33. The Andamanese Language Family

by George Weber

In the 1800-page "International Encyclopaedia of Linguistics" (Oxford University Press, 1992), where even obscure or extinct and little-known single languages receive many full pages and dozens of references, the Andamanese language family received one reference and four sentences of text plus a list of member languages - and that is it!

Andamanese is indeed the most obscure of all still living language families - by far!

The Andamanese language family is usually described as a complete isolate, i.e. it does not seem to have any linguistic connections to any other language or language family. However, during the last few decades, a few possible, if shadowy, connections have been made.

When the linguist Joseph H. Greenberg in 1971 published the results of his research into links between Papuan, Tasmanian and Andamanese languages, only a few linguistic specialists took note. Small wonder: linguistics had not contributed to the knowledge of the remoter human past before. Moreover, the languages investigated are among the least-known in the world. Greenberg called his new hypothetical linguistic grouping of around 750 languages the Indo-Pacific phylum and arranged it as follows:

1. Andamanese family (fourteen languages of which three are still living
2. Tasmanian family (classification controversial, up to nine languages, all extinct)
3-13. eleven Papuan language families (727 languages)

A language family is a very high-level classification and one that is rarely obvious. Greenberg used methods he had employed in his earlier work on high-level African (1950s) and Amerindian (1960s) language classification, compiling long lists of basic vocabularies along with whatever grammatical information he could lay hands on. Sifting through this material, he thought he had found 35 cognates that he thought connected Andamanese to Tasmanian. The discovery remains controversial, not least because our knowledge of the extinct Tasmanian is even more incomplete and confused than that of Andamanese. It must also be admitted that 35 cognates are not much to go on and Nicolas Burenhult has recently questioned the lack of systematic correspondences among Greenberg's cognates. However, his analysis (see Reprints) of the Andamanese languages has led him to suspect a far deeper relationship that, incidentally, re-admits the Andamanese-Tasmanian connection by another, much wider and more theory-oriented door. Burenhult himself says:

The most important point to be made concerning Andamanese... is that it is very different - not only genetically but also typologically - from neighbouring language families and may represent a trace of what Southeast Asia was like linguistically a few thousand years back.

Work recently done - but not yet published - on the mitochondrial DNA of the Andamanese and many other Asian and African populations also seems to point into the same direction. DNA results cannot say anything directly about a language, only about the people who speak it. Apart from the controversial 35 cognates and the even vaguer link to the Dravidian language family recently claimed by Indian researchers, it still shows no relationship with any other language grouping and remains an enigma.

What about the languages of other Negrito and possibly related people?

With the sole exception of the Andamanese, all other Negrito, Vedda and Veddoid groups have lost their original languages. The Vedda (from a Sinhala word meaning "hunter") of Sri Lanka have been speaking dialects of their neighbours' Indo-European Sinhala language for a very long time. Possible remnants of the prehistoric Vedda language have been reported in the late 19th century by the Sarasin cousins, who also expressed the hope that someone would do a systematic analysis of their evidence. Nobody has done so. The Sarasins noted a number of local terms used by some (but not all) Vedda groups that could well be survivors. Among them are three synonyms tambela, galrekki and malakedde for "axe" but also words used by some Vedda groups for "bow" and "arrow." The words vary from group to group which may reflect the existence of several original Vedda languages or, as the Sarasins themselves warned, the words could be local creations of ultimately Sinhala origin. Until someone does that rigorous analysis more than a century after it has been suggested, we have no way of telling.

There are also two Veddoid groups speaking unclassified languages that may not be related to the languages of their neighbours: the inaccessible Shompen of Great Nicobar speak a language about which virtually nothing is known (even if it is usually classified as Nicobarese, for lack of somewhere else to put it) while the unclassified Lom language from the interior of Bangka island in Indonesia, (off the eastern coast of Sumatra) has last been reported in the 19th century and has not been heard of since. Off the west coast of Sumatra on the tiny island of Enggano, a language is spoken that undoubtedly belongs to the Austronesian family but is said to be "extremely aberrant." If all this looks like there are some serious gaps in our knowledge - appearances are not deceiving.

As far as surviving traces of Negrito languages are concerned, things do look a little brighter. Traces of what must be the original Negrito language have been reported among the Semang of Malaya and the Philippine Negrito. That their change of language happened during prehistoric times is hinted at by the fact that the Malaysian Negrito speak languages of the Aslian branch of the Austro-Asian family, a family that dominated the area until two thousand years ago but has since been replaced on the peninsula except for isolated pockets by Malay and other Austronesian languages.

Among the Malay Semang Negrito words that cannot be traced to any other language have been reported. There are jebeg ("bad"), chog or seneng ("bag"), lebeh ("bamboo"), boo ("big"), kawod ("bird"), herpai ("coconut"), keto ("day"), kam ("frog"), chas ("hand"), napeg ("pig"), jekob ("snake"), wayd ("squirrel"), takob ("yam") and many others.

Recent painstaking research by Lawrence A. Reid has also uncovered a wealth of evidence for the existence of several prehistoric non-Austronesian Negrito languages in the Philippines. Reid has even found evidence to give a likely sequence for the beginning in the change of languages by the Negritos: around 5000 years ago the first Austronesian agriculturalists appeared in the Philippines probably from Taiwan and somehow persuaded local Negrito hunter-gatherers to labour in their new fields. Out of the need for a means of communications between immigrants and Negritos a pidgin language arose quickly which later, in the course of passing centuries, was creolised to such an extent that the modern Negrito languages of the Philippines came to bear a close resemblance to the neighbouring Austronesian languages. A similar sequence very likely also took place in Sri Lanka, the Malay peninsula and elsewhere whewre Negritos were overrun by later immigrants.

Reid has found many unique (i.e. non-Austalasian) terms. The record holder among Filipino Negrito languages is a term shared by no less than four modern Negrito languages: lati means "rattan" in the North Agta, Central Agta, Alta and Arta Negrito languages. Other unique sample words are shared among three Negrito languages each: litid ("vein"), tapur ("bury," "inter") and babak ("snake) and still many others occur in only one or two languages.

Traces of ancient and extinct Negrito languages found so far not surprisingly show no obvious relationship with Andamanese. No cognates have been found. It would indeed be asking rather a lot: even 5000 years ago the Negritos of Malaysia and the Philippines must have been out of contact with their Andamanese brethren for a very long time - their languages, if they were ever related would have drifted apart into profound mutual unintelligibility long before. Yet the hope remains that a painstaking analysis of the available evidence will one day reward a hardworking linguist with a surprising "aha!" event - and the invevitably following controversy.

[ Go to HOME ] [ Go to ANDAMAN Table of Contents ] [ Go to THE GREAT MIGRATION Table of Contents ] [ Go to APPENDICES Table of Contents ]

Last changed 3 May 2005