Roger D. WoodardRezensionen

Contains a chapter summarizing the historical context, writing system(s), phonology, morphology, and syntax of about every reasonably well-known language attested before about AD 500, as well as a number of more fragmentarily known ones. Being written by different authors, the various chapters differ a bit in style, focus, and depth, but all maintain a similar format.

More-or-less all the information is, I expect, available online, and in some cases significant advances in our understanding have occurred since publication, but while getting all the information for free may be possible, it is certainly not easy - the WP articles on many of the languages covered are, for example, rather sorry compared to the treatments here. So if an encyclopaedia of ancient languages sounds like your sort of thing, and you can find a copy for a reasonable price, I do recommend it.

http://nwhyte.livejournal.com/2527510.html

the European chapters extracted from the Cambridge Encyclopedia of the World's Ancient Languages, with a foreword explaining that the languages treated here are those with written records from before the fall of the Western Roman Empire (Cambridge University Press have generously put the whole thing online). That gives a shorter list than I would have thought, the chapters of the book covering Attic Greek, other Greek dialects, Latin, other Italic languages, Etruscan, continental Celtic, Gothic and ancient Nordic. I had not realised that written Irish was later than that. Obviously the chapters on Attic Greek and Latin have the most to say, but they are reasonably disciplined and establish a framework for the other languages that the reader may be less familiar with.

My discovery here is the weirdness of Etruscan, the only language on the list which is not from the Indo-European family. I'm intrigued by the numbers from one to ten - θu; zal; ci; huθ (or śa); maχ; śa (or huθ); semφ; cezp (probably); nurφ; śar - we don't even know whether huθ or śa is four or six. It's fascinating that the Etruscan word "zatlaθ", meaning axe carrier, became Latin "satelles" meaning bodyguard and is the origin of our word "satellite". I'm interested that like some Finno-Ugric languages, nouns take a lot of suffixes but have no gender. (Wikipedia says that the nouns did have gender, but Helmut Rix in this book says not.) And this language, long extinct, is a substratum for Latin which in turn has influenced every European language spoken today.

It is impressive that we have been able to reconstruct as much as we have, and I would have liked to read more about the process by which the ancient scripts were decoded. Most of them are at least vaguely related to the Latin and Greek alphabets which survive today, but only vaguely; if I were trying to decode them, I wouldn't know where to start. Some mysteries remain; the Gaulish letter known as the Tau Gallicum could have been pronounced st, ts, θ, or perhaps an emphatic t' like the Georgian ტ. Or possibly different Gauls pronounced it in different ways at different times.

And all of these languages are a melancholy reminder that life is short, and we have no idea what will survive. Many of the few surviving inscriptions in the lost language of Venetic are dedications to the goddess Reitia. Among other things, she is supposed to have been a goddess of writing, which is just as well as the other Venetic gods have been forgotten, as has any speaker of the language who did not leave their name in writing. And these languages, spoken by hundred of thousands who we could not now understand, are the exceptions rather than the rules. Humans have used language for hundreds of thousands of years, and the earliest European writing is the Linear B referred to in the extract above, from 3500 years ago, and the earliest Egyptian hieroglyphs are a thousand years older. So more than 95% of the thoughts ever thought, the stories ever told, the songs ever sung, are forgotten and cannot be retrieved.