Open main menu

Popular Science Monthly/Volume 54/March 1899/The Racial Geography of Europe: The Peoples of the Balkan Peninsula - The Greek, the Slav, and the Turk XVIII



THE significant geography of the Balkan Peninsula may best be illustrated by comparing it with the other two south European ones, Italy and Spain. The first point to notice is that it is divided from the mainland by rivers and not by a well-defined mountain chain. Iberia begins definitely at the Pyrenees, and Italy proper is cut off from Europe by the Apennine chain. On the other hand, it is along the line of the Danube and of its western affluent, the Save (see map between pages 614 and 615), that we find the geographical limits of the Balkan Peninsula. This boundary, as will be observed, excludes the kingdom of Roumania, seeming to distinguish it from its trans-Danubian neighbor Bulgaria. This is highly proper, viewed from the standpoint of geography and topography. For Roumania is, for the most part, an extensive and rich alluvial plain; while the Balkan Peninsula, as soon as you leave the Bulgarian lowlands, is characteristically rugged, if not really mountainous.

From Adrianople west to the Adriatic, and from the Balkan Mountains and the Save River south to the plains of Epirus and Thessaly, extends an elevated region upward of two thousand feet above the sea, breaking up irregularly into peaks often rising above five thousand feet. There is no system in these mountains. The land is rudely broken up into a multitude of little "gateless amphitheaters," too isolated for union, yet not inaccessible enough for individuality. As White observes, "If the peninsula, instead of being the highly mountainous and diversified district it is, had been a plateau, a very different distribution of races would have obtained at the present day." Nor can one doubt for a moment that this disordered topography has been an important element in the racial history of the region.

In its other geographical characteristics this peninsula is seemingly more favored than either Spain or Italy. More varied than the former, especially in its union of the two flora of north and south; far richer in contour, in the possession of protected waters and good harbors than Italy; the Balkan Peninsula, nevertheless, has been, humanly speaking, unfortunate from the start. The reason is patent. It lies in its central or rather intermediate location. It is betwixt
PSM V54 D635 Peoples of the balkan peninsula.png

and between; neither one thing nor the other. Surely a part of Europe, its rivers all run to the east and south. "By physical relief it turns its back on Europe," continually inviting settlement from the direction of Asia. It is no anomaly that Asiatic religions, Asiatic institutions, and Asiatic races should have possessed and held it; nor that Europe, Christianity, and the Aryan-speaking races should have resisted this invasion of territory which they regarded in a sense as their own. In this pull and haul between the social forces of the two continents we finally discover the dominant influence, perhaps, which throughout history has condemned this region to political disorder and ethnic heterogeneity.

As little racial as of topographical system can we discover in this Balkan Peninsula. Only in one respect may we venture upon a little generalization. This is suggested by the preliminary bird's-eye view which we must take as to the languages spoken in the peninsula. This was a favorite theme with the late historian Freeman. It is developed in detail in his luminous writings upon the Eastern question. The Slavs have in this part of Europe played a role somewhat analogous to, although less successful than, that of the Teutons in the west. They have pressed in upon the territory of the classic civilizations of Greece and Rome, ingrafting a new and physically vigorous population upon the old and partially enervated one. From some center of dispersion up north toward Russia, Slavic-speaking peoples have expanded until they have rendered all eastern Europe Slavic from the Arctic Ocean to the Adriatic and Ægean Seas. Only at one place is the continuity of Slavdom broken; but this interruption is sufficient to set off the Slavs into two distinct groups at the present day. The northern one, of which we have already treated,[2] consists of the Russians, Poles, Czechs, and Slovaks. The southern group, now before us, comprises the main body of the Balkan peoples from the Serbo-Croatians to the Bulgars, as shown upon the accompanying map. Between these two groups of Slavs—and herein is the significant point—is a broad belt of non-Slavic population, composed, of the Magyars, linguistically now as always Finns; and the Roumanians, who have become Latin in speech within historic times. This intrusive, non-Slavic belt lies along or near the Danube, that great highway over which eastern peoples have penetrated Europe for centuries. The presence of this water way is distinctly the cause of the linguistic phenomenon. Rome went east; and the Finns, like the Huns, went west along it, with the result as described. Linguistically speaking, therefore, the boundary of the southern Slavs and that of the Balkan Peninsula, beginning, as we have said, at the Danube, are one and the same.

We may best begin our ethnic description by the apportionment of the entire Balkan Peninsula into three linguistic divisions, viz., the Greeks, the Slavs, and the Tatar-Turks. Of these the second is numerically the most important, comprising the Serbo-Croatians, the Albanians, and, in a measure, the Bulgarians. Their distribution is manifested upon our map, to which we have already directed attention. These Slavic-speaking peoples form not far from half the entire population. Next in order come the Greeks, who constitute probably about a third of the total. As our map shows, this Greek contingent is closely confined to the seacoast, with the exception of Thessaly, which, as an old Hellenic territory, we are not surprised to find Greek in speech to-day. The Slavs, contrasted with the Greeks, are primarily an inland population; the only place in all Europe, in fact, where they touch the sea is along the Adriatic coast. Even here the proportion of Greek intermixture is more considerable than our map would seem to imply. The interest of this fact is intensified because of the well-deserved reputation as admirable sailors which the modern Dalmatians possess. They are the only natural navigators of all the vast Slavic world. Everywhere else these peoples are noted rather for their aptitude for agriculture and allied pursuits. There is still another important point to be noted concerning the Greeks. They form not only the fringe of coast population in Asiatic as well as in European Turkey; they, with the Jews, monopolize the towns, devoting themselves to commerce as well as navigation. Jews and Greeks are the natural traders of the Orient. Thus is the linguistic segregation between Greek and Slav perpetuated, if not intensified, by seemingly natural aptitudes.

Perhaps the most surprising feature of our map of Turkey is the relative insignificance of the third element, the Turks. There were ten years ago, according to Couvreur, not above seven hundred and fifty thousand of them in all European Turkey. Bradaska estimated that they were outnumbered by the Slavs seven to one. Our map shows that they form the dominant element in the population only in eastern Bulgaria, where they indeed constitute a solid and coherent body. Everywhere else they are disseminated as a small minority among the Greeks or Slavs. Even about Constantinople itself the Greeks far outnumber them. In this connection we must bear in mind that we are now judging of these peoples in no sense by their physical characteristics, but merely by the speech upon their lips. Nowhere else in Europe, as we shall soon see, is this criterion so fallacious as in the Balkan states. Religion enters also as a confusing element. Sax's original map, from which ours is derived, distinguishes these religious affiliations as well as language. He was indeed the first to employ this additional test. The maze of tangled languages and religions upon his map proved too complicated for our imitative abilities. We were obliged to limit our cartography to languages alone. The reader who would gain a true conception of the ethnic heterogeneity of Turkey should consult his original map.

The word Turk was for several centuries taken in a religious sense as synonymous with Mohammedan,[3] as in the Collect for Good Friday in its reference to "Jews, Turks, infidels, and heretics." Thus in Bosnia, where in the fifteenth century many Slavs were converted to Mohammedanism, their descendants are still known as Turks, especially where they use the Turkish speech in their religion. Obviously in this case no Turkish blood need flow in their veins. It is the religion of Islam, acting in this way, which has served to keep the Turks as distinct from the Slavs and Greeks as they are to-day. Freeman has drawn an instructive comparison in this connection between the fate of the Bulgars, who, as we shall see, are merely Slavonized Finns, and the Turks, who have steadily resisted all attempts at assimilation. The first came, he says, as "mere heathen savages (who) could be Christianized, Europeanized, assimilated," because no antipathy save that of race and speech had to be overcome. The Turks, in contradistinction, came "burdened with the half-truth of Islam, with the halfcivilization of the East." By the aid of these, especially the former, the Turk has been enabled to maintain an independent existence as "an unnatural excrescence" on this corner of Europe.

Even using this word as in a measure synonymous with religious affiliations, the Turks form but a small and decreasing minority in the Balkan Peninsula. Couvreur affirms that not over one third of the population profess the religion of Islam, all the remainder being Greek Catholics. This being so, the query at once suggests itself as to the reason for the continued political domination of this Turkish minority, Asiatic alike in race, in speech, and in religion. The answer is certain. It depends upon that subtle principle, the balance of power in Europe. Is it not clear that to allow the Turk to go under, as numerically he ought to do, would mean to add strength to the great Slavic majority, affiliated as it is with Kussia both by speech and religion? This, with the consent of the Anglo-Saxon and other Teutonic rivals of the Slav, could never be allowed. Thus does it come about that the poor Greek is ground between the upper Turkish and the nether Slavic millstone. "Unnatural disunion is the fate of the whole land, and the cuckoo-cry about the independence and integrity of the Ottoman Empire means, among the other evil things that it means, the continuance of this disunion." Let us turn from this distressing political spectacle to observe what light, if any, anthropology may shed upon the problem.


From the relative isolation of the Greeks at the extreme southern point of the peninsula and especially in the Peloponnesus, it would seem that they might be relatively free from those ethnic disturbances which have worked such havoc elsewhere in the Orient. Nevertheless, Grecian history recounts a continuous succession of inroads from the landward north, as well as from the sea. It would transcend the limits of our study to attempt any detailed analysis of the early ethnology of the country.[4] Examination of the relationship of the Pelasgi to their contemporaries we leave to the philologists. Positively no anthropological data on the matter exist. We are sufficiently grateful for the hundred or more well-authenticated ancient Greek crania of any sort which remain to us. It is useless to attempt any inquiry as to their more definite ethnic origin within the tribal divisions of the country,[5] The testimony of these ancient Greek crania is perfectly harmonious. All authorities agree that the ancient Hellenes were decidedly long-headed, betraying in this respect their affinity to the Mediterranean race, which we have already traced throughout southern Europe and Africa.[6] Whether from Attica; from Schliemann's successive cities excavated upon the site of Troy; or from the coast of Asia Minor; at all times from 400 b. c. to the third century of our era; it would seem proved that the Greeks were of this dolichocephalic type. Stephanos gives the average cranial index of them all as about 75.7, betokening a people like the present Calabrians in head form; and, for that matter, about as long-headed as the Anglo-Saxons in England and America. More than this concerning the physical traits of these ancient Greeks we can not establish with any certainty. No perfect skeletons from which we can ascertain their statures remain to us. Nor can we be more positive as to their brunetness. Their admiration for blondness in heroes and deities is well known. As Dr. Beddoe ('93) says, almost all of Homer's favorites were blond or chestnut-haired, as well as large and tall. Lapouge[7] seems inclined to regard this as proof that the Greeks themselves were of this type, a deduction which appears to us in no wise well founded.[8] As we shall see, every characteristic in their modern descendants and every analogy with the neighboring populations leads us to the conclusion that the classical Hellenes were distinctly of the Mediterranean racial types, little different from the Phœnicians, the Romans, or the Iberians.

Since the Christian era, as we have said, a successive downpour of foreigners from the north into Greece has ensued.[9] In the sixth century came the Avars and the Slavs, bringing death and disaster. A more potent and lasting influence upon the country was probably produced by the slower and more peaceful infiltration of the Slavs into Thessaly and Epirus from the end of the seventh century onward. A result of this is that Slavic place names to-day occur all over the Peloponnesus in the open country where settlements could readily be made. The most important immigration of all is probably that of the Albanians, who, from the thirteenth century until the advent of the Turks, incessantly penetrated the land. As a result the Albanian language is spoken to-day over a considerable part of the Peloponnesus, especially in its northeastern corner, where it attaches to the mainland. Only one little district has preserved, it may be added, anything like the original classical Greek speech. The Tzakons, in a little isolated and very rugged district on the eastern coast, include a number of classical idioms in their language. Everywhere else, either in the names of rivers, mountains, and towns, or in borrowed words, evidence of the powerful influence of the Slavic infiltration occurs. This has induced Fallmerayer, Philippson, and others to assert that the Slavs have in fact submerged the original Greeks entirely,[10] Explicit rebuttal of this is offered by Hopf, Hertzberg, and Tozer, who admit the Slavic element, but still declare the Greeks to be Greek. This is a matter concerning which neither philologist nor geographer has a right to speak; the anthropological testimony is the only competent one. To this we turn.

The modern Greeks are a very mixed people. There can be no doubt of this fact from a review of their history. In despite of this, they still remain distinctly true to their original Mediterranean ancestry. This has been most convincingly proved in respect of their head form.[11] The cephalic index of modern living Greeks ranges with great constancy about 81. This, it should be observed, betokens an appreciably broader head than in the case of the ancient Hellenes. Stephanos, who has measured several hundred recruits, finds dolichocephaly to be most prevalent in Tkessaly and Attica; while broadheadedness, so characteristic, as we shall see, of the Albanians and other Slavs, is more accentuated toward the north, especially in Epirus. About Corinth also, where Albanian intermixture is common, the cephalic index rises above 83. The Peloponnesus has probably best preserved its early dolichocephaly, as we should expect. In Thessaly alone are the modern Greeks as purely Mediterranean as in classic times. There can be no doubt that in Asia Minor at least, the word Greek is devoid of any racial significance. It merely denotes a man who speaks Greek, or else one who is a Greek Catholic, converted from Mohammedanism. Greek, like Turk, has become entirely a matter of language and religion, as these people have intermingled. Thus in the southwest of Asia Minor, where Semitic influences have been strong, von Luschan[12] makes the pregnant observation that the Greeks, in the main, look like Jews and speak Turkish. Hero, then, is proof positive that no Greeks of pure Mediterranean descent remain to represent the primitive Hellenic type in that region. But it is equally certain that in the main body of the Greeks at home in Greece, the original racial traits are still in the ascendant. The smoothly oval and long faces in our two Greek portraits are surely of Mediterranean type. To this, the ideal form, the purest elements in the nation still tend to revert.

Whatever may be thought of the ancients, the modern Greeks are strongly brunet in all respects. Ornstein ('79) found less than ten per cent of light hair, although blue and gray eyes were characteristic of rather more than a quarter of his seventeen hundred and sixty-seven recruits. This accords with expectation, for among the Albanians, next neighbors and most intrusive aliens in Greece, light eyes are quite common. Weisbach's ('82) data confirm this, ninety-six per cent of his Greeks being pure brunets.[13] In stature these people are intermediate between the Turks and the Albanians and Dalmatians, which latter are among the tallest of Europeans. In facial features Nicolucci's early opinion seems to be confirmed, that the Greek face is distinctively orthognathous—that is to say, with a vertical profile, the lower parts of the face being neither projecting nor prominent. The face is generally of a smooth oval, rather narrow and high, especially as compared with the round-faced Slavs. The nose is thin and high, perhaps more often finely chiseled and straight in profile. The facial features seem to be well demonstrated in the classic statuary, although it is curious, as Stephanos observes, that these ideal heads are distinctly brachycephalic. Either the ancient sculptors knew little of
PSM V54 D643 Peoples of the balkan states greek people.png
PSM V54 D643 Peoples of the balkan states romanian people.png
Roumanians. County Hunyad, Hungary.
PSM V54 D643 Peoples of the balkan states bulgarian people.png
Bulgarians. County Temes, Hungary.

Balkan States.


anthropology, or else we have again a confirmation of our assertion that, however conscious of their peculiar facial traits a people may be, the head form is a characteristic whose significance is rarely recognized.


Linguistically the pure Slavs in the Balkan states comprise only the Serbo-Croatians and the Albanians (see map), dividing between them the ancient territory of Illyria. This western half of the peninsula, rugged and remote, has been relatively little exposed to the direct ravages of either Finnic or Turkish invaders. Especially is this true of Albania. Nearly all authorities since Hahn are agreed in identifying these latter people—who call themselves Skipetars, by the way—as the modern representatives of the ancient Illyrians. They are said to have been Slavonized by the Serbo-Croatians, who have been generally regarded as descendants of the settlers brought by the Emperor Heraclius from beyond the Save. This he is said to have done in order to repopulate the lands devastated by the Avars and other Slavs who, Procopius informs us, first appeared in this region in the sixth century of our era. The settlers imported by Heraclius came, we are told, from two distant places: Old Servia, or Sorabia, placed by Freeman in modern Saxony; and Chrobatia, which, he says, lies in southwestern Poland. According to this view, the Serbo-Croatians are an offshoot from the northern Slavs, being divided from them to-day by the intrusive Hungarians, while the Albanians alone are truly indigenous to the country.

The recent political fate of these Illyrian peoples has been quite various, the Albanians alone preserving their independence continually under the merely nominal rule of the Turks. Religion, also, has affected these Slavs in various ways. Servia owes much of its present peace and prosperity to the practical elimination of the Moslems. Bosnia is still largely Mohammedan, with about a third of its people, according to White ('86), still professing that religion. The significance of this is increased, since it was mainly the upper classes in Bosnia, according to Freeman, who embraced the religion of Islam in order to preserve their power and estates. The conversion was not national, as in the case of the Albanians. Thus social and religious segregation work in harmony to produce discord. With multitudes of Jews monopolizing the commerce of the country and the people thus divided socially as well as in religion, the political unrest in Bosnia certainly seems to require the strong arm of Austrian suzerainty to preserve order.

Whatever the theory of the historians as to origins may be, to the anthropologist the modern Illyrians—Serbo-Croatians and Albanians alike—are physically a unit. Two characteristics render this ethnic group distinctive: first, that it comprises some of the tallest men in the world, comparing favorably with the Scotch in this respect; and, secondly, that the Illyrians tend to be among the broadest-headed people known. In general, it would appear that the people of Herzegovina and northern Albania possess these traits to the most notable degree, while both in the direction of the Save and Danube and of the plains of Thessaly and Epirus they have been attenuated by intermixture. Presumably also toward the east among the Bulgarians in Macedonia and Thrace these characteristics diminish in intensity. Thus, for example, while the Herzegovinians, measured by Weisbach, yielded an average stature of five feet nine inches, the Bosnians were appreciably shorter; and the Dalmatians and Albanians were even more so. Nevertheless, as compared with the Greeks, Bulgars, Turks, or Roumanians, even the shortest of these Slavs stood high. From this specific center outward, especially around the head of the Adriatic Sea, over into Venetia, spreads the influence of this giantism. It confirms, as we have said, the classical theory of an Illyrian cross among the Venetians, extending well up into the Tyrol.

As for the second trait, the exaggerated broad-headedness, it too, like the tallness of stature, seems to center about Herzegovina and Montenegro. Thus at Scutari, in the corner of Albania near this lastnamed country, Zampa[14] found a cranial index of 89; in Herzegovina the index upon the living head ranges above 87. It would be difficult to exceed this brachycephaly anywhere in the world. The square foreheads and broad faces of the people correspond in every way to the shape of the heads. Its significance appears immediately on comparison with the long oval faces of the Greeks.

One more trait of the Balkan Slavs remains for us to note. The people are mainly pure brunets, as we might expect, but they seem to be less dark than either the Greeks or the Turks. Especially among the Albanians are light traits by no means infrequent. In this respect the contrast with the Greeks is apparent, as well as with the Dalmatians along the coast and the Italians in the same latitude across the Adriatic. Weisbach found nearly ten per cent of blond and red hair among his Bosnian soldiers, while about one third of the eyes were either gray or blue. The Herzegovinians are even lighter than the Bosnians, almost as much so as the Albanians. From consideration of these facts it would appear as if the harsh climate of these upland districts had been indeed influential in setting off the inland peoples from the Italian-speaking Dalmatians along the coast. For among the latter brunetness certainly increases from north to south, conformably to the general rule for the rest of Europe. In the interior, blondness apparently moves in the contrary direction, culminating in the mountain fastnesses of northern Albania and the vicinity. On the whole, we find also in this trait of brunetness competent evidence to connect these Illyrians with the great body of the Alpine race farther to the west. We have another illustration of its determined predilection for a mountainous habitat, in which it stoutly resists all immigrant tendencies toward variation from its primitive type.


The Osmanli Turks, who politically dominate the Balkan Peninsula, notwithstanding their numerical insignificance, are mainly distinctive among their neighbors by reason of their speech and religion. Turkish is the westernmost representative of a great group of languages, best known, perhaps, as the Ural-Altaic family.[15] This comprises all those of northern Asia even to the Pacific Ocean, together with that of the Finns in Russian Europe. Its members are by no means unified physically. All varieties of type are included within its boundaries, from the tall and blond one which we may call Finnic, prevalent about the Baltic; to the squat and swarthy Kalmucks and Kirghez, to whom we have in a physical sense applied the term Mongols. The Turkish branch of this great family of languages is to-day represented in eastern Europe by two peoples, whom we may roughly distinguish as Turks and Tatars,[16] The term Tatar, it should be observed, is entirely of European invention, like the similar word Hungarian. The only name recognized by the Osmanli themselves is that of Turk. This, by the way, seems quite aptly to be derived from a native root meaning "brigand," according to Chantre. They apply the word Tatar solely to the north Asiatic barbarians. By general usage this latter term, Tatar, has to-day become more specifically applied by ethnologists to the scattered peoples of Asiatic descent and Turkish speech who are mainly to be found in Russia and Asia Minor.

Of the two principal physical types to-day comprised within the limits of the Ural-Altaic languages, the Turks and Tatars seem to be affiliated with the Mongol rather than the Finn, not physically alone, but in respect of language as well. As a matter of fact, they are racially nearer the Aryan-speaking Europeans than most people imagine, in everything except their speech. Their nearest relatives in Asia seem to be the Turkoman peoples, who, to the number of a million or more, inhabit the deserts and steppes of western Asia. It was from somewhere about this latter region, as we know, that the hordes of the
PSM V54 D647 Uzbek from ferganah of the turkoman peoples.png
Uzbeq. Ferghanah.
PSM V54 D647 Kipchak people of the Turkoman peoples.png
PSM V54 D647 Kara khirgiz people of the turkoman people.png

Turkoman Types.


Huns under Attila, and those of Genghis Khan and Tamerlane, set forth to the devastation of Europe. The physical type of these inhabitants of Turkestan has been fairly well established by anthropologists. It persists throughout a great multitude of tribes of various names, among whom the Kara-Kirghez, Uzbegs, and Kiptchaks are prominent.[17] On page 625 we have portraits of these Turkoman types. The most noticeable feature of the portraits is the absence of purely Mongol facial characteristics. Except in the Kara-Kirghez the features are distinctly European. There is no squint-eye; the nose is well formed; the cheek bones are not prominent, although the faces are broad; and, most important of all, the beard is abundantly developed, both in the Uzbeg and the Kiptchak. The Kara-Kirghez, on the other hand, betrays unmistakably his Mongol derivation in every one of these important respects. One common trait is possessed by all three—to wit, extreme brachycephaly, with an index ranging from 85 to 89. The flatness of the occiput is very noticeable in our portraits in every case, giving what Hamy calls a "cuboid aspect" to the skull. These portraits, if typical, should be enough to convince us that the Turkoman of the steppes about the Aral and Caspian Seas is far from being a pure Mongol even in his native land, although a strain of Mongol blood is apparent in many of their tribes.

The fact is that the Asiatic Turkomans, whence our Osmanli Turks are derived, are a highly composite type. A very important element in their composition is that of certain brachycephalic peoples of the Pamir, the Galchas and mountain Tadjiks. These are for all practical purposes identical with the Alpine type of western Europe. In their accentuated brachycephaly, their European facial features, their abundance of wavy hair and beard, and finally in their intermediate color of hair and eyes,[18] these latter peoples in the Pamir resemble their European prototypes, or perhaps we had better say, congeners. So close is this affiliation that the occurrence of this type in western Asia is the keystone in any argument for the Asiatic origin of the Alpine race of Europe. The significance of it for us in this connection is that it explains the European affinity of many of the Turkoman tribes, who are more strongly European than Mongol in their resemblances. It is highly important, we affirm, to fix this in mind, for the prevalent opinion seems to be that the Turks in Europe have departed widely from their ancestral Asiatic type, because of their present lack of Mongol characteristics, such as almond eyes, lank black hair, flat noses, and high cheek bones.
PSM V54 D649 Ivervec nomad of the turkic peoples.png
Nomad Ivervek. Lycia, Asia Minor.
PSM V54 D649 Turk of the turkic peoples 1.png
Turk. Lycia, Asia Minor.
PSM V54 D649 Turk of the turkic peoples 2.png
Turk. Lycia, Asia Minor.



Either the Osmanli Turks were never Mongols, or they have lost every trace of it by intermixture. Our portraits on the opposite page give little indication of Asiatic derivation except in their accentuated short-and broad-headedness. This is considerably more noticeable in Asia Minor than in European Turkey.[19] West of the Bosporus the Turks differ but little from the surrounding Slavs in head form. They have been bred down from their former extreme brachycephaly, which still rules to a greater degree in Asia Minor. In our portraits from this region the absence of occipital prominence is very marked. In addition to this, the Turks are everywhere, as Chantre observes, "incontestably brunet." The hair is generally stiff and straight. The beard is full. This latter trait is fatal to any assumption of a persistence of Kirghez blood, or of any Mongolic extraction, in fact. The nose is broad, but straight in profile. The eyes are perfectly normal, the oblique Mongol type no more frequent than elsewhere. In stature tallness is the rule, judging by Chantre's data, but in this respect social conditions are undoubtedly of great effect. On the whole, then, we may consider that the Turks have done fairly well in the preservation of their primitive characteristics. Chantre especially finds them quite homogeneous, considering all the circumstances. They vary according to the people among whom their lot is cast. Among the Armenians they become broader-headed, while among the Iranian peoples—Kurds or Persians—the opposite influence of intermixture at once is apparent.

The Bulgarians are of interest because of their traditional Finnic origin and subsequent Europeanization. This has ensued through conversion to Christianity and the adoption of a Slavic speech. Our earliest mention of these Bnlgars would seem to locate them between the Ural Mountains and the Volga.[20] The district was, in fact, known as Old Bulgaria till the Russians took it in the fifteenth century. As to which of the many existing tribes of the Volga Finns represent the ancestors of these Bulgarians, no one is, I think, competent to speak. Pruner Bey seems to think they were the Ostiaks and Voguls, since emigrated across the Urals into Asia; the still older view of Edwards and Klaproth made them Huns; Obédénare, according to Virchow, said they were Samoyeds or Tungus; while Howorth and Beddoe claim the honor for the Chuvashes. These citations are enough to prove that nobody knows very much about it in detail. All that can be affirmed is that a tribe of Finnic-speaking people crossed the Danube toward the end of the seventh century and possessed themselves of territory near its mouth. Remaining heathen for two hundred odd years, they finally adopted Christianity and under their great leaders, Simeon and Samuel, became during the tenth century a power in the land. Their rulers, styling themselves "Emperors of the Slavs," fought the Germans; conquered the Magyars as well as their neighbors in Thrace, receiving tribute from Byzantium; became allies of Charlemagne; and then subsided under the rule of the Turks. Since the practical demise of this latter power they have again taken courage, and in their semi-political independence in Bulgaria and northern Roumelia rejoice in an ever-rich and growing literature and sense of nationality.

Bulgarian is spoken, as our map at page 614 indicates, far outside the present political limits of the principality—indeed, over about two thirds of European Turkey. Gopčević has made a brilliant attempt to prove that Macedonia, shown by our map and commonly believed to be at bottom Bulgarian, is in reality populated mainly by Serbs.[21] The weakness of this contention was speedily laid bare by his critics. Political motives, especially the ardent desire of the Servians to make good a title to Macedonia before the disruption of the Ottoman Empire, can scarcely be denied. Servia needs an outlet on the Mediterranean too obviously to cloak such an attempted ethnic usurpation. As a fact, Macedonia, even before the late Greco-Turkish war, was in a sad state of anarchy. The purest Bulgarian is certainly spoken in the Bhodope Mountains; there are many Roumanians of Latin speech; the Greeks predominate all along the sea and throughout the three-toed peninsula of Salonica, while the Turks are sparsely disseminated everywhere. And as for religion—well, besides the severally orthodox Greeks and Turks, there are in addition the Moslem and apostate Bulgarians, known as Pomaks, who have nothing in common with their Greek Catholic fellow-Bulgars, together with the scattering Pindus Roumanians and Albanians in addition. This interesting field of ethnographic investigation is, even at this late day, practically unworked. As Dr. Beddoe writes—and his remarks are equally applicable to Americans—"here are fine opportunities for any enterprising Englishman with money and a taste for travel and with sufficient brains to be able to pick up a language. But, alas! such men usually seem to care for nothing but 'killing something.'"

The Roumanians, or Moldo-Wallachians, are not confined within the limits of that country alone. Their language and nationality cover not only the plains along the Danube and the Black Sea, but their speech extends beyond the Carpathian Mountains over the entire southeastern quarter of Hungary and up into the Bukovina. Transylvania

PSM V54 D652 Peoples of hungary and transylvania.png

is merely a German and Magyar islet in the vast extent of the Roumanian nation. There are more than a third as many Roumanians as there are Magyars in the Hungarian kingdom, according to the census of 1890. Politically it thus happens that these people are pretty well split up in their allegiance. Nor can this condition be other than permanent. For the Carpathian Mountains, in their great circle about the Hungarian basin, cut directly through the middle of the nation as measured by language. This curious circumstance can be accounted for only on the supposition that the disorder in the direction of the Balkan Peninsula, incident upon the Turkish invasion, forced the growing nation to expand toward the northwest, even over the natural barrier interposed between Roumania proper and Hungary. Geographical law, more powerful than human will, ordains that this latter natural area of characterization—the great plain basin of Hungary—should be the seat of a single political unit. There is no resource but that the Roumanians should in Hungary accept the division from their fellows over the mountains as final for all political purposes.[22]

The native name of these people is Vlach, Wallach, or Wallachian. Various origins for the name have been assigned. Lejean asserts that it designates a nomad shepherd, in distinction from a tiller of the soil or a dweller in towns. Picot voices the native view as to ethnic origins by deriving the word Wallach from the same root as Wales, Walloon, etc., applied by the Slavs and Germans to the Celtic peoples as "foreigners." This theory is now generally discountenanced. Obédénare's attempt to prove such a Celtic relationship has met with little favor,[23] The western name Roumanian springs from a similarly exploded hypothesis concerning the Latin origin of these people. To be sure, Roumanian is distinctly allied to the other Romance languages in structure. It is an anomaly in the eastern Slavic half of Europe. The most plausible explanation for this phenomenon, and one long accepted, was that the modern Roumanians were descendants of the two hundred and forty thousand colonists whom the Emperor Trajan is said to have sent into the conquered province of Dacia. The earlier inhabitants of the territory were believed to have been the original Thracians. Since no two were agreed as to what the Thracians were like, this did not amount to much. Modern common sense has finally prevailed over attempts to display philological erudition in such matters. Freeman expresses this clearly. Roumania, as he says, lay directly in the pathway of all invasions from the East; the hold of the Romans upon Dacia was never firm; the province was the first to break away from the empire; and finally proof of a Latinization only at the late date of the thirteenth century is not wanting. The truth seems to be that two forces were contending for the control of eastern Europe. The Latin could prevail only in those regions which were beyond the potent influence of Greece. Dacia being remote and barbarian, this Latin element had a fighting chance for survival, and succeeded.

Our ethnic map at page 614 shows a curious islet of Roumanian language in the heart of the Greek-speaking territory of Thessaly. There is little sympathy between the two peoples, according to Hellène. The occurrence of this Roumanian colony, so far removed from its base, has long puzzled ethnographers. Some believe the peoples were separately Romanized in situ; others that they were colonists from Dacia in the ninth and tenth centuries. At all events, these Pindus Roumanians are too numerous—over a million souls—to be neglected in any theory as to the origin of their language.[24] Another islet of quasi-Roumanian speech occurs in Istria, on the Adriatic coast. Its origin is equally obscure.[25]

It is no contradiction that, in spite of the fact of our exclusion of Roumania from the Balkan Peninsula owing to its Latin affinities, thereby seeming to differentiate it sharply from Bulgaria, the latter of Finnic origin; that we now proceed to treat of the physical characteristics of the two nationalities, Roumanian and Bulgarian, together. Here is another example of the superficiality of language, of social and political institutions. They do not concern the fundamental physical facts of race in the least. At the same time we again emphasize the necessity of a powerful corrective, based upon purely natural phenomena, for the tendency of philologists and ethnographers to follow their pet theories far afield, giving precedence to analogies of language and customs over all the potent facts of geographical probability. Let us look at it in this light. Is there any chance that, on the opposite sides of the Danube, a few Finns and a few Romans respectively interposed among the dense population which so fertile an area must have possessed, even at an early time, could be in any wise competent to make different types of the two? There is nothing in our confessedly scanty anthropological data to show it, at all events. We must treat the lower Danubian plain as a unit, irrespective of the bounds of language, religion, or nationality.

It was long believed that the Bulgarians were distinctive among the other peoples of eastern Europe by reason of their long-headedness. All the investigations upon limited series of crania pointed in that direction. This naturally was interpreted as a confirmation of the historic data as to a Finnic Bulgarian origin very distinct from that of the broad-headed Slavs. Several recent discoveries have put a new face upon the matter. In the first place, researches by Dr. Bassanovitch, of Varna, upon several thousand recruits from western Bulgaria prove that in the west these Bulgarians even outdo many of the Balkan Slavs in their broad-headedness.[26] At the same time it appears that the older authorities were right, after all, in respect of the eastern Bulgarians. Among them, and also over in eastern Roumelia, long heads are still the rule. The oval-faced Bulgarians among our portraits are probably of this dolichocephalic type. Their contrast facially with the broad-headed Roumanians is very marked. Thus it is established that the Bulgarian nation is by no means a unit in its head form. We should add also that, although not definitely proved as yet, it is highly probable that similar variations occur in Roumania. In the Bukovina brachycephaly certainly prevails. Our square-faced Roumanians on page 621 may presumably be taken to represent this type. This broad-headedness decreases apparently toward the east as we leave the Carpathian Mountains, until along the Black Sea it seems, as in Bulgaria, to give way to a real dolichocephaly.[27]

How are we to account for the occurrence of so extended an area of long-headedness all over the great lower Danubian plain? Our study of the northern Slavs has shown that no such phenomenon occurs there among the Russians. It certainly finds no counterpart among the southern Slavs or the Turks. The only other people who resemble these Bulgars in long-headedness are the Greeks. Even they are far separated; and, in any event, but very impure representatives of the type. What shall we say? Two explanations seem to be possible, as Dr. Beddoe observes.[28] Either this dolichocephaly is due to the Finnicism of the original Bulgars, or else it represents a characteristic of the pre-Bulgarian population of the Danube basin. He inclines with moderation to the former view. The other horn of the dilemma is chosen by Anutchin[29] in a brilliant paper at the late Anthropological Congress at Moscow. According to his view—and we assent most heartily to it—this dolichocephaly along the Black Sea represents the last survival of a most persistent trait of the primitive inhabitants of eastern Europe. Referring again to our study of Russia,[30] we would call attention to the occurrence of a similar longheaded race underlying all the modern Slavic population. We are able to prove also that such a primitive substratum occurs over nearly all Europe. It has been unearthed not far from here, for example, at Glasinac in Bosnia. When archælogical research is extended farther to the east, new light upon this point may be expected. It will be asked at once why this primitive population should still lie bare upon the surface, here along the lower Danube, when it has been submerged everywhere else in Central Europe. Our answer is ready. Here in this rich alluvial plain population might, expectedly, be dense at a very early period. As we have observed before, such a population, if solidly massed, opposes an enormous resistance to absorption by new-comers. A few thousand Bulgarian invaders would be a mere drop in the bucket of such an aggregation of men. We are strengthened in this hypothesis that the dolichocephaly of the Danubian plain is primitive, by reason of another significant fact brought out by Bassanovitch.[31] Long-headedness is overwhelmingly more prevalent among women than among men. The former represent more often what Bassanovitch calls the "dolichocephalic Thracian type." The oval-faced Bulgarian woman among our portraits would seem to be one of these. Now, in our treatment of the Jews,[32] we have sought to illustrate the principle that in any population the primitive type persists more often in the women. The bearing of such a law in the case of the Bulgars would seem to be definite. Their long-headedness, where it occurs, must date from a far more remote period than the historic advent of the few thousand immigrants who have given the name Bulgaria to the country.

As for the other physical traits of the Bulgarians and Roumanians there is little to be added. It goes without saying that they are both deep brunets. Obédénare says the Roumanians are very difficult to distinguish from the modern Spaniards and Italians. This is probably true in respect of brunetness. The Oriental cast of features of our portraits, on the other hand, can not fail to attract attention. More than two thirds of Bassanovitch's nineteen hundred and fiftyfive Bulgarians were very dark-haired. Light eyes were of course more frequent, nearly forty per cent being classed as blue or greenish. A few—about five per cent—were yellow or tawny-haired, these individuals being at the same time blue-eyed. This was probably Procopius's excuse for the assertion that the Bulgars were of fair complexion. He also affirmed that they were of goodly stature. This is not true of either the modern Roumanians or Bulgars. They average less than five feet five inches in height,[33] being considerably shorter than the Turks, and positively diminutive beside the Bosnians and other southern Slavs. The Bulgarians especially are correspondingly stocky, heavily boned and built. We may also affirm a real difference in temperament between the two nationalities, built up, as we assert, from the same foundation. The Wallachians are said to be more emotional and responsive; the Bulgarians inclined to heaviness and stolidity. Both are pre-eminently industrious and contented cultivators of the soil, with little aptitude for commerce, so it is said. We hesitate to pass judgment upon either in respect of their further aptitudes until fuller data can be provided than are available at the present time.


  1. Advance sheets from The Races of Europe, in press of D. Appleton and Company, many footnotes and detailed references being here omitted.
  2. Popular Science Monthly, October, 1898.
  3. Consult Taylor, 1890, p. 48; Von Luschan, 1889, p. 198; Sax, 1863, p. 91.
  4. Consult Fligier, 1881. Stephanos, 1884, p. 430, gives a complete bibliography of the older works. Cf. also Reinach, 1893 b, in his review of Hesselmeyer; and on the supposed Hittites, the works of Wright, De Cara, Conder, etc.
  5. Stephanos, 1884, p. 432, asserts the Pelasgi to have been brachycephalic, while Zampa, 1886 b, p. 639, as positively affirms the contrary view.
  6. Nicolucci, 1865 and 1867; Zaborowski, 1881; Virchow, 1882 and 1893; Lapouge, 1896 a, pp. 412-419; and Sergi, 1895 a, p. 75, are best on ancient Greek crania.
  7. 1896 a, p. 414.
  8. Stephanos, 1884, p. 439.
  9. Philippson, Zur Ethnographie des Peloponnes. Petermann, xxxvi, 1890, pp. 1-1 1, 33-41, with map, gives a good outline of these. Consult also Stephanos, 1884, pp. 422 et seq.
  10. Cf. Couvreur, 1890, p. 514; and Freeman, 1877 d, p. 401.
  11. Weisbach, 1882; Nicolucci, 186*7; Apostolides in Bull. Soc. d'Anth., 1883, p. 614; Stephanos, 1884; Neophytos, 1891; Lapouge, 1896 a, p. 419. Von Luschan, 1889, p. 209, illustrates the similarity between the Greek and the Bedouin skull.
  12. 1889, p. 209.
  13. Neophytos finds 82.5 per cent of dark-brown or black hair, only five per cent blond or red; while seventeen per cent of the eyes were dark among two hundred individuals.
  14. 1886 b, p. 637.
  15. Vambéry, 1885, divides the Ural-Altaic family into five groups—viz., (1) Samoyed, (2) Tungus, (3) Finnic, (4) Mongolic, (5) Turkish or Tatar.
  16. On terminology consult Vambéry, 1885, p. 60; Chantre, 1895, p. 199; Keane, 1897, p. 302.
  17. Complete data on these people will be found in Ujfalvy, 1878-'80, iii, pp. 7-50; Les Aryens, etc., 1896, pp. 385-434; Bogdanof, 1888; Yavorski, 1897.
  18. Ujfalvy (Les Aryens, etc., 1896, p. 428) found chestnut hair most frequent, with twenty-seven per cent of blondness, among some of the Tadjiks. The eyes are often greenish gray or blue (Ujfalvy, 1878-80, iii, pp. 23-33, tables).
  19. On the anthropology of European Turks, Weisbach, 1873, is the only authority. He found an average cephalic index of 82.8 in 148 cases. Elisyeef, 1890-'91, and Chantre, 1895, pp. 206-211, have worked in Anatolia, with indices of 86 for 143 individuals, and 84.5 for 120 men, respectively. Both von Luschan and Chantre give a superb collection of portrait types in addition.
  20. Read Pruner Bey, 1860 b; Howorth; Obedenare, and especially Kauitz, 1875, for historic details.
  21. 1889 a, with map, in Petermann, 1889 b. Cf. criticism of his contention by Oppel, 1890; Couvreur, 1890, p. 523; and Ghennadieff, 1890, p. 663.
  22. Auerbach, 1898, p. 286, gives a full summary of the rival controversy between Roumanians and Hungarians as to priority of title in Transylvania.
  23. Cf. Picot, 1883, in his review of Tocilescu; and Rosny, 1885, p. 83.
  24. Picot, 1875, pp. 390 et seq.
  25. Auerbach, 1898, p. 211.
  26. 1891, p. 30. Dr. Bassanovitch has most courteously sent me a sketch map showing the results of these researches. Deniker, 1897, p. 203, and 1898 a, describes them also.
  27. Deniker, 1898 a, p. 122; Weisbach, 1877, p. 238; Rosny, 1885, p. 85.
  28. 1879, p. 233.
  29. 1893, p. 282.
  30. Popular Science Monthly, October, 1898, p. 734.
  31. 1891, p. 31. Women dolicho-, twenty-five per cent; meso-, forty-two per cent; brachy-cephalic, thirty per cent; while among men the percentages are 3, 16, and 81 ± per cent respectively.
  32. Popular Science Monthly, January, 1899, p. 350.
  33. Bassanovitch's series of 1,955 individuals averages only 1.638 metre. Op. cit., p. 30. Auerbach, 1898, p. 259, gives an average of 1.63 metre for 880 Wallachians in Transylvania. Obédénare, 1876, p. 374, states brown eyes to be most frequent in Roumania.