I’ve been awash in nostalgia reading Angelo’s reports on the LSA Institute at Sauvage Noble (1.1, 1.2, 1.3); ah for the days of splashing around in long-dead Anatolian languages and Tocharian transliteration! He reflects on “the basic nature of the divide between, e.g. Indo-Europeanists and ‘historical’ O[ptimality]T[heor]ists”:
Traditional historical linguistics is concerned with getting the description down, establishing the data, i.e. answering the question “what were the changes?”, and post-generative historical linguistics is concerned with accounting for “how and why were the changes?”. The former can be impressionistic regarding “how and why”, and the latter can play fast and loose with the “what”… (It’s striking to see how sparse the traditional classes are vs. the post-generative ones, that there’re more of us attending their classes than the other way around, at least based on my schedule and who I recognize.)
As I said in his comments, a thousand times better to have solid facts with insufficient theory than brilliant theory with undependable facts. (And I regret to say I’m not at all surprised at his parenthetical remark.)
He has this nice remark on the Hittites:
I forgot to mention my favorite item of Hittite coolness: the Hittites were inclusive, open to other religions. Their capital Hatti was known as “Land of a Thousand Gods”. Not only this, but they believed gods must be worshipped in their own language. So the non-Hittite gods accommodated into the Hittite pantheon were worshipped in their respective non-Hittite languages, as textual remains indicate. A sensible policy that optimized the coexistence of conquerors and conquered.