Maklew is spoken in Welbuti village.[1] The former two villages are located in Ilwayab district, and the latter three in Tubang district.
Classification
The two languages are transparently related.
Ross (2005) tentatively included them in the proposed Trans-Fly – Bulaka River family, but Usher, who reconstructs that family, does not connect Bulaka River to any other language family.[2]
Proto-language
Phonology
Usher (2014) reconstructs the consonant inventory as follows.[3]
Although the modern inventories of Yelmek and Maklew are nearly identical, they lack a one-to-one correspondence. Maklew in particular has been heavily influenced by Marind, and participates in a number of sound changes that occurred in that language. Usher posits:
j for Yelmek j ~ Maklew s (→ [z] in the Jab dialect of Yelmek; also found in loans from Marind /j/, which in some dialects is [hʲ])
w for Yelmek w ~ Maklew h (also found in loans from Marind /w/, which in some dialects is [hʷ])
ɣ for Yelmek ŋ ~ Maklew h (→ [g] in the Jab dialect of Yelmek; Makelew /h/ also found in loans from Marind /ɣ/, which in the central dialects becomes [h])
and, in loan words, mostly from Marind,
s for Yelmek t ~ Maklew s (→ [ts] is Jab).
In addition, there is a set of correspondences between alveolars in Yelmek and velars in Maklew (n~ŋ, t~k, d~g). Usher transcribes these as a series of palatal consonants (*ɲ *c *ɟ), but this is merely a typographic convenience. The phonetic forms are not easily recoverable, but most instances (8 out of 10) are followed by *e, suggesting that there was a vocal component. Usher suggests that *ɲ *c *ɟ might actually have been *niV *tiV *diV or *ŋiV *kiV *giV, none of which occur in the reconstructions despite the high frequency of the sequence *iV otherwise. The expected sequences *itV and *ikV also do not occur, so it's possible that *ɲ *c *ɟ reflect all three of these series, rather than a fourth place of articulation.
*m
*n
*ɲ (= *{n|ŋ}i/_V?)
*ŋ
*p
*t
*c (= *{t|k}i/_V?)
*k
*b
*d
*ɟ (= *{d|g}i/_V?)
*g
[*s]
*w
*l
*j
*ɣ
*i
*u
*e
(*ə)
*o
*a
The reconstruction of *ə is not firm, at least partly because the transcribed data is often unreliable.
There are vowel sequences of *iV and *uV. These might have been reconstructed as **jV and **wV, with no vowel sequences in the proto-language, but that analysis would require changing *w and *j in the consonant table above to **β and **ʝ, distinct from **w and **j, resulting in a larger set of consonants and an odd inventory of fricatives.
^Evans, Nicholas (2018). "The languages of Southern New Guinea". In Palmer, Bill (ed.). The Languages and Linguistics of the New Guinea Area: A Comprehensive Guide. The World of Linguistics. Vol. 4. Berlin: De Gruyter Mouton. pp. 641–774. ISBN978-3-11-028642-7.
