Sunday, January 01, 2012 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Gell-Mann on evolutionary linguistics

Murray Gell-Mann gave a very interesting one-hour talk at CERN:

The topic is a favorite one for Murray Gell-Mann as well as for Tom Vonk and perhaps other TRF readers. In Santa Fe, Gell-Mann's favorite (mostly Russian) scholars have been working hard to reconstruct the tree of languages spoken on Earth.

The history of languages of the kind we consider human goes something like 10,000 years into the past. The "branches" of languages (e.g. the Slavic branch) are relatively recent and we know nearly everything about them and their common ancestors; the evidence for families (a greater unit including several branches, such as the Indo-European family) is extensive.

The evidence gets weaker and the reconstructions become less certain when you get to superfamilies and hypothetical super-superfamilies. Gell-Mann mentioned that North American linguists don't consider the research into "super-superfamilies" to be science – and I may understand where they're coming from. There may only be one such a super-superfamily. And maybe several of them. At all steps, one must be careful about the difference between vertical, "genetic" relationships between languages and horizontal ones, "borrowings".

One encounters some important biases. Languages or groups that are very important or widespread today or in a relatively recent history may have come from a relatively small subgroup of a small subgroup of some subset of a subset of a family (something that the linguists focusing on these languages may hate to hear because it diminishes their apparent importance!): but the resulting language could have spread for various reasons over much of the world, nevertheless. English, a provincial West Frisian dialect, is a major modern example. ;-)

So one needs to use totally different "weights" when he talks about the composition of languages in the past. The composition of the "survivors" we see today has a very different measure. Some previously minor if not generic dialects seem excessively important today while others that are extinct or nearly extinct today may have been much more important, widespread, and internally structured in the past.

As a guy educated in a relatively provincial Central European education system, I am of course much more familiar with the Indo-European languages: the European nations are my model used to visualize the cultural diversity of the world. Some examples of basic words in proto-languages that Gell-Mann mentioned are rather consistent with the idea that proto-Czech was the ultimate language of the first humans who were placed on Earth by the extraterrestrial aliens. In particular, the apparent similarity between Czech and Sanskrit has been pretty striking to see. ;-)

Nevertheless, it was refreshing to learn something about the vast diversity of the other, non-Indo-European languages, too.

The colloquium probably took place in June 2004 but I haven't seen it before.

Experimental proof of matrix string theory

This video with Veniamin Oprea inside is an experimental proof of the fact that slinkies, technically known as screwing string theory (or matrix string theory) and co-discovered by your humble correspondent, contain everything one needs for life:

You don't need fancy buzzwords such as "orbifold conformal field theory" to understand the video above. The only thing that the human slinky above doesn't explain sufficiently well is the DVV interaction. He would have to cut his body.

This blog entry was scheduled a few hours ago – in a different year.

Add to Digg this Add to reddit

snail feedback (2) :

reader Brian G Valentine said...

The only thing that the human slinky above doesn't explain sufficiently well is the DVV interaction. He would have to cut his body.

Or otherwise change his cobordism class.

On another note, I was amazed to learn recently how close Sanskrit was to Farsi.

reader Frank Ch. Eigler said...

The slinky video is predated by work by Mummenschantz back in the 70s.