Semitic, Caucasian, and Uralian, and gather all of these, together with Indo-
European, into a super-family dubbed ‘Nostratic’. This shimmering construct
is of no consequence for the present study. But it is good to bear in mind that
Indo-European was not a unique, original entity like the primal cosmic atom
before the Big Bang. As a historical reality, it necessarily existed in a historical
context.^3
If there was an Indo-European language, it follows that there was a people
who spoke it: not a people in the sense of a nation, for they may never have
formed a political unity, and not a people in any racial sense, for they may
have been as genetically mixed as any modern population defined by
language. If our language is a descendant of theirs, that does not make them
‘our ancestors’, any more than the ancient Romans are the ancestors of the
French, the Romanians, and the Brazilians. The Indo-Europeans were a
people in the sense of a linguistic community. We should probably think of
them as a loose network of clans and tribes, inhabiting a coherent territory
of limited size. It has been estimated that in prehistoric conditions the largest
area within which a single language could exist without dividing into
mutually unintelligible tongues (as Indo-European, of course, eventually did)
might be of up to a million square kilometres –– roughly the size of Ontario ––
but was probably a good deal less.^4
A language embodies certain concepts and values, and a common language
implies some degree of common intellectual heritage. Within the original
common territory,^5 which we may call Eurostan, there no doubt existed local
diversities: differences of material culture, of dialect, of cult and custom. But
so long as the dialects remained mutually intelligible and there was easy
communication across the whole area, we might suppose there also to have
been a measure of shared tradition in such spheres as religion, storytelling,
and general ideology. If the evidence assembled in the present work is not
illusory, this theoretical expectation is fulfilled.
Indo-European studies have long spilled beyond the confines of purely
linguistic analysis and reconstruction. By the middle of the nineteenth
century some scholars –– the pioneer was Adalbert Kuhn –– had started to
make inferences from the linguistic evidence about the people who spoke
the proto-language: about their habitat, their conceptual world, their social
(^3) Typological similarities with other language families are reviewed by B. Comrie in Ramat
(1998), 74–97.
(^4) Mallory (1989), 145 f., cf. 64.
(^5) It should be understood that ‘original’ here does not mean ‘occupied from the beginning
of time’, but refers to the initial area from which the later diversification of Indo-European
languages proceeded; in other words, the territory occupied by the Indo-Europeans in the last
phase of development before they and their one language began to divide.
2 Introduction