- The Early Celts -
available for each sector and in part from the fact that the Insular evidence, by and
large - and this holds true even for the earliest sources - belongs to a considerably
later date than the bulk of the primary Continental sources, the earliest of which take
us back several centuries Be.^17 Also a lot of the Continental evidence (especially
recently discovered inscriptions such as those of Botorrita, Peiialba de Villastar,
Chamalieres, Lezoux and Larzac) provides a fuller range of evidence, for all the
uncertainty still surrounding many readings, formations and patterns, than has
hitherto emerged from the very earliest Insular records, and this information relates
to syntax as well as phonology and morphology.18
Professor Karl Horst Schmidt has on several occasions sought to establish the
order of emergence of the individual Celtic languages and indeed tried to demon-
strate the means whereby one can exploit the evidence of individual Celtic languages
in order to reconstruct some of the features of a notional Proto-Celtic.^19 Relying
especially on his view of the well-known development of the Indo-European
labio-velar '~kw (and the cluster 'rkw) and the more controversial patterning of the
development of the Indo-European syllabic nasals in Celtic languages he has
assumed that Goidelic (the parent language of Irish) and Celtiberian broke away
from Proto-Celtic before the change of the labio-velar kw to the bilabial p in other
parts of the Celtic linguistic domain. Evidence of the sporadic preservation of the
labio-velar in Gaulish is thought of as 'basically archaic'. He also concedes that this
is in part a mark of the 'dialectal diversity of ancient Gaul'. He sees an opposition
between what he believes to be a change of the Indo-European syllabic nasals ,rill, '~1J
into Goidelic *em, 'ren, ("im, 'rin) as distinct from Gallo-Brittonic and Celtiberian
'ram, 'ran. In this highly selective analysis, controversial in itself, it is claimed
that the supposed 'unique development of the vocalic nasals is Goidelic' means that
'this language was the first to break away from Proto-Celtic'. Moreover, Schmidt
affirms that we can recognize the character of Proto-Celtic, up to a point, insofar as
'any Indo-European archaism preserved in one or other of the Cel~ic languages
must have been a feature of Proto-Celtic as well' and that the identification of inno-
vations shared by Celtic and other Indo-European languages mark them off as
features of Proto-Celtic. Part of the trouble with this analysis is that it does not
question the validity of the concept of a reconstructed Proto-Celtic language or
model, and, more seriously, that it makes too few more or less controversial analyses
of certain phonological features carry too much weight in the urge to perceive the
'order of emergence of the individual Celtic languages'.2o
V.P. Kalygin and A.A. Korolev have taken a different stance on this front in their
consideration of what they have termed 'The classification of the Celtic languages'.21
They stress that 'the differences between the Insular and Continental Celtic
languages cannot be explained by built-in differences in an original linguistic
structure', that these differences do not 'enable us to judge whether the Insular
Celtic languages had a single source or several' and that 'it is difficult to find any
individual criterion on whose foundation one might build a single classification for
all Celtic languages'. Although they declare their interest in Ursprachen and
Common Languages they rightly reject the concept of a monolithic kind of early
Celtic linguistic unity (,without any signs of dialectal articulation'). On the other
hand they, to my mind, strangely favour a panorama in which 'Disagreement over
I I