| Ongan | |
|---|---|
| Jarawa–Onge | |
| Geographic distribution | Andaman Islands |
| Linguistic classification | One of the world's primarylanguage families |
| Proto-language | Proto-Ongan |
| Subdivisions | |
| Language codes | |
| Glottolog | jara1244 |
 Distribution of the Ongan languages prior to 1850 (Fig. 1) and in 2005 (Fig. 2) | |
Ongan, also calledAngan,[1]Jarawa–Onge, or ambiguouslySouth Andamanese, is alanguage family which comprises two attestedAndamanese languages spoken in the southernAndaman Islands.
The two known extant languages are:
The extinct language of theJangil and that of the uncontactedSentinelese may be Ongan, but there is no linguistic information for either of them.
The label 'Southern Andamanese languages' is also used for the closely related southern pair of theGreat Andamanese languages.
The attestedAndamanese languages fall into two clear families,Great Andamanese and Ongan.
The similarities between Great Andamanese and Ongan are mainly of atypological andmorphological nature, with little demonstrated common vocabulary. Linguists, including long-range researchers such asJoseph Greenberg, have expressed doubts as to the validity of Andamanese as a family.[2]
It has since been proposed (byJuliette Blevins 2007) that Ongan (but not Great Andamanese) is distantly related toAustronesian in a family calledAustronesian–Ongan.[3] However, the proposal of a genealogical connection between Austronesian and Ongan has not been well-received by other linguists.George Van Driem (2011) considers Blevins' evidence as "not compelling", although he leaves the possibility open that some resemblances could be the result of contact/borrowing, a position also held by Hoogervorst (2012).[4][5]Robert Blust (2014) argues that Blevins' conclusions are not supported by her data, and that of her first 25 reconstructions, none are reproducible using the comparative method. Blust concludes that the grammatical comparison does not hold up, and also cites non-linguistic (such as cultural, archaeological, and biological) evidence against Blevins' hypothesis.[6]
| Proto-Ongan | |
|---|---|
| Reconstruction of | Ongan languages |
The two attested Ongan languages are relatively close, and the historical sound reconstruction mostly straightforward:
| Proto-Ongan | *p | *b | *t | *d | *kʷ | *k | *ɡ | *j | *w | *c | *ɟ | *m | *n | *ɲ | *ŋ | *l | *r |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Jarawa | p, b | b | t | d | hʷ, h | h | ɡ, j | j | w | c | ɟ | m | n | ɲ | ŋ | l | r |
| Onge | b | b | t, d | d, r | kʷ, h | k, ɡ | ɡ, Ø | j | w | c, ɟ | ɟ | m | n | ɲ | ŋ | l, j | r/j/l, Ø |
| Proto-Ongan | *i | *u | *a | *e | *o | (*ə) |
|---|---|---|---|---|---|---|
| Jarawa | i | u | a | e, ə, o | o | (ə) |
| Onge | i | u | a | e, ə, o | o | (ə) |
*ə appears to be allophonic for *e before a nasal coda.
The Ongan languages areagglutinative, with an extensive prefix and suffix system.[8][9] They have anoun class system based largely on body parts, in which everynoun andadjective may take aprefix according to which body part it is associated with (on the basis of shape, or functional association).[10] Another peculiarity of terms for body parts is that they areinalienably possessed, requiring apossessive adjectiveprefix to complete them, so one cannot say "head" alone, but only "my, or his, or your, etc. head".[10]
The Ongan pronouns are here represented by Önge:
| I, my | m- | we, our | et-, ot- |
| thou, thy | ŋ- | you, your | n- |
| he, his, she, her, it, its | g- | they, their | ekw-, ek-, ok- |
There is also an indefinite prefixən-, on- "someone's". Jarawa does not have the plural series, but the singular is very close:m-, ŋ- orn-, w-, ən-. From this, Blevins reconstructs Proto-Ongan *m-, *ŋ-, *gw-, *en-.
Judging from the available sources, the Andamanese languages have only twocardinal numbers:one andtwo and their entire numerical lexicon is one, two, one more, some more, and all.[9]
Nevertheless, it is conceivable that some of the given forms are genuinely related, though better explained as loans than common inheritance (p. 91).