November 23, 2000

8:45 PM

The Polynesians were, undoubtedly, the greatest navigators of the ancient world. Using outrigger canoes, they were able to colonize lands spread as far apart as Madagascar and Easter Island and as far south as New Zealand. But where did they originally come from? Jared Diamond demonstrates how, by using linguistic and archaeological evidence, it's possible to reconstruct their journey from China and Taiwan to the Philippines, from there on to Borneo, Indonesia, Malaysia, New Guinea and out to the Pacific one way and Madagascar in the other. As an exercise, try comparing the numbers 1 to 10 in all Polynesian and Indonesian languages, to see how the language gradually changed as they hopped from island to island.
posted by lagado (4 comments total)

This is OK as far as it goes, since he's depending on Roger Blust, a respectable source. But there's some stuff in here to be wary about. Foremost is the claim, taken from Merritt Ruhlen, that "The 260 or so Aboriginal Australian languages are usually considered to belong to a single language family". "Usually considered" is a vast overstatement. Ruhlen and his teacher Joseph Greenberg, are isolated extremists in this regard, and are the object of much controversy and downright nastiness in historical linguistics. Ruhlen in particular is fond of ex cathedra pronouncements like this. Non-experts like his work, because it makes big sweeping claims about the ultimate relatedness of all human languages, but much of the field considers him a loose cannon (at best).

Diamond's linking of Australian with Austronesian in this article might be confusing. They're completely different areas, despite the resemblance in names (there's also "Austric" and "Austro-Asiatic", just to make it even worse). The fact that Australian languages are "similar in their sounds but diverse in their vocabularies" is a red herring. No mainstream historical linguist would use "similarity of sounds" as a criterion for relatedness, and "diversity in vocabularies" is exactly what you'd expect for languages with long separate histories. The way you assess relatedness is by figuring out etymologies for words, and the way you do that is by finding patterns of regular correspondences between sounds.
posted by rodii at 10:44 AM on November 24, 2000

Yeah, Australian languages have a bunch of striking similarities--and not just in sound systems but in grammar and in more abstract areas like taboo vocabularies ("mother-in-law languages"). They're one of the great "linguistic areas"--areas in which languages with only a distant genetic relationship have striking similarities. Diamond is right in *not* seeing this as evidence of genetic relationship.

Languages tend to diverge *and* converge over time. They innovate by borrowing and by "internal" language change in ways that can take them away from once closely related varieties, but usually move them toward nearby "prestige" varieties. Actual geographic expansion is rarely rapid enough to lead to actual "steamroller"-style replacement (though your examples above are good ones)--it usually leads to diglossia, in which one groups speaks two varieties, that of the conquering group and the conquered (usually symbolized H for high and L for low). Over time this can lead to substantial convergence between the two. There are many, many examples of this, one of the best known being English. Given enough time and scope, it can be a major factor in the creation of linguistic areas--but it's doubtful that all LAs come from this.

One famous case of an areal resemblance that doesn't seem to have this explanation is the French/German "uvular /r/", where a particular pronunciation of /r/ is shared across language boundaries (northern France, the Low Countries, Germany) but isn't even universal within the languages in question (southern French != northern). However this feature spread, it probably wasn't due to political dominance of one group over another. Given the sort of highly multilingual situation Diamond points out (very common in Australian languages, btw), you can see how this kind of thing could operate on a nearly continent-wide scale.

The point is that lots of processes can lead to language convergence. Some people even think that the famous breakup of Indo-European was due less to the individual dialects naturally drifting apart than to their being influenced by various local languages. Untangling genetic relationships would be much less tricky if we didn't have all that to deal with.

(And, oops, "Roger" Blust is someone I went to high school with--apologies to Robert Blust.)
posted by rodii at 7:00 PM on November 25, 2000

As usual it's far more compex (and interesting) than first meets the eye.

posted by lagado at 4:06 AM on November 27, 2000