INDO-EUROPEANS. The term Indo-European appears to have been invented by the well-known physician and physicist Thomas Young. In an article in the loth volume of the Quarterly Review (Oct. 1813) he uses the word without any remark as to its being a new coinage. The word was devised especially to indi cate a family of languages, but from the beginning it has been found very difficult to keep apart the use of the word as indi cating languages and the application of the term not only to the speakers of the languages at any given period but also to the supposed original speakers of these languages. The effect of this confusion has often been serious. It is obvious that a lan guage may be spoken by persons who have no blood relationship with the general body of speakers of the language, and that this body itself may not be all of one and the same origin. Thus the negro in Liberia or in the Southern States of America who speaks English as his mother tongue may have no drop of English blood in his veins, and even 200 years ago Daniel Defoe was certain that there was no such person as the "true-born Englishman." Ever since it was possible to identify a separate family of lan guages as Indo-European, continual efforts have been made to ascertain who were the original speakers of the language from which the surviving members of the family, or those which are recorded in literature or other monuments, have sprung, where was their home, what were their surroundings and their manner of life. The first of the elaborate statements of these problems was that published by Adolphe Pictet in 1859. The more im portant attempts of the same kind which have appeared since are enumerated in the bibliography at the end of this article. It is hardly necessary to say that the majority of such attempts deal mainly or altogether with the evidence derived from language. This is necessarily so, because, when the question arose, the lin guistic scholars were the first who were ready to carry on the in vestigation; but the problem can be attacked in various other ways, and it is only by an attack converging from all quarters that a certain result will ever be attained.
From language it is clear what were the animals, what were the plants and what were the seasons in the country whence this language was derived. The animals still familiar amongst us were well known ; the sheep, the cow, the horse, the dog and the pig. The names of the goat vary much more from language to language; but it does not necessarily follow from this that the goat was not known until a later time. Animals for which there are no early terms in the Indo-European languages are the lion, the tiger, the elephant, the camel and the ass. It may therefore be safely concluded that these animals did not exist in the original habitat. The lion, at one time, certainly roamed over a much larger area than it does now. According to Herodotus lions were found as far to the west as Macedonia and the prevalence of the representation of the lion in monuments of Asia Minor and Persia is proof to some extent, that the animal was well known. It still survives in Cutch but has lost its mane and is on the way to ex tinction in Asia. As there is no word for the elephant and no representation of it in the more western part of Asia or in Europe it is clear that this animal also was unknown. The same is true of the tiger and the camel and the ass. In Asia the common beast of burden was the ass. The horse, though known in Meso potamia from the time of Hammurabi before 2000 B.C., owes its introduction to the Kassite invasion of 1926 B.e., and was called the mountain ass. Palaeolithic man was able to represent in his drawings two species or varieties of horse. The name of the cow has the same origin in almost all the languages and the same is true of the sheep, the pig, the dog and the mouse ; but in their wild state the same environment is not suitable to all. The horse is an animal of the prairie ; it lives in the open plain and the foal is able from the day of its birth to accompany the mare. The cow, on the other hand, in the wild state lives in countries with open spaces well provided with woods or clumps of trees. When the cow travels afield to find food, the calf is unable to accompany her and has to be hidden in some brake or thicket, as it can walk but feebly and its eyesight is still imperfect. Unless it were so hidden it would fall a prey to animals like wolves or eagles, both of which seem to have been known to the first Indo-Europeans. The grass which suits horses or cows is unsuitable for sheep which prefer the short crisp grass of upland pasture. If, therefore, all these animals were well known to the early Indo-Europeans they must have lived in an area where the landscape was diversified. The country seems to have no connection with the sea ; the more widely spread meanings of the word for the sea are either moor or marsh. That open water was known is shown by the fact that there were words for water birds, the swan, the goose and prob ably the duck, but all of these could find a resting place in very small sheets of water. From what has been said, it will be clear that the Indo-Europeans must originally have lived in the tem perate zone. The bear was known, but which species we cannot tell. It is well known that the name "bruin" meant the brown bear, being cognate with the Lithuanian word beaas, for the Lithu anian word is used as an adjective meaning brown. Other animals known were the beaver and the hare, the word for the beaver being generally a reduplicated form from the same stem as the word for the bear and the hare, meaning originally the grey beast. Its origin was the same as that of the Latin adjective canals from an older cas-nos.
The trees and other plants are not so easy to identify because it is found that when men migrate to a new area, they carry with them the names of plants and, to some extent, also of animals and apply them to plants or animals in the new area which bear some resemblance to the plants or animals they have left behind. Thus in England the oriental plane, so well known in Asia Minor and Persia as the Chinar, grows only with difficulty north of the Trent and in northern Scotland not at all. But the name plane is preserved, notwithstanding, and applied to the sycamore which resembles the plane in being an umbrageous tree and having a leaf somewhat similar. In the same way, the Eng lish in America have given the name of robin to a much larger bird, because it has a red breast. The trees most widely spread in the temperate zone are the willow and the birch. The birch extends from the British Isles to the farthest Himalayas. Some variety of willow is the first tree to grow near streams in a country like Siberia which has not long been freed from surface ice. On the other hand, some of the commoner trees, like the beech, do not penetrate far into the southern peninsulas of Europe. The consequence is that in Greece the word correspond ing to beech in English and to fagus in Latin, means oak or some times possibly chestnut. In the same way in Latin the word cog nate with the English birch is used for the ash, fraxinus. The pine in some of its numerous forms was also certainly known.
More important, perhaps, for the his tory of the race, is the existence of cereals. Within historical times, only two new cereals have been introduced into Europe, rice and maize, though a plant of another kind, buckwheat. was brought in by the Tatars and has received its name because its fruit was ground, and treated as a cereal. Many attempts have been made to connect the Greek oryza with the Sanskrit vrihi, which in modern Persian appears as birinj. But traffic down the Red Sea was mostly with southern India and the simpler explana tion is to connect the Greek word with the Tamil, arishi. The early names for grain are very ambiguous. The word corn connotes in England wheat, in Scotland oats, in Sweden barley, in Germany rye and in America maize. The most widely spread name for a cereal is that which appears in Sanskrit as yava, in Greek as zea or zeid, the latter a derivative of the former, in Lithuanian jawai. It may have meant originally barley, but is also used for spelt, a simple and ancient form of wheat, which, as being such, was long used in religious ritual.
The names for the seasons of the year help us somewhat. There are words widely spread for winter and for snow. The stem pima is found in the name of the Himalayas ("the abode of Snow") and in the Greek klieima with its adjective duskhimos, the Latin hiems and many others. We do not use a word from its root for winter ourselves ; but notwithstanding we have a word borrowed from the Norse and much used by Cumberland shep herds, gimmer, a winterling or sheep one year old. There is less agreement about the other seasons but there is a fairly widely spread word for spring. In many countries summer is simply the hot season. For harvest the languages do not agree very well; the Latin auctumnus is simply the "season of increase," which we have introduced into English as autumn. The word harvest is probably the season of fruit gathering, not of corn harvest, the two being still distinguished in German as herbst, the fruit har vest, and ernte, the reaping time. Snow was certainly known be cause the stem is found in so many languages, but with lofty mountains in the neighbourhood that might well be, even in a hot climate ; from the names of the animals and plants, however, as we have seen, the climate was temperate and if it was a con tinental climate as, from the absence of words for sea,' it appar ently was, the climate might be severe, even if it were tolerably far south. The areas that will satisfy the conditions require a land with a temperate climate, remote from the sea and shut off from other areas, for otherwise it seems impossible that languages with so complicated a grammatical system as the Indo European could have developed, bearing so close a resemblance to one another and on the whole so strongly differentiated from other languages. The only area which will satisfy the conditions postu lated by the languages is the great area in Europe which includes practically the former empire of Austria-Hungary. In this area are found rich corn land, great prairies for the production of horses, pasture for cattle in the plains and for sheep on the mountain slopes, while the mountains themselves contain a large supply of minerals which were later utilised by man. But when the migrations from this centre began, the use of metals had not advanced very far, the only word for metal common to many languages being that which appears in Sanskrit as ayas, in Latin as aes and in English .as ore. The word at the time of the migra tion meant either copper or bronze which sometimes, as in Greek, are not distinguished, the first bronze probably being a natural alloy. From what has been said it is clear that much could be done on this subject by the geographers.
Even when we have found a people using a cer tain vocabulary, it is necessary to discover, if they migrated from their original habitat, in what manner it was possible or probable that they did migrate. It is, for example, a common belief that migration from Europe to Asia or the reverse, was carried on largely by the steppes of Southern Russia, the shores of the Cas pian and Turkistan. In ancient times, however, progress in this direction would not have been so easy as it is to-day. In the Caspian area very large and important changes in the earth's surface have taken place. The eastern end of the Caucasus, for example, has sunk till it is now far below the level of the Caspian. The northern Caspian is steadily being filled up by silt from the Volga. In earlier times, it extended much further north than it does at present, while eastwards its extent was very much greater, including within its area not only the present Sea of Aral and what is now the intervening desert, but also passing still further to the east to a distance which is not yet clearly determined. Scholars often talk lightly also of great hordes passing the Cau casus. This also would be very difficult to achieve. According to early Greek notions the Caucasus was a barrier without openings through it and in fact at the present day there is but one, the Pass of Dariel. Historical migrations, in fact, have taken place much more readily across Asia Minor than across the Steppes. The main movement from Asia into Europe which went north of the Caucasus was in early times that of the Scythians and in mediaeval times that of the Hungarians. That the migration of the Indo Europeans was through Asia Minor is proved by the discovery in 1906-07 of a large mass of records dating from the 15th and i4th centuries B.e., at Boghaz Keui, an important site east of the river Halys. Here were discovered the archives of the ancient Hittite empire, recorded in eight languages of which two have a certain Indo-European element in them. The most important is called Kanisian by Dr. Emil Forrer, who has been very success ful in their decipherment. The grammar of this language, its noun and verb forms, are clearly Indo-European ; but only a small part of its vocabulary is so, the main part coming from an un known tongue. The first records of this language which were dis covered came not from this find at Boghaz Keui, but from two letters discovered many years before, in the archives of Ikhnaton (q.v.), the heretic king of Egypt at Tell-el-Amarna. The mixture of two languages might be compared to what has happened in English where the pronouns, the numerals and the few remaining inflections of noun and verb are Teutonic but the bulk of the vocabulary is of Latin, or of Greek origin. The forms of this language, particularly in the names of certain deities, and of the numerals were so like those of Sanskrit that the first investi gators were of opinion that either the records proceeded from Indians advancing westwards, or that the original home of the Indo-Europeans must once more be carried back to central Asia, where it had been located by some of the early authorities. A difficulty was found in the fact that the existing remains of the ancient languages of Persia—Zend and old Persian—as repre sented in the dialects of the Avesta and the kindred tongue of Darius' inscriptions at Behistun were so much less archaic in sounds than either the forms of these records or of Sanskrit. The explanation, however, is simple. When the speakers of Sanskrit passed over the mountains into a new country, they found them selves faced by powerful tribes more numerous than themselves and speedily realized that if they were not to be absorbed, they must stand aloof from mixture with the native tribes. Hence the beginning of the system of caste in India, for the Sanskrit word for "caste" means colour. The changes found in the language left behind in Persia must have taken place after the migration of the emigrants to India. This is not at all unusual. In the i6th cen tury important changes took place in Spanish, but these changes were not shared by those who had already established Spanish in America. The English settlers of New England carried with them the forms noos for news and Toosday for Tuesday, because the y sound before a had not developed when these settlers left England.
If the original home of the Indo-Europeans is really located in Austria-Hungary, we should expect a considerable number of the Indo-Europeans to have been pile-dwellers. For this country had at one time much more water in it than it has now, when the Platten See, a shallow marshy lake, is all that remains of what was at one time a vast expanse of slowly drying morass. That there were amongst them many pile-dwellers, we can be certain, be cause they carried into Switzerland and into Italy for the western migrations, the same form of habitation, and gradually extended it further to the West.
Anthropology.—What manner of men were these emi grants? In Germany many scholars contend that they were tall men with fair hair and blue eyes and they attribute to the Indo Europeans all the characteristics of the ideal German. For this, however, there is no solid foundation. It is said by these scholars that when Homer talks of xanthos Menelaos, this should be trans lated "yellow haired Menelaos." This translation, however, is not accurate. To a Greek xanthos did not mean blond, but brown, as is clearly shown from the use of the verb, which signifies "to change the colour of meat in roasting," so that the colour would not have been lighter than auburn. In fact, when the Greeks came in contact with German children they did not know how to describe the colour of their hair and said they had hair like old men and described them as polioi, "with grey hair." The anthropologists at one time maintained that the Indo Europeans had long heads as distinct from other people with short heads. The mixture, however, of long and short heads is not a new thing. As Sir Arthur Keith points out (Antiquity of Man, vol. i. p. I 10  ), the antiquary, Dr. R. R. Schmidt, found at Of net, fifty miles south-west of Nuremberg, a large number of skulls of the Azilian epoch (a period linking up palaeo lithic with neolithic times), some of which were long and others short. There is no reason to suppose that the Indo-Europeans had only one type of skull. No records lead us to suppose that we can trace the Indo-Europeans back more than ten thousand years. In earlier Europe there were peoples of different types, some of a physical frame now confined to the natives of South Africa, others with features akin to those of native Australians, and others again of types which no longer exist. The Basque language in the -Pyre nees is a last relic of languages which preceded the Indo-European in the west of Europe; and the characteristics of colouring, hair and eyes of natives of western Britain and western Ireland, which Huxley called Iberian, are probably the relics of this population which has been almost absorbed. In Persia, on the other hand, a short, very dark people (negrito), which survives only in the extreme south-west of Asia, apparently preceded the Indo-Euro pean stock. The value of the relation of length to breadth of head, to which the elder Retzius first referred, has been much exaggerated, for the total content must be of more importance than the relation of length to breadth. As has been emphasized in recent years by Mr. Griffith Taylor, the height of the brain pan is worthy of consideration.
Pottery.—Still more recently the subject has been attacked from another point of view. In 1902 Professor Kossinna of Berlin undertook to show that pottery would be a more certain index to the history of peoples than any of the other methods pro posed. But in the quarter of a century which has since elapsed, Professor Kossinna has withdrawn almost all the statements which he then made and his proposed new treatment of the Indo-Euro peans has stuck fast after one part was published. That, however, evidence will be forthcoming from this side is certain. But a good deal more has still to be done in identifying the makers of par ticular types of pottery before this clue can be regarded as of more value than any of the others. At present the best chance of progress lies in the possibility of dating positively, and linking up with the linguistic evidence, the results which can be derived from the "Bell-beaker Civilization" and the Megalithic graves which extend very nearly over the same area which was covered by the early migrations of the Indo-Europeans. The probability is that the Indo-Europeans and their civilization arose from the mixture of two earlier types, a Mediterranean and a Northern type. But Schuchhardt in the second edition of his Alteuropa (1926) goes further and boldly assigns (p. 282) the Indo-Europeans to a mix ture of the Germans coming from the north and the Celts from the Danube area, a stock in Thuringen possibly first "indoger manizing" north Germany by settlement and south Germany by conquest. To this he has added (Sitzungsberichte der preuss ischen Akademie, Feb. 16, 1928) that it was the conquest by the Thiiringians of South Germany and Switzerland that created the Celtic stock; the Thuringians afterwards wandered to the Danube and were thus the original stock of the Indo-European people.
Leaving these archaeological points to be definitely decided by future investigation, we may classify the Indo-European languages according to their connec tions. The Indo-European languages fall into two groups distin guished from one another by their treatment of certain guttural sounds. In one group, lying mainly in Europe, these sounds remain as k, g, kh and gh, whereas in the other group they change into some form of sibilant c, z and the aspirated sounds either drop the aspiration and are treated as k or g, or appear as h. The causes for this distinction are not certainly known and the distinction did not mark a deep difference between the languages at an earlier period, because the Greek, which is a k language, resembles in its syntax Sanskrit, which is a c language, more than any other. The two series are generally named from the word for i oo, in Latin centum (kentum), in Zend satem. The languages which preserve the guttural are centum languages, those which change it into a sibilant are
Thus the Greek kuon "dog," Latin canis, appears in Sanskrit as cvii (n) and in Lithuanian as szuo. The chief centum languages are (I) Greek, in three distinct groups of dialects, Aeolic, Ionic and Doric, with more archaic forms in Arcadian and Cyprian. (2) Latin and the other Italic languages of Italy, Oscan in Campania and the South, Umbrian on the East of the Apennines. (3) The Celtic languages, which fall into two groups according to their treatment of an original combination of a guttural with a w sound like the English qu. One group changes qu into p (ancient Gaulish, Welsh, Cornish, Breton), the other represents qu by c (Irish and its descendants, Scotch Gaelic and Manx of the Isle of Man). Thus, corresponding to Latin quis is Irish cia and Welsh pwy. (4) The Teutonic or Germanic lan guages which fall into three groups: (a) Gothic, (b) Norse (Dan ish, Norwegian, Icelandic, Swedish), (c) West Germanic, including English, Frisian, Dutch and Flemish, Low and High German. The most distinctive feature of this group is the "sound shifting" which the stop consonants and s have undergone, and which are catalogued under the action of "Grimm's Law" and "Verner's Law," in the treatment of Teutonic languages. (5) Very remote from these four sections which are all in Europe, is Tocharish, presumably the language of migrants from somewhere much far ther to the west, though the records of this now extinct language are found in Chinese Turkestan. (6) The Indo-European section of the Hittite languages (see above).
satem languages are (I) in northern India Sanskrit and its descendants, and in Persia and its adjacent countries the ancient dialects called Iranian, viz., Zend (of the Avesta), Old Persian, Sogdian and North Iranian ; Sanskrit and Iranian together form the Aryan branch properly so-called, though in England the word has often been applied wrongly to mean Indo-European, (2) Arme nian, (3) Slavonic in its numerous dialects and Baltic (Lithu anian, Lettish, and Old Prussian, which is now extinct), (4) Al banian, more probably the descendant of the Ancient Thracian than of the Ancient Illyrian which occupied in part the same area. The relation of Illyrian to the Ancient Venetic, once spoken north of the Po and to the Ancient Messapian spoken in the heel of Italy, is not yet definitely ascertained. The other important language of Italy, Etruscan, did not belong to this family, but seems to have been related to Ancient Lydian and other languages of the Levant now extinct. (See also articles on HITTITES ; AR YANS; MIGRATIONS, and separate articles on various languages.) BIBLIOGRAPHY —(i.) Indo-European Antiquities: O. Schrader, Sprachvergleichung and Urgeschichte (3rd ed., 1907) ; Reallexikon der indogermanischen Altertumskunde (2 vols., 2nd ed., 1917-26) ; H. Hirt, Die Indogermanen (2 vols., 1905, 1907) ; S. Feist, Kultur, Ausbreitung and Herkunft der Indogermanen. (1913), Indogermanen and Germanen (3rd ed., 1924), and Germanen and Kelten in der antiken Ueberlieferung (1927) ; A. Carnoy, Les Indo-Europeens (1921) ; V. Gordon Childe, The Dawn of European Civilization (1925), and The Aryans (1926) ; Carl Schuchhardt, Alteuropa (2nd ed., 1926) ; Sir G. A. Grierson, Linguistic Survey of India, vol. i. part i. (1927). The best book of reference for prehistoric antiquities is Max Ebert, Reallexikon der Vorgeschichte (1924 etc., S—Z still unpublished).
Indo-European Languages: K. Brugmann, Grundriss der vergleichenden Grammatik der indogermanischen Sprachen (2nd ed., 1897-1916), of which a small additional part has been published since the author's death in 1919 ; a shorter work by the same author is Kurze vergleichende Grammatik (in 3 parts, 19o2-04), dealing only with the more important languages; H. Hirt, Indogermanische Gram matik (the first four vols., only, are published, 1921-28) ; A. Meillet, Introduction a l'etude comparative des langues Indo-Europeenes (1903, now in the 5th ed.) , and Les dialectes indo-euro peens (1908) . The comparative grammar of Latin and Greek is often treated sep arately, as by A. Meillet et J. Vendryes, Traite de grammaire comparee des langues classiques (1924) ; P. Giles, A Manual of Comparative Phil ology for Classical Students (3rd ed. in preparation) . A complete bibliography of recent publications appears annually in the Indoger manisches Jahrbuch begun by A. Thumb and W. Streitberg in 1913.