Koch is particularly circumspect within his appraisal of the origins of Celtic (the language), but what he suggests is that Celtic is a much deeper stratum within Europe. You must be aware of the number of languages now extant in Europe - and when one counts the sub-national languages (Catalan, Alsace, Frisian etc.), let alone the dialects, there is a huge distinction number of them. In the Italian peninsula in the first millenium BC there was Latin, Etruscan, Umbrian, Oscan, Venetic, Lepontic, Cisalpine Gaulish, Raetic, Ligurian, Messapic, Faliscan and Greek being spoken. Within such a compact area, with strong contact between the groups, still there were so many disparate languages - which were not mutually intelligible. Yet, we are supposed to believe that there was a united language, lasting over 4 millenia, that was spoken across the European continent - even when we can deduce limited contact between these groups. It simply does not make any sense.
What Koch suggests means that we must re-address the language families and their relationships. If the Celtic language is to be placed early in the development of European languages then it pre-dates Italic and Germanic. It also means that the languages developed, much later, in the Western Atlantic zones and those that developed (separately) in the Danube basin are only distantly related, probably no more related than they are to Italic and Germanic languages.
This actually would address many of the problematic issues of the currently sprawling Celtic branch. It would explain why Celtic seems to be related in many ways to Italic and to Germanic languages (and also the possible relationship between Italic and Germanic). P-Celtic shares that shift with Germanic and Italic, for instance, and perhaps the many Celtic 'loanwords' in Germanic are because Germanic is derived from this much deeper stratum.
Bookmarks