The Early History

of Indo-European Languages
The common ancestor of these languages has been traced to Asia
rather than to Europe, the authors say. The once-clear distinction
between the family's Eastern and Western branches is now blurred

by Thomas V. Gamkrelidze and V. V. Ivanov

tic and Slavic families. Affinities be-

B iguistics, the scientific study of rasia, seeking at the origin of those

inguage, can reach more deep- paths the homeland of the protolan- tween these and the "Aryan" languag-
ly into the human past than guage itself. The early investigators es spoken in faraway India were no-
the most ancient written records. It placed the homeland in Europe and ticed by European travelers as early as
compares related languages to re- posited migratory paths by which the 16th century. That they might all
construct their immediate progenitors the daughter languages evolved into share a common ancestor was first
and eventually their ultimate ances- clearly defined Eastern or Western proposed in 1786 by Sir William Jones,
tor, or protolanguage. The protolan- branches. Our work indicates that the an English jurist and student of East-
guage in turn illuminates the lives of protolanguage originated more than ern cultures. He thus launched what
its speakers and locates them in time 6,000 years ago in eastern Anatolia came to be known as the Indo -Euro-
and place. and that some daughter languages pean hypothesis, which served as the
The science developed from the must have differentiated in the course principal stimulus to the founders
study of the Indo-European superfam- of migrations that took them first to of historical linguistics in the 19th
ily of languages, by far the largest in the East and later to the West. century.
number of languages and number of The reconstruction of ancient lan-
speakers. Nearly half of the world's
population speaks an Indo-European
language as a first language; six of the
10 languages in which Scientific Ameri-
can appears—English, French, Ger-
guages may be likened to the method
used by molecular biologists in their
quest to understand the evolution of
life. The biochemist identifies molec-
ular elements that perform similar
I n their reconstruction of the an-
cestral Indo -European language,
the early linguists relied heavily
on Grimm's law of Lautverschiebung
("sound shift"), which postulated that
man, Italian, Russian and Spanish—be- functions in widely divergent species sets of consonants displace one an-
long to this superfamily. to infer the characteristics of the pri- other over tune in predictable and
Over the past 200 years, linguists mordial cell from which they are pre- regular fashion. The law was posed in
have reconstructed the vocabulary sumed to have descended. So does 1822 by Jacob Grimm, who is more
and syntax of the postulated Indo- the linguist seek correspondences in widely famed for the anthology of
European protolanguage with increas- grammar, syntax, vocabulary and vo- fairy tales he wrote with his broth-
ing confidence and insight. They have calization among known languages in er, Wilhelm. Grimm's law explained,
tried to unravel the paths by which order to reconstruct their immediate among other things, why in the Ger-
the language broke into daughter lan- forebears and ultimately the original manic languages certain hard conso-
guages that spread throughout Eu- tongue. Living languages can be com- nants had persisted despite their uni-
pared directly with one another; dead versal tendency to yield to soft ones.
languages that have survived in writ- The set of softer, "voiced" consonants
THOMAS V. GAMKRELIDZE and V. V. ten form can usually be vocalized by "b," "d," "g" (followed by momentary
IVANOV are the authors of The Indo- inference from internal linguistic evi- vibration of the vocal cords), posited
European Language and the Indo-Euro- dence. Dead languages that have never in the protolanguage, had apparently
peans, a two-volume work published in been written, however, can be recon- given way to the corresponding hard
Russian in 1984; an English version will structed only by comparing their de- set "p," "t," "k." According to Grimm's
be published this fall by Mouton de law, this had come about by "devoic-
Gruyter. Gamkrelidze directs the Tseret- scendants and by working backward
eli Institute of Oriental Studies in Tbilisi according to the laws that govern ing" those consonants ("p," for exam-
and is a professor of linguistics at Tbili- phonological change. Phonology—the ple, is unaccompanied by vocal vibra-
si State University. Ivanov is professor study of word sounds—is all-impor- tion). Thus, the Sanskrit dhar is seen
of linguistics and chair of the depart- tant to historical linguists because as an archaic form of the English
ment of Slavic languages at the Institute sounds are more stable over the cen- "draw," which is itself more archaic
for Slavic and Balkan Studies in Moscow. turies than are meanings. than the German tragen (all of which
The authors wish to thank Gerard Piel, Early studies of Indo-European lan- mean "to pull").
chairman emeritus of Scientific Ameri-
can, for helping to prepare this article guages focused on those most famil- These rules were used to recon-
for publication. iar to the original European research- struct an Indo -European vocabulary
ers: the Italic, Celtic, Germanic, Bal- that implies how its speakers lived.
110 SCIENTIFIC AMERICAN March 1990
The words described a landscape and
climate that linguists originally placed
More recent evidence now places
the probable origin of the Indo-Euro-
in modern Turkey and as far east as
Tocharia, in modern Turkestan. Their
'
in Europe between the Alps in the pean language in western Asia. Three observations, together with new ideas
south and the Baltic and North seas generations of archaeologists and lin- in pure linguistic theory, have made it
in the north [see "The Indo-European guists have thus far excavated and necessary to revise the canons of lin-
Language," by Paul Thieme; SCIENTIFIC deciphered manuscripts in close to a guistic evolution.
AMERICAN, October, 1958]. dozen ancient languages from sites The landscape described by the

FAMILY TREE of the Indo-European languages can be traced ties to Celtic, an ancient European tongue. Similarities between
back to a protolanguage that flourished more than 6,000 years the Balto-Slavic and Indo-Iranian families indicate that they
ago. The protolanguage split into dialects, which evolved into influenced each other before their speakers moved north and
distinct languages; these then fissioned into generations of south, respectively. Dead languages are shown in italics; lan-
daughter languages. Tocharian, a dead language of Asia, has guages that left no literary remains are enclosed in brackets.

SCIENTIFIC AMERICAN March 1990 111

protolanguage as now resolved must 1200 B.C., after the fall of the Hittite Hattusas archives show that by the
lie somewhere in the crescent that Empire. To this emerging family of middle of the second millennium B.C.
curves around the southern shores of Anatolian languages linguists added the Indo-Iranian group had given rise
the Black Sea, south from the Balkan Lydian (closer to Hittite) and Lycian to a language spoken in the Mitanni
peninsula, east across ancient Ana-- (closer to Luwian), known from in- kingdom on the southeast frontier of
tolia (today the non-European territo- scriptions dating back to late in the Anatolia that was already different
ries of Turkey) and north to the Cauca- first millennium B.C. from ancient Indian (commonly called
sus Mountains [see illustration below]. Sanskrit) and ancient Iranian. Cretan-
Here the agricultural revolution creat-
ed the food surplus that impelled the
Indo-Europeans to found villages and
city-states from which, about 6,000
years ago, they began their migrations
T he appearance of Hittite and
other Anatolian languages at
the turn of the third to the sec-
ond millennium B.C. sets an absolute
chronological limit for the breakup
Mycenaean texts from the same eras
as Mitanni, deciphered in the early
1950's by the British scholars Michael
G. F. Ventris and John Chadwick,
turned out to be in a previously un-
over the Eurasian continent and into of the Indo-European protolanguage. known dialect of Greek. All these lan-
history. Because the Anatolian protolanguage guages had gone their separate ways
Some of the migrants invaded Ana- had already fissioned into daughter from Armenian.
tolia from the East around 2000 B.C. languages by that point, investigators Tocharian was another language
and established the Hittite kingdom, estimate that it departed from the family that diverged from the Indo-
which held all of Anatolia in its power parent Indo-European no later than European protolanguage quite early.
by 1400 B.C. Its official language was the fourth millennium B.C. and possi- Tocharian is one of the more recently
among the first of the Indo-European bly much earlier. discovered Indo-European languages,
languages to find its way into writing. This inference is supported by what first recognized in the early decades
Early in this century, Bedfich Hrozny, is known about the portion of the of the 20th century in texts from Chi-
a linguist at Vienna University and lat- Indo-European community that re- nese Turkestan. The texts were com-
er at Charles University in Prague, de- mained after the Anatolian family had paratively easy to decipher because
ciphered Hittite inscriptions (written broken away. From that communi- they were written in a variant of the
in cuneiform, the ancient writing sys- ty came the languages that persist- Brahmi script and were mainly transla-
tem based on wedge-shaped symbols) ed into written history. The first to tions from known Buddhist writings.
on tablets that had been found in branch off was the Greek-Armenian- Not long ago, the British scholar
the library of the capital at Hattusas, Indo-Iranian language community. It W. N. Henning suggested that the To-
200 kilometers east of modern Ankara. must have begun to do so in the fourth charians be identified with the Guti-
The library also contained cuneiform millennium B.C. because by the middle ans, who are mentioned in Babylonian
tablets in two related languages: Lu- of the third millennium B.C. the com- cuneiform inscriptions (in Akkadian,
wian and Palaic. The evolution of Lu- munity was already dividing into two a Semitic language) dating from the
wian could be traced in later hier- groups, namely, the Indo-Iranian and end of the third millennium B.C., when
oglyphic inscriptions made around the Greek-Armenian. Tablets in the King Sargon was building the first

TOCHARIA

E G Y P T I N D I A

MIGRATIONS AND CULTURAL DIFFUSION carried the Indo-Eu- to Iran and India. Most Western languages stem from an East-
ropean protolanguage from the homeland, which the authors ern branch that rounded the Caspian Sea. Contact with Semit-
place in the Transcaucasus, and fragmented it into dialects. ic languages in Mesopotamia and with Kartvelian languages
Some spread west to Anatolia and Greece, others southwest in the Caucasus led to the adoption of many foreign words.

112 SCIENTIFIC AMERICAN March 1990

great Mesopotamian Empire. If Hen-
VOICED ASPIRATES VOICELESS
ning's views are correct, the Tochari-
ans would be the first Indo-Europeans
to appear in the recorded history of (b) bh
the ancient Near East. Lexical affinities d dh t
of Tocharian with Italo-Celtic give ev-
idence that the speakers of the two g gh k
language families had associated in
the Indo-European homeland before GLOTTALIZED VOICED/ASPIRATES
the Tocharians began their migration
eastward.
The diverging pathways of linguistic b/bh p/ph
transformation and human migration f d/dh t/th
may now be traced back to a conver- k' g/gh k/kh
gence in the Indo-European protolan-
guage and its homeland. This has fol-
lowed from the revision in the canons THREE SERIES OF STOPS (consonants produced when the air passage is closed)
of phonology we mentioned above. An characterized proto-Indo-European. The classical model (top) posits that one series
uncontested peculiarity of the sound was voiced (followed by a vibration of the vocal cords, as in the "g" in "tiger"),
system of the protolanguage, for ex- another was voiced and aspirated (with an "h" sound not found in English) and
ample, is the near absence, or suppres- the third was voiceless (like the "k" in "disk"). In the authors' model (bottom), the
sion, of one of the three consonants first series was globalized (formed voicelessly by constricting the passage near the
vocal cords, as in the Cockney pronunciation of the "t" in "bottle"), the second had
"p," "b" or "v," which are labials (con- voiced and voiced-aspirated forms and the third had voiceless and voiceless-aspi-
sonants sounded with the lips). Tradi- rated forms. Glottalized stops bear a prime; absent stops are within parentheses.
tionally, it had been thought that "b"
was the suppressed consonant. Subse-
quent studies in phonology indicated, sure of the throat at the vocal cords the classical conception that the for-
however, that if one of the three labial that prevents the outward flow of mer languages had undergone a sys-
consonants is lacking in a language, it breath. Here the voiceless labial stop tematic sound shift, whereas Sanskrit
is least likely to be the one sounded as ("p"') appears suppressed, followed had faithfully conserved the original
"b" in English and other living Europe- by "t"' and "k'." As ("p"') is to ("b"), sound system.
an languages. voiceless and voiced, respectively, so The transformation of consonants
"t"' is to "d" and "k"' is to "g." Glottal- from parent to daughter languages

O n that basis we decided to re-

examine the entire system of
consonants posited for the
protolanguage, and as early as 1972,
ized stops occur in many different
language families, particularly those
of northern Caucasian and southern
Caucasian (Kartvelian) provenance.
The glottalized stop—which hardens
may be illustrated by the word "cow"
in English and Kuh in German; in San-
skrit the word for "ox" is gduh, and in
Greek it is bous. All have long been
recognized as descending from a com-
we proposed a new system of conso-
nants for the language. Our proposal a consonant—tends to weaken and mon Indo-European word for "ox," or
remains in the crucible of debate from disappear in most languages of the "cow." The word has different forms,
which consensus forms in every sci- world. So we surmised that—among however, in the glottalic and classical
ence. The debate now focuses more the labial stops—it was the "p"' rath- systems. In the glottalic it has the
strongly on features that relate the er than the "b" that most likely had voiceless consonant *k'wou- (the as-
Indo-European protolanguage to oth- been suppressed in the rndo-Europe- terisk before a word designates it as
er major language families and that an protolanguage. a word in the protolanguage), which
have at last begun to bring their com- Our so-called Indo-European glot- makes it phonetically closer to the
mon ancestor into view. talic system, which has been con- corresponding words in English and
According to classical theory, the structed by comparing the phonology German than to those in Greek and
"stop" consonants—those that are of the living and the historically attest- Sanskrit.
sounded by interruption of the out- ed Indo-European languages, appears In the classical system the word is
ward flow of the breath that excites more probable than the classical one. *gwou, which is practically the same
the vibration of the glottis, or vocal The near absence of the labial pho- as that in Sanskrit. In accordance with
cords—are divided into three cate- neme ("p"') finds a natural phonologi- Grimm's law, the transformation of
gories [see top of illustration on this cal explanation in relation to the evo- *gwou to the German would require
page]. The labial stop consonant "b" lution of the other two glottalized devoicing of the first consonant from
appears in the first column as a voiced stops and to the entire system of "g" to "k." And so the glottalic system
consonant; the parentheses enclosing stops shown above. seems to make the most sense: it
it there indicate its supposed suppres- eliminates the need for devoicing and
sion. It is associated with two other
voiced stop consonants: "d" (stopped
by the forward part of the tongue
against the palate) and "g" (stopped
by the back of the tongue against the
I n revising the consonant system of
the Indo-European protolanguage,
we have also called into question
the paths of transformation into the
historical Indo-European languages.
correlates the voiceless stops in the
Germanic languages (German, Dutch,
Scandinavian and English) with voice-
less glottalized stops in the ancestral
Indo-European protolanguage. In this
palate). Our reconstruction of the protolan- respect the Germanic languages are
In the scheme we have developed guage's consonants shows them to be more archaic than Sanskrit and Greek.
[see bottom of illustration on this page], closer to those of the Germanic, Arme- The glottalic system is seen, corre-
the corresponding consonants are nian and Hittite daughter languages spondingly, as more conservative than
sounded with a globalized stop: a clo- than to Sanskrit. This neatly reverses the classical system. It has brought the
SCIENTIFIC AMERICAN March 1990 113
 PROTOLANCUACE DAUGHTER LANGUAGES protolanguage closer to some of its
daughter languages without resorting
to such difficult phonological trans-
formations as that from "g" to "k."
We can learn more about the earliest
Indo-Europeans from other aspects of
their reconstructed vocabulary. Some
words, for example, describe an agri-
cultural technology whose existence
dates back to 5000 B.C. By that time
the agricultural revolution had spread
north from its origins in the Fertile
Crescent, where the first archaeologi-
cal evidence of cultivation dates back
to at least 8000 B.C. From this region
agriculture also spread southward to
sustain the Mesopotamian civilizations
and westward to Egypt. The Indo-Euro-
pean words for "barley," "wheat" and
"flax"; for "apples," "cherries" and
their trees; for "mulberries" and their
bushes; for "grapes" and their vines;
and for the various implements with
which to cultivate and harvest them
describe a way of life unknown in
northern Europe until the third or sec-
ond millennium B.C., when the first
archaeological evidence appears.
The landscape described by the re-
constructed Indo-European protolan-
guage is mountainous—as evidenced
by the many words for high moun-
tains, mountain lakes and rapid rivers
flowing from mountain sources. Such
a picture cannot be reconciled with
either the plains of central Europe or
the steppes north of the Black Sea,
which have been advanced as an alter-
native homeland for the Indo-Europe-
ans. The vocabulary does, however, fit
the landscape of eastern Anatolia and
Transcaucasia, backed by the splen-
dor of the Caucasus Mountains. The
language clothes its landscape in the
flora of this region, having words
for "mountain oak," "birch," "beech,"
"hornbeam," "ash," "willow" or "white
willow," "yew," "pine" or "fir," "heath-
er" and "moss." Moreover, the lan-
guage has words for animals that are
alien to northern Europe: "leopard,"
"snow leopard," "lion," "monkey" and
"elephant."

T he presence of a word for "beech

tree," incidentally, has been cit-
ed in favor of the European
plains and against the lower Volga as
the putative Indo-European homeland.
Beech trees, it is true, do not grow east
of a line drawn from Gdansk on the
Baltic to the northwest corner of the
Black Sea. Two species of beech (Fagus
orientalis and F. sylvatica) flourish,
WORD GENEALOGIES are traced as far as literary records go and then are reconstruct- however, in modern Turkey. Oppos-
ed, for the preliterate period, on the basis of laws governing the evolution of sounds. ing the so-called beech argument is
Reconstructed words are marked with an asterisk. Many Indo-European languag- the oak argument: paleobotanical evi-
es derive words for "man" or "earth" from *dheghom-, a root in the protolanguage. dence shows that oak trees (which are
114 SCIENTIFIC AMERICAN March 1990
listed in the reconstructed language's
lexicon) were not native to postglacial
northern Europe but began to spread
there from the south as late as the
turn of the fourth to the third millen-
nium B.C.
Another significant clue to the iden-
tification of the Indo-European home-
land is provided by the terminolo-
gy for wheeled transport. There are
words for "wheel" (*rotho-), "axle"
(*hakhs-), "yoke" (*iuk'om) and asso-
ciated gear. "Horse" is *ekhos and
"foal" *pholo. The bronze parts of the
chariot and the bronze tools, with
which chariots were fashioned from
mountain hardwoods, furnish words S4^
that embrace the smelting of met-
als. Petroglyphs, symbols marked on
stone, found in the area from the
Transcaucasus to upper Mesopotamia
between the lakes Van and Urmia are
the earliest pictures of horse-drawn
chariots [see illustration at right].
The postulated homeland of the
Indo-Europeans is, if not the only re-
gion, certainly one of the regions in
which the horse completed its domes-
tication and was harnessed as a draft PETROGLYPHS from the Uzbek Soviet Socialist Republic (dating from the second or
third millennium B.C.) provide archaeological corroboration of linguistic evidence
animal in the fourth millennium B.C. that the Indo-Europeans had chariots. Wheeled vehicles, such as those drawn here, fa-
From here wheeled vehicles spread cilitated agriculture and the migrations that resulted from a growing hunger for land.
with the migration of the Indo-Europe-
ans in the third and second millennia
B.C. eastward to central Asia, westward
to the Balkans, and in a circular mo- languages. Nickolai I. Vavilov, a promi- sia, eastern Anatolia and northern Iran.
tion around the Black Sea and thence nent Soviet plant geneticist, found a In the 2,000 years before the Indo-
to central Europe. vivid instance of such an exchange: Europeans who remained in the home-
The chariot provides significant evi- the Russian vinograd ("grape"), the land began to write history, the suc-
dence of cultural mixing, for chariots Italic vino and the Germanic wein cess of the agricultural revolution
figured in the funerary and other reli- ("wine"). These all reach back to the brought a population explosion to the
gious rites of both the Indo-European Indo-European *woi-no (or *wei-no), Indo-European community. The pres-
peoples and the Mesopotamians. Con- the proto-Semitic *wajnu, the Egyp- sure of population, we may surmise,
tact with other western Asiatic cul- tian *\vns, the Kartvelian *\vino and compelled the migration of successive
tures is also evidenced in the sharing the Hittite *wijana. waves of Indo-Europeans to fertile ar-
of various mythological subjects—for eas that were not yet cultivated.
example, the theft of the Hesperian
apples by Hercules and similar tales in
Norse and Celtic. Moreover, the Semit-
ic and Indo-European languages each
identify man with the earth. In He-
W e concede that in the broad ter-
ritory in which we have placed
the homeland of the Indo-Eu-
ropeans there is no archaeological evi-
dence of a culture that can be positive-
The linguistic translocation of the
Indo -European homeland from north-
ern Europe to Asia Minor requires
drastic revisions in theories about the
migratory paths along which the Indo-
brew, adam means "man" and ada- ly linked to them. Archaeologists have European languages must have spread
mah means "earth"; both were derived identified, however, a number of sites across Eurasia. Thus, the hypothetical
from a root in the Semitic protolan- that bear evidence of a material and Aryans who were said to have borne
guage (cf. Genesis 2:7,"... God formed spiritual culture similar to the one im- the so-called Aryan, or Indo-Iranian,
man from the dust of the ground"). plied by the Indo-European lexicon. language from Europe to India—and
"Human" and "humus" came to En- The Halanan culture of northern Mes- who were conscripted into service as
glish through Latin (homo, humus) opotamia decorated its vessels with the Nordic supermen of Nazi myth-
from *dheghom-, the word for "earth" religious symbols—bulls' horns and ology—turn out to be the real Indo-
and "man" (etymologically, "earthly sometimes rams' heads, which are Iranians who made the more plausi-
creature") in the Indo-European pro- masculine symbols, and ritual imag- ble migration from Asia Minor around
tolanguage. The rooting of the Indo- es of leopard skins—that are shared the northern slopes of the Himalaya
European languages in eastern Anato- by the somewhat later Catal Htiyiik Mountains and down through modern
lia is also suggested by the frequency culture of the seventh millennium Afghanistan to settle in India. Europe
of words borrowed from a number of B.C. in western Anatolia. Both cultures is seen, therefore, as the destination,
languages that flourished there: Se- have affinities with the later Transcau- rather than the source, of Indo -Euro-
mitic, Kartvelian, Sumerian and even casian culture in the region embraced pean migration.
Egyptian. Conversely, Indo-European by the Kura and the Araks rivers, Speakers of the Hittite, Luwian and
contributed words to each of those which includes southern Transcauca- other Anatolian languages made rel-
SCIENTIFIC AMERICAN March 1990 115
 atively small migrations within the in a new light. The colonies may now lithic structures, such as Stonehenge,
homeland, and their languages died be considered as very early settle- which they built near the periphery of
there with them. The more extensive ments that were established when the the continent.
migrations of speakers of the Greek- Greeks began migrating to their final The languages of the previous in-
Armenian-Indo-Iranian dialects began home in the Aegean. habitants of Europe, with the excep-
with the breakup of the main Indo- The historical European languag- tion of Basque—a non-Indo-European
European language community in the es—those that left literary remains- language with possible remote rela-
third millennium B.C. Two groups of provide evidence that the dialects tives in the Caucasus—were crowded
Indo -Iranian speakers made their way from which they descended had found out by the Indo-European dialects.
East during the second millennium their way into central Asia along with Nonetheless, those languages made
B.C. One of them, speakers of the Ka- the Tocharians. These languages have contributions to the historical Europe-
firi languages, survives to this day in many words in common. An exam- an language families that account for
Nuristan, on the southern slopes of ple is the word for "salmon," once re- certain differences among them. In his
the Hindu Kush in northeast Afghani- garded as a weighty argument for a study of the megalithic cultures and
stan. In Five Continents, a posthumous homeland in northern Europe. Salmon their disappearance, as well as of the
book recounting his many botanical abounded in the Baltic rivers of Eu- spread of farming from the ancient
expeditions between 1916 and 1933, rope, and the word /ox (German Lachs) Near East, the British archaeologist
Vavilov speculated that the Kafirs in the Germanic languages is perhaps Colin Renfrew has reached conclu-
might perpetuate some "original rel- echoed by lak- in Hindu, for a lacquer sions about the coming of the Indo-
ics" of Indo-Iranian. of a pink color that evokes the color of Europeans that agree well with ours
The second group of Indo-Iranians, salmon flesh. One species of salmon, [see "The Origins of Indo -European
who followed a more southerly path Salmo trutta, is found in the streams Languages," by Colin Renfrew; SCIEN-
into the Indus Valley, spoke a dialect of the Caucasus, and the lak-s- root TIFIC AMERICAN, October, 1989].
from which the historical languages of denotes "fish" in earlier and later Our deductions, resting so prepon-
India are descended. Their earliest lit- forms of Tocharian as well as in the derantly on linguistic evidence, must
erary ancestor is embodied in the Rig- ancient European languages. find confirmation in archaeological in-
Veda hymns, written in an ancient var- The migration of the speakers of vestigations that remain to be done.
iant of Sanskrit. The indigenous peo- some of the early Indo-European dia- Undoubtedly, the counting of base-
ples of the Indus Valley, known from lects into central Asia is established by pair substitutions in the DNA of hu-
the archaeological discoveries at their loan words from the Finno-Ugric lan- man cells will contribute to the family
capital Mohenjo-Daro, were apparent- guage family, which gave rise to mod- tree of the speakers of the Indo-Euro-
ly displaced by the Indo-Iranians. Af- ern Finnish and Hungarian. Under the pean languages and to the mapping of
ter the separation of the Indo-Iranians influence of Finno-Ugric, Tocharian un- their migrations. Anthropometry and
and their departure for the east, the derwent a complete transformation of history also will contribute to the ulti-
Greek-Armenian community remained its system of consonants. Words in the mate picture. Pending the elaboration
for a time in the homeland. There, ancient European languages that are and correction of our work, we may
judging by the numbers of loan words, clearly borrowed from the Altaic and state with a high order of certainty
they had contact with speakers of other languages of central Asia give that the homeland of the Indo-Europe-
Kartvelian, Tocharian and the ancient further testimony to the sojourn of ans, the cradle of much of the world's
Indo-European languages that later their speakers there. civilization, was in the ancient Near
evolved into the historical European Circling back to the west, the ancient East: "Ex oriente lux\"
languages. One such borrowing from Europeans settled for a time north
the Kartvelian became the Homeric of the Black Sea in a loosely federat-
koas, "fleece." ed community. Thus, it is not entire- FURTHER READING
ly wrong to think of this region as a INDO-EUROPEAN AND THE INDO-EUROPE-
A bilingual cuneiform tablet found second homeland for these peoples. ANS: A RECONSTRUCTION AND HIS-
/\in the Hattusas archives records From the end of the third through TORICAL TYPOLOGICAL ANALYSIS OF A
JL \. the mythological tale of a hunt- the first millennium B.C., speakers of PROTOLANGUAGE AND PROTO-CULTURE.
er in the then already dead Hum- ancient European languages spread Parts I and H. Thomas V. Gamkrelidze
an language along with a translation gradually into Europe. Their coming and Vjacheslav V. Ivanov. Tbilisi State
University, 1984.
into Hittite. This remarkable discov- is demonstrated archaeologically by ARCHAEOLOGY AND LANGUAGE: THE PUZ-
ery gave us the Human word ashi the arrival of the seminomadic "pit ZLE OF INDO-EUROPEAN ORIGINS. Colin
from which Homer's askos, for "hide" grave" culture, which buried its dead Renfrew. Cambridge University Press,
or "fur," apparently stemmed. Before in shafts, or barrows. 1988.
their migration to the Aegean, the RECONSTRUCTING LANGUAGES AND CUL-
Greeks borrowed the Hittite word kur- A nthropometry, which is the scien- TURES: ABSTRACTS AND MATERIALS
sa, which by a familiar phonological / \c measurement of the human FROM THE FIRST INTERNATIONAL INTER-
DISCIPLINARY SYMPOSIUM ON LANGUAGE
shift became bursa, another synonym J. JLbody, has begun to chart the AND PREHISTORY, ANN ARBOR, NOVEM-
for "fleece." These words seem to con- imposition of the Hittite physiogno- BER 8-12,1988. Edited by Vitaly Shevo-
firm the Greeks' belief that their an- my, typified in Hittite reliefs, on cer- roshkin. Studienverlag Dr. Norbert
cestors had come from western Asia, tain European populations. The blue- Brockmeier, 1989.
as recounted in the myth of Jason and eyed, blond-haired Nordic must still IN SEARCH OF THE INDO-EUROPEANS: LAN-
the Argonauts, who sought the Golden be regarded as the product of inter- GUAGE, ARCHAEOLOGY AND MYTH. J. P.
Fleece in Colchis, on the eastern shore breeding between the Indo-Europe- Mallory. Thames and Hudson, 1989.
of the Black Sea. The evidence that the an invaders and their predecessors in WHEN WORLDS COLLIDE: INDO-EUROPE-
ANS AND PRE-INDO-EUROPEANS. Edited
Greeks came thence to their historical the settlement of Europe. The culture by John Greppin and T. L Markey. Karo-
homeland puts the Greek "colonies" of the indigenous populations of Eu- ma Publishers, Inc., 1990.
on the northern shore of the Black Sea rope is memorialized by the mega-
116 SCIENTIFIC AMERICAN March 1990