CHAPTER TWO
THE EARLY CELTS
The evidence of language
--.... --
D. Ellis Evans
T
he study of the testimony of language concerning the early Celts has in recent
years reached a new and exciting stage of vitality, for all the difficulty and un-
certainty surrounding the interpretation of a relatively high proportion of the extant
sources. Work on the earliest Celtic linguistic data, both Continental and Insular,
has been carried forward by a distinguished array of leading scholars. This interest
reflects in part the continuing attraction of the earliest evidence available concerning
the languages of the Celts of antiquity, the precursors of the modern Celtic
languages. This interest has also been fired anew in part through the discovery of
valuable new sources (especially fairly long inscriptions) which, by and large, have
helped scholars to gain a better understanding of Old Celtic overall. We have had
several fresh attempts at tracing the emergence and interrelationship of Celtic
languages, both in general and in detail. We have also had a remarkable spate of
scholarly writing reflecting the abiding concern for establishing new archaeo-
linguistic models. Several of these have been applied to the world of the Celts, in the
enduring quest for a clearer understanding of Celtic origins and the celticization of
lands which came to be parts of that wide world.
I will admit at once that this scholarly activity has, by and large, not produced
results that are generally acceptable and enlightening. The labyrinthine and frustrat-
ing nature of the subject discussed here must not be denied or disguised, for all the
new insights gained from caring concentration on it. Its fascination and importance
remain undiminished for both Celtic and non-Celtic scholars alike. It is, for all its
hazards, significant for the study of linguistics and the early history (also to some
extent the prehistory) and culture of Europe.
In this short section we can briefly describe only some of the relevant evidence and
some of the ways in which it has been handled as part of the testimony concerning the
early Celts.l I wish to stress at the outset that we would do well to heed the insistence
of the social anthropologist, Edwin Ardener,2 on how difficult it is to identify a
people as 'part of an imaginary world' and, in relation to Celtic and other peoples,
how 'the expropriation of an image of another is a puzzling thing', and also how
(in Kirsten Hartrup's words) 'peoples are named and defined according to selected
criteria, reflecting a semantic density centring around only a minority of the actual
population'.^3