RENDALL
THE CRADLE OF THE ARYANS
OL
N

THE UNIVERS
JERSITY OF M
T
CO
CIENTIA
A·VER!
GAN.
IGAN
"LIBRARIES ·
OF MIC
"
1
:
THE
CRADLE OF THE ARYANS.
BY
GERALD H RENDALL, M.A.,
LATE FELLOW OF TRINITY COLLEGE, CAMBRIDGE,
PRINCIPAL AND PROFESSOR OF GREEK, UNIVERSITY COLLEGE, LIVERPOOL.
London:
MACMILLAN AND CO.
AND NEW YORK.
1889.
The Right of Translation and Reproduction is Reserved.
GN
539
R397
011 380
006
५
PREFACE.
THE following Essay is a paper originally communicated,
necessarily in abridged form, to the members of the Liver-
pool Literary and Philosophical Society, and is an attempt
to give a clear account of present controversies regarding
the origin, local and racial, of the Aryan stock. It makes
no claim to original discoveries in the field of language;
but it embodies the results of somewhat extensive reading,
mainly in German authorities; it gives a much fuller account
than as yet exists in English of the latest Continental
theories; and it presses some conclusions and inferences to
what appears to me their legitimate issue, more closely than
has been done by previous writers. I give it separate
publication, thinking it may be of use to other students in
the field of language. The better to fulfil this end, in the
sections dealing with vocabulary I have systematically
inserted in brackets some familiar representative of the term
discussed; where this is not done, it is to be understood that
the English word itself represents the stem in question.
This, with the comparative tables introduced, will make it
easy to check etymological statements, without uncertainty
and loss of time. To gain compactness, I have refrained
from not a few tempting digressions, and confined myself to
categorical statement, with or without supporting reference
in the notes. I have not needlessly multiplied citations,
423966
iv
but every one acquainted with the literature of the subject
will perceive my obligations to Max Müller, especially in
his Biographies of Words and The Home of the Aryas;
to Penka, Origines Ariaca and Herkunft der Arier; to
Schrader Sprachvergleichung und Urgeschichte; and in
somewhat less degree to Spiegel Die Arische Periode, I
hope no material point advanced by Benfey, Geiger, Hehn,
Lenormant, Van den Gheyn, or among English writers by
Latham, Sayce, Whitney and others, has escaped me. Much
of Pictet is by general consent obsolete. For comparative
etymology I have leaned, as must every student, on Fick's
Vergleichendes Wörterbuch der Indo-Germanischen Sprachen,
using the 3rd ed. (1876), and checking results by the useful
lists in Max Müller's Biographies of Words, and by the
stricter canons of modern etymology.
TABLE OF CONTENTS.
I.
THE INDO-EUROPEANS AND THEIR HABITAT
...
II. INDO-EUROPEAN CIVILISATION
...
...
III. CLIMATE, FAUNA AND FLORA...
IV. PRIMITIVENESS OF TYPE
:
...
V.
DIFFERENTIATION OF STOCKS
:
VI. HISTORY
VII. LANGUAGE AND RACE ...
VIII. CONCLUSION
::
PAGE
7
9
14
...
23
28
36
...
...
39
62
:
:
THE CRADLE OF THE ARYANS.
I. The Indo-Europeans and their Habitat.
WHEN, in the closing years of the last century, Sir W.
Jones discovered to the Western world the secret of
Sanskrit and taught them that in the far East, the accepted
cradle of the human race, there existed a language of
immemorial antiquity, the germs of which showed unmis-
takable identity with European speech at large, it seemed for
a moment as though the primitive, perhaps the universal,
language of mankind had been unearthed, and India was its
home. Study, analysis, and reflection soon dispelled the
dream; and the true character of the great Indo-European
unity, which Schlegel was perhaps the first to apprehend,
was established by the genius and industry of Bopp, who
mapped out for all time the broad general demarcations
which define the separate groups of closely correlated
families of speech. Students of language have ever since.
that time been busily engaged in cross-questioning the
evidence of language, and trying to extort thence the secret
of the prehistoric movements, culture, and dwelling-places of
Indo-European man.
The first crude hypothesis of an Indian origin was soon
found untenable. It was exchanged for the theory, pro-
pounded by Pott, and brilliantly elaborated by Pictet, that
the true home of the stock was to be found in the Bactrian
plains of Central Asia; and this general theory, in forms
variously modified, may be said to have held the field with-
out serious dispute until quite recent years. The theory had
much to commend it. It gave a compact and simple scheme
8
THE CRADLE OF THE ARYANS.
to philologists, already wedded on the strength of historical
and biblical* prepossessions to the assumption of an Eastern
descent; it brought the Eranian and Indian stocks, whose
languages seemed to stand closest to the parent speech,
locally nearest to the cradle of the race; it assumed the con-
tinuous migration westward, which, while admitting simple
and intelligible diagrammatical presentment, tallied agreeably
with accredited beliefs; it could be brought without much
violence to explain the seeming ratio of divergence shown by
the derivative languages from the parent stock; and it con-
veniently located the primitive Aryan in districts inaccessible
and unknown. It is still maintained in general outline by
such veteran authorities as Fick, Hehn, Kiepert, and Max
Müller, as well as by a crowd of lesser lights; but within the
last ten years the new heresy that advocates a European
origin has everywhere gained ground, and now threatens to
chase its adversary from the field. The somewhat random
guess first hazarded by Latham † in 1854, can no longer‡ be
set aside as the fancy of a hare-brained Englishman of "the
land of curiosities," or the adopted whim of a Göttingen
Professor. One may smile at eccentricities or extravagances
of argument in Geiger or Cuno, or at Pösche's strange pre-
ference for the Rokitno swamps traversed by the Dnieper,
the Beresina, and the Pripet; but it is impossible to set
aside such names as Benfey, Spiegel, Schrader, or in our
own language, Whitney, Sayce, and Taylor. Members of
this Society may welcome a clear account of the controversy
as it now stands.
The Semitic tradition is valueless for the determination of Aryan
descent, but how strong a hold it still has may be seen in Van den Gheyn,
L'Origine Europ. des Aryas, p. 17-19.
First, in his Native Races of the Russian Empire, 1854; more
deliberately in Elements of Comparative Philology (1862).
As in 1874 by Hehn, Culturpflanzen u. Hausthiere, referring to Latham
and Benfey, and in a subsequent sentence to Geiger. The 1883 ed., 424 pp.,
treats the problem in a different tone.
i
THE CRADLE OF THE ARYANS.
9
II. Indo-European Civilisation.
Indecisive as for our purpose the results will be, it seems
needful to rehearse the inferences gained from language
regarding Indo-European culture prior to the separation of
stocks. The material has been thrashed by many flails, *
and brief recapitulation will suffice. Between the dashing
constructiveness of Pictet, and the guarded scepticism of
Hehn, there is a wide gulf; I shall content myself with
results that seem secure. It will certainly be right to
accept Max Müller's criterion that to vindicate its claim to
Indo-European rights a word must find representation in at
least one section of the Asiatic and one of the European
groups, a distinction the validity of which Spiegel's Die
Arische Periode has so irrefragably enforced. Over and
above this the strictest canons demand-(1) that there should
be correspondence between suffix as well as root, word-
formation, that is to say, as well as derivation; otherwise
mere prominence of some defining attribute may be miscon-
strued into a guarantee for its exemplification in some form
really belonging to a later stage; e.g. Sk. ragatá beside
apy-upos is a poor warrant for the Indo-European acquaint-
ance with silver. (2) Even complete correspondences are
sometimes insecure, and must never be pushed to implica-
tions belonging to a later period. Sk. puri by móλis will
prove only rude mound embankments, not the construction
of walled cities, which are unknown to the Vedas; still less,
to take verbal roots, does community in the root VAP
prove weaving with the loom, or NA the invention of the
needle. (3) Full allowance must be made for borrowed
terms, and what is hardly less important, the borrowed
application of existing terms. This caution applies specially
* For English statements see in particular Sayce, Introduction to Science
of Language; Lenormant, Manual of Oriental History.
10
THE CRADLE OF THE ARYANS.
in the consideration of implements, maritime, for instance,
or agricultural. Such safeguards minimise necessarily the
range allowed to Indo-European civilisation, and must be
balanced by a frank recognition that deficient evidence can
never prove ignorance or absence of the idea or thing de-
siderated. Because no common term has chanced to survive
for river, or butter, or egg, or the human skin, no one would
argue that any one of these was either unknown or unnamed.
Building on such canons of evidence, and omitting much
that is unessential to the present enquiry, the following
results seem secure. At the Indo-European stage of lan-
guage, man had advanced beyond the savage predatory stage,
which depends for existence on the bounties of nature and
the spoils of river, plain, and forest. He was more than
hunter and fisher. Animals were already distinguished into
wild and tame. Already the sheep, the dog, par excellence
the cow, and in all reasonable probability the horse, the pig,
and the goat* had become the friends of man. To these
the Eastern section, before their differentiation, added the
mule, the ass, and possibly the camel, and the North Euro-
pean the reindeer.
Of birds, the goose and duck were
named, and perhaps domesticated, and the Easterns knew.
and possibly prized the hen. The general cast of the
vocabulary implies (though negatively, it is true) an era
of peace, and even at the much later stage, suggested by a
comparison of Greek and Latin, there is a singular dis-
crepancy between the terms that deal with war. The few
common names for weapons may well represent nothing but
man's primitive warfare with nature and wild beasts. The
* Hehn seems hypercritical in contesting the domestication of the horse
and the goat, the more as archæology attests the fact at very early stages;
but it matters little to the general controversy. Schrader excludes the pig
and the horse from the ranks of the domesticated (Sprachvergleichung, &c.,
IV, cap. ii, with comparative table at pp. 350–1). Hehn satisfactorily traces
the ass, mule, and fowl to the east. Spiegel supplies the common terms.
THE CRADLE OF THE ARYANS.
11
*
social instinct was developed, and men clustered in commun-
ities, owning forms of government, resting apparently on a
patriarchal and monogamic basis; family relationships,
alike of blood and marriage, to the second and third degree,
had attained a precision of definition that modern civilisation
has virtually dispensed with. The mode of life was pastoral,
the people tending herds and living on their produce.
"Daughters" of the family, according to an approved etymo-
logy,t took their name from their daily work as milkers of
the herd. How far boviculture was advanced, and how
universal it was, is proved by the completeness of the voca-
bulary. Bull, ox, cow, heifer, calf, are all distinguished, and
the equation Sk, vádhri = Opis proves the systematic prac-
tice of castration. In two of the derived languages cattle
supplied the measure of wealth. Flesh was eaten roast,
marrow was gathered from the bones, and wild honey sup-
plied a drink that anticipated the more intoxicating efficacies
of wine and beer. Agriculture was in its infancy; the
culture of fruit trees is unattested, but one cereal at least
was named, and in all probability cultivated.§ Ploughing
* So Benfey infers from nóσis and Tórvia. Schrader, p. 198, demurs.
+ The derivation from DUH, suck," has rather better phonetic warrant
(Schrader, p. 195), but Max Müller, in his Biog. of Words, still upholds the
old etymology.
‡ Sk. madhu. Ale or beer is European, and the proof that wine is Indo-
European is doubtful, resting on vinum, etc., compared with Arm. gini,
Georg. ghwino. The word may be Semitic, see p. 35, as was the plant.
§ The one admitted equation is Sk. Zd. Slav. Lit. jáva
Šelá, &c.
Both Hehn and Mommsen demur to the inference of systematic agri-
culture: but it is quite unlikely that the name of any wild product would
have secured such widespread and persistent identity, and the weight
of reason lies irresistibly with Benfey and Schrader, Sprachv. u. Urge-
schichte, pp. 354–67, in vindicating a knowledge of agriculture for Indo-
European man. The cumulative evidence of the words collected under
Agriculture, in Max Müller's Biographies of Words, is decisive. As regards
Zd. gurt-ak κριθή hordeum, and Sk. sasá Zd. hahya =ña, it may
with perfect right be urged that pastoral folk would have names for fodder,
grass, and the like, prior to any deliberate agriculture.
=
=
12
THE CRADLE OF THE ARYANS.
and sowing were recognised operations; man, that is to say,
had learned that seed should be scattered and sods turned;
ruder implements at least, such as the waggon with roller
wheels and axle, and the yoke, appropriate it will be observed
to oxen as the animals for draught, the reaping-hook, and
probably the mill-stone, were at command. Further ad-
vances in agriculture are certified only for the European
section, where much, that Mommsent has worked out for
Greeks and Italians in particular, is applicable to the group
as a whole.
*
Building, weaving, planting, and navigation, however
rudimentary their forms, were numbered among the arts:
but 'building' may be limited to cave-dwellings, roofed
burrows among the rocks, or rude huts of wood or wattle;
'weaving' to mere twisted network of osier, fibre or other
raw material. For weapons, we are certified of the use of
the spear, the axe, the sword, the bow and arrow, and in the
East the club; but it is of special interest to know that
more than one name explicitly records, and none refutes, its
pre-metallic fabrication from wood, or bone, or stone. It is
interesting to find the razor § (i.e., "scraper ") already in
vogue. Copper (or, less probably, iron) had attained dis-
tinction, but neither gold nor silver can be legitimately
* Sk. malana mola, etc., the doubtful kûrna
= γρῦ granum, dhána
Slav. duna, establish a presumption in favour of Indo-European grinding,
but "
quern" is specialised from a more general term for stone, and mill
and its cognates from the generalised MAR = crush. As for reaping-hook,
λâcov and O.N. lẽ by Sk. lavi seems decisive, though äρ, S1. skupu is Euro-
pean only.
† Mommsen, Hist. Rome, ch. ii.
† So in wood kŵλov and dópʊ by Sk. dru, in bone wédekus by Sk. parasú
(rib), and in stone äкwv and äкµwv (cf. German hamar and sahs).
Denmark
§ Kshurá, ŝvpóv, to which Benfey has devoted special essays.
has yielded a razor from the early Bronze age. Montelius, Civilisation of
Sweden, p. 60.
On this see Max Müller, Biographies &c., pp. 180-1, with App. V,
252 pp.
THE CRADLE OF THE ARYANS.
13
traced back to the Indo-European stage. For science, if
the term is not too pretentious, numeration was definite up
to 100; sun, moon, and stars were observed, and the Great
Bear had received his name; broad divisions of the year
into seasons and months had gained recognition, though
language has nothing to suggest that the simple and obvious
lunar month had been adjusted to any maturer form of
annual calendar. The brighter colour contrasts were dis-
tinguished, speech had stepped into rhythm (an ingenious
inference of Westphal from the metrical use of 'foot'),† and
conscious appreciation of sounds foretold the charms of
melody. Religion, in the form of mythological beliefs, had
attained prolific growth.
The civilisation thus depicted may obviously have had its
home in Europe or in Asia. It stops far short of Pictet's
idyllic reconstructions; it represents man emerging out of
neolithic savagery into the less intolerable disciplines of the
age of bronze.
Evidence so limited and fragmentary might not unna-
turally awaken some prima facie distrust of the results.
Interpretative imagination, it might be felt, the true his-
torian's gift, is needed to supplement methods too grudging
and too sceptical. But, on the whole, archæology and com-
parative anthropology remarkably confirm the order and pro-
portion of development that the linguistic evidence suggests.
The range, for instance, of domestic animals, and even the
order of domestication, the priority of pastoral and nomadic
life to agricultural, the proportionate advance in building,
weaving, and the other arts, are all remarkably confirmed by
facts of archæology or savage life. A specific instance may
* To this vexed question Schrader has given exhaustive discussion,
Sprachvergleichung, Part III.
† Schrader, p. 199, protests, but the figure is too much out of the com-
mon for mere coincidence in Skt., Zd., and Gk.
14
THE CRADLE OF THE ARYANS.
i
*
be given in the fact that, while the primitive language
distinguishes the more violent colours--red, yellow, white,
and black-the finer grades (sc. of shorter wave-length), the
recognition of which indubitably comes later, find no Indo-
European names. Throughout, there is coherence and pro-
portion; and if the evidence were far in arrear of the actual
truth, some paradox of language would doubtless crop out,
and inferentially reveal the unattested interspace.
III. Climate, Fauna, and Flora.
Of the results established by the earliest enquirers, none
have proved more permanent than their conclusion regarding
climatic environment. They disposed finally of the Indian
supposition. The one defined season perpetuated in the
vocabularies of all branches of Indo-European folk is winter
(Sk. himá), and this, moreover, used in Eastern tongues
to designate time, a winter season, not merely winter-cold.
With it range the companion stems that designate snow and
ice, shared by Eranian with European tongues. India has,
indeed, lost names which lay outside her experience in
things, but still exhibits the snow (snih) stem in the mean-
ing shine.'
For 'spring' also (ver) there is a common
term, but alike for summer and for autumn' the
stems that can claim to be Indo-European shew far more
vagueness of definition, and in the Eastern group denote
year or season in a general rather than specific sense. The
fact that winter dominates the yearly reckoning, and firmly
holds its place, even in torrid India, of necessity relegates
the founders of Indo-European speech to some climate
familiar with ice and snow, and winter's binding grip. To
this condition Northern Europe and Central Asia conform
with equal ease.
* The argument remains sound, if (as with some African tribes) recog-
nition was extended only to the colours found in cattle, those of natural
phenomena being unimportant and ignored.
THE CRADLE OF THE ARYANS.
15
Vocabulary next claims critical examination.
The term for 'sea' deserves preliminary handling.
There is no common name. One word is European,*
*
endorsed by Latin, Celtic, Slavonic, Lithuanian, and Teu-
tonic stocks. The Eastern languages use different stems,
and those distinct from one another, the Eranian daraya
or zaraya only finding representation in India as denoting a
flat plain. This may be mere accident; Greek has lost the
word, and turned to independent sources: and why not
therefore Zend likewise and Sanskrit, just as no common
term for river' is shared by East and West? Still
the fact is striking. The sea is one of the phenomena likely
to make a permanent impression on the mind, and win a
name defying change or oblivion. And assuming some his-
toric circumstance to underlie the divergence, what is the
interpretation? The upholders of the Asiatic hypothesis
set some store by the fact. The race, they say, was cradled
in the interior. One branch, moving southwards, first
found the sea on coasts of Persia and of India, and there
devised new names for the new wonder; and so likewise the
westward migration for itself, on gaining Europe and the
sea. But the fact that all the European brotherhood (the
Greeks excepted) share the same word, implies that the term
was framed in common, not independently therefore on
European shores, and the further inference is drawn that it
was the Caspian Sea that supplied the new experience and
the new name. The argument savours of perilous precision,
and the unanimity of the European races in conservation of
the term, as they pursued their distant inland routes, tempts
to semi-credulous surprise. The alternative interpretation,
say on Penka's reading, teaches that the word is an heirloom
* Mare, O.G. more, O.S. morje, Goth. marei, Lit. márè, Ir. muir,
Sl. morje. Fick and others find this in Sk. mira, ocean, but Max Müller
denies the identity.
+ But possibly preserved in 'Aµpí-uapos.
16
THE CRADLE OF THE ARYANS.
of the stock, first taught on Baltic shores; that it survived
in all the European tongues, but that, in the long pilgrimage
and sojourn in the East, it was lost to the Eranian vocabu-
lary, and a new term borrowed or devised by the Indians,
and the first explorers of the Persian coast. In such a
statement there is little forced or unnatural, though some
small balance of plausibility rests with the former view.
One solid inference remains, viz., that the language was
not framed in the interior of Europe. The word must
derive from some joint home upon some sea-board, whether
of Europe, or peradventure of the Caspian or Aral seas.
The same may be said, though less forcibly, of the dis-
crepancies in terms for fish. Teutonic, Celtic, Latin are
at one; Sanskrit and Zend diverge in company. But
here the argument is feeble, inasmuch as rivers breed
fish equally with the sea. Oriental diet, or nomad pastoral
life on the wide plains of Asia, may well have proved fatal to
the word.
The names of fauna and flora next invite consideration.
The subject has been thrashed almost ad nauseam without
carrying conviction to disputants on either side, yet I cannot
hold with Schrader that it is doomed to sterility. It is true
that with the disappearance of an animal or tree the name
assigned to it drops spontaneously out of the vernacular, and
that there is little likelihood of a common name surviving
the removal of all occasion for its use; and the recognition of
this fact will relieve us from much fruitless and prolonged
discussion. But nevertheless the consideration of the
vocabulary in this respect may yield general inferences of no
small value, which it will be the object of this section to
deduce.
To begin with quadrupeds. The domestic animals give
no results of value. The dog, the cow, the sheep, the pig,
the goat and the horse all bear Indo-European names, and
THE CRADLE OF THE ARYANS.
17
were known therefore, and all probably domesticated, prior to
the separation of tongues. Whether the species originated,
or whether they were first domesticated, in Asia or in
Europe, has no bearing on the present question.
Wild animals deserve closer attention. The lion opens a
difficult controversy. The main facts are unquestioned — that
no common term connects the Asiatic and the European
groups; that Indian and Eranian are themselves out of
accord; while the same name is preserved in every member
of the western branch. Upon the actual root employed
opinions differ; some regard it as a loan-word from Semitic,
which in laish approaches closely to λís. But it seems
impossible, without violence, to explain all current forms
from this original, so that high authorities are disposed to
trace their origin to some Indo-European root, as RAV to
roar, LU or LEV destroy, or LIV dun-coloured.
The tangle
is perplexing. The lion is certainly associated with the East
(though valid evidence attests the existence of lions in
Greece into historic times), and one would have expected
that so picturesque and commanding a creature once named
would not have lost its title; that Asiatic emigrants would
have preserved its memory, and by adaptation, if not other-
wise, retained its title, even if they passed out of its range of
habitation; still less would the Aryans proper have lost or
changed the name. The difficulties do not end here. Assum-
ing Semitic borrowing, it is hard to explain the uniformity
of European terminology; assuming Indo-European identity
acquired in Asia, it is strange that the name should have
been conserved by all the stocks, except those to whom the
lion himself remained familiar. The Tiger, whose range
extends northwards beyond Bactria, is etymologically im-
ported from India, and so too the panther, Tάpdos and távonp
both coming of late borrowing. Camel is once again exotic,
and pilu of Sanskrit, from the Persian pil, has no European
B
18
THE CRADLE OF THE ARYANS.
congener. These are the great characteristic quadrupeds of
the East, for size and make and mark more notable than
any other in the animal kingdom. Being non-European all
may have dropped into oblivion in western speech, but taken
collectively they constitute a negative argument of some
small weight against the Asiatic derivation of the Indo-
European family. And the wild-ass, the jackal, and the ape
reinforce the argument from silence.
*
Passing from negative to positive evidence, we will first
group the quadrupeds with names of Indo-European pedi-
gree. They are the Bear (ursus), Beaver, Boar (wild or
tame), Fox,↑ Hare, Mouse, Otter, Squirrel (viverra), Wolf.
Though for the most part indecisive, the general complexion
of the list is European rather than Asiatic, and two animals,
the Beaver and the Otter, seem specifically European.‡ In the
latter case there is perfect correspondence of form, Lit. udra
being absolutely identical with Skt. and Zd. udra, but in
meaning, Greek, Sanskrit and Zend differ. It is possible,
though the chance is rare, that distinct stocks fashioned
independently words (identical in root and suffix) for distinct
animals; it is more likely that in districts where the otter
was unknown, the designation was transferred to another
animal of corresponding habitat. If so, assuredly the
European otter was first owner of the superfluous name, not
vice versa. The beaver stem shows like divergence of mean-
ing. The beaver known to Teutonic, Lithuanian, and Latin
vocabulary-though the animal seems now confined to
Russia and Poland-appears in Zend as bawri, and in
* To transcribe comparative lists of words, which controversy has made
miliar, seems superfluous. Where English does not give the key, I have
put down one representative term for guidance.
+ Fox is doubtful, and in spite of Np. rôbât, Arm. aloüés, Curtius and
Spiegel both reject the equation ảλúŋš = Sk. lõpāça; Lit. lápè and rubà
are added as European cognates. Also Sk. khinkira appears by kípa.
The wolf, I understand, does not now frequent India or Persia, but the
Rigveda and Zend Avesta abound in invocations against his ravages.
THE CRADLE OF THE ARYANS.
19
Sanskrit as babhru, denoting an animal of the large
ichneumon kind. Now this is precisely what might be
expected, the transference of a superfluous name (for the
beaver is strange to India) to an animal bearing a rough
general resemblance to the original. Precisely in this way
the jackal of the East supplies the modern Greek with his
name for badger (Taxáλns). Those who reject this account
content themselves with questioning the etymological
identity.
Snakes (anguis), worms (xáµπη, vermis), ants (Zd.
maoiri), flies (musca), and other Indo-European vermin
afford no topographical clue; they are ubiquitous.
Before leaving the mammals, it will be well to consider
the list common to the European group, recognised in the
south by Greek or Latin, in the north by Teutonic, Slavonic,
or Lithuanian. The following, as a glance at the appended
table will show, comply with these conditions:-
Greek. Latin.
Celtic. Slavonic. Lettic. Germanic.
taxus* (late, prob. borrowed)
aper
Badger
Boar
[καπρost caper
Hedgehog
ἐχῖνος
Lynx
λύγξ
Marten
martes
O.G. dahs
v-epri
Eber
O.N. hafr]
jezi
ezys
O.G. igil
luszis
O.G. luhs
O.G. marder
Rabbit coney) Kóvikλos cuniculus coinin
Caninchen
σέλαχος
O.G. selah
ἔλαφος
Cym. eilon
jel-eni el-nis
Kepaós ‡ cervus Cam. karw
kravu karve Hirsch
Seal
Stag
In this branch of the discussion the importance of
European consensus
consensus has been strangely underrated or
*For the suspicions of borrowing that attach to this, and not less to the
Coney, see Hehn (ed. Stallybrass), Wanderings of Plants, &c., pp. 493-4,
and p. 343.
+
Unspecified, of the male, boar or buck. Cf. verres.
The horned,' and the specification is not complete, as in both Slav.
and Lit. the word signifies' cow.'
20
THE CRADLE OF THE ARYANS.
ignored. These animals are known to all the West, though
possibly unknown as well as unpreserved by languages of the
East. Now the upholders of an original Asiatic home are
virtually at one in assuming the separation of stock dialects
in the Asiatic period,* in the regions east or south of the
Caspian. Thus in the case of names common to the whole of
Europe it is inferred that the animals were known and
named on the far side of the Caspian, and the names.
faithfully preserved in after years by the various emigrant
stocks. There is nothing in the present list to refute such
a hypothesis, for the seal is a denizen of the Caspian as well
as the Northern Seas, but it combines with the previous
group to strengthen the impression of a European rather
than Asiatic cradle of experience, and it is highly significant
that every noticeable European quadruped, tame or wild, is
included either in the Indo-European or the European unity.
Birds teach little. Besides the duck (anas) and goose
(anser) which were perhaps domesticated, and omitting
onomato-poetic names (such as cuckoo) which are valueless
as evidence, the crane, crow, owl,† pigeon (columba), thrush,
quail, and probably vulture (glede) and falcon (ixTivos and
perhaps hvn) are Indo-European, a few species, that is,
singled out by size or cry or plumage or flight in the days
before birds were much available for food.
If there were
proof that the duck and goose were not domesticated, the
prominence of water-fowl would be worthy of remark, for to
this category belongs Sk. marala (merulus), while both in
Sanskrit and Greek the xoλuußós designates the grey diver.
For the group of water animals is more fertile in sugges-
tion. The discrepancy between the generic European terms
for fish (piscis and ixús) and the Eastern matsya has been
paraded with some satisfaction by the advocates of Bactria.
* On this see further at p. 33.
† Ulucus by Sk. uluka, and bubo by Pers. trum are the comparisons.
THE CRADLE OF THE ARYANS.
21
"People starting from that central home in Asia," writes.
Max Müller, "ought to have little knowledge of fishes."
The inference is flimsy. Wherever the Indo-Europeans
were cradled, by river, lake, or sea, fish must have been
known to them, and the discrepancy must be reckoned
among the thousand casualties that have befallen speech.*
More valid inferences await us in this sphere, of which the
name for eel yields the most vivid illustration. Anguilla,
ZyXEλUS, Lit. ungurys, Sl. agulja, show a common term for
an animal that appears not to exist in the Black Sea or Cas-
pian or their tributary streams. Here then, at last, we are
driven to a corner, and must accept the inference that the
European unity was maintained at a point clear of the water-
sheds of the Euxine and the Caspian, that the ancestors of
Greeks and Italians were in touch with ancestors of Slavs
and Lithuanians at some point in Central or Western Europe
which the evidence forbids us to place far east of Russia's
western frontier. This, if tenable, will prove decisive to
the whole controversy.
Among water animals attesting an Indo-European unity
is the crab (Sk. karka, xápxa, krebs, Sl. (k)raků); while
European unity appears in the lobster (xáuapos, [homarus],
Hummer), the seal (σéλaxos, O.N. sel-r), perhaps the mus-
sel† (musculus), and the oyster. The latter (oστρɛov, ostrea,
Cel. oestren, Sl. ostrei), Max Müller imputes to borrowing,
claiming Greek for the original: but of the rest it is
difficult to credit that they were carried from the Caspian
as a common heritage by stocks so widely divergent as the
* The three groups—
Sk. matsya, Zd. masya, N.P. mãhi,
piscis, Got. fisks, Ir. iasc,
ixoús, Arm. tzukn, ? Lit. zuvis, (? GHU)
suggest derivation from specific fish, familiar or favourite in given areas;
none of them represents a common verbal root, 'swim' or the like.
† SK. çankha, κóyxos, congius is of precarious meaning.
22
THE CRADLE OF THE ARYANS.
Teutonic and the Greek. On the opposite assumption, that
of European origin, their disappearance (except the crab,
which has land representatives) was inevitable in the Eastern
tongues. In so far they corroborate the evidence supplied
by the eel.
A
Passing from fauna to flora, we find the beech giving rise
to vehement and frequently mistaken controversy.*
common name (onyós, fagus, buocha, beech, and the unproved
pers. buk) appears throughout the Western group, and is
in many of them applied to the beech. The Eastern limit
of the tree in Europe appears to be marked by a line drawn
roughly from Königsberg past the East Polish frontier to
the Crimea. Thus the community of name, assuming its
primary association with the beech, becomes of great impor-
tance to the secondary issue in showing that the united.
Aryans, if of European origin, came not from Rokitno
marshes or Russian steppes, but from some more Westerly
district but seeing that it reappears in Asia Minor, † and
skirts the South shores of the Caspian, it can contribute
nothing to the main issue of European or Asiatic origin.
The birch and probably the pine (Tírus) are genuinely Indo-
European, but grow equally in Asia and Europe.
evidence is quite insufficient to show that either the oak!
(dpũs, Sk. dru=wood, Zd. dru=shaft, tree), ash (ornus, Sk.
arna) or fir attained (or retained) identification in the
Asiatic group, while the common name for withy (vitis),
and probably too populus, by Sk. pippala, is descriptive
rather than generic. Common to widely severed European
stocks, Northern and Southern, are names for the alder
(alnus), apple, ash, elm (ulmus), fir (kλátŋ), hasel (corulus),
* See letters in Times of October 10, 12, 14, 15, 18 (1887).
The
† Schow, Botanical Atlas; Drude, Atlas der Pflanzen-Verbreitung. Pl. IV.
Dru is clearly inconclusive, and ßáλavos, glans, Arm. kalin, does not
define a particular tree.
THE CRADLE OF THE ARYANS.
23
maple,* pine (TEÚxn), willow (salix), yew (taxus), and other
trees. As with the quadrupeds, so here, the list goes far
towards exhausting the common and conspicuous trees, truly
indigenous to Europe.t Once more is it likely, is it
credible, that the unity is due to a hypothetical period of com-
mon speech to South or East of the Caspian? It has all the
marks of North or Central European derivation. One of the
number, śʊxŋ, is of special interest; for while in Greek,
Lithuanian, and Old German (fiuh-ta, whence Fichte), it
represents fir, it is applied in Sanskrit to the betel-nut palm.
How naturally this transference of name would come about
upon the theory of European origination is obvious.
As for cereals, rye and barley, Sk. and Zd. yava by Čɛά,
Lit. java, Celt. corna, and again Zd. gurtak by xpion,
hordeum, gersta, are the only Indo-European terms, and the
latter is unspecialised to barley in the East. Those who
are content with Geiger to build on the evidence of silence,
argue that for the cradle of the speech we must select an
area suitable for barley and rye, but not for wheat, and urge
North Europe upon this ground. But the conclusion cannot
stand upon its own merits, and is more than usually precari-
ous in making no allowance for agricultural as well as
climatic conditions. That it accords with this hypothesis is
all that may be said. To extend the argument further to the
vetch (ervum and cicer), xáμapos, malva, and other Indo-
European or European plants would add nothing to our
main results. The term for hemp (A.S. haenep, xávvaßıç)
shows complete Indo-European equivalence, but structure
and form of correspondence indicate borrowing rather than
primitive identity, and its origin is probably exotic.
* Ahorn seems borrowed from Acer, but klen stem connects Slavonic,
Lithuanian, German, Celtic and Greek. Hehn, Kulturpflanzen, Anm. § 71.
† Heyn shows how many of our common trees, chesnut, sycamore,
cypress, laurel, and others, are imported from the East within historic
times.
‡ Development of Human Race, p. 145–6.
24
THE CRADLE OF THE ARYANS.
IV. Primitiveness of Type.
Primitiveness of type is another argument upon which
advocates of the Asiatic theory, and conspicuously Professor
Max Müller, have strongly insisted in support of their case.
But in truth the argument as usually put is fallacious, the
inferences in large measure unwarrantable, and the facts are
now passing over to the enemy. The argument may be stated
thus:-The Eastern languages, particularly Sanskrit, are
the most primitive, that is to say, exhibit the closest
fidelity to the original type. Primitiveness of form implies
proximity to source of derivation. Therefore the cradle of
Indo-European speech is to be looked for in the East.
First for the main assertion, that Sanskrit is the most
primitive in type. Of morphology it still holds true that
Sanskrit exhibits a larger number of the inflectional forms,
and closer adherence to the original type, than any of the
cognate languages. It preserves intact for instance the seven
cases of the Indo-European noun. But recent years have
greatly weakened the argumentative force even of this
plea. Schleicher and his school largely reconstructed the
inflectional system of Indo-European out of Sanskrit
materials, with the result that the primitiveness of Sanskrit
was exaggerated out of all just proportion. The revision of
their results, necessitated by the progress of research, has
revealed that there is hardly a language, if indeed one at all,
which does not at important points adhere more closely to
the archetype than the Sanskrit of 3,000 years ago.
There are philologists of repute who claim for Celtic, and
still more for the Lithuanian group, a primitiveness ex-
ceeding that of Sanskrit itself. What is true of morphology,
*
Whitney for instance, Language and the Study of Language, p. 203,
writes:-"Of all the existing tongues of the whole great family, the
Lithuanian, on the Baltic, retains by far the most antique aspect;" and in
the same sense Pösche, Die Arier, and in the main Cuno.
THE CRADLE OF THE ARYANS.
25
might with justice be repeated of sematology. But when we
come to the all important sphere of phonology, the tables are
completely turned. Schleicher crudely constructed his vowel-
scale upon a Sanskrit basis, and regarded the European
e and o as modifications of the original and uniform a con-
served by Sanskrit. The doctrine is now disproved, and it
is matter not of conjecture but of demonstrated fact that the
European stocks have adhered more closely to the Indo-
European system of vocalisation than either of the great
Eastern groups. In fidelity to the triple vowel Ablaut, Greek
stands immeasurably nearer to the archetype than Sanskrit ;
so too to a lesser extent does Lithuanian. Of the
guttural series the like is true. * Sanskrit, even in
preserving traces of original differences of utterance, has
travelled further from the archetypal sounds than almost any
European group. These phonological facts are far-reaching
and fundamental; though to some extent obscure and
difficult in their bearing upon the controversy in hand, they
go deeper than morphological divergence or correspondence;
for this reason-that they, if any, rest on physiological
modifications. It is hard to believe that climate or physical
structure acts very directly on morphology; in the sphere of
phonology on the other hand it cannot but be active, and a
revolutionised system of articulation is a far weightier
argument for changed localisation than any tenacity of
grammatical or morphological instinct can counterpoise.
A further claim to primitiveness on behalf of Sanskrit
rests on the simplification or clearer definition of the Root
elements in words. With those who can still hold to a
belief in the genesis of language from monosyllabic roots,
verbal and pronominal, by aid of agglutinative processes,
this plea will have weight. But that chimera is doomed;
accumulating evidence shows that the progress of language is
* A fuller statement of this is reserved for p. 59.
26
THE CRADLE OF THE ARYANS.
a development from confusion into system, from multiplicity
towards simplicity, from caprice into conformity. So is it
with sounds, with word-building, with roots. Language is
the outcome of slow selective instinct, a residuum of number-
less experiments and ventures. The Sanskrit alphabet
is but a remnant of Indo-European sounds; suffixes and
inflections are survivors of a discarded host of superfluities;
and so too the logical development of a system of roots and
stems, in so far as it is not the retrospective analysis of the
grammarian, implies distance rather than proximity to the
parent type, in which it is matter of demonstration that no
fixed forms had crystallised.
But further, and not less important, qualifications of the
argument are necessary. The conditions of comparison
must be tested; and among the most important of these is
the relative age of the languages compared, the distance in
time that separates them from the archetype, and the corrup-
tive forces to which they have been exposed. Between
Sanskrit and old Celtic, or between Zend and Slavonic
no just comparison can be instituted. They represent
periods of dialectic variation separated by centuries, and
realised under very different circumstances. The language
of the Vedas anticipates by centuries the first recorded
accents of Celtic or Teutonic, of Slavic or Italic stocks, and
serves only to give some measure of the changes wrought by
the corrosion and denudation of those centuries in the
unwritten vernaculars of kindred tongues. It is not even
possible to establish any equation between growth in intel-
ligence and conservatism in linguistic habit. The savage
Celt retained much of morphological elaboration that the
cultured Greek lost. Concurrently with the most rapid
development of national culture, English divested itself of
the whole system of inflections which was its morphological
heritage. The one truth that seems to emerge is that con-
THE CRADLE OF THE ARYANS.
27
servatism in language rests far more upon the creation of a
recognised norm-of a literature, that is, whether oral or
written than upon any mere lapse of time or change of
domicile, or even of physiological habit. This is the prime
antiseptic, as shown in the history of Greek, of Slavonic, of
Latin, of English, of the Romance Languages, one and all.
Sanskrit is most faithful to primitive forms, in so far
as it first and nearest to its source attained literary
permanence.
But assuming the primitiveness of Sanskrit, what deduc-
tion can it warrant ? Schleicher, basing his theory on
purely linguistic and mainly morphological lines of argu-
ment, and assuming an Asiatic derivation, maintained that
throughout the Indo-European languages the more easterly
the geographical position of a language, the closer was its
adherence to primitive forms and sounds, and that, conversely,
the further west a language had migrated, the earlier was its
detachment from the common original. Further study of.
Celtic revealed to him the formidable infringements of this
generalisation, which the phenomena of that group present;
and modern phonology has annihilated it. It does not even
rest upon any sound general principle. There is nothing to
show that a tribe upon the move modifies its language faster
or slower than a tribe at rest. There is no fixed equation
between language-change and place-change. Language-
structure as yet gives us no information, whether Sanskrit
or Celtic has travelled furthest, or what circumstances or
ethnical characteristics induced greater fidelity to type in
one case than in the other. Assuming that language de-.
pends ultimately on physiological causes, dependent in their
turn upon stability of environment and type, there is a
probability that the race which remains in undisturbed
possession of its first home will deviate least from the
archetypal sounds and forms of speech. But even here
28
THE CRADLE OF THE ARYANS.
unknown disturbances, such, for instance, as the infusion of
new race elements, must render the argument precarious;
and how cautiously it should be used, Sanskrit itself
attests with emphasis. If one thing is more certain than
another, it is that the forefathers of Sanskrit-speaking folk
framed their language in presence of winter colds unknown
to torrid India; another is that the Indian stock entered
India by migration from the north-west, and that the
Eranian or Persian stocks-whatever the original habitat of
Indo-European speech-stood nearer to the locality and the
environment of the original race, than the occupants of
India. In spite of which, the Zend of the Persian diverged
far more widely from the archetype than the Sanskrit of
the Indian. Linguistic primitiveness is ineffectual to prove
proximity of local origin.
V. Differentiation of Stocks.
The next head of the argument concerns the dispersion
of the original Indo-European into clearly marked linguistic
stocks, the relation of the different stocks, the order, and (so
far as they can be traced) the routes of migration.
Our starting point in this enquiry must be the proved
linguistic relationship between the different families of
speech. All minor details, and all modern developments
that fall within the ken of history may be ignored. It will
be enough to consider the great main divisions of speech—
Indian, Eranian, Armenian, Greek, Italic, Celtic, Germanic,
and Baltic-Slavonic. In comparing these, clearness of
differentiation is the first striking feature. Familiarity robs
it of strangeness, but the absence of connecting links between
type and type is a curious phenomenon; separation of types
must have been gradual, following the ordinary course of
dialectic variation, and the extreme divergence attained
clearly implies the extinction or absorption of many inter-
THE CRADLE OF THE ARYANS.
29
:
mediate types, such as Macedonian or Phrygian might, if
surviving, have exhibited. The relationship between the
survivors presents a problem full of perplexity. Phonology,
morphology, vocabulary, yield distinct and often opposed
results. Different resemblances have impressed different
inquirers, and the doctors differ irreconcilably concerning the
true order of affinity. Max Müller, Fick, Spiegel, content
themselves with the broad division into Asiatic (Aryan) and
European Lottner, Schleicher, Justi, insist upon the
divergence of the North from the South European group,
coupling the latter with the Eastern stocks: while Hübsch-
mann's association of the Letto-Slavonic group with Armenian
and the Aryan branch throws an apple of discord between these
two broad groupings. To enumerate hypotheses is tedious.
Greek may be selected as a single example. Schleicher
upheld a Græco-Italo-Keltic period, and the morphological
resemblances of the Italian and Celtic types unite them
closely. Mommsen and Fick give the weight of their high
authority to the existence of a distinct Græco-Italic stage,
with progress in agriculture for one of its specific marks.*
On the other hand, the alliance between Greek and Armenian
is undeniable, and at many points, in religion and mytho-
logy even more conspicuously than etymology, Greek stands
in far closer relationship with Sanskrit than with any of
the Italian dialects. To classical students, the intimate
correspondence of Greek and Sanskrit in their inflectional
systems, alike of noun and verb, must appeal with command-
ing force in the verb for instance, the preservation of
- forms of conjugation, the distinction of Thematic and non-
Thematic formations, the survival of the dual, with etymologi-
cal correspondence even in minute vowel-variations, the
*
* Mommsen, Hist. Rom., ch. ii. Hehn and Helbig contest this theory,
and regard the resemblances as European rather than specifically Græco-
Italian.
30
THE CRADLE OF THE ARYANS.
entire system of middle personal terminations, the retention
of the augment, &c., &c.—features wholly, or almost wholly,
obliterated in Latin and other European stocks. In truth it
is absolutely impossible to deny the reality of the observed
affinities. Aryo-Greek, Slavo-Iranian, Germano-Slavic, Ger-
mano-Celtic, Italo-Celtic combinations, all rest on solid
evidences. What is the key to the puzzle?
The old idea of lineal descent and relationship must be
abandoned. Schleicher, under the sway of biological pre-
conceptions, regarded the differentiation of languages as
analogous to that of species, assuming a common ancestor,
fission of a dialect or group of dialects at a given point,
and ultimate estrangement by process of continuous diver-
gence. Were this conception true, it could be traced and
verified the line of succession and the order of severance
determined. The facts necessitate a readjustment of view.
There is no relationship of direct lineal succession between
the stocks; this indeed was long ago perceived: but beyond
this, there was no divergence, or rather divarication, at a
given point in time or place. Rather there was progressive
differentiation, realised under conditions of intermittent con-
tact and reaction. This theory is rather fancifully described
by Schmidt as a 'wave-theory' of language.* It conceives
the problem thus. Indo-European speech does not (within
linguistic purview) begin as a single homogeneous dialect
spoken at a single centre, and radiating lineally thence; the
original Indo-European of the philologists is rather a widely
distributed body of speech, extending over a large area of
country, possessed of the strong general resemblance due to
common descent, such as that exhibited by the yet more
widely severed members of the Turko-Tataric family. In
common with all languages, it was subject to dialectical
variation of every degree and kind. Distinct areas, due to
* Verwantschaftsverhältmisse der Indo-Germanischen Sprachen, p. 27.
THE CRADLE OF THE ARYANS.
31
geographical configurations, or tribal relationships, or other
more fortuitous centripetal influences, gradually formed
themselves, obeying different laws of dialectical variation, and
acted on by different determinants in the form of climate,
surroundings, occupation. and, in all probability, alien forms
of aboriginal speech. And in this connection it is important
to observe the large allophylic element incorporated into the
various Aryan tongues, as though each branch had made its
individual appropriations from foreign sources. In Greek,
for instance, Wharton estimates that out of more than
twenty-seven hundred primary words fifteen hundred only
can be with any probability referred to Indo-European
origin. For long these distinct areas, destined to develop
the different families of speech, were not severed by hard
and fast barriers of speech. There was overlapping of area
on area, play and counter-play of phonetic tendency, inter-
change or unconscious co-operation in the work of word-
manufacture. The rapprochement between group and group
was sporadic, variable, and discontinuous, influences coming
sometimes from one side, sometimes from another. Such is
the sure inference that comparative philology supplies.
*
Thus viewed, the relations of the various groups become
fairly explicable, and it seems possible to determine approxi-
mately which areas of speech came within range of mutual
influence. A disposition somewhat like the following leaves
no important problem of affinity unexplained. Minor
readjustments are possible nach Belieben; if any one, for
instance, believes strongly in a specific Germano-Aryan or
Italo-Aryan connection, it would be possible to introduce it,
but either lacks linguistic warrant.
While in general accord with the geographical distribution
Sayce, Address to British Association, 1887. It would be a laborious
and delicate task to ascertain the extraneous element in each of the great
stocks, but would yield important results in determining relative fidelity to
the parent speech and essential purity of strain.
32
THE CRADLE OF THE ARYANS.
of stocks, the diagram satisfies all the salient affinities that
language attests. It explains the broad separation between
the northern and southern European types; it reconciles
with that the proximity of Celtic to Italic. It places Greek
in right proportionate relation to the Italic, Armenian, and
Aryan (especially Indian) groups, and it yet shows how the
Aryan, and more particularly the Armenian and Iranian type,

CELTIC.
GERMANIC.
BALTIC-SLAVONIC
TRANC
*
ARMENIAN:
ITALIC.
GREEK.
NDIAN
may have shared with the Letto-Slavonic a large phonetic
tendency or contagion, such as that which has resulted in
the parallel sibilation of the guttural series, which is common
to the two groups.†
What bearing has this reconstruction upon the original
habitat of Indo-European? It makes strongly and almost
conclusively for a European habitat, situate west of the
* Armenian very probably represents a later migration, derivative from
Græco-Phrygian stock., by emigration perhaps not earlier than 700 B.C.
+ This and other instances of telling phonological correspondence are
treated more completely, p. 59-69. For proportionate correspondence in
Vocabulary between East-Aryan and the separate European stocks, tables
in Schmidt's Verwantschaftsverhältnisse der Indog. Spr. give useful data.
THE CRADLE OF THE ARYANS.
33
Caspian and north of the Caucasus. Great physical obstruc-
tions such as these interpose natural linguistic barriers,
which would check free interaction between the separated
areas. Not one of those who uphold an Asiatic origin has
proposed any arrangement that satisfies the linguistic con-
ditions of the problem. Take Lenormant's as one upon
which it would be difficult to improve. The cradle of
the Eastern stocks he places to north-eastward, on the
borders of Sogdiana, whence they passed respectively to
Persia, and, via Kabulistan, to India. Prior to this migration
the European sections had felt the westward impulse. In
Bactria lay the Graeco-Italic tribes, who, clinging for a while
together, passed by Herat, Khorassan, and Asia Minor to
their European peninsulas. The Celtic branch, issuing from
a more northern cradle, say Margiana, and moving south
of the Caspian, crossed the Caucasus, skirted the Black Sea,
and so passed on to the Danubian basin and Central Europe.
The Germano-Slavic sections, fording the Oxus, and keeping
to latitudes above the Caspian, traversed the Scythian steppes
into N. Europe. The whole of this is ingenious speculation,
devoid of corroborative evidence from mythology, history or
archæology. And from the linguistic side it signally breaks
down. For instance, it contradicts the broad European unity
that exists in the general fabric of language, and more specifi-
cally in the nomenclature of trees, animals, and agriculture.
It gives no account of the unique relationship between Greek
and Sanskrit, and it implies a tie far closer than exists.
between Greek and Latin. It gives no explanation of Latin
and Celtic correspondences. It puts the German stock
entirely out of touch with the Celtic, and it would naturally
imply German participation in all phonetic idiosyncrasies
shared by Letto-Slavonic with the Aryan (i.e., Asiatic) types.
The breakdown could not be more complete.
There remains a single point. In whichever direction the
C
34
THE CRADLE OF THE ARYANS.
migration took effect, the barrier of the Caucasus and Cas-
pian would be likely to impose a specific unity upon the
tribes who crossed it in company. Of this there is very
little to be traced in the European family regarded collect-
ively. High authorities consider southern European stocks
to stand in closer relation with Aryan than with north
European. But the Aryan family do exhibit a close resem-
blance, and, moreover, just of that kind which active com-
panionship would supply. The correspondences with their
Western kin are broad and fundamental, and consist in
ingrained habits, phonetic or morphological; their corre-
spondence with one another is in the sphere of vocabulary,
as though at some later stage they had lived joint lives, and
experienced common vicissitudes, and together moved in new
surroundings.* Assuming that their Indo-European contact
lay in the West, and that at some later date they passed the
Caucasus together, or in one another's train, the specific
unity of the united Aryan stock falls into its natural place.
There are incidental corroborations of these views upon
the linguistic side, negative indeed and to some extent
precarious, but deserving of mention as giving cumulative
support and coherence to the argument. The theory of
Asiatic origin generally, if not in all hands, postulates an
advance through Asia Minor of at least the Græco-Italic
stock. But all traces of Aryan speech recovered from Asia
Minor seem clearly derived from European immigrations,
such as the traditional passage of the Phrygians from Thrace,
and not a trace is left of any pristine Aryan names or ver-
nacular. Of Armenia, Media, Susiana, the very countries
* Cf. common terms for Eastern animals, and common losses, as of
sea,"
""fish," etc. Any chapter of Spiegel, Arische Periode, supplies
numerous and telling illustrations.
†This is the one factor not fairly met and elucidated by the foregoing
diagram, which shews the prior stage only.
Should Hittite, as is now whispered, turn out to be of Aryan lineage,
this section of the argument will need readjustment.
THE CRADLE OF THE ARYANS.
35
to which the ancestors of the great European stocks are
assigned, the same is true. The evidence is positive that
the population and speech of these districts was Turanian.
Neither Accadian nor any of the allied languages appears
to exhibit a trace of Indo-European influence. The Aryan
languages that do make their appearance in this quarter
show Iranian, not European affinities. Thus, in its accepted
form, the Asiatic hypothesis requires us to believe that many
centuries B.C., at a time, indeed, not so very far distant from
the clear differentiation of stocks, the language whose repre-
sentatives were vigorous and numerous enough to overspread
the whole of Europe, had not only abdicated possession of
its native soil, but had perished absolutely without leaving a
trace among the rivals to which it succumbed. So for-
midable is the assumption involved, that Sayce,* in the days
when he still felt tied to the accepted theory of Bactrian
descent, fell back upon the desperate alternative of carrying
the whole European migration round by a route north of the
Caspian Sea.
Semitic joins hands with Turanian in rebutting the
Asiatic theory. The latest authorities are disposed to assign
to the prehistoric movements of the Semitic race a direction
from the North, instead of from Arabia or the South, and
Kremer derives the first migrations from the very plains,
west of the Pamir plateau, between the Oxus and Jaxartes
streams, which are claimed as the cradle of the Aryans..
If this be so, or even had Aryan and Semitic been
neighbour languages in Asia, the probability is that marks
of relationship or obligation would have established them-
selves in the two languages. Hommel maintains that such
exist, and rests his proof on the following analogues :
App. I to Principles of Comparative Philology. In the Preface to his
latest edition, the Professor unreservedly adopts the theory of European
origin, advocated in his Address at the British Association, 1887.
36
THE CRADLE OF THE ARYANS.
Ind.-Eur. ...staura karna laiwa, liw
Semitic......tauru karnu
(0x) (horn)
gharata sirpara, waina
labi'atu, libatu harûdu țarpu wainu
(lion) (gold) (silver) (wine)
ence.*
but neither the class of words adduced, nor their phonetic
correspondences, are able to sustain so momentous an infer-
Van den Gheyn recurs with frequent emphasis to
primitive Aryan and Semitic intercourse, and asserts their
anthropological unity. But nowhere does he buttress his
assertions with solid arguments, which would be invaluable
for the settlement of the controversy.
VI. History.
To this prehistoric controversy, history can make but
indecisive contributions: first, the hearsay evidence of
national tradition, in the form of legend and mythology,
unauthoritative, yet entitled to a hearing; secondly, those
broad analogies which may tend to support, or to discredit,
views advanced on independent grounds.
Chinese history was once alleged, in recorded movements
of the Hiongnu and the Yeta tribes, to chronicle the latest
westward migrations of our Indo-European progenitors; but
the identification of these tribes with Huns and Getai (fore-
runners of the Gothic stock) appears to be generally aban-
doned, and they are now regarded as Thibetan tribes,
without the smallest claim to Indo-European connection.
Spiegel, the first living authority upon the subject, is
equally positive in rejecting the view first popularised by
Lassen, that in its Airyanem vaejo the Vendidad records a
fixed tradition of descent from the wintry uplands of Central
Asia. No hint connects it with a belief in previous national
migration. Geographical bearings it has none, but Ariana,
* For discussion, see Max Müller, Biographies of Words, &c., 111 pp.
† Spiegel, drische Periode, p. 5. Arische Periode, 122 pp.
THE CRADLE OF THE ARYANS.
37
when located, lies to north-west, not to north-east. It is a
land of pure myth, parallel to that of the Hyperboreans of
Greek fable, but with localisation more vague, and associations
more fabulous. History, when first it finds voice, shews the
Persian races advancing northwards, and gives no hint of
prior movements in the converse direction. From that day
to this, Eranian dialects appear to have been pushing slowly
northwards, and now range across Turkoman territory to
Khiva and Bokhara, with feelers extended north and north-
eastward of the Oxus towards the great Pamir plateau.
Here their modern limit is attained, and there is not the
smallest indication of tradition, of history, or of linguistic
inference, that their infancy was nursed amid the inhos-
pitable uplands of Pamir. Explorers have at length, in
recent years, partially penetrated these unknown wilds. So
far from furnishing corroborations of the conjectures of the
old school, their reports shew that few places in the globe
could furnish a more unlikely cradle for the prolific race,
with which it was in ignorance accredited.
There are other members of the Indo-European family,
whose earliest memories are of some worth, particularly the
Indian and the Greek. The hymns of the Rig Veda, like
the chapters of the Avesta,* are the voice of an invading race
winning its way forward among strangers of alien race and
religion. India (whatever hypothesis of prior habitat is
favoured) was entered by its north-western passes, and these
earliest utterances prove that, at a period ranging from.
1000-600 B.C., the memory of migration was still fresh, if
not still in course of accomplishment. One other literature,
* On the Aryans of the Avesta as a conquering and usurping race, and,
as it would seem, a race of pastoral habits, as opposed to the nomad hordes
of the steppe, see Geiger's Civilisation of Eastern Iranians, pp. 19-22,
25 pp.
+ "For myself, I have grave doubts whether the invasion was earlier than
the eighth or even the seventh century B.c." Sayce, British Association, 1887.
38
THE CRADLE OF THE ARYANS.
that of Greece, carries us back to a like date. Apart from
all controversies as to date of final composition, the Homeric
poems carry us back to an era not posterior to 1000 в.c.
Though the existence of an aboriginal (Pelasgic, or other)
population is implied, there survives no distinct conscious-
ness of national immigration. The Greeks are already in
possession of their eventual home; and there is little doubt.
that they entered it, as Kiepert holds, from the European
north, breaking through and dispossessing Illyrian prede-
cessors, rather than creeping along the Aegean coasts from
the Hellespont and Asia Minor. Curtius' theory, that
Greece was colonised from Asia Minor, has gained no favour,
and general credence is given to the established tradition*
that Thracian immigrants occupied Asia Minor, and that the
Phrygians were an offshoot of the same stock, by emigration
from the confines of Macedonia. It is a conjecture, favoured
by tradition and corroborated by the scanty remains of lan-
guage, that the Armenians likewise represent an earlier pre-
historic emigration from the European continent. However
this may be, we attain at least a valuable result of relative
chronology. The Greeks were in immemorial possession of
their peninsula at the time when the Indian Aryans were
still annexing theirs. Facts must be tortured to reconcile
this with the hypothesis of Asiatic origin. The evidence
from Italy, though less early and less precise, makes in the
same direction.
Further historical indications, or, to speak
more strictly, analogies, are submitted in presenting Dr.
Penka's theories (p. 54). Here it is enough to insist on the
broad fact that neither the location nor the movements of the
European stocks, when emerging into the twilight of history,
give the slenderest support to the theory of an immigration
from the East. In the time of Herodotus, the Slavs were
* Herod. vii. 73.
+ Schrader, Sprachvergleichung &c., ch. ix.
THE CRADLE OF THE ARYANS.
39
in settled occupation of Gallicia or West Russia; German
stems, somewhat later, are found occupying their ancestral
territory from the upper Rhine to the borders of Dacia. The
Celts, prior to the time of Cæsar, are in occupation of
Gaul, and the notices that survive testify to expansion east-
ward or south-eastward into Switzerland and the adjacent
lands, not to recession westward, such as the theory of
Asiatic immigration is accustomed to assume. Pressure and
movement are uniformly from the North, not from the East.
The backwave of Celtic expansion south-eastward probably
reflects pressure upon their own north-eastern frontiers.
Here would naturally follow the evidences from mytho-
logy. But comparative mythology is still an uncleared
waste of multitudinous controversy. Regarding the broadest
issues there is no approach yet to agreement.
The con-
sanguinity of Greek and Sanskrit gods, the almost complete
divorce between Greek and Italian, the relation of Teutonic
and Norse mythology to that of the South European or the
Asiatic groups, are riddles still unsolved.* Invaders who
imposed their language, may nevertheless have adopted the
gods of the people whom they subjugated. Which mythology
is primitive, which derivative, which borrowed and posterior,
not even experts can affirm. It seems rather droll to find
the time-honoured dragon taken seriously in this controversy.
M. Löher claimed his support for the European theory, but
Van Gheyn says, that since the quaternary period Europe
has disagreed with the greater saurians, which however
throve in the Caspian and Central Asia, and are chronicled
by the annalists of China!
VII. Language and Race.
Such then are the conclusions attainable from the study
of linguistic phenomena alone. Is it possible to define still
* Wolzogen tries to prove Sanskrit mythology primitive, Norse derivative.
40
THE CRADLE OF THE ARYANS.
more closely the cunabula of the original stock? The hope
of doing so depends upon the possibility of co-ordinating the
results of linguistic study with the conclusion reached by
independent lines of research, pursued by the ethnologist or
palæontologist. For any such combination Max Müller
emphatically holds that the time is not yet ripe; that for the
present philologist and anthropologist must work on in isola-
tion, following and determining their separate clues, trusting
that at last their separate strands may converge in one.
Yet that an eventual synthesis exists between philology and
ethnology is certain; that it is discoverable is at least
possible; and the time seems to have come to attempt con-
structive co-ordination. Even a false synthesis has its
advantages, for its demolition will pave the way for some
better substitute. A working hypothesis serves as a useful
criterion, though it can only command assent by the cumu-
lative support of collateral proofs.
There is much to encourage the attempt. On the one
hand, there is the established unity of the European brother-
hood of languages. Familiarity alone has dulled the sense
of wonder at that marvellous phenomenon which a century
ago would have been dismissed with incredulous derision.
On the other hand, anthropologists and archæologists have.
been steadily reducing the number of racial varieties, which
finally underlie the inextricable medley of European stocks.
Few prominent anthropologists of the present day accept
more than four or five at the most, and from that small total
would subtract one or more as sporadic or obviously
imported into some isolated corner, that for the present
question may be fairly left out of account. The evidence
tends to bring us back at no very distant antiquity to a
Europe sparsely peopled by a very few well-differentiated
types. This being so, it seems a scarcely insoluble problem
to identify with approximate certainty the founders of Indo-
THE CRADLE OF THE ARYANS.
41
European speech - if once a European origin can be
established from the internal evidence of language—with
one out of the few alternatives open for choice. It is true
that no general correlation between language and race can be
maintained, and upon this ground able philologists, Oppert
for instance and Hovelacque, contend that community in
language affords no valid inference for asserting original
affinity of race. Up to a certain point their position is
incontestable. The child, of whatever race, acquires the
tongue of those among whom it is brought up, and hereditary
predisposition to a particular language, or even a particular
phonology is, if not non-existent, so faint as to count for
nothing against environment and education.
It is true,
further, that entire nations have in historical times adopted
the tongue of foreign conquerors. In Europe, the Romance
languages furnish the most conspicuous wholesale example,
and the coincidence that the displaced tongue was of the
same Indo-European lineage as the usurping Latin, cannot
be pressed to a contention that such interchange is possible
only under like conditions of affinity. In India, in Jamaica,
in South America, in the Pacific Isles, the yet more radical
change is being freely realised on hardly, if at all, less large.
a scale. But this is due to the cosmopolitan conditions of
modern civilisation, on which it is needless to expatiate at
length, and there is no evidence that it can be realised
except where the encroaching language achieves annexation
by the innate ascendency that belongs to a high civilisation
in presence of an inferior, to which it is practically able to
dictate. Aboriginal languages yield only to virtual exter-
mination before a more powerful physical type, or to com-
plete incorporation in some more highly organised form of
civilisation. Now in prehistoric times there is no ground for
admitting incorporation parallel to those supplied by the
history of the Roman Empire, or of Islam, or of the Colonial
42
THE CRADLE OF THE ARYANS.
enterprise of modern Europe. Evidence and analogy, as
well as a priori considerations, make strongly against such a
view. Savages do not on a large scale subjugate neighbour
savages, build up organic empires, and impose on them their
customs and their speech. The advance is gradual, and
mainly physical, by slow and creeping supersession, not
energetic annexation. And there is every reason to agree
with Whitney that, for prehistoric times, correspondence of
language gives the very strongest presumption, if not valid
proof, of correspondence of descent, and that the unity of
Indo-European speech implies underlying unity of stock
present among the early peoples who used it. Upon this
showing, it will be legitimate to use Aryan or Indo-European
as a race-term, if only it is possible to decide to what race
(pure or composite) it properly belongs. Against such an
application Max Müller hotly, and not without some reason,
protests. Aryan, to the currency of which as a linguistic
term his own writings have in truth contributed so power-
fully, is in his eyes the property of the philologist, minted
and circulated by him with a fixed value and connotation.
It has nothing to do with race; it implies nothing as to
colour, tribe, or habitat. Aryans-whether black or white,
dolichocephalic or brachycephalic, Asiatic, European, African,
or American-are those, and those only, who use Aryan,
that is, Indo-European speech. To talk of an Aryan race,
or an Aryan skull, or an Aryan country, in any other sense.
than this is a misappropriation of terms, an infringement, so
to say, of the philologist's copyright, that inevitably tends to
mischievous confusions. Still, the invention of new termino-
logy is troublous and for the most part forbidding work,
beset with not a few disadvantages; and it is more important
to be clear as to the sense intended than to quarrel with the
form of expression. In dealing with Penka's theory, to
which I now turn, it will be most convenient to adopt his own
THE CRADLE OF THE ARYANS.
43
terminology, and understand by Aryan that blond dolicho-
cephalic North-European race with whom he endeavours to
associate the origination of Indo-European speech. Penka's
two principal works, the Origines Ariacae and Die Herkunft
der Arier* offer the most coherent and comprehensive
synthesis of racial and linguistic descent yet propounded, and
English readers may welcome a compendious statement of
conclusions that have not yet been summarised in English
books. It may be said in advance that his philological
assertions are often rash and partisan, and at times even
ridiculous; but these incidental trips do not invalidate his
main argument, while on the anthropological side there is
great fulness and lucidity of statement, supported by an
impressive array of authorities and facts, and commanding
the eager assent of distinguished workers in that field. ‡
Recent classifications of mankind reduce the distinct
types or races of man to four or three. The triple division
recognises—
* To these, for brevity, I refer in later notes as O.A. and H. A. respectively.
+ It is hardly possible for instance to read him seriously, when he finds
(H. A., p. 23) the " Aryan" title ready embodied in Chatu-arii, Ripu-arii,
Boio-arii, and the other numerous appellatives derived from the -ware
( = men) element. Equally rash etymologising occurs in connection
with the same name in O. A., p. 34, where ȧpyós, ardere, Armenian, Arii,
Iron, Alemanni, Ermanrik and its congeners, and even Romani (for
Aramani), Romanes, and erus are preposterously lumped together upon a
common AR or AL denoting the white or bright skin and hair of the
European blond. This derivation of tribe names from colour amounts to
almost a craze with him-cf. e.g. O.A., p. 98, on “Eλanves; p. 122-4 on
Britanni, Belgae, Celtae.
+
‡ I have not adhered to the order of presentation adopted by Penka, nor
refrained from free addition or omission, where it seemed advisable. The
prefatory anthropological classification is from other sources, and I have
only attempted to give in a few pages the gist of arguments elaborated
through two volumes. Not a little of the material he uses has already been
handled in earlier portions of this essay. I should perhaps add, that I am
not competent to check Penka's anthropology, and space does not permit
the statement of counter theories, many of which Penka himself fully states
and discusses.
44
THE CRADLE OF THE ARYANS.
1. The
Ethiopian or Negroid, which in this large
grouping includes the Australioid division, which
Huxley groups as a distinct fourth type, and thus
covers the Negroes of Africa (Hottentots and Bush-
men), the Oceanic and Melanesian Negroes, and
the Negrites of the Pacific and Andaman Islands.
2. The Mongolian, including the natives of Northern
and Eastern Asia, the Malays, the comprehensive
American group, and (in an exaggerated form) the
Eskimo.
3. The so-called Caucasian, which embraces the Xantho-
chroic and Melanochroic types discriminated by
Professor Huxley.
This so-called Caucasian race, which alone is of interest
for the present discussion, parts broadly into two divisions, the
Northern blond type (the Xanthochroic), having for its most
representative samples the people of Scandinavia, Denmark,
and Iceland, and the Southern or darker (Melanochroic) type,
appearing most typically in N. Africa and S. Western Asia.
Between these two, and occupying all but isolated corners of
Europe, are representatives of the race, showing every variety
of modification that intervenes between the Scandinavian of
the North, and the Arab or Persian of the South.
One somewhat too facile hypothesis is disposed to rest
content with this broad triple subdivision, and to hold that
in between the Negroid of the South and the Mongolian of
the North, was intruded wedgewise some more favoured and
highly-selected stock, provisionally styled Aryan, which, by
obstinacy of physical vitality, and superiority of mental
endowment, has pushed its precursors northward and south-
ward, shouldering its way irresistibly into occupation of all
Europe, and a broad fringe of Africa, and destined in the
ages wholly to supersede its other human competitors. The
earlier habitat of this people is conjecturally placed on or else
THE CRADLE OF THE ARYANS.
45
eastward of the Asiatic and European confines of modern
Russia, say the steppes northward of the Caucasus in the
region of Astrakhan.
The main shortcomings of this hypothesis are twofold:—
first, it fails to give an adequate explanation of the marked
variations between the Xanthochroic and Melanochroic, the
white and dark types of Caucasian derivation; and secondly,
it fails egregiously to explain the presence of the purest
specimens of the white type in N. Europe, especially Scandi-
navia. These two points deserve accentuation.
The White and the Dark types diverge so widely, and
there is so much fixity withal in the characters of the
divergence, as in the eyes of leading anthropologists to
justify and even necessitate sub-racial differentiation. To
account for this it is assumed, with rather easy vagueness,
that the stronger infusion of Negroid* elements from the
South, mingling with the pure Aryan strain, produced in
varying degree and form the Melanochroic resultant; while
in the North, the original Aryan retained his purity, though
here and there betraying parallel Mongolian influences,
tending to produce something of assimilation towards
Melanochroic types. These assumptions are vitiated by fatal
flaws, the most convincing of which is derived from compari-
son of cranial characteristics. The skull-index is in the
eyes of the anthropologists the most persistent of racial
mint-marks. Now the Aryans are by the hypothesis dolicho-
cephalic in skull, like the blond whites of N. Europe, and
once again the Negroid race is yet more markedly dolicho-
cephalic; no infusion of these stocks could possibly tend
towards that brachycephalism which is characteristic of the
dark whites. Subsidiary objections, deduced from compari-
son of hair-structure, lip-formation, and facial angle may be
* Professor Huxley is disposed to regard the Melanochroi as a fusion
between Xanthochroic and Australioid elements.
46
THE CRADLE OF THE ARYANS.
passed by. It is enough to say that there is no justification
for believing that the blond white of the North, with his
dolichocephalic skull-index, was ever, within approximately
historic and language-making* eras, identical in race with
his brachycephalic brother-white of more Southern Europe.
The second great difficulty lay in the presence of the
pure blond white upon the Baltic shores. Assuming him to
represent most faithfully the primal stock, it is inexplicable
that starting from some distant centre, he should have
conserved in a far off home, in changed environment of food
and climate and ways of life, a type unable to perpetuate
itself and issue victoriously from the struggle for existence in
its native cunabula.
Neither however is it admissible, as a refuge from this
difficulty, to regard the Melanochroic or dark-skinned white
as the normal Caucasian type, and to say that the Xantho-
chroic white is a modification produced through the lapse
of centuries by the Northern environment-for (1) there
is no proof and no likelihood that environment or latitude
can in a measurable time induce such racial modifications,
not only in physiognomy and tint, but also in cranial
characteristic. Lapps, Finns, Eskimo, the Samoyedic tribes,
in a word the whole Northern Mongolian stock, are pre-
dominantly dark, and also brachycephalic. This might be
set down to greater obstinacy of hereditary physique, or to
special idiosyncrasy of the Mongolian stock. But not only
there, but also amongst the Welsh, the Irish, and the High-
land Scotch, there is the permanent element of the dark
*
*By language-making era I allude particularly to the time to which the
Aryan Ursprache may fairly be attributed. Without pretending to precise
chronological definition, philologists generally are disposed to assign this
stage to a period ranging from 1500 B.c. to 3000 в.C. Whitney, Lang. and
Study of Lang., p. 201, writes 2000 в.c. as an approximation. 1500 B.c. is
the extreme inferior limit that the facts admit; the extension of this back-
ward must be determined by analogy and divination rather than evidence,
THE CRADLE OF THE ARYANS.
47
brachycephalic type, with the anatomical peculiarities un-
changed, traceable from neolithic times, and which, so far
from showing signs of yielding, compromising, or disappear-
ing, rather gains than loses upon its fair-skinned corrival.
And (2) the theory fails wholly to account, even to
attempt to account for the sporadic appearance of the fair-
skinned type in almost every country or continent peopled by
Aryan-speaking populations. These significant appearances
become mere random "sports," frolics of nature, springing
from no assignable cause. True, they appear usually in
mountain uplands-but so sporadically and capriciously, that
it is impossible to find in upland life a vera causa for their
development.* Nay, precisely to the upland conditions of
Bavarian and Swiss life have been ascribed the counter-
process towards dark brachycephalism, which characterises
the populations of those districts.
†
Recognising the full intricacy of the anthropological
problem, and assuming it certain that somewhere amid.
Aryan-speaking peoples exists the racial type, which de-
veloped the language now common to all-assuming, that is,
that Aryan is not Negroid, or Chinese, or Dravidian, or
Mongolian by derivation-Penka first asks, What are the
existent types? They are seven in all-Indian, Iranian,
S. European, Slavic, two Celtic, and the Teutonic, or more
specifically, Germano-Scandinavian. ‡ The Indian and Ira-
nian do not come in question where European ancestry is
under consideration, and in no case could they be regarded
as the racial progenitors of the Aryan-speaking populations
of Europe. The south European type may be ignored, for
* H. A., p. 138-41.
† H. A., p. 12; O. A., p. 13.
+
+
+
This classification, be it observed, is anthropological, not linguistic, and
it is instructive to observe how closely it accords with the grouping upon
linguistic grounds; it attests the power of racial idiosyncrasy in determining
and individualising its proper casts of speech, whether borrowed or original.
48
THE CRADLE OF THE ARYANS.
there is conclusive evidence that the aborigines of Greece
and Italy, from whom this type indubitably derives, Etrus-
cans, Iapygians, Pelasgians or otherwise, were not of Aryan
speech, but succumbed before the Italic and Hellenic immi-
grants. Of the two Celtic stocks, one (styled sometimes
Milesian) seems associated with Cro-Magnon pedigree, and
thus may be discarded. The second, in skull-index, in skin
colour, and in general build, shows such marked affinities with
the Slavic, that the two may be grouped together. Thus the
choice practically narrows itself to the full blond dolicho-
cephalic Teutonic on the one hand, and the shorter darker
brachycephalic man of Celto-Slavonic type on the other.
Of these two, one, and one only, exhibits traces of itself
everywhere among the various populations for which philo-
logy or archæology attest Aryan antecedents. It is the
last-named, the blue-eyed, fair-haired men of the Ger-
mano-Scandinavian family. Everywhere, and throughout all
history, it confronts us, and challenges explanation. It
appears pictorially on Egyptian monuments two thousand
years before Christ. In the pages of the Rig Veda the
white skins of the invading and triumphant Aryans are
expressly contrasted with the black-skinned vanquished
Dasyu. The earliest European historians, from Strabo to
Jordanes, one after another describe the type in their
portraits of Cimbrians and Teutons, Gauls and Franks,
Goths and Visigoths. To-day we find it not only throughout
all northern and central Europe and the British Isles, but
it arrests the observer's eye among the northern ranges of the
Iberian peninsula, in Italian Piedmont, in the mountains of
the Peloponnese, in remarkable purity among the Sphakiots of
Crete, and, passing to Africa, reveals itself in the hill-country
of Algiers* and Morocco, as well as upon the flanks of the
* On the racial identity of the Algerian Kabyles with the red Celt, see
Sayce's emphatic testimony at British Association, 1887.
THE CRADLE OF THE ARYANS.
49
Aures and the Atlas range. It crops out freely in the Ossets.
and other tribes on either side the great Caucasian range; it
has its offshoots among the Tartars of the Kirghiz steppes.
It reappears among the (Iranian) Galtschas of Persia, in the
hill folk of Afghanistan, in the Siah Posch* or Kaffirs of the
Hindu-Koosh. No other stock can claim the same ubiquity,
or ubiquity of like sporadic kind.
This then as a first presumption might be regarded
as the Aryan stock. It is the one stock which gives unity
upon the physiological side to Aryan speaking people: is
there not a likelihood that from it too derived the philo-
logical unity? With a view to determining this it is impor-
tant to trace its pedigree, history and fortunes, and see how
far they favour such a hypothesis.
Alone among races it can claim continuity of European
development. The cranial index, which is the chief court of
appeal for the identification of prehistoric European races,
is dolichocephalic. This carries us back to the quaternary
period. Already in the glacial epoch, when all north-
eastern Europe, from Mecklenburg to Russia, lay an
uninhabitable wilderness of ice, palæolithic man makes his
appearance in central Europe, and his cranial index, as
attested by the Engis skull,—and, indeed, all skulls securely
attributable to the quaternary period—is dolichocephalic. It
is to the stern environment of paleolithic man through-
out these immemorial ages that Penka would attribute the
original differentiation of the blond type. That primitive
man was not blond is certain; that he was produced by racial
differentiation, not by independent creation, men of science
are now agreed; that the long-headed type may be dated
back to paleolithic times, and precedes the short-headed in
* Van d. Gheyn, p. 26. Presence of blonds among the Siah Posch is
denied by Ujfalvy, but rests on unassailable evidence.
† H. A., p. 65-8, 91-2. O. A., p. 82-5.
D
50
THE CRADLE OF THE ARYANS.
Europe, is the verdict of the evidence as it now stands.
That it arose out of the conditions of the quaternary epoch
is the natural inference. It is the outcome of the long
struggle with an ice-bound world, by which European man
was slowly inured to the strength of frame and the hardy
resourcefulness of mind which, late in time, has secured to
him the leadership of the race.
At the close of the glacial epoch, man, it would seem,
like the flora which helped him to eke out subsistence, and
the reindeer which was his one animal friend, moved north-
wards.* Such at least is the general inference drawn from
the difficult problem known as the Hiatus. Between palæo-
lithic man and neolithic lies a gulf. † Man of the quaternary
stage, the so-called epoch of Madeleine, is the hunter and the
fisherman, without domestic animals, without agriculture,
without utensils or the rudest architectural device, as yet
unable to grind or polish the split stones with which he
waged his precarious struggle with the mammoth, the ice-
bear, and the larger felines, which still ranged the plains and
woods of Europe. With man of the neolithic period, the
epoch so-called of Robenhausen, all this is changed. There
is an advance that means a gap of centuries.
In some
localities, in parts of France for instance, in much, if not all,
of Germany and Switzerland, in Austria, and as some too
will have it, in Britain, the gap seems absolute; in others
it is imperfectly bridged, as for instance by the Cro-Magnon
men. When neolithic man appears, not only his acquisi-
tions and his habits, but he too himself is of a different
type. The long-headed skull of the quaternary drift is
replaced by the short-headed of neolithic times. Avoiding
long discussion on this head, and leaving on one side the
* O. A., p. 81–2; H. A., p. 62, 79.
+
† H. A., 96 pp., and cf. Boyd Dawkins, Early Man, p. 244, etc.
‡ H. A., p. 82.
THE CRADLE OF THE ARYANS.
51
Cro-Magnon men,* who, whether an original or immigrant
stock, seem to have advanced from the south-west to the
the occupation of France, Belgium, and the British Isles,
we discern two new invading types; first, the Iberian, or
Ibero-Semitic, moving upwards by way of Spain; and-of
far more import for language and for anthropology-the
Turanian stock, advancing from Asia and the East, bringing
with them the products and arts of Eastern civilisation, and
peopling eastern and central Europe with the dark-skinned
short-headed type, known as the Melanochroic, which to-day
still retains the numerical superiority in Europe.
In one region of Europe, and one alone, there is evidence
of a continuous development-in Scandinavia. This country
presents the archæologist with problems to which different
solutions have been given. From grey antiquity we find
there an intermixture of skulls, partly of brachycephalic,
partly of dolichocephalic formation. There was a mixed
population—the one a small minority, corresponding to the
Mongolian type of the modern Ugro-Finnish tribes, the
other in all essential characteristics to that of the modern
Swede. The older interpretation assumed that the brachy-
cephalic remains belonged to the first occupants, progenitors
of the modern Lapps or Finns. But sounder research con-
cludes that the dolichocephalic population can claim equal
antiquity that looking backward, they are of the same order
as the dolichocephalic skulls that belong to the paleolithic
age; looking forward, the indubitable representatives of the
mediæval Frank and the modern Swede. The theory that a
*O.A., pp. 89-94. It is disputed whether the Cro-Magnon men are
referable to the paleolithic or neolithic stage. Boyd Dawkins, Early Man,
pp. 206, 229, finds his most direct and unadulterated descendant in the
Eskimo, a theory which if true gives an instructive parallel to the retreat
of the blond type northward.
H. A., pp. 1-10, summarise the results attained by Nilsson, Hildebrand,
Montelius, Worsaae, Von Düben, and others.
52
THE CRADLE OF THE ARYANS.
1
population of earlier Lapps was pushed northward by the
advance of the fair whites has not been confirmed; on
the contrary, it appears that the Lapps eventually entered
Scandinavia from the north, and that between them and
southern Scandinavia lay, in these prehistoric times, an
impassable frozen barrier of hill, morass, and plain, which
yields no evidence of having been occupied or traversed by
man at this primitive stage. The second brachycephalic
order of skulls * (in percentage not exceeding ten per cent.)
must represent then a pristine population that died out, or
a Mongolian infusion that, as serfs or associates, formed a
part of the blond Aryan community. This dolicho-cephalic
race then shows here, through the stone age into the
bronze age and the iron, a unique instance of continuous
development. The famous kitchen middens found on the
south-eastern coasts of the Scandinavian peninsula are the
imperishable record of their slow advance. Steenstrup
assigns for these massed accumulations of bones and shells
and other refuse, periods amounting to 10,000 or 12,000
years, through which advance is barely, if at all, perceptible.
Then, whether from impulse communicated from without,
or as others (Steenstrup, Engelhardt, Evans, Torrell,
Montelius ), prefer to think, by self-development resulting
from improved conditions, advance begins. The ruder
implements mingle with others of superior form and finish,
bearing however the stamp of a self-developing transition,
not of mere importation, as elsewhere in Europe, at the
hands of a more gifted invading race. The mesolithic passes.
to the neolithic. The kitchen-midden men become the
Dolmen-builders, and the period of advance begins.
This gradual and unbroken continuity of development,
* O.A., p. 69–70.
† H. A., p. 53, 60–4, 84–6.
Montelius' Civilisation of Sweden has recently become accessible in
English through Rev. F. H. Wood's translation,
THE CRADLE OF THE ARYANS.
53
nowhere else traceable or admissible, carries with it an
important inference. Had there been a great subjugation, or
some overwhelming inroad of a superior race capable of
supplanting and exterminating every trace of the older
language, it is almost certain that here, as elsewhere, it
would reveal its traces to the archæologist. But if these
were Aryan-speaking men, or if development of language
shared the continuity of development in civilisation and the
arts, then it is certain that the direction of Aryan migration
was not from Russia northwards, but in the converse direc-
tion; for language and archæology combine to prove that a
higher stage than that of the kitchen-midden period had
been attained before the great migration and separation of
the Aryan stocks.*
Next comes, untraceable in detail whatever theory be
adopted, the period of the great expansion, with its develop-
ment into distinct Aryan stocks and languages. Archæology
traces the Dolmen-builders of the North passing southward
through France and Spain to Africa, and extending the
milestones of their march even to the borders of Egypt,
while in language the story of their expansion survives.
in the distribution of the Aryan tongues. Everywhere it
was carried by the blond dolichocephalic race that was
cradled on the Baltic shores; everywhere, as we have seen,
in Europe, Africa, or Asia, he has left representatives per-
petuating his physical characteristics. The key is found
to the sporadic appearance of the type in all regions that
betray Aryan influence. The hypothesis of so prolific and
exuberant an expansion of a single race from this centre
* H. A., pp. 34-6. As a side issue, the argument rebuts Pösche's hasty
hypothesis of origination near the Rokitno swamps, between the Dnieper and
the Bug. Anthropologically, it is quite untenable to attribute the persistent
blond type of the North to local depigmentation observed over so small a
district. And the theory finds no independent corroboration from either
language or early remains.
2
54
THE CRADLE OF THE ARYANS.
seems startling. Yet there is much to support it. Always,
from the very dawn of history, we are faced by the same
phenomena of countless hordes of northern men streaming
eastward, southward, westward, first vanquishing and then
by gradual absorption coalescing with the indigenous popula-
tion. The multitudinous irruptions of the Cimbrians and
Teutons are but historic repetitions of the great prehistoric
movements which spread the Aryan tongue and features over
the European world. The records of Egypt vouch for the
invasion of the country by a great European coalition,* with
contingents, as it would appear, from Sicily, Italy, and
Greece, in the reign of Menephtah I, which falls probably
into the fourteenth century B.C. The Galatai of Asia Minor
are but one less evanescent sample of myriad precursors,
and themselves anticipate the Crusaders of later date.
Always the early chroniclers, from Strabo and Tacitus to
Ammian, Procopius and Jordanes, trace German or Celt,
Lombard or Vandal, Franks or Juts, back to Scandinavia as
the hive from which they swarmed. It was the vagina
et officina gentium, the sheath and factory of nations. Move-
ments are from the North, southward and westward, not
vice versa.
The process still continues. Since the days of
Gustavus Adolphus, peaceable forms of expansion have
indeed superseded the periodical inundations of armed war-
riors. But in new channels the incessant stream of emigra-
tion still runs on, crossing the Atlantic instead of flooding
populous Europe. Always too, as a long chain of evidence
shows, it is the Aryan's fate to succumb even in conquering.
Only in the Scandinavian north does he retain his pristine
vigour. In Norway, at this present day, the blond type
remains universal, corresponding to the Germans and
Franks of the earlier Christian centuries. To-day the Nor-
* On this, in its European connexions, see Gladstone's Homeric Syn-
chronism, p. 141.
THE CRADLE OF THE ARYANS.
55
*
In
wegians not only retain the old physical characteristics, but
exhibit the highest average stature, the most prolific pro-
ductiveness, the lowest rate of child mortality, the highest
average longevity of any European race. The further we
recede from Baltic shores, the less stable it becomes.
Denmark, North Germany, and the British Isles, it best
maintains itself. Elsewhere it succumbs before the Tura-
nian black-haired, short-headed type of European man.†
The isolated districts where it survives are habitually high
table lands or exposed sea-coasts, which most nearly approxi-
mate in climatic condition to that of the Scandinavian
peninsula. High altitudes do not produce, but can conserve
the blond type that has found its way there. Its capacity
for successful acclimatisation is small; immeasurably infe-
rior, for instance, to that of the Jewish stock. The hundreds
of thousands of Northern barbarians who poured into Spain
and Italy like a deluge, from the days of the Roman Empire
to the close of the Middle Ages, have left there scarcely a
trace. In France, the old Frank type is fast yielding to the
Melanochroic, and is practically confined to the north.
The pictures of old masters shew how far more common,
alike in Italy and Germany, was the blond type only four or
five centuries ago than it now is. And still the process of
retrogression is visibly continuing.
In the light of these impressive facts, a theory, which
novelty puts at a disadvantage, grows more credible.
To the alternative hypothesis, namely, that the Aryan
language was the property of the Slavo-Celtic stock, the
objections are far more formidable; they seem, indeed,
insuperable. This melanochroic type appears in Central
* H. A., 115 pp., supply the statistics.
† O. A., pp. 100-4, 118; H. A., pp. 120-4.
‡0. 4., pp. 90-1; cf. H. A., 20 pp., 31.
56
THE CRADLE OF THE ARYANS.
Europe in the neolithic, not the quaternary period. It repre-
sents a wave of Asiatic immigration comparable to that
of the Huns in the fifth century, or the Mongols in the
thirteenth. The proof of this rests not only on the
evidence of physical conformation, but on the unassailable
testimony of Nephrit axes and other implements which Asia
only could have furnished. These immigrants brought
with them a higher civilisation than that as yet attained by
the Northern whites, and in all probability introduced into
Europe the various orders of domesticated animals,* of
which the dog alone appears to have been known to early
kitchen-midden men. But it is almost impossible to sup-
pose that they likewise imposed upon them their language.
Not only because their racial affinities, so far as that is
a trustworthy clue, imply some form of Turanian speech,
but for the far more solid reason that this brachycephalic
invasion never effected a lodgement in the northern regions
occupied by the blond Aryans. On the contrary, whenever
and wherever the two types do eventually intermingle, the
long skulls appear to denote the conquering, the short,
broad skull the subjugated race.t Such is the evidence
gleaned from the graves, where, as a rule, the remains
of chiefs are of the long-headed type. As another signifi-
cant fact, strongholds become most numerous where the two
types are most evenly matched, in Gaul, for instance, in
Britain, and in the Dnieper district, plainly attesting the
defensive struggle of an indigenous race against a dreaded
invader. In the seats of the blond Aryan, and throughout
North Germany, cities, strongholds or walled defences were
(as Tacitus, in later times, bears witness) unknown. In the
Middle Ages it still remained true, far more markedly
than is now the case, that throughout Germany and Switzer-
land and other European countries the blond type was
* Boyd Dawkins, Early Man, pp. 295–300. † O. A., pp. 20–8.
THE CRADLE OF THE ARYANS.
57
proportionately far more prevalent among the aristocracy,
while the serfs remained predominantly brachycephalic.
That a prolific aristocracy of immigrant warriors should have
gradually established their own language in vanquished
territory is conceivable; that a race of serfs should have
linguistically annexed regions to which they never pene-
trated is an impossibility.
Such then, mainly from the racial side, is the case in
favour of Scandinavia and North Germany as the cradle of
Aryan speech. It finds speaking corroborations from the
linguistic side.
The northern cast of the common Indo-
European and European vocabulary in respect to fauna and
flora, and the somewhat large infusion of marine names
become intelligible. It is said that every single animal,
bird, or tree that belongs to the common speech is a native
of the Scandinavian peninsula. The bone accumulations
show remains of the Bear, Beaver, Boar, Deer, Dog,
Duck, Fox, Goose, Lynx, Mouse, Otter, and Wolf, to which
with some force may be added the Eagle and the Swan.
The cow, goat, horse, and sheep all appear in the neolithic
period, to which the development of the mother speech
(Ursprache) must be assigned, and the civilisation of which
is in striking accord with the demands of Indo-European
lexicography.*
And passing from vocabulary to language-structure,
certain phenomena find new and suggestive elucidation.
It is well known that Ugro-Finnish dialects, though of the
agglutinative order, and classed with the other agglutinative
forms of Mongolian speech, show remarkable approximations
to the inflectional methods of Aryan in their treatment of
declensional and conjugational suffixes. The problem has
been a puz le to philologists, some of whom have actually
H. A., pp. 34-40; and compare Montelius and Boyd Dawkins, Early
Man, ch. viii.
58
THE CRADLE OF THE ARYANS.
regarded Finnish as attaining the inflectional stage. As we
pass eastwards to the allied Ural-Altaic, Samoyedic, or
Turko-Tataric populations these peculiarities disappear.
Alongside of this morphological approximation, there are
arresting resemblances in vocabulary, which lie deeper than
mere borrowing, and affect pronominal stems, numerals, and
primary verb roots. Now anthropology shows an exact
coincidence to this state of things. From very early times
there must have been a strong infusion of men of the blond
type among the Ugro-Finnish population, that extends east-
ward from Scandinavia. In Finland, the blond type appears
side by side with the dark brachycephalic almost as freely as
is seen for instance in Scotland or in Wales. As we move.
eastward the blond strain dwindles, and ere Asia is reached
wholly disappears. The testimony of language and of
anthropology combine to prove an early and a large infusion
of the Scandinavian blonds amid the neighbouring tribes,
imprinting a lasting stamp upon the physiognomy alike
of feature and of speech. And the contact seems fairly
traceable to a period earlier than that of the Aryan disper-
sion.
The reversed direction of linguistic movement involves
an entire reconsideration of established views upon phono-
logy. Within the last ten or fifteen years the phonetic
assumptions of Schleicher, Curtius, and inquirers of the
same school, have been revolutionised on internal evidence
of unanswerable force, by the completer researches of Osthoff,
Brugmann, Verner, and other workers in the same field. The
false views were in many cases due to the assumption, then
practically unquestioned, that Sanskrit and Zend represented
the primitive phonology more faithfully than the Western
languages. Penka claims that the hypothesis of an opposite
* For English summary see
I
Taylor's paper at British Association,
1887, and on whole argument O.A., pp. 63-8, H.A., pp. 24-6.
THE CRADLE OF THE ARYANS.
59
The consonantal groups
interesting coincidences
sibilation of an original
movement explains much that has hitherto seemed dark. In
Zend, and in Sanskrit still more completely, the primitive
vowel-gradations have been more irrevocably confused and
obliterated than in any of the European languages; A E O
have been merged into a uniform A.
offer more perplexing problems, and
of treatment. In the guttural series
k and g (palatal explosives) to s and z is found alike in the
Letto-Slavonic and the Indo-Iranian groups. Penka sees in
this an evidence of these two groups maintaining relation,
or else being exposed to like conditions subsequently to the
differentiation of groups. It finds a natural explanation in
the adoption of Aryan speech by a population, whose
phonetic idiosyncrasies naturally produced this modification.
It is parallel to the French representation of Latin c by the
soft c or ç; and the asserted sibilation of k in loan-words by
the Ugro-Finnic tribes is a speaking analogy, which suggests
that the change is due throughout to Turanian proclivities.
Similarly, in the phenomena known as labialisation of the
gutturals, as may be seen in Brugmann's t careful classifica-
tion, the Eastern languages again form with Armenian,
Albanian, Lithuanian and Slavonic, a
and Slavonic, a distinct group
contrasted with the European orders, Greek, Latin, and
Teutonic. The assumption of eastward progress renders
natural and intelligible a community of treatment which
otherwise baffles explanation.
In the complicated question of the Aspirates,‡ Penka
upholds a similar explanation. Philologists know that
Greek, Sanskrit, and Iranian constantly exhibit the aspirated
tenuis, where the tenuis simple appears in Celtic, Latin,
* O. A, pp. 139-43; H. A., pp. 50-1. For the exact representation, as
contrasted with treatment in the Western groups, see Brugmann, Compara-
tive Grammar, § 380.
+ Comparative Grammar, § 417.
‡ O.A., 158 pp.
60
THE CRADLE OF THE ARYANS.
Lithuanian, and Slavic. Curtius and others maintained that
the aspirate was a later development imported into the
Eastern group. Penka casts in his lot with those who hold
that the original language knew only aspirated mutes, and
observes that to this day all North-German dialects retain in
articulation the aspirated tenuis, which in the current spelling
appears as tenuis simple.* He attributes the disappearance
of the aspirate in the Western group to the phonetic habit of
the populations Aryanised over the European area. The
controversy is far too complicated for examination in these
pages, and Brugmann decisively vindicates the unaspirated
tenuis for the original speech, but upon any showing there
remains the salient fact that Greek here groups itself with
the Eastern division, a phenomenon which it is extremely
hard to reconcile with any theory of Asiatic origin, which
inevitably associates Greek with Italic, and exposes it to
similar environment long after its separation from Indian
and Iranian.
Finally, the series of sound-shiftings comprehended under
"Grimm's law," the explanation of which involves such
grave difficulties, is thus re-read by light of the new theory.
In the Ugro-Finnic tongues neither the aspirated tenues
(ph, kh, th) nor the aspirated media (bh, gh, dh) find a place,
while the simple mediæ, though existent, appear to have
been difficult and alien sounds, in origin derivative from
their corresponding tenues. This being so, it is intelligible
enough that in adopting Aryan speech Turanian populations
(like modern Hungarians in borrowing German terms) should
have represented Aryan media by the corresponding tenues,
while the unfamiliar aspirates, gh, bh, dh, naturally enough
fell to g, b, d. The predominance of the Turanian strain
* Saying for instance, Khind, Khunst. Ujfalvy's contradiction of this,
Berceau des Aryas, p. 30, arises from phonological ignorance.
† Penka's more elaborate account of the kh, ph, th modification I omit,
as it rests in great part on insecure hypotheses.
THE CRADLE OF THE ARYANS.
61
supplies the true explanation of the phonetic habitudes of the
South-German dialects. The second great sound-shifting,
affecting the High German dialects, commenced during the
earlier Christian centuries, and passed from the south north-
wards, representing the ground gradually gained by the
Turanian element on the pure German (or Aryan) stock. It
was achieved through monastic and other civilising influences,
by which the darker white has gradually inoculated and
superseded the blond, and in point of sound-change the
supersession never became complete, taking most effect
among the Bavarians and the Southern tribes most strongly
permeated with a Turanian element, and least where the
primitive blond stock experienced least of allophylic infusion.
Lastly, Penka appeals to mythology and legend in
support of his hypothesis. In Greek literature he finds clear
traces of the pristine northern home, not merely in the
Odyssey at large, the structure of which Müllenhoff identifies
with the German tale of Orendel, the historic hero of the
Northern Sea, but more specifically still in the conception
and legend of Oceanus, and in the notice of Cimmerian
folk-etymologically "the men of darkness," and in history
appearing as the Cimbri-"beside the bounds of swift-flowing
ocean, shrouded in mist and cloud, and never does the
shining sun look down upon them with his rays, neither
when he climbs the vault of starry heaven, nor when he turns
again from heaven's height towards earth, but deathly night
is spread o'er miserable mortals*"; while the tall long-
locked progenitors of the race are reproduced in the doughty
Laistrygones, “a host past number, like to the giants, not
men,"† of whom is written the strange record :—“ There
shepherd cries to shepherd as he drives home the flock, and
he that drives afield answers the call there, sleep foregone,
might a man earn double wage, one for herding the cows, the
* Od. xi. 14-19. † Od. x. 119.
62
THE CRADLE OF THE ARYANS.
other for pasturing white-fleeced flocks, so near are the out-
goings of the night and of the day." Other unconscious
obligations of Homer to Northern lore, to the phenomena of
icebergs in his description of the isle of Aeolus, and to the
northern legend of the ferrymen of the dead in his account
of the Phæacians, would similarly be explained as confused
reminiscences rather than as garbled reports gained from
Phoenician mariners.
A similar origin may underlie the Persian description of
the Aryas' land in the first chapter of the Vendidâd, where
'the winter months are ten, and the months of summer two,
and these cold for the waters, cold for the earth, cold for the
trees; and winter falls there, with the worst of its plagues.'
In the essay in which Van den Gheyn undertakes to tra-
verse the arguments of Penka, apart from partial demurrers
upon the anthropological side, which Penka himself con-
siders and overrules, there is little to invalidate the cumu-
lative strength of the case which he has made out, and
which at present offers the most complete and satisfactory
synthesis that exists of problems which are scarcely likely
to attain complete historic certitude.
Conclusion.
(
Our results may be thus summarised. The theory of
the Asiatic origin of Aryan speech is devoid of solid
evidence, and at present rests mainly on tradition and
unfounded prepossessions. Its one solid support, the claim
to higher primitiveness advanced on behalf of Sanskrit and
Iranian, has broken down; partly that the progress of
philology, especially in the province of phonetics, has inva-
lidated the claim itself, partly that primitiveness of type is
irrelevant to local origination. The theory is beset with
objections and difficulties: the absolute estrangement of
Aryan from Semitic, Accadian, or other forms of indigenous
* Od. x. 81-6.
THE CRADLE OF THE ARYANS.
63
Asiatic speech, constitutes a most formidable argument from
silence; the suggestions of history and the facts of geo-
graphical exploration are both adverse; the evidences of
Indo-European vocabulary almost fatal; the theory precludes
any rational account of the general unity of Western speech,
and leaves the observed relationships between the various
stocks unexplained and inexplicable. The theory of Euro-
pean origin elucidates all these difficulties, without creating
a single new one. It tallies with the evidences of vocabu-
lary; it explains the European unity; it allows for the
gradual differentiation of stocks, and gives scope for that
continued interrelation which the linguistic facts attest; it
admits the observed affinities; it has the support of phono-
logy; it is vindicated by large analogies in history; it
deserves, therefore, confident acceptance. If European
origin be conceded, the evidence derived from Indo-European
climate, trees and fish, especially the beech and the eel,
points decisively to the large northern plateau extending
from the German Ocean to the confines of the Black Sea and
Caspian watersheds, and this alone provides space for the
wide expansion implied by the broad differentiation of dia-
lects into separate stocks; vocabulary further implies fami-
liarity with marine animals and the sea, doubtless the North
Sea or the Baltic. So much seems well assured by successive
steps of argument. Beyond this, from the nature of the
materials, reasoning becomes precarious. At first sight it
must appear idle guesswork to attempt the racial identifica-
tion of the founders of Aryan speech. But examination
shows that the possible alternatives are few, and the col-
lateral evidences of anthropology at once more striking and
more harmonious than might have been expected; and
Penka has gone far towards establishing an association
between Aryan speech and the race of blond whites, whose
central and immemorial home is found in Scandinavia.
LIVERPOOL:
D. MARPLES AND CO. LIMITED,
LORD STREET.

UNIVERSITY OF MICHIGAN
3 9015 01138 0006
DO NOT REMOVE
OR
MUTILATE CARD