George Starostin: Interests
My main area of occupation is historical linguistics, which is, quite naturally, a field where you can rarely see one end from the other, so if you split it into sub-topics, here are the ones that occupy me the most:
Although the comparative method, invented and perfected by linguists in the XIXth century, today remains the main driving engine of all historical linguistics (and is probably bound to remain in that status just like simple arithmetics will always remain at the core of mathematical studies), there's still lots of room for further improvements. Importance of linguistic typology for reconstruction, issues of relative and absolute dating for prehistorical changes, problems of semantic reconstruction - none of these questions have yet been resolved to general satisfaction. I am not all that interested in the theory of historical linguistics per se, but looking at how various proposed theoretical hypotheses fare with particular language families can be quite fascinating - and profitable.
I count myself as belonging to the Moscow school of comparative linguistics, and as such, insist that research on distant language relationships, which aims at classifying language families on high levels and reconstructing proto-languages going beyond 6,000 years and maybe even much further, can - and should - be as rigorously scientific as research on traditionally accepted families (e. g. Indo-European, Uralic, or Semitic). The acceptance of this paradigm, however, comes with a whole lot of methodological baggage: one has to state one's methods and objectives very clearly in order not to fall in with the crowd of mystics and maniacs who claim their native language to have been the original idiom of the first humans - or, at the very least, that it was the language of the Phaistos Disc. Thus, although my work on long-range comparison is mostly practical, methodology plays a large part in it.
One of my main duties as participant of the EHL program is to summarize and re-test currently available hypotheses on the overall taxonomy of the languages of Africa, including such large and/or notorious language families as Nilo-Saharan, Niger-Kordofanian, and Khoisan. To do that, I am using a variety of methods, ranging from a refined and formalized version of 'wordlist comparison based on similarity' (vaguely reminiscent of the late Joseph Greenberg's mass comparison, but restricted to small amounts of lexics with no semantic drift and more formal definition of phonetic similarity) to a rough preliminary application of the comparative method to the analyzed material, as well as an automated statistical algorithm of data analysis. With more than 1,500 languages at stake here, I am not sure if I will ever manage to cover even a third of that distance, but there's no harm in trying.
Khoisan languages (formerly known as Bushman-Hottentot) form one of the most unique unities on the planet. This is primarily due to their predilection for click sounds, which are not found in such extreme abundance in any other language on the planet (and where they are found in other African languages, it is only because the Khoisan people gave them away). However, there are many more noteworthy, complex, and rare features to be met among the various Khoisan idioms. Unfortunately, most of the languages are already extinct, with their speakers assimilated by waves of first Bantu and then European intruders into South Africa, and in order to classify them, analyze their structure and reconstruct their prehistory, one has to strike a very delicate balance between relying on old, frequently poorly transcribed data and newer, better represented data which is only to be found for a small handful of languages. Still, it's only through careful work with these sources that we may hope to confirm or disprove some of the fascinating rumours on Khoisan - for instance, the idea that click sounds represent part of their inheritance from the "primordial language". My research currently focuses on reconstruction of the proto-languages for smaller subbranches of Khoisan (such as Zhu or !Ui-Taa), but I'm also probing the ground for higher level relationships.
In my PhD ("candidate of philological sciences") thesis, I have proposed several significant modifications to the traditional model of the phonology of Proto-Dravidian - the ancestral language of present day Tamil, Malayalam, Kannada, Telugu, Tulu, and about twenty more "minor" Dravidian languages mostly spoken in South and Central India. Time will tell if all of these modifications were an adequate interpretation of the available data, but in the meantime I am slowly continuing my research on this important language family, expanding and explicating the system of etymological databases and also trying to properly integrate Dravidian data into the general framework of Nostratic.
My only fieldwork was done in 1993 with the Ket language in Siberia - one of the most grammatically complicated languages of the world and also, as it turns out, one of the important linguistic links between the languages of the Caucasus and the Sino-Tibetan family. I have written both on the synchronic structure of Ket and the reconstruction of the verbal system of Proto-Yeniseian (the language family to which it belonged, along with Kott, Arin, and Pumpokol, three by now completely extinct languages of the same area). Today, I am primarily interested in Yeniseian as part of the Sino-Caucasian macrofamily, sometimes also called Dene-Caucasian if the Na-Dene family in North America is to be included. (See Dene-Yeniseic for the currently most promoted version of the external relations of Yeniseian - it should be noted, however, that relationship between Ket and North Caucasian, as well as Sino-Tibetan languages, is even more evident than that between Yeniseian and Na-Dene once proper methodology is applied).
Old Chinese is one of the most fascinating languages in the world. With a complicated theory of how to reconstruct the way it was read, a unique grammar that not only relies completely on syntax rather than any noteworthy amount of morphological means, but also discards with such a basic notion as parts of speech, and a history that goes back more than three thousand years and includes literary works of great cultural and historical significance, there is a whole sea of problems to be solved in its study. My primary concerns when working with Old Chinese, as well as teaching it, are with trying to weed out the chaff, i. e. sort the "pure", spoken old Chinese from layers of "dead" literary weight accumulated on top of it when it was no longer a living language. Most existing translations of Old Chinese literature pay little attention to that problem, and frequently end up distorting the original meaning through anachronistic interpretations or ones not based on true textual evidence - hopefully, the situation will be rectified some day. I am also interested in the status of Chinese as a Sino-Tibetan language - the reconstruction of its ancient pronunciation and the importance it holds for our overall understanding of Sino-Tibetan taxonomy and prehistory.
The original hypothesis that several of Eurasia's large macro-families constitute a huge "macrofamily" called Nostratic, proposed by V. M. Illich-Svitych, has stood the test of time (no matter what the hyperskepticists may object), but, of course, the details are undergoing constant revision and improvement, since reconstructions for lesser families are also undergoing constant revision. Work on a more precise phonetic correspondence system, reconstruction of grammar and lexicon, statistical evaluation of the data, and also establishing which lesser families of Eurasia are Nostratic and which ones are not, is still going on, and I try to take part in it whenever possible.
As mentioned above, the Sino-Caucasian theory, originally put forward by S. Starostin, proposes to demonstrate that North Caucasian, Yeniseian, and Sino-Tibetan languages all spring from one common ancestor. To these three main groups may also be added such language isolates as Basque in Spain and Burushaski in Pakistan; the addition of Na-Dene languages (Eyak-Athapaskan, Tlingit, and possibly also Haida) in North America also seems very likely. The demonstration of relationship between these groups is a rather complicated task, since most of the languages either have tremendously complex phonological systems (North Caucasian, Na-Dene) or, on the contrary, have undergone radical simplification of the original system (Sino-Tibetan); likewise, they either have very complex morphologies that are hard to trace back to each other (North Caucasian, Yeniseian, Na-Dene) or have stripped away morphology almost completely (Sino-Tibetan again!). Still, at the moment much progress has been made, and I am currently editing comments to S. Starostin's database on Sino-Caucasian to try and make things easier for those not well versed in the data.
I don't have too many of those, to be honest, let alone ones that could be bragged about on an Internet site. I do have a relatively large record collection of popular music, mainly from the Sixties and upwards, much of which I have reviewed on my music-dedicated site, Only Solitaire, because I couldn't figure out what else to do with it. Strange enough, some people actually liked my reviews even though I can't read music, much less write it; so it's only fair that many more people hated them. But I did get a few gentlemen into Bob Dylan's Selfportrait and Paul McCartney's Ram, so, in my eyes at least, there was some practical use to the site. I don't update it much any more, but I still occasionally write a little bit on music - my Marginal Observations.
Another odd interest of mine is classic Sierra adventure games; I have a weird theory that this type of games represented the pinnacle in computer entertainment, and that, since this pinnacle is no longer standing, computer entertainment nowadays isn't really worth a damn thing. As a small, humble tribute to said pinnacle, I am occasionally adding little touches to my Sierra-dedicated site - hoping that it might get a few people to try out those old classics and, perchance, to recognize true gaming bliss when they see it.