siamese or Japanese. In intellectual ability, and what we may call the artistic faculty, they are inferior to the Bengali. Most Europeans, however (or is it, therefore?) find them among the most congenial of Indian races. They are social, good-natured, straightforward people. In the western Himalayas, there has been intermixture with Aryan invaders, as in the Kangra Valley and Nepal, and the ruling dynasties claim Rajput origin, for the Indo-Aryans loved to settle in the cool hills, much as the Anglo-Indian does to this day. But on the mountainous frontiers of North-East Bengal and Assam, the Mongoloid peoples have remained undisturbed till our own time. Linguistically, this group is peculiarly interesting, since they speak many tongues, many of which still remain to be recorded and studied by European scholars.

(7) The Dravidian type, which extends from Ceylon to the valley of the Ganges and covers all South-Eastern India. It is found in Madras, Hyderabad, the Central Provinces, most of Central India, and Chota Nagpore. Its purest representatives dwell on the Malabar coast and in Chota Nagpore. Here we have probably the original inhabitants of India, now modified in some degree by an infiltration of Aryan, Scythian and Mongoloid elements. "The stature is short or below mean; the complexion very dark, approaching black; hair plentiful, with an occasional tendency to curl; eyes dark; head long; nose very broad, sometimes depressed at the root, but not so as to make the face appear flat."

Plate II

Kayasthas—the writer caste
(Mirzapur district)

It must, of course, be understood, that these types and the names allotted to them merely show that in certain areas the average characteristics of the peoples dwelling there can be sufficiently separated to be recognisable not only by eye but by the callipers of the anthropologist. The names, it will be noticed, in some cases, imply theories as to the origin of the races thus grouped together. These theories are partly based on measurements, partly on tradition, partly on linguistic considerations. It remains for me to state, very rapidly, what these theories are.

That the Dravidians are the oldest race in India is rendered prim facie probable by the fact that they inhabit the southernmost part of the peninsula, between races who can with some certainty be called invaders—and the deep sea. There is a remarkable uniformity of physical characteristics among the lower specimens of this type. They have in common an animistic religion, their distinctive language, their peculiar stone monuments, and a primitive system of totemism. They do not resemble Europeans on the one hand, or the races of the Far East on the other. Until proof to the contrary is forthcoming they may well be regarded as the autochthones of India.

There is more room for difference of opinion as to the origins of the brilliant and highly civilised Indo-Aryans of the Panjab and Rajputana. As I have said before, we have here a population closely resembling that of modern Europe in many respects. I might have added that it still more closely resembles the Europe of the Roman empire. Nowhere else in Hindu India does caste sit so lightly, or approach so nearly to the social classes of Europe. Though there are rules, or rather customs, forbidding intermarriage between different castes, yet these are mitigated by the custom, not unknown to ourselves, of hypergamy. This simply means that a man may take a wife from a lower caste, but will not give his daughters to men of that caste. The result is a uniformity of physical type found nowhere else in India. Moreover these people speak a language of the Indo-European family, and have many words and idioms in common with ourselves. The present theory of their origin is simply that they are in the bulk immigrants into India, immigrants who came into the land from the north-west with their herds and families, as the Jews entered into and possessed Palestine.

One chief objection to this theory is that the lands through which they must have passed are in no way fitted to be an officina gentium, being now dry, barren, and all but deserted. But abundant indications remain to show that the climate of South-Eastern Persia and the tracts to the north has changed within comparatively recent times. The relics of crowded populations and ancient civilisations abound in regions now sandy desert, and there is evidence in the tales told by Greek and Chinese travellers that the Panjab itself, most of it comparatively arid, was once well wooded. The theory then is that the homogeneous and handsome population of the Panjab and Rajputana represents the almost pure descendants of Aryan settlers, who carried the Indo-European languages now prevailing over Northern India, just as our own emigrants took the English language to America.

But we have also to account for the Aryo-Dravidians who inhabit the sacred "midland" country of Hinduism, and here we have Dr Hoernle's now famous theory, remarkably confirmed by the researches of Sir George Grierson's Linguistic Survey. This theory supposes that a second swarm of Aryan-speaking people, perhaps driven forward by the change of climate in central Asia, entered India through the high and difficult passes of Gilgit and Chitral, and established themselves in the fertile plains between the Ganges and the Jumna. They followed a route which made it impossible for their women to accompany them. They took to themselves wives from the daughters of dusky Dravidian aborigines. Here, by contact with a different, and in their sentiment, inferior race, caste came into being. Here most of the Vedic hymns were composed. Here, by a blending of imported and indigenous religious ideals, the ritual and usages of Hindu religion came into being, to spread in altered forms east and west and south. The necessity for this second hypothesis is twofold. It accounts for the marked ethnical barrier which separates western from eastern Hindustan. Elsewhere the various types melt imperceptibly into one another. Here alone is a definite racial border line. Again, the theory accounts for the fact that the Vedic hymns contain no description whatever of the earlier Aryan migration, and for the fact that the inhabitants of the middle land always felt a dislike for the early immigrants as men of low culture and barbarous manners. For the present, at all events, and perhaps for all time, Dr Hoernle's ingenious theory holds the field.

No special theory is required to account for the physical and mental qualities of the Mongolo-Dravidians of Bengal. No doubt the original population was Dravidian with a strong intermixture of Tibeto-Burmese blood, especially in the east and north-east. But the Hindu religion, developed in the sacred Midlands round Benares, spread to Bengal, bringing with it the Indo-European speech which in medieval times became the copious and supple Bengali tongue. From the west too came what we in Europe would call the gentry, the priestly and professional castes. These have acquired most of the local physical characters, dusky skin, low stature, round heads. But in nearly all cases, the fineness and sharp outline of the nose shows their aristocratic origin, and in some instances a Bengali Brahman has all the physical distinction of a western priest or sage.

When we turn to the Scytho-Dravidian group we have again to fall back on records of ancient invasions from the north. Ancient some of them were, but far less ancient than the settlement of the Aryans in the north-west. The Sakas have provided India with one of its many chronological eras; they founded dynasties which have left coins behind them, they have left vague but widely spread traditions. They were what we Europeans call Scythians. They were known to the Persians, the Parthians, and the Chinese. Their original home seems to have been in the south of China, a land of pre-eminently round-headed races. We know that they established their dominion over portions of the Panjab, Sind, Gujarat, Rajputana and Central India. If they have left traces of their settlement on their descendants we may reasonably expect to find round-headed races and tribes in regions mostly surrounded by long-headed peoples. Such a zone of broad-headed people does in fact extend from the western Panjab right through the Deccan, till it finally ends in Coorg. Sir H. H. Risley's theory is that the Scythians first occupied the great grazing country of the western Panjab, and finding their progress eastwards blocked by the Indo-Aryans, turned southwards, mingled with the Dravidians, and became the ancestors of the warlike Maratha race. Such an origin forms a tempting explanation of the well-known predatory habits of the Maratha hordes, and of their frequent raids all over the peninsula under the decaying administration of the later Mogul Emperors. It is an interesting and fascinating speculation, since it accounts not only for the physical aspect of the Marathas but for their characteristic political genius, for their wide-ranging forays, their guerilla warfare, their unscrupulous dealings, their inveterate love of intrigue, their clannish habits.

I must here boldly borrow Sir H. H. Risley's summary of the historical record of Scythian invasions into India, since that is the main justification for his theory. "In the time of the Achaemenian kings of Persia," he says, "the Scythians, who were known to the Chinese as Sse, occupied the regions lying between the lower course of the Sillis or Jaxartes and Lake Balkash. The fragments of early Scythian history which may be collected from classical writers are supplemented by the Chinese annals, which tell us how the Sse, originally located in southern China, occupied Sogdiana and Trans-oxiana at the time of the establishment of the Graeco-Bactrian monarchy. Dislodged from these regions by the Yueh-chi, who had themselves been put to flight by the Huns, the Sse invaded Bactriana, an enterprise in which they were frequently allied with the Parthians. To this circumstance, Ujfalvy says may be due the resemblance which exists between the Scythian coins of India and those of the Parthian kings. At a later period, the Yueh-chi made a further advance, and drove the Sse or Sakas out of Bactriana, whereupon the latter crossed the Paropamisus and took possession of the country called after them Sakastan, comprising Segistan, Arachosia, and Drangiana. But they were left in possession only for a hundred years, for about 25 B.C. the Yueh-chi disturbed them afresh. A body of Scythians then emigrated eastwards, and founded a kingdom in the western portion of the Panjab. The route they followed in their advance upon India is uncertain; but to a people of their habits it would seem that a march through Baluchistan would have presented no serious difficulties.

"The Yueh-chi, afterwards known as the Tokhari, were a power in Central Asia and the north-west of India for more than five centuries, from 130 B.C. The Hindus called them Sakas and Turushkas, but their kings seem to have known no other dynastic title than that of Kushan. The Chinese annals tell us how Kitolo, chief of the Little Kushans, whose name is identified with the Kidara of the coins, giving way before the incursion of the Ephthalites, crossed the Paropamisus, and founded, in the year 425 of our era, the kingdom of Gandhara, of which, in the time of his son, Peshawar became the capital. About the same time, the Ephthalites or Ye-tha-i-li-to of the Chinese annals, driven out of their territory by the Yuan-yuan, started westward, and overran in succession Sogdiana, Khwarizan (Khiva), Bactriana, and finally the north-west portion of India. Their movements reached India in the reign of Skanda Gupta (452-80) and brought about the disruption of the Gupta empire. The Ephthalites were known in India as Huns. The leader of the invasion of India, who succeeded in snatching Gandhara from the Kushans and established his capital at Sakala, is called by the Chinese Laelih, and inscriptions enable us to identify him with the original Lakhan Udayaditya of the coins. His son Toramana (490-515) took possession of Gujarat, Rajputana, and part of the Ganges valley, and in this way the Huns acquired a portion of the ancient Gupta kingdom. Toramana's successor, Mihirakula (515-44), eventually succumbed to the combined attack of the Hindu princes of Malwa and Magadha."

I now come to the ethnography as distinguished from the ethnology of India. Of anthropometry and the lessons to be learnt from it, I have no personal experience, and have had to borrow my materials at second-hand. But with the great system of caste, its workings, its manifold ramifications, everyone who has lived in India has come into more or less close contact. How important caste is in the social life of the country may be easily inferred from this little fact. I once asked the late Navin Chandra Sen, then the most popular of Bengali poets, if he would attempt a definition of what a Hindu is. After many suggestions, all of which had to be abandoned on closer examination, the poet came to the conclusion that a Hindu is (1) one who is born in India of Indian parents on both sides, and (2) accepts and obeys the rules of caste. Hinduism is, roughly speaking, the religion of the Aryo-Dravidians, the upper and fairer classes among whom regarded the aborigines, matrimonially, much as white Americans regard their negro fellow citizens. It has spread over nearly the whole of India and is still spreading, usually but not always, carrying with it one of the Indo-European languages of India. It is the religion and social system of races and classes which consider themselves intrinsically superior, and practise a traditional kind of eugenics, of race preservation. Humbler or more barbarous races are admitted on various conditions into caste, sometimes into higher, sometimes into lower positions. The process is one of that kind of "legal fiction" with which students of Roman law are familiar. It is a process of unification and, at the same time, of social segregation. I have already alluded to the suggestion that caste-divisions are horizontal, as it were, compared with the geographical divisions of races. But it is always dangerous to make general statements about three hundred millions of people scattered over so large an area as India. There are Brahmans in every part of India, and these usually trace their origin back to the sacred midland where Hinduism came into being. They may be, and probably are, the descendants of the missionaries by whom the religion of the Hindus is, imperceptibly and without open proselytism, spread abroad. Something corresponding to a warrior caste and a caste of scribes is to be found in most provinces, and many of these either claim to be migrants, or have been admitted by adoption into the privileges of warrior or writer blood.

But there are many castes which are purely local, even in name, and are not found elsewhere than in the places where they were admitted into the Hindu community. Many closely printed pages in the Census Reports of each province and state enumerate and describe the thousands of castes revealed by the numbering of the people. It is, of course, only possible to give a very vague and general idea of some of the classes into which the castes of India may conveniently be divided.

I am tempted here to borrow Sir Herbert Risley's definition of caste. But it is a highly abstract definition, and one that cannot be easily carried in the head, even by those who have a practical and familiar acquaintance with members of Indian castes. Roughly a caste is a group of human beings who may not intermarry, or (usually) eat, with members of any other caste. There are also sub-castes which are also endogamous. Very frequently, especially in the parts of India where caste is already an institution of immemorial antiquity, a caste has allotted to it a profession or occupation.

Before we discuss castes properly so called, it is convenient to speak of the tribes of India, since tribes have a tendency to become castes when they come under the pervasive influence of Hindu social ideas. In the south of India are Dravidian tribes, of which the best example are the tribes of Chota Nagpore. These are divided into a number of exogamous groups or clans, calling themselves by the name of an animal or plant, which may be regarded as their totem. The Khonds of Orissa, who once bore an evil name for their practice of human sacrifices to propitiate the earth-goddess, are divided into fifty gochis or exogamous clans, each of which bears the name of a village, and believes itself to be descended from a common ancestor. These gochis are the nearest known approach to the local exogamous tribe which Mr McLennan and the French sociologists believe to be the earliest form of human society.

The Mongoloid tribes of Assam are much of the same kind, but in many cases, as among the head-hunting Nagas, live at perpetual warfare with one another. In such cases they usually capture their wives in war. It is interesting to note that when population grows too dense for the profitable pursuit of the chase, their principal means of livelihood, such a tribe breaks up into two or more "villages," which immediately begin waging war with one another, which is quite what a French sociologist would expect them to do. I can tell of a case within my own experience in which the headman of a parent village invited the chief of a colony village (his own nephew) to a feast and palaver with his young warriors. The guests were all treacherously put to the sword, as a means of acquiring heads and concubines. I could not get the headman to see that he had been guilty of an atrocious crime. For him, it was lawful strategy. And indeed Naga warfare is merely a series of artfully planned ambushes in which not a few of our own officers perished before we undertook the direct administration of the Naga Hills. Sir H. Risley remarks of this group of tribes that "no very clear traces of totemism have been discovered among them." Subsequent enquiries, however, show that totemistic clans do exist in some of the Assam tribes.

Plate III

Dharkars
(Mirzapur district)

Of the Turko-Iranian tribes of the north-western frontier I need not speak at any length, since these tribes are all sturdy followers of the Prophet, and save that they are under British rule can hardly be said to belong to India at all. There is no likelihood that they will ever be received into the tolerant bosom of Hinduism, since, to the Indian proper, the Baloch and the Afghan are disagreeable and swaggering caterans, who have an innate scorn for the typical Hindu hierarchy of caste. Among these tribes it is martial ability and valour that win a man consideration and wives.

Let us now turn to caste properly so called, the traditional social divisions of the Hindus. And first it is necessary to say something of the ancient Hindu theory of what caste is, and how it came into existence.

As with the Hebrews, the religious literature of India contains a vast mass of what can only be called law, and perhaps, the most famous of Indian law books is the Institutes of Manu, a compilation of rules relating to magic, religion, law, custom, ritual and metaphysics. Even to this day, these branches of speculation and enquiry, so distinct to western imaginations, are apt to be confused together as a result of the pantheistic feeling which pervades Hinduism. The Institutes is a comparatively modern book, but it repeats ideas which are found in a more or less explicit form in early authorities[1]. In this book we are told that in the beginning of things the Pan-theos who "contains all created things and is inconceivable" produced by effort of thought a golden egg, from which he himself was born as Brahma, the creator of the known universe. From his mouth, his arms, his thighs, and his feet respectively he created the four great leading castes, the Brahman, the Kshatriya, the VaiÇya, and the Sudra. These were, briefly, the priests, the warriors and gentlefolk, the traders, and the servile classes of human society. The other castes were gradually formed, the theory states, by intermarriages between these. The three higher castes were allowed to take wives from lower castes. When the caste of the mother was next below that of the father, the child took the caste of his mother and no new caste was formed. But where the difference of condition was greater than this, new castes were formed, lower than those of either parent. Some discrepancies of rank produced unions which were regarded as peculiarly offensive to human feelings and as tantamount to incestuous intercourse. These resulted in very degraded castes. Where the father married beneath him, the marriage was described as anuloma or "with the hair." When a woman was guilty of a mÉsalliance, the marriage was called pratiloma or "against the hair." The most disgraceful union of this kind was that between a Brahman woman and a Sudra man, the resulting offspring being relegated to the caste of Chandal. The unfortunate Chandal is described as "that lowest of mortals," and is condemned, as Sir H. Risley says, to live outside the village, to clothe himself in the garments of the dead, to eat from broken dishes, to execute criminals, and to carry out the corpses of friendless men.

The most superficial acquaintance with existing caste divisions shows that this theory is not so much a hypothesis as a fanciful fiction. In eastern Bengal, for instance, the Chandal is evidently a Mongoloid aboriginal, with a considerable strain of Dravidian and perhaps even of Aryan blood. Yet the fiction shows plainly enough the estimation in which one of the numerically largest divisions of local society is held. Some thirty years ago, when I was a young magistrate, a comely Chandal girl appeared before me, her face streaming with blood from a scalp wound. She asserted gravely that a Sudra of higher caste had struck her on the head with a stick, because he had found her reading a book as she sat in the doorway of her father's cottage. I was disinclined to believe this story, but her assailant was promptly sent for, and being brought straight to me, admitted the truth of the charge, and seemed surprised at my indignation at a cowardly assault.

As an attempt to account for the origin and explain the nature of caste the theory of Manu is obviously a failure. But it contains a picture of the early castes. It is also interesting because the idea of four original varnas or "colours" of men may have been borrowed from the old Persian social organisation. The early scriptures, the Vedas, show that this conception of four original castes was not brought to India by Aryan immigrants. But when caste came into being as a result of the contact of Aryan settlers with Dravidian aborigines, this mythological explanation, which gave such conspicuous eminence to priests and warriors, an eminence already conceded to them on account of the importance of their functions, was readily accepted as a convincing explanation of the hereditary differences between men in society, a difference not merely of function, but of colour, aspect, gesture, speech, breeding, and intelligence. It is necessary to mention this theory, however briefly, since it still holds ground, except among those Indians who have had a European education and even among them has the interest of early and sacred associations which, in Europe, belongs to the cosmological speculations of the book of Genesis.

What, next, are castes as they appear to the eye of the European ethnologist, free from preconceived prejudice, and only anxious to come as near the truth as is possible in his dealings with ancient institutions round which has gathered a vast mass of venerable superstition and religious speculation? In the first place, castes are often still recognisably tribes. Sometimes the leading men of an aboriginal tribe will acquire sufficient wealth and social consideration to wish to obtain the stamp of recognition as reputable Hindus. They will call themselves, for example, and induce their neighbours and the priests of these to call them, Rajputs. They may not at first succeed in intermarrying with true hereditary Rajputs, but in time they will be just Rajputs like any other Rajputs. Or, again, a number of non-Hindus, animists, will join one of the many Hindu sects or fraternities and will intermarry with Vaishnavas, Lingayats, Ramayats, or other devotees of some favourite deity. Or again, a whole tribe or a considerable portion of a tribe, usually one of some political importance, will enter Hinduism by means of some plausible fiction. The instance quoted by Sir H. Risley is that of the Koches of north-eastern Bengal. These people are Tibeto-Burmans and until recent times spoke a dialect of the agglutinative Bodo language. They now call themselves Rajbansis, "of royal birth," or Bhanga Kshatriyas, "broken warriors," names which enable them to claim an origin from the traditional dispersion of the Aryan warrior caste by the hero Parasu Rama, "Rama of the battle axe." They claim descent from the epic monarch Dasarath, father of Rama, have their own Braahmins, and have begun to adopt the Brahminical system of exogamous gotras. But, as Sir H. Risley remarks, they are in a transitional state, since they have all hit upon the same gotra, and are therefore compelled to marry within it, except in the rare instances in which they contract unions with Bengali women.

A still more interesting, because more recent, instance of this sort is that of the Meithei, now known to Hindus as Manipuris. In the Mahabhaarata is told the tale of how the hero Arjuna wandered from his brethren into Southern and Eastern India, and, among other adventures, met (as Æneas with Dido) with Chitrangada, the fair daughter of the King of Manipur, somewhere near the eastern coast. Some 150 years ago, the then king of the beautiful valley of Imphal, between Assam and Burma, was thinking of becoming a Muhammadan, by way of courting the favour of the Muhammadan rulers of Bengal. But Hindu priests persuaded him that a better way of linking his fortunes with those of India, rather than with Ava (with whose royal family his dynasty had usually intermarried), was by becoming Hindu with all his people. Imphal was identified with Manipur, and many of the Meithei race became Vishnuvite Hindus with their ruler, though they retain their primitive Tibeto-Burman language. I may mention a little personal reminiscence to show how completely the change by fictitious adoption was accepted in Bengal. In 1891, my old friend and chief, Mr Quinton, with all his staff, was treacherously murdered at Manipur. Subsequently when I was magistrate of Chittagong, I found that my head clerk, an extremely mild and intelligent Bengali Kayastha, had celebrated the easily suppressed mutiny at Manipur by writing a drama based on the ancient legend of Arjuna's amours with Chitrangada!

Sometimes an aboriginal tribe will become a Hindu caste without losing its old tribal designation. They will worship Hindu gods without daring wholly to neglect tribal deities, which, as might perhaps be expected, are left chiefly to the women of the tribe. Such a tribe will rapidly assimilate itself to the beliefs and practices of Hindu neighbours, and finally only its name and (except in case of occasional intermarriage with other castes) its physical aspect will remain to testify to its origin.

Castes are at present classified as follows:

(1) What Sir H. Risley calls the tribal type, instances of which have been given above. Such tribal castes abound in all parts of India. It is not improbable that the great Sudra division of Hindu tradition was originally the whole mass of Dravidian aboriginals as they came into contact with Aryan immigrants, and were conceded a subordinate place in their social system. It would be useless to give a list of the names of such castes, but I cannot refrain from mentioning the excellent Doms of the Assam Valley, whose name unfortunately associates them with very different people in India proper. They are obviously of Tibeto-Burman origin, and deserve closer study than they receive. Their long thatched places of worship, true synagogues for meeting together and curiously unlike the tiny cellÆ of Hindu temples, are among the most conspicuous features of Assam villages. They have no idols, and place a puthi, a holy book, on what may pass for the village altar. They are vaguely Hinduised, but will humbly declare "ami hindu na hÔ," "we are not Hindu folk." Yet they are well on their way towards acceptance into caste, and have already a strong infusion of Hindu blood.

Other border races, though they are still too savage and independent to become Hindu, are marked down for absorption. Such, for instance, are the Daflas of the northern border of Assam, cousins of the Abors to whom attention has been drawn by recent events. The Daflas are still frankly animistic; their love of strong spirits and other intoxicants, their addiction to their favourite diet of roast pork, their extremely uncleanly habits and barbarous speech, all make them very offensive to the gentle vegetarian Hindus their neighbours. But it happens that the tribal costume closely resembles the traditional dress of Mahadeva, the Destroyer, the most active and formidable member of the Hindu Trinity, and already some Hindus speak of these genial Highlanders as Siva-bansa, as "of Siva's race." Many other examples, with interesting details of fictional methods, will be found in Mr E. A. Gait's admirable History of Assam.

(2) The functional or occupational type of caste. This is the form of caste best known to Europeans, because, since the first European missionaries and traders visited those parts of India where the caste system has had the longest opportunity to evolve, they came most into contact with this, which is probably the oldest and most elaborated form of caste. The Hindu theory of caste encouraged the adoption of special occupations, and now the evolution has proceeded so far that change of occupation may usually result in a change of caste. A remarkable instance of this is found in the Marathi districts of the Central Provinces. Here is a separate and newly formed caste of village servants called Garpagari, "hail-averters," whose business it is to protect the village crops from hailstorms. Shepherds who take to tillage break away from their pastoral brethren, and so on. Even those who retain their traditional occupations are wont to adopt more seemly-sounding names than those that belong to their trade. I have known barbers who called themselves Chandra-vaidyas[2], which is a promotion more subtle than a mere ascent to the status of "hair-dresser," and washermen who have followed suit by dubbing themselves Sukla-vaidya, a word of which "white-worker" is a crude but sufficiently suggestive translation.

(3) The sectarian type is a singularly interesting example of the strong social influence of Hindu sentiment. Nearly all new Hindu sects begin by renouncing caste in the enthusiastic following of some single deity, some new explanation of the mysteries of life, and love, and death. These sects are usually the followers of some reforming theorist, whose leadership is apt to become hereditary. Such sects almost always believe that all men are equal, or at all events, that all who accept their doctrines are equal. One of my most interesting recollections is of a now distant interview with a buxom middle-aged lady, the hereditary leader of the Karta-bhajas of Central Bengal. She sat unveiled, and was accessible to all who, like myself, were interested in the community over which she exercised a firm but good-natured control. It is a picturesque detail that her chosen seat when receiving visitors was an ancient European four-poster bedstead. Her followers (and revenues) were growing rapidly, increased chiefly by the democratic instinct which, even in India, revolts against social prestige. But it would seem that when such a sect grows and spreads, the old separatist ideas reassert themselves, and the sect breaks up into smaller endogamous communities, whose status depends on the original position of the members in Hinduism. The most remarkable instance of this kind is furnished by the great Lingayat caste of Bombay, which contains over two and a half millions of members. In the twelfth century the Lingayats were a sect who believed in the equality of all men. In Mr P. J. Mead's Bombay Census Report for 1911 is a very interesting account of the present condition of the Lingayats, an account which shows how the scholar, the linguist, and the administrator can work together to find materials for the anthropologist. Dr Fleet's examination of ancient inscriptions has thrown much light on the origin of the sect, but the author of the Report holds that there may be some reason to think that the sect is much older than is commonly supposed. In any case, they are already divided into three great groups, comprising many subdivisions.

(4) Castes formed by crossing come aptly to show that there was some basis for Manu's theory of caste after all. Castes, nowadays, increase by fission, by throwing off sub-castes, and one species of these sub-castes is created by mixed marriages. This tendency, curiously enough, is most evident in Dravidian tribes, such as the Mundas, which are not yet wholly Hinduised, but have been affected by Hindu example. So far as I know, these mixed castes do not occur among the Mongoloid peoples, and I have come across cases where a member of an aboriginal tribe has been accepted into the caste of a Hindu girl he has married. In one case, within my own experience, the bridegroom had begun as an animist, had become Christian, and finally entered by marriage into the quite respectable Koch caste. One interesting caste in Bengal, that of the Shagirdpeshas, owes its origin to concubinage with the so-called slaves, the women of tenants surrounding a homestead who pay their rent in service. This, it will be observed, is a caste of illegitimacy, in which the relationship between the legitimate and illegitimate children of a man of good caste is recognised, but the two are not allowed to eat together. The classical instance of a mixed caste is the Khas of Nepal, said to be the result of very ancient intermarriages between Rajput or Brahman immigrants and the Mongolian "daughters of men."

Plate IV

Banjara women
(Mirzapur district)

(5) Castes of the national type. This somewhat daring title we owe to the great authority of Sir H. Risley. As one instance, he mentions the Newars, a Mongoloid people, who were once the ruling race in Nepal, till the Gurkha invasion in 1769, and have now become a caste. Other instances might be found on the north-eastern frontier. But the people Sir Herbert Risley had in mind when he invented this term was undoubtedly the remarkable Maratha race, once the most daring warriors and freebooters in India, and now the rivals of the Bengalis in intellectual ability, and probably more than their equals in political sagacity. Sir Ramkrishna Gopal Bhandarkar is our authority for the statement that the Rattas were a tribe who held political supremacy in the Deccan from the earliest days. In time they became Maha-rattas, "Great Rattas," and the land in which they lived was called Maharattha, which, by a common linguistic habit of mankind, was Sanskritised into Maha-rashtra. Their marriage customs show marked traces of totemistic institutions. An extremely interesting account of the present condition of this warlike and enterprising race will be found at pp. 289, 290 of the Bombay Census Report for 1911. It neither supports nor discourages Sir H. Risley's ingenious theory of the Scythic origin of the Marathas, which is at least a theory which recognises the respect in which our ancestors held their martial prowess and talents[3].

(6) Castes formed by migration. These are new castes which serve to enforce the warning against a too ready acceptance of the definition of caste as a "horizontal" division of humanity. It is a method of forming new communities of Hindus which is very easily intelligible to us, seeing that our own race is split into sections only differing from castes in not being strictly endogamous, such as Anglo-Indians, Australians, New Zealanders, and so forth. Members leave home and settle among strangers. They are assumed to have formed foreign habits, eaten strange food, worshipped alien gods, and have a difficulty—an expensive difficulty—in finding wives in the parent caste. After a time they marry only among themselves, become a sub-caste, and are often known by some territorial name, Barendra, Rarhi, or what not. Such seemingly are the remarkable Nambudri Brahmans of Malabar, and the Rarhi Brahmans of Bengal. Sometimes change of habitat brings about loss of rank, sometimes promotion. These are matters on which the Census Reports now being published are full of interesting details. But they are matters which are not easily summarised. No doubt Mr Gait's Report on the combined results of Census operations in India will show the progress of castes of this type during the last ten years.

(7) Castes formed by changes of custom. This is a fruitful cause of new divisions of Hindu society. It is, for the moment, more than usually operative, owing to the spread of education, and often represents a difference of social opinion which corresponds, more or less closely, to Conservative and Radical ideas among ourselves. It evidently was always a cause of fissiparous tendencies. The most notable instance is the distinction between Jats and Rajputs, both apparently sprung from the same stock, but separated socially, amongst other causes, by the fact that the former practise and the latter abjure infant marriages.

This is a very rapid and highly summarised account of the races and castes of India. There are many obvious omissions. Nothing has been said of the Sikhs, little or nothing about the numerous races of the north-eastern frontier. But enough has been said to give a fair general impression of what the physical characters of the Indian peoples are, and what kind of institution caste is in its practical working. More might have been said about totemistic clans, but on this subject those who would pursue their studies further have only to turn to Dr J. G. Frazer's work on the subject. In the next chapter, I have to borrow my materials from Sir G. A. Grierson, and show how the peoples of India are divided by differences of language. On the whole, those linguistic divisions correspond with remarkable accuracy to the orographical and climatic structure of the country and the racial divisions which we owe to the learning and ingenuity of Sir H. H. Risley. Where there are great open plains, watered and fertilised by mighty rivers, we get large populations speaking the great literary languages of India. In the rugged recesses of the mountains we find small communities, divided from one another by physical obstacles which have produced rigid local patriotisms and enmities, and a wonderful variety of savage speeches. The linguist has usually worked independently of the ethnologist, and has come to his own unprejudiced conclusions. It is interesting to find how closely the results of their separate enquiries agree.

Postscript.

Sir H. H. Risley's theory as to the Scythian origin of the Marathas has not passed unquestioned, and those who wish to see a brief and clear account of the latest theories on the subject should read Mr Crooke's paper on "Rajputs and Marathas" in Vol. XL. (January—June, 1910) of the Journal of the Royal Anthropological Institute. Mr Crooke, who gives copious references to the latest literature on the subject, holds that "the theory that a Hun or Scythian element is to be traced in the population of the Deccan is inconsistent with the facts of tribal history, so far as they can now be ascertained." Mr Crooke thinks that the anthropometrical facts can be explained otherwise than by Saka invasion and an infusion of Scythian blood. "The presence of a brachycephalic strain," he says, "in Southern and Western India need not necessarily imply a Mongoloid invasion from Central Asia. The western coast was always open to the entry of foreign races. Intercourse with the Persian Gulf existed from a very early period, and Mongoloid Akkads or the short-headed races from Baluchistan may have made their way along the coast or by sea into Southern and Western India. But it is more probable that this strain reached India in prehistoric times, and that the present population is the result of the secular intermingling of various race types, rather than of events within the historical period." Mr Crooke's view is supported by the recently issued Census Report of the Bombay Presidency, which says, "the term Maratha is derived by some from two Sanskrit words, maha, 'great,' and rathi, 'a warrior.'" According to Sir Ramkrishna Gopal Bhandarkar it is derived from Rattas, a tribe which held political supremacy in the Deccan from the remotest time. "The Rattas called themselves Maha Rattas or Great Rattas, and thus the country in which they lived came to be called Maharattha, the Sanskrit of which is Maha-rashtra."

Indigenous names are frequently Sanskritised, much as we turn French chaussÉe into "causeway." Sometimes the change is so complete that the original cannot be identified. In some cases the alteration is easily recognised. In Northern Bengal, for instance, is the river Ti-sta, a name which belongs to a large group of Tibeto-Burman river names beginning with Ti-, or Di-, such as Ti-pai, Di-bru, Di-kho, Di-sang, etc., etc. Hindus say the name Ti-sta is either a corruption of Sanskrit Tri-srotas, "having three streams," or of T???a, "thirst." Etymology and legend, in fact, give but doubtful guidance to the ethnologist, and the best hope of acquiring some real knowledge of Rajput and Maratha origins lies in the possible discovery of coins and inscriptions in the absence of direct historical records.