Please enable JavaScript to view the comments powered by Disqus. blog comments powered by Disqus
Of Interest:

A more considered response...

To follow up on yesterday’s post regarding the latest theory in the field of historical linguistics, Sam had posted this comment in a thread elsewhere on the web:

“[...] There is no such thing as a constant rate of change when we talk about language. Contemporary Icelandic is, for all intents and purposes, exactly the same as Old Norse, such so that a schoolchild in Iceland can read the thousand year old eddas and sagas in the original with no issues. Compare English, and how we almost need a dictionary for Shakespeare, written only four hundred years back. "Glottochronology" is pseudoscience and woo. Comparative Historical Linguistics, as an academic pursuit, relies on actual evidence when reconstructing protoforms, and thus can only make claims about the common ancestor of two or more languages of which we have actual evidence, usually of the written variety. This pretty much only goes back about 5,000 years, with the Sanskrit of the Rig Veda being the oldest attested Indo-European language we have, therefore closest to the protoform. In the same way that one can reconstruct Latin if one knows about French, Italian, Spanish, Romanche, etc., it is possible to reconstruct I-E with knowledge of Old Church Slavonic, Greek, Latin, Sanskrit, Hittite, etc. I know of no proof of efficacy for any attempt to posit relations between the larger language families like Semitic, Altaic, Bantu, Sino-Tibetan, and I-E, etc., and call anyone out on the verifiability of any such claims.


We have an awful lot more than a few words in proto-Indo-European, actually, and a very good idea of the society and belief system in which they were spoken. Even some of the laws and religious beliefs and practices. Here's a good cocktail fact for example: the words 'naked' and 'night' are, in fact, etymologically related -so we can posit that the Indo-European speakers probably wore no pyjamas...


The methodology used for the study is also suspect, in my opinion. The claim is that the phonemic inventory (i.e., the set of allowable sounds within a given language) shrinks the further one gets from southwest central Africa, but they've only sampled 504 languages. It is generally agreed that there are about 5,000 to 7,000 languages in the world, so why is the sample so small? The grammars of most all of these languages have been published over the years, so the data are available...


Nonetheless, the article is making some ripples in the unusually staid waters of the field. Closer inspection is warranted... “


We’d be glad to have your thoughts on the matter.
blog comments powered by Disqus