Chapter XV: THE CAUCASIC PEOPLES (continued) (1)

THE PEOPLES OF ARYAN SPEECH--European Trade Routes--"Aryan"
Migrations--Indo-European Cradle--Indo-European Type--Date of
Indo-European Expansion--Origin of Nordic Peoples--The _Cimbri_
and _Teutoni_--_The Bastarnae_--_The Moeso-Goths_--Scandinavia--
Modification of the Nordic Type--THE CELTO-SLAVS: Their Ethnical
Position defined--Aberrant _Tyrolese_ Type--_Rhaetians_ and
_Etruscans_--Etruscan Origins--The Celts--Definitions--Celts
in Britain--The Picts--Brachycephals in Britain--Round Barrow
Type--Alpine Type--Ethnic Relations--Formation of the English
Nation--Ethnic Relations in Ireland--Scotland--and in Wales--Present
Constitution of the British Peoples--The English Language--_The
French Nation_--Constituent Elements--Mental Traits--_The Spaniards
and Portuguese_--Ethnic Relations in Italy--_Ligurian_, _Illyrian_,
and _Aryan Elements_--The Present _Italians_--Art and Ethics--_The
Rumanians_--Ethnic Relations in Greece--_The Hellenes_--Origins and
Migrations--The _Lithuanian_ Factor--_Aeolians_; _Dorians_;
_Ionians_--The Hellenic Legend--The Greek Language--THE SLAVS--
Origins and Migrations--_Sarmatians_ and _Budini_--_Wends_,
_Chekhs_, and _Poles_--The Southern Slavs--Migrations--_Serbs_,
_Croats_, _Bosnians_--_The Albanians_--_The Russians_--
Panslavism--Russian Origins--_Alans_ and _Ossets_--Aborigines
of the Caucasus--THE IRANIANS--Ethnic and Linguistic Relations--
_Persians_, _Tajiks_ and _Galcha_--_Afghans_--Lowland and Hill
Tajiks--The Galchic Linguistic Family--Galcha and Tajik Types--
_Homo Europaeus_ and _H. Alpinus_ in Central Asia--THE HINDUS--
Ethnic Relations in India--Classification of Types--_The Kols_--
_The Dravidians_--Dravidian and Aryan Languages--The Hindu
Castes--OCEANIA--_Indonesians_--_Micronesians_--_Eastern
Polynesians_--Origins, Types, and Divisions--Migrations--
Polynesian Culture.

As the result of recent researches there is an end of the theory that bronze came in with the "Aryans," and it is from this standpoint that the revelation of an independent Aegean culture in touch with Babylonia and Egypt some four millenniums before the new era is of such momentous import in determining the ethnical relations of the historical, _i.e._ the present European populations.

Some idea of cultured relations in prehistoric times may be obtained from a review of the trade communications as indicated by archaeology during the Bronze Age which lasted through the whole of the third millennium down to the middle of the second. As we have seen, in the Nile valley, in Mesopotamia and in the Aegean area, remains characteristic of Bronze Age culture rest on a neolithic substratum, and a transitional stage, when gold and copper were the only metals known, often connects the two. From the time of this dawning of the Age of Metals, the inhabitants of the Nile Valley, of Crete, of Cyprus and of the mainland of Greece freely exchanged their products. Navigation was already flourishing, and the sea united rather than divided the insular and coastal populations. Gradually Egeo-Mykenaean civilisation extended from Crete and the Greek lands to the west, influencing Sicily directly, and leaving distinct traces in Southern Italy, Sardinia and the Iberian peninsula, while Iberia in its turn contributed to the development of Western Gaul and the British Isles. The knowledge of copper, and, soon after, that of bronze, spread by the Atlantic route to Ireland, while Central Europe was reached directly from the south. Thanks to the trade in amber, always in demand by the Mediterranean populations, there was a continuous trade route to Scandinavia, which thus had direct communication with Southern Europe. As civilisation developed, the lands of the north and west became exporters as well as importers, each developing a distinct industry not always inferior to the more precocious culture of the south[1194].

With trade communications thus stretching across Europe from south to north, and from east to extreme west, it would seem not improbable that movements of peoples were equally unrestricted, and this would account for the appearance on the threshold of history of various peoples formerly grouped together on account of their language, as "Aryan." J. L. Myres, however, is inclined to attribute "the coming of the North" to the same type of climatic impulse which induced the Semitic swarms described above (p. 489). After referring to the earliest occurrence of Indo-European names[1195], he continues "Before the time of the Eighteenth Dynasty of Egypt there had been a very extensive raid of Indo-European-speaking folk by way of the Persian plateau, as far as the Syrian coastland and the interior of Asia Minor." These raids coincide with a new cultural feature of great significance. "It is of the first importance to find that it is in the dark period which immediately precedes the Eighteenth Dynasty revival--when Egypt was prostrate under mysterious 'Shepherd Kings,' and Babylon under Kassite invaders equally mysterious--that the civilized world first became acquainted with one of the greatest blessings of civilisation, the domesticated horse. The period of Arabian drought, which drove forth the 'Canaanite' emigrants, may have had its counterpart on the northern steppe, to provoke the migration of these horsemen." He adds, however, "our knowledge both of the extent of these droughts and of the chronology of both these migrations, is too vague for this to be taken as more than a provisional basis for more exact enquiry[1196]."

The attempt has often been made to locate the original home of the Indo-European people by an appeal to philology, and idyllic pictures have been drawn up of the "Aryan family" consisting of the father the protector, the mother the producer, and the children "whose name implied that they kept everything clean and neat[1197]." They were regarded as originally pastoral and later agricultural, ranging over a wide area with Bactria for its centre. With advancing knowledge of what is primitive in Indo-European this circumstantial picture crumbled to pieces, and Feist[1198] reduces all inferences deducible from linguistic palaeontology to the sole "argumentum ex silencio" (which he regards as distinctly untrustworthy in itself), that the "Urheimat" was a country in which in the middle of the third millennium B.C. such southern animals as lion, elephant, and tiger, were unknown. It was commonly assumed that the "Aryan cradle" was in Asia, and the suggestion of R. G. Latham in 1851 that the original home was in Europe was scouted by one of the most eminent writers on the subject--Victor Hehn--as lunacy possible only to one who lived in a country of cranks[1197]. But since this date, there has been a shifting of the "Urheimat" further and further west. O. Schrader[1199] places it in South Russia, G. Kossinna[1200] and H. Hirt[1201] support the claims of Germany, while K. Penka and many others go still further north, deriving both language and tall fair dolichocephalic speakers (proto-Teutons) from Scandinavia[1202].

F. Kauffmann[1203], noting the contrast between the cultures associated with pre-neolithic and with neolithic kitchen-middens, is prepared to attribute the former to aboriginal inhabitants, Ligurians, and, further north, Kvaens (Finns, Lapps), and the neolithic civilisation of Europe to Indo-Europeans. "Thus the neolithic Indo-Europeans would already have advanced as far as South Sweden in the Litorina period of the Baltic, during the oak-period."

On the other hand the discovery of Tocharish has inclined E. Meyer[1204] to reconsider an Asiatic origin, but the information as to this language is too fragmentary to be conclusive on this point. After reviewing the various theories Giles[1205] concludes "in the great plain which extends across Europe north of the Alps and Carpathians and across Asia north of the Hindu Kush there are few geographical obstacles to prevent the rapid spread of peoples from any part of its area to any other, and, as we have seen, the Celts and the Hungarians etc. have in the historical period demonstrated the rapidity with which such migrations could be made. Such migrations may possibly account for the appearance of a people using a _centum_ language so far east as Turkestan[1206]."

More acrimonious than the discussion of the original home is the dispute as to the original physical type of the Indo-European-speaking people. It was almost a matter of faith with Germans that the language was introduced by tall fair dolichocephals of Nordic type. On the other hand the Gallic school sought to identify the Alpine race as the only and original Aryans. The futility of the whole discussion is ably demonstrated by W. Z. Ripley in his protest against the confusion of language and race[1207]. Feist[1208] summarises our information as follows. All that we can say about the physical type of the "Urvolk" is that since the Indo-Europeans came from a northerly region[1209] (not yet identified) it is surmised that they belonged to the light-skinned people. The observation that mountain folk of Indo-Germanic speech in southern areas, such as the Ossets of the Caucasus, the Kurds of the uplands of Armenia and Irania, and the Tajiks of the western Pamirs not infrequently exhibit fair hair or blue eyes supports this view. Nevertheless, as he points out, brachycephals are not hereby excluded. His own conclusion, which naturally results from a review of the whole evidence, is that the "Urvolk" was not a pure race, but a mixture of different types. Already in neolithic times races in Europe were no longer pure, and in France "formed an almost inextricable medley" and Feist assumes with E. de Michelis[1210] that the Indo-Europeans were a conglomerate of peoples of different origins who in prehistoric times were welded together into an ethnic unity, as the present English have been formed from pre-Indo-European Caledonians (Picts and Scots), Celts, Roman traders and soldiers and later Teutonic settlers[1211].

The evidence that Indo-Europeans were already in existence in Mesopotamia, Syria and Irania about the middle of the second millennium B.C. has already been mentioned. About the same time the Vedic hymns bear witness to the appearance of the Aryans of Western India. The formation of an Aryan group with a common language, religion and culture is a process necessarily requiring considerable length of time, so that their swarming off from the Indo-European parent group must be pushed back to far into the third millennium. At this period there are indications of the settling of the Greeks in the southern promontories of the Balkan peninsula at latest about 2000 B.C., while Thracian and Illyrian peoples may have filled the mainland, though the Dorians occupied Epirus, Macedonia, and perhaps Southern Illyria. Indo-European stocks were already in occupation of Central Italy. It would appear therefore that the period of the Indo-European community, before the migrations, must be placed at the end of the Stone Ages, at the time when copper was first introduced. Thus it seems legitimate to infer that the expansion of the Indo-Europeans began about 2500 B.C. and the furthest advanced branches entered into the regions of the older populations and cultures at latest after the beginning of the second millennium[1212]. About 1000 B.C. we find three areas occupied by Indo-European-speaking peoples, all widely separated from each other and apparently independent. These are (1) the Aryan groups in Asia; (2) the Balkan peninsula together with Central and Lower Italy, and the Mysians and Phrygians of Asia Minor (possibly the Thracians had already advanced across the Danube); and (3) Teutons, Celts and Letto-Slavs over the greater part of Germany and Scandinavia, perhaps also already in Eastern France and in Poland. The following centuries saw the advance of Iranians to South Russia and further west, the pressing of the Phrygians into Armenia, and lastly the Celtic migrations in Western Europe.

From the linguistic and botanical evidence brought forward by the Polish botanist Rostafinski[1213] the ancestors of the Celts, Germans and Balto-Slavs must have occupied a region north of the Carpathians, and west of a line between Koenigsberg and Odessa (the beech and yew zone). The Balto-Slavs subsequently lost the word for beech and transferred the word for yew to the sallow and black alder (both with red wood) but their possession of a word for hornbeam locates their original home in Polesie--the marshland traversed by the Pripet but not south or east of Kiev.

Although, owing to the absence of Teutonic inscriptions before the third or fourth century A.D. it is difficult to trace the Nordic peoples with any certainty during the Bronze or Early Iron Ages, yet the fairly well-defined group of Bronze Age antiquities, covering the basin of the Elbe, Mecklenburg, Holstein, Jutland, Southern Sweden and the islands of the Belt have been conjectured with much probability to represent early Teutonic civilisation. "Whether we are justified in speaking of a Teutonic race in the anthropological sense is at least doubtful, for the most striking characteristics of these peoples [as deduced from prehistoric skeletons, descriptions of ancient writers and present day statistics] occur also to a considerable extent among their eastern and western neighbours, where they can hardly be ascribed altogether to Teutonic admixture. The only result of anthropological investigation which so far can be regarded as definitely established is that the old Teutonic lands in Northern Germany, Denmark and Southern Sweden have been inhabited by people of the same type since the neolithic age if not earlier[1214]." This type is characterised by tall stature, long narrow skull, light complexion with light hair and eyes[1215].

During the age of national migrations, from the fourth to the sixth century, the territories of the Nordic peoples were vastly extended, partly by conquest, and partly by arrangement with the Romans. But these movements had begun before the new era, for we hear of the _Cimbri_ invading Illyricum, Gaul and Italy in the second century B.C. probably from Jutland[1216], where they were apparently associated with the _Teutoni_. Still earlier, in the third century B.C., the _Bastarnae_, said by many ancient writers to have been Teutonic in origin, invaded and settled between the Carpathians and the Black Sea. Already mentioned doubtfully by Strabo as separating the Germani from the Scythians (Tyragetes) about the Dniester and Dnieper, their movements may now be followed by authentic documents from the Baltic to the Euxine. Furtwaengler[1217] shows that the earliest known German figures are those of the Adamklissi monument, in the Dobruja, commemorating the victory of Crassus over the Bastarnae, Getae, and Thracians in 28 B.C. The Bastarnae migrated before the Cimbri and Teutons through the Vistula valley to the Lower Danube about 200 B.C. They had relations with the Macedonians, and the successes of Mithridates over the Romans were due to their aid. The account of their overthrow by Crassus in Dio Cassius is in striking accord with the scenes on the Adamklissi monument. Here they appear dressed only in a kind of trowsers, with long pointed beards, and defiant but noble features. The same type recurs both on the column of Trajan, who engaged them as auxiliaries in his Dacian wars, and on the Arch of Marcus Aurelius, here however wearing a tunic, a sign perhaps of later Roman influences. And thus after 2000 years are answered Strabo's doubts by modern archaeology.

Much later there followed along the same beaten track between the Baltic and Black Sea a section of the Goths, whom we find first settled in the Baltic lands in proximity to the Finns. The exodus from this region can scarcely have taken place before the second century of the new era, for they are still unknown to Strabo, while Tacitus locates them on the Baltic between the Elbe and the Vistula. Later Cassiodorus and others bring them from Scandinavia to the Vistula, and up that river to the Euxine and Lower Danube. Although often regarded as legendary[1218], this migration is supported by archaeological evidence. In 1837 a gold torque with a Gothic inscription was found at Petroassa in Wallachia, and in 1858 an iron spear-head with a Gothic name in the same script, which dates from the first Iron Age, turned up near Kovel in Volhynia. The spear-head is identical with one found in 1865 at Muenchenberg in Brandenburg, on which Wimmer remarks that "of 15 Runic inscriptions in Germany the two earliest occur on iron pikes. There is no doubt that the runes of the Kovel spear-head and of the ring came from Gothic tribes[1219]." These Southern Goths, later called Moeso-Goths, because they settled in Moesia (Bulgaria and Servia), had certain physical and even moral characters of the Old Teutons, as seen in the Emperor Maximinus, born in Thrace of a Goth by an Alan woman--very tall, strong, handsome, with light hair and milk-white skin[1220], temperate in all things and of great mental energy.

Before their absorption in the surrounding Bulgar and Slav populations the Moeso-Goths were evangelised in the fourth century by their bishop Ulfilas ("Wolf"), whose fragmentary translation of Scripture, preserved in the _Codex Argenteus_ of Upsala, is the most precious monument of early Teutonic speech extant.

To find the pure Nordic type at the present day we must seek for it in Scandinavia, which possesses one of the most highly individualised populations in Europe. The Osterdal, and the neighbourhood of Vaage in Upper Gudbrandsdal in Norway, and the Dalarna district in Sweden contain perhaps the purest Teutonic type in all Europe, the cephalic index falling well below 78. But along the Norwegian coasts there is a strong tendency to brachycephaly (the index rising to 82-3), combined with a darkening of the hair and eye colour (the type occurs also in Denmark), indicating an outlying lodgement of the Alpine race from Central Europe. The anthropological history of Scandinavia, according to Ripley, is as follows: "Norway has ... probably been peopled from two directions, one element coming from Sweden and another from the south by way of Denmark. The latter type, now found on the sea coast and especially along the least attractive portion of it, has been closely hemmed in by the Teutonic immigration from Sweden[1221]." Brachycephalic people already occupied parts of Denmark in the Stone Age[1222], and, according to the scanty information available, the present population is extremely mixed. One-third of the children have light hair and light eyes, and tall stature coincides in the main with fair colouring, but in Bornholm where the cephalic index is 80 there is a taller dark type and a shorter light type, the latter perhaps akin to the Eastern variety of the Alpine race[1223].

The original Nordic type is by no means universally represented among the present Germanic peoples. From the examination made some years ago of 6,758,000 school children[1224], it would appear that about 31 per cent. of living Germans may be classed as blonds, 14 as brunettes, and 55 as mixed; and further that of the blonds about 43 per cent. are centred in North, 33 in Central and 24 in South Germany. The brunettes increase, generally speaking, southwards, South Bavaria showing only about 14 per cent. of blonds, and the same law holds good of the long-heads and the round-heads respectively. To what cause is to be attributed this profound modification of this branch of the Nordic type in the direction of the south?

That the Teutons ranged in considerable numbers far beyond their northern seats is proved by the spread of the German language to the central highlands, and beyond them down the southern slopes, where a rude High German dialect lingered on in the so-called "Seven Communes" of the Veronese district far into the nineteenth century. But after passing the Main, which appears to have long formed the ethnical divide for Central Europe, they entered the zone of the brown Alpine round-heads[1225], to whom they communicated their speech, but by whom they were largely modified in physical appearance. The process has for long ages been much the same everywhere--perennial streams of Teutonism setting steadily from the north, all successively submerged in the great ocean of dark round-headed humanity, which under many names has occupied the central uplands and eastern plains since the Neolithic Age, overflowing also in later times into the Balkan Peninsula.

This absorption of what is assumed to be the superior in the inferior type, may be due to the conditions of the general movement--warlike bands, accompanied by few women, appearing as conquerors in the midst of the Alpines and merging with them in the great mass of brachycephalic peoples. Or is the transformation to be explained by de Lapouge's doctrine, that cranial forms are not so much a question of race as of social conditions, and that, owing to the increasingly unfavourable nature of these conditions, there is a general tendency for the superior long-heads to be absorbed in the inferior round-heads[1226].

The fact that dolichocephaly is more prevalent in cities and brachycephaly in rural areas has been interpreted in various ways. De Lapouge[1227] contended that in France the restless and more enterprising long-heads migrated from the rural districts in disproportionate numbers to the towns, where they died out. For the department of Aveyron he gives a table showing a steady rise of the cephalic index from 71.4 in prehistoric times to 86.5 in 1899, and attributes this to the dolichos gravitating chiefly to the large towns, as O. Ammon has also shown for Baden. L. Laloy summed up the results thus: France is being depopulated, and, what is worse, it is precisely the best section of the inhabitants that disappears, the section most productive in eminent men in all departments of learning, while the ignorant and rude _pecus_ alone increase.

These views have met with favour even across the Atlantic, but are by no means universally accepted. The ground seems cut from the whole theory by A. Macalister, who read a paper at the Toronto Meeting of the British Association, 1897, on "The Causes of Brachycephaly," showing that the infantile and primitive skull is relatively long, and that there is a gradual change, phylogenetic (racial) as well as ontogenetic (individual) toward brachycephaly, which is certainly correlated with, and is apparently produced by, cerebral activity and growth; in the process of development in the individual and the race the frontal lobes of the brain grow the more rapidly and tend to fill out and broaden the skull[1228]. The tendency would thus have nothing to do with rustic and urban life, nor would the round be necessarily, if at all, inferior to the long head. Some of de Lapouge's generalisations are also traversed by Livi[1229], Deniker[1230], Sergi[1231] and others, and the whole question is admirably summarised by W. Z. Ripley[1232].

But whatever be the cause, the fact must be accepted that _Homo Europaeus_ (the Nordics) becomes merged southwards in _Homo Alpinus_ whose names, as stated, are many. Broca and many continental writers use the name _Kelt_ or _Slavo-Kelt_, which has led to much confusion. But it merely means for them the great mass of brachycephalic peoples in Central Europe, where, at various times, Celtic and Slavonic languages have prevailed.

It is remarkable that in the Alpine region, especially Tyrol, where the brachy element comes to a focus, there is a peculiar form of round-head which has greatly puzzled de Lapouge, but may perhaps be accounted for on the hypothesis of two brachy types here fused in one. To explain the exceedingly round Tyrolese head, which shows affinities on the one hand with the Swiss, on the other with the Illyrian and Albanian, that is, with the normal Alpine, a Mongol strain has been suggested, but is rightly rejected by Franz Tappeiner as inadmissible on many grounds[1233]. De Ujfalvy[1234], a follower of de Lapouge, looks on the hyperbrachy Tyrolese as descendants of the ancient Rhaetians or Rasenes, whom so many regard as the parent stock of the Etruscans.

But Montelius (with most other modern ethnologists) rejects the land route from the north, and brings the Etruscans by the sea route direct from the Aegean and Lydia (Asia Minor). They are the Thessalian Pelasgians whom Hellanikos of Lesbos brings to Campania, or the Tyrrhenian Pelasgians transported by Antiklides from Asia Minor to Etruria, and he is "quite sure that the archaeological facts in Central and North Italy ... prove the truth of this tradition[1235]." Of course, until the affinities of the Etruscan language are determined, from which we are still as far off as ever[1236], Etruscan origins must remain chiefly an archaeological question. Even the help afforded by the crania from the Etruscan tombs is but slight, both long and round heads being here found in the closest association. Sergi, who also brings the Etruscans from the east, explains this by supposing that, being Pelasgians, they were of the same dolicho Mediterranean stock as the Italians (Ligurians) themselves, and differed only from the brachy Umbrians of Aryan speech. Hence the skulls from the tombs are of two types, the intruding Aryan, and the Mediterranean, the latter, whether representing native Ligurians or intruding Etruscans, being indistinguishable. "I can show," he says, "Etruscan crania, which differ in no respect from the Italian [Ligurian], from the oldest graves, as I can also show heads from the Etruscan graves which do not differ from those still found in Aryan lands, whether Slav, Keltic, or Germanic[1237]." Perhaps the difficulty is best explained by Feist's suggestion that the Etruscans were merely a highly civilised warlike aristocracy, spreading thinly over the conquered population by which they were ultimately absorbed[1238].

The migrations of the Celts preceded those of the Teutonic peoples to whom they were probably closely related in race as in language[1239]. At the beginning of the historical period Celts are found in the west of Germany in the region of the Rhine and the Weser. Possibly about 600 B.C. they occupied Gaul and parts of the Iberian peninsula, subsequently crossing over into the British Isles. In Italy they came into conflict with the rising power of Rome, and, after the battle of the Allia (390 B.C.) occupied Rome itself. Descents were also made into the Danube valley and the Balkans, and later (280 B.C.) into Thessaly. At the height of their power they extended from the north of Scotland to the southern shores of Spain and Portugal, and from the northern coasts of Germany to a little south of Senegaglia. To the west their boundary was the Atlantic, to the east, the Black Sea[1240].

Unfortunately the indiscriminate use of the term Celt has led to much confusion. For historians and geographers the Celts are the people in the centre and west of Europe referred to by writers of antiquity under the names of _Keltoi_, _Celtae_, _Galli_ and _Galatae_. But many anthropologists, especially on the continent, regard Celts and Gauls as representing two well-determined physical types, the former brachycephalic, with short sturdy build and chestnut coloured hair (Alpine type), and the latter dolichocephalic with tall stature, fair complexion and light hair (Nordic type). Linguists, ignoring physical characters, class as Celts those people who speak an Indo-European language characterised in particular by the loss of p and by the modifications undergone by mutation of initial consonants, while for many archaeologists the Celts were the people responsible for the spread of the civilisation of the Hallstatt and La Tene periods, that is of the earlier and later Iron Age[1241].

It is not surprising therefore that it has been proposed to drop the word Celt out of anthropological nomenclature, as having no ethnical significance. But this, says Rice Holmes[1242], "is because writers on ethnology have not kept their heads clear." And in particular one point has been overlooked. "Just as the French are called after one conquering people, the Franks; just as the English are called after one conquering people, the Angles; so the heterogeneous Celtae of Transalpine Gaul were called after one conquering people; and that people were the Celts, or rather a branch of the Celts in the true sense of the word. The Celts, in short, were the people who introduced the Celtic language into Gaul, into Asia Minor, and into Britain; the people who included the victors of the Allia, the conquerors of Gallia Celtica, and the conquerors of Gallia Bel['g]ica; the people whom Polybius called indifferently Gauls and Celts; the people who, as Pausanius said, were originally called Celts and afterwards called Gauls. If certain ancient writers confounded the tall fair Celts who spoke Celtic with the tall fair Germans who spoke German the ancient writers who were better informed avoided such a mistake.... Let us therefore restore to the word 'Celt' the ethnical significance which of right belongs to it."

It is not certain at what date the Celtic tribes effected settlements in Great Britain, but it is held by many that the earliest invasions were not prior to the sixth or possibly even the fifth century. At the time of the Roman conquest the Celts were divided into two linguistic groups, _Goidelic_, represented at the present day by Irish, Manx and Scotch Gaelic, and _Brythonic_, including Welsh, Cornish and Breton. These groups must have been virtually identical save in two particulars. In Brythonic the labial velar q became p (a change which apparently took place before the time of Pytheas), whilst in Goidelic the sound remained unaltered. q is retained in the earlier ogham inscriptions, but by the end of the seventh century it had lost the labial element, appearing in Old Irish as c. Thus O. Irish _cenn_, head, as in Kenmare, Kintyre, Kinsale, equates with Brythonic _pen_, as in Penryn (Cornwall), Penrhyn (Wales), Penkridge (Staffordshire), Penruddock, Penrith and many others. The two groups are therefore distinguished as the Q Celts and the P Celts[1243]. From the fact that Goidelic retained the q it has been commonly assumed that the Goidels were separated from the main Celtic stock at a time before the labialisation had taken place, but many scholars maintain that the parent Goidelic was evolved in Ireland, and was carried from that island to Man and Scotland in the early centuries of our era[1244].

From an anthropological point of view, the Picts are if possible more difficult to identify than the Celts. But the question is not between tall fair long-heads and short dark round-heads, but between short dark long-heads (neolithic aborigines) and Celts. The Pictish question is summed up by Rice Holmes[1245] and the various theories have been more recently reviewed by Windisch[1246] giving a valuable summary of earlier writings. On the one hand it is maintained as "the most tenable hypothesis that the Picts were non-Aryans, whom the first Celtic migrations found already settled here ... descendants of the Aborigines[1247]." Windisch[1248] at the other extreme, regards them as late comers into North Britain, when Scotland was already occupied by Brythonic tribes. But the geographical distribution of the Picts in historical times suggests rather a people driven into mountainous regions by successive conquerors, than the settlements of successful invaders. Also it is not improbable that the language of the Bronze Age lingered in these wilder districts, and this would account for the fact that St Columba had to employ an interpreter in his relations with the Picts; though this is explained by others on the assumption that Pictish was Brythonic. The linguistic evidence is however extremely slight, only a few words presumably Pictish having survived and these through Celtic writers. "The one absolutely certain conclusion to which the student of ethnology can come is that the name of the Picts has not been proved to be of pre-Aryan origin[1249]." "For me," continues Rice Holmes (p. 417), "the Picts were a mixed people comprising descendants of the neolithic aborigines, of the Round Barrow Race, and of the Celtic invaders--a mixed people who [or at least whose aristocracy] spoke a Celtic dialect."

Before attempting a survey of the ethnology of Britain it is necessary to ascertain what ethnic elements the area contained before the arrival of the Celts. The neolithic inhabitants, the short, dark dolichocephals of Mediterranean type have already been described (Ch. XIII.). Their remains are associated with the characteristic forms of sepulchral monuments the dolmens and the long barrows. But towards the end of the Stone Age a brachycephalic race was already penetrating into the islands. This appears to have been a peaceful infiltration, at any rate in certain districts, where remains of the two types are found side by side and there is evidence of racial intermixture. The brachycephals introduced a new form of sepulture, making their burial mounds circular instead of elongated, whence Thurnam's convenient formula, "long barrow, long skull; round barrow, round skull." But the earlier view that there was a definite transition from long heads, neolithic culture and long barrows, to round heads, bronze culture and round barrows can no longer be maintained. "It is often taken for granted that no round barrows were erected in Britain before the close of the Neolithic Age, and that the earliest of the brachycephalic invaders whose remains have been found in them landed with bronze weapons in their hands[1250]." But there is abundant evidence that the brachycephalic element preceded the knowledge of metals, and a number of round barrows in Yorkshire and further north show no trace of bronze.

Nevertheless the majority of the round barrows belong to the Bronze Age, and the physical type of their builders is sufficiently well marked. The stature is remarkably tall, attaining a height of 1.763 m. or over 5 ft. 9 ins. The skull is brachycephalic with an average index of about 80. It is also characterised by great strength and ruggedness of outline, with (often) a sloping forehead, prominent supraciliary ridges, and a certain degree of prognathism.

According to Rolleston's description "The eyebrows must have given a beetling and probably even formidable appearance to the upper part of the face, whilst the boldly outstanding and heavy cheekbones must have produced an impression of raw and rough strength. Overhung at its root, the nose must have projected boldly forward." And Thurnam adds "the prominence of the large incisor and canine teeth is so great as to give an almost bestial expression to the skull[1251]."

Although this type is conveniently called the Round Barrow type, or even the Round Barrow Race, the round barrows also contain remains of a different racial character. The skull form shows a more extreme brachycephaly, with an index of 84 or 85, and exhibits none of the rugged features associated with the true Round Barrow type. On the contrary, of the two typical groups, one from round barrows in Glamorganshire, and the other from short cists in Aberdeenshire not one of the skulls is prognathous, the supraciliary ridges are but slightly developed, the cheek bones are not prominent, the face is both broad and short and the lower jaw is small. But the greatest contrast is in the height, which averages in the two groups, 1.664 m. and 1.6 m. respectively, _i.e._ 5 ft. 5-3/4 ins. and 5 ft. 3 ins. All these characters connect this type closely with the Alpine type on the continent.

These round-headed peoples have been the subject of much discussion ably summarised and criticised by Rice Holmes, whose conclusion perhaps best represents the view now taken of their affinities and origins.

"The great mistake that has been made in discussing the question is the not uncommon assumption that the brachycephalic immigrants who buried their dead in round barrows arrived in Britain at one time, and came from one place. Some of them certainly appeared before the end of the Neolithic Age: others may have introduced bronze implements or ornaments; others doubtless came, in successive hordes, during the course of the Bronze Age. Some of those who belonged to the Grenelle race [Alpine type], who certainly came from Eastern Europe and possibly from Asia, and whose centre of dispersion was the Alpine region, may have started from Gaul; others could have traced their origin to some Rhenish tribe; and I am inclined to believe that those who belonged to the characteristic rugged Round Barrow type crossed over, for the most part, from Denmark or the out-lying islands[1252]."

After the passage of the Romans, who mingled little with the aborigines and made, perhaps, but slight impression on the speech or type of the British populations, a great transformation was effected in these respects by the arrival of the historical Teutonic tribes. Hand in hand with the Teutonic invasions went a lust for expansion on the part of the peoples in Ireland. Settlements were effected by them in South Wales and Anglesey, the Isle of Man and Argyll, probably also in North Devon and Cornwall. For many generations the south and east of England were the scenes of fierce struggles, during which the Romano-British civilisation perished. Only in more inaccessible districts, such as the fen country, may a British population have survived, though Celtic languages are not yet dislodged from their mountain strongholds in Wales and Scotland, and lingered for many centuries in Strathclyde and Cornwall. After the strengthening of the Teutonic element by the arrival of the Scandinavians and Normans, all very much of the same physical type, no serious accessions were made to this composite ethnical group, which on the east side ranged uninterruptedly from the Channel to the Grampians. Later the expansion was continued northwards beyond the Grampians, and westwards through Strathclyde to Ireland, while now the spread of education and the development of the industries are already threatening to absorb the last strongholds of Celtic speech in Wales, the Highlands, and Ireland.

Thanks to its isolation in the extreme west, Ireland had been left untouched by some of the above described ethnical movements. It is doubtful whether Palaeolithic man ever reached this region, and but few even of the round-heads ranged so far west during the Bronze Age[1253]. The land oscillations during post-Glacial times appear to have been practically identical over an area including northern Ireland, the southern half of Scotland, and northern England. There was a period of depression followed by one of elevation. The Larne beach-deposits prove that Neolithic man was in existence from almost the beginning of the deposition of that series until after its conclusion. The estuarine clays of Belfast Lough correspond to the depression, and the Neolithic period extended from at least near the top of the lower estuarine clay to the beach-deposit of yellow sand which overlies it, or possibly till later. It is to this period of elevation that the Neolithic sites among the sand dunes of North Ireland belong; those of Whitepark Bay and Portstewart, for example, extend to the maximum elevation. A slight movement of subsidence of about five feet in recent times has left the surface as we now find it. The implements found in the Larne gravels correspond to some extent with those of Danish kitchen-middens; this was not a dwelling site but a quarry-shop or roughing-out place, the serviceable flakes being taken away for further manipulation; it thus belongs to the earliest phase of neolithic times. The sandhill sites were occupied, continuously and occasionally, during neolithic times, through the Bronze Age, and into the Iron and Christian periods[1254]. Nina F. Layard has recently studied the Larne raised beach and exposed a new section. She states that "Taken as a whole the flints certainly do not correspond at all closely either to the Palaeoliths or Neoliths so far found in England.... Some are strongly reminiscent of well-known drift type.... Again, there are shapes that bear a closer resemblance to some of the earliest Neolithic types[1255]." She believes that, from their rolled condition, they were derived from another source.

F. J. Bigger[1256] described some kitchen-middens at Portnafeadog, near Roundstone, Connemara, which yielded stone hammers but no worked flints, pottery or metal-ware. The chief interest of this paper is due to the fact that it is the first record of the occurrence of vast quantities of the shells of _Purpura lapillus_, all of which were broken in such a manner that the animal could easily be extracted. There can be no doubt that the purple dye was manufactured here in prehistoric times[1257]. W. J. Knowles[1258] suggests from the close resemblance--in fact identity--of a great number of neolithic objects in Ireland with palaeolithic forms in France (Saint-Acheul, Moustier, Solutre, La Madeleine types), that the Irish objects bridge over the gap between the two ages, and were worked by tribes from the continent following the migration of the reindeer northwards. These peoples may have continued to make tools of palaeolithic types, while at the same time coming under the influence of the neolithic culture gradually arriving from some southern region. The astonishing development of this neolithic culture in the remote island on the confines of the west, as illustrated in W. C. Borlase's sumptuous volumes[1259], is a perpetual wonder, but is rendered less inexplicable if we assume an immense duration of the New Stone Age in the British Isles. The Irish dolmen-builders were presumably of the same long-headed stock as those of Britain[1260], and they were followed by Celtic-speaking Goidels who may have come directly from the continent[1261], and there is evidence in Ptolemy and elsewhere of the presence of Brythonic tribes from Gaul in the east. Since these early historic times the intruders have been almost exclusively of Teutonic race, and Viking invaders from Norway and Denmark founded the earliest towns such as Dublin, Waterford and Limerick. Now all alike, save for an almost insignificant and rapidly dwindling minority, have assumed the speech of the English and Lowland Scotch intruders, who began to arrive late in the 12th century, and are now chiefly massed in Ulster, Leinster, and all the large towns. The rich and highly poetic Irish language has a copious medieval literature of the utmost importance to students of European origins.

In Scotland few ethnical changes or displacements have occurred since the colonisation of portions of the west by Gaelic-speaking Scottic tribes from Ireland, and the English (Angle) occupation of the Lothians. The Grampians have during historic times formed the main ethnical divide between the two elements, and brooklets which can be taken at a leap are shown where the opposite banks have for hundreds of years been respectively held by formerly hostile, but now friendly communities of Gaelic and broad Scotch speech. Here the chief intruders have been Scandinavians, whose descendants may still be recognised in Caithness, the Hebrides, and the Orkney and Shetland groups. Faint echoes of the old Norrena tongue are said still to linger amongst the sturdy Shetlanders, whose assimilation to the dominant race began only after their transfer from Norway to the Crown of Scotland.

Since 1901 the researches of Gray and Tocher[1262] on the pigmentation of some 500,000 school children of Scotland have increased our information as to racial distribution. The average percentage of boys with fair hair is nearly 25 for the whole of the country, and when this is compared with 82 in Schleswig Holstein "we are driven to the conclusion that the pure Norse or Anglo-Saxon element in our population is by no means predominant. There is evidently also a dark or brunette element which is at least equal in amount and probably greater than that of the Norse element" (p. 380). Pure blue eyes for the whole of Scotland average 14.7 per cent., which may be compared with 42.9 in Prussia. The greatest density for fair hair and eyes is to be found in the great river valleys opening on to the German Ocean, and also in the Western Isles. The Tweed, Forth, Tay and Don all show indications of settlements of a blonde race "probably due to Anglo-Saxon invasions," but the maximum is to be found at the mouth of the Spey. The high percentage here and in the Hebrides and opposite coasts, the authors trace to Viking invasions. The percentage of dark hair for boys and girls is 25.2 as compared with 1.3 in Prussian school children, the maximum density as we should expect being in the west. Jet black hair (1.2%) has its maximum density in the central highlands and wild west coast. Beddoe[1263] commenting on Gray and Tocher's results calculates an even higher percentage of black hair (over 2%) "either within or astride of the Highland frontier. Except Paisley, there is not a single instance south of the Forth, nor one between the Spey and the Firth of Tay. Surely there is something 'racial' here." Beddoe's map, constructed from Gray and Tocher's statistics, clearly indicates the distribution of racial types.

The work carried on in Wales for a number of years by H. J. Fleure and T. C. James[1264] has produced some extremely interesting results. The chief types (based on measurements and observations of head, face, nose, skin, hair and eye colour, stature, etc.) fall into the following groups.

1. "The fundamental type is certainly the long-headed brunet of the moorlands and their inland valleys. He is universally recognised as belonging to the Mediterranean race of Sergi and as dating back in this country to early Neolithic times." The cephalic index is about 78, with high colouring, dark hair and eyes, and stature rather below the average. A possible mixture of earlier stocks is shown in a longer-headed type (c.i. about 75), with well-marked occiput, very dark hair and eyes, swarthy complexion, and average stature (about 1690 mm. = 5 ft. 6-1/2 ins.). Occasionally in North Wales the occurrence of lank black hair, a sallow complexion and prominent cheekbones suggests a "Mongoloid" type; and a type with small stature, black, closely curled hair and a rather broad nose has negroid reminiscences. The Plynlymon moorlands contain a "nest" of extreme dolichocephaly and an unusually high percentage of red hair.

2. Nordic-Alpine type, with cephalic index mainly between 76 and 81. This group includes (_a_) a "local version of the Nordic type" occurring at Newcastle Emlyn and in South and South-West Pembrokeshire with fair hair and eyes, usually tall stature and great strength of brow, jaw and chin; (_b_) a heavier variant on the Welsh border, often with cephalic index above 80, and extremely tall stature; (_c_) the Borreby or Beaker-Maker type, broad-headed and short-faced with darker pigmentation, probably a cross between Alpine and Nordic, characteristic of the long cleft from Corwen _via_ Bala to Tabyllyn and Towyn.

3. Dark bullet-headed short thick-set men of the general type denoted by the term Alpine or more exactly perhaps by the term Cevenole are found, though not commonly, in North Montgomeryshire valleys.

4. Powerfully built, often intensely dark, broad-headed, broad-faced, strong and square jawed men are characteristic of the Ardudwy coast, the South Glamorgan coast, Newquay district (Cardiganshire) and elsewhere.

The authors observe that Type 1 with its variations contributes "considerable numbers to the ministries of the various churches, possibly in part from inherent and racial leanings, but partly also because these are the people of the moorlands. The idealism of such people usually expresses itself in music, poetry, literature and religion rather than in architecture, painting and plastic arts generally. They rarely have a sufficiency of material resources for the latter activities. These types also contribute a number of men to the medical profession.... The successful commercial men, who have given the Welsh their extraordinarily prominent place in British trade (shipping firms for example) usually belong to types 2 or 4, rather than to 1, as also do the majority of Welsh members of Parliament, though there are exceptions of the first importance. The Nordic type is marked by ingenuity and enterprise in striking out new lines. Type 2 (_c_) in Wales is remarkable for governmental ability of the administrative kind as well as for independence of thought and critical power" (p. 119).

We have now all the elements needed to unravel the ethnical tangle of the present inhabitants of the British Isles. The astonishing prevalence everywhere of the moderately dolicho heads is at once explained by the absence of brachy immigrants except in the Bronze period, and these could do no more than raise the cephalic index from about 70 or 72 to the present mean of about 78. With the other perhaps less stable characters the case is not always quite so simple. The brunettes, representing the Mediterranean type, certainly increase, as we should expect, from north-east to south-west, though even here there is a considerable dark patch, due to local causes, in the home shires about London[1265]. But the stature, almost everywhere a troublesome factor, seems to wander somewhat lawlessly over the land.

Although a short stature more or less coincides with brunetteness in England and Wales, and the observations in Ireland are too few to be relied on, no such parallelism can be traced in Scotland. The west (Inverness and Argyllshire), though as dark as South Wales, shows an average stature of 1.73 m. to 1.74 m. (5 ft. 8 ins. to 5 ft. 8-1/2 ins.), which is higher than the average for the whole of Britain. And South-west Scotland, where the type is fairly dark, contains the tallest population in Europe, if not in the world. Ripley suggests either that "some ethnic element of which no pure trace remains, served to increase the stature of the western Highlanders without at the same time conducing to blondness; or else some local influences of natural selection or environment are responsible for it[1266]"; and he hints also that the linguistic distinction between Gaels and Brythons may have been associated with physical variation.

The English tongue need not detain us long. Its qualities, illustrated in the noblest of all literatures, are patent to the world[1267], indeed have earned for it from Jacob Grimm the title of _Welt-Sprache_, the "World Speech." It belongs, as might be anticipated from the northern origin of the Teutonic element in Britain, to the Low German division of the Teutonic branch of the Aryan family. Despite extreme pressure from Norman French, continued for over 200 years (1066-1300), it has remained faithful to this connection in its inner structure, which reveals not a trace of Neo-Latin influences. The phonetic system has undergone profound changes, which can be only indirectly and to a small extent due to French action. What English owes to French and Latin is a very large number, many thousands, of words, some superadded to, some superseding their Saxon equivalents, but altogether immensely increasing its wealth of expression, while giving it a transitional position between the somewhat sharply contrasted Germanic and Romance worlds.

Amongst the Romance peoples, that is, the French, Spaniards, Portuguese, Italians, Rumanians, many Swiss and Belgians, who were entirely assimilated in speech and largely in their civil institutions to their Roman masters, the paramount position, a sort of international hegemony, has been taken by the French nation since the decadence of Spain under the feeble successors of Philip II. The constituent elements of these Gallo-Romans, as they may be called, are much the same as those of the British peoples, but differ in their distribution and relative proportions. Thus the Iberians (Aquitani, Pictones, and later Vascones), who may perhaps be identified with the neolithic long-heads[1268], do not appear ever to have ranged much farther north than Brittany, and were Aryanised in pre-Roman times by the P-speaking Celts everywhere north of the Garonne. The prehistoric Teutons again, who had advanced beyond the Rhine at an early period (Caesar says _antiquitus_) into the present Belgium, were mainly confined to the northern provinces. Even the historic Teutons (chiefly Franks and Burgundians) penetrated little beyond the Seine in the north and the present Burgundy in the east, while the Vandals, Visigoths and a few others passed rapidly through to Iberia beyond the Pyrenees.

Thus the greater part of the land, say from the Seine-Marne basin to the Mediterranean, continued to be held by the Romanised mass of Alpine type throughout all the central and most of the southern provinces, and elsewhere in the south by the Romanised long-headed Mediterranean type. This great preponderance of the Romanised Alpine masses explains the rapid absorption of the Teutonic intruders, who were all, except the Fleming section of the Belgae, completely assimilated to the Gallo-Romans before the close of the tenth century. It also explains the perhaps still more remarkable fact that the Norsemen who settled (912) under Rollo in Normandy were all practically Frenchmen when a few generations later they followed their Duke William to the conquest of Saxon England. Thus the only intractable groups have proved to be the Basques[1269] and the Bretons, both of whom to this day retain their speech in isolated corners of the country. With these exceptions the whole of France, save the debateable area of Alsace-Lorraine, presents in its speech a certain homogeneous character, the standard language (_langue d'oil_[1270]) being current throughout all the northern and central provinces, while it is steadily gaining upon the southern form (_langue d'oc_[1270]) still surviving in the rural districts of Limousin and Provence.

But pending a more thorough fusion of such tenacious elements as Basques, Bretons, Auvergnats, and Savoyards, we can scarcely yet speak of a common French type, but only of a common nationality. Tall stature, long skulls, fair or light brown colour, grey or blue eyes, still prevail, as might be expected, in the north, these being traits common alike to the prehistoric Belgae, the Franks of the Merovingian and Carlovingian empires, and Rollo's Norsemen. With these contrast the southern peoples of short stature, olive-brown skin, round heads, dark brown or black eyes and hair. The tendency towards uniformity has proceeded far more rapidly in the urban than in the rural districts. Hence the citizens of Paris, Lyons, Bordeaux, Marseilles and other large towns, present fewer and less striking contrasts than the natives of the old historical provinces, where are still distinguished the loquacious and mendacious Gascon, the pliant and versatile Basque, the slow and wary Norman, the dreamy and fanatical Breton, the quick and enterprising Burgundian, and the bright, intelligent, more even-tempered native of Touraine, a typical Frenchman occupying the heart of the land, and holding, as it were, the balance between all the surrounding elements.

In Spain and Portugal we have again the same ethnological elements, but also again in different proportions and differently distributed, with others superadded--proto-Phoenicians and later Phoenicians (Carthaginians), Romans, Visigoths, Vandals, and still later Berbers and Arabs. Here the Celtic-speaking mixed peoples mingled in prehistoric times with the long-headed Mediterraneans, an ethnical fusion known to the ancients, who labelled it "Keltiberian[1271]." But, as in Britain, the other intruders were mostly long-heads, with the striking result that the Peninsula presents to-day exactly the same uniform cranial type as the British Isles. Even the range (76 to 79) and the mean (78) of the cephalic index are the same, rising in Spain to 80 only in the Basque corner. As Ripley states, "the average cephalic index of 78 occurs nowhere else so uniformly distributed in Europe" except in Norway, and this uniformity "is the concomitant and index of two relatively pure, albeit widely different, ethnic types--Mediterranean in Spain, Teutonic in Norway[1272]."