South Ramu River

Timothy Usher, Santa Fe Institute

The South Ramu River family consists of three languages spoken along and around the middle Ramu river in the South Ramu subdistrict of Papua New Guinea's Madang province. One of the languages, Rao, is spoken also across the border in East Sepik Province as far as the upper Keram River (Z'graggen 1971: 83-85.) It is called the Annaberg Group in the works of Z'graggen after the name of a mission station in Rao .

The internal classification of South Ramu River is as follows:

South Ramu River

Rao

Aram-Aren

Aram (Anor)

Aren (Aiome)

Rao and Aram-Aren are very different from one another, but more similar than either is to any other Ramu River family. Z'graggen (1971: 86) gives lexicostatistical relationships between the languages of the Annaberg Group as follows:

It's possible that Z'graggen's figures understate the degree of resemblance between Rao and Anor and Aiome respectively, as the shapes of many roots have undergone substantial transformation in Rao.

Z'graggen (1971: 83-86) includes Banaro in his Annaberg group, giving 9% and 5% resemblance between Banaro and Anor and Aiome respectively and 11% with Rao.

Note that the term Atemble or Atemble, the name of a village, has been used to designate both Aram and Aren as well as the Mand language of the Sogeram River subgroup of Madang immediately upriver to the east (Z'graggen 1971: 60, 1975: 141, q.v. Capell 1951-1952: 182.)

[under construction]

Capell (1951-1952: 182) grammar sketches of and (pp. 201-203) 79 comparative terms for Anaberg of Tjigurubu village and Atemble (Aram)

Z'graggen (1969, 1971) …

Z'graggen (1972) 431 comparative terms forf Anor (Aram,) Aiome (Aren) and Rao

Kaspruś (n.d.) vocabulary of Atemple (Aram) (unobtained)

Kaspruś (1973) …

Davies and Comrie (1985: 283-411) 46 comparative terms for Rao of Gokto village elicited by Lyle Scholtz

Stanhope (1980) grammar of Rao including (pp. 13-17) 368-term vocabulary and 34 inflected verb stems

[under construction]

Proto-South Ramu River had at least 15 consonants and perhaps 6 vowels as follows:

…

…

Initial consonants correspond as follows:

Subinitial consonants correspond as follows:

Medial consonants correspond as follows:

Vowels correspond as follows:

These correspondences are exemplified as follows, with Rao attestations drawn from Stanhope (1980: 13-17,) Rao, Anor (Aram) and Aiome (Aren) from Z'graggen (1972) and Anaberg (Rao) and Atemple (Aram) from Capell (1951-1952: 201-203.)

… initial consonants …

Initial bilabial nasal /*m/ … It is retained as such in all three languages:

One example of initial /*m/ is a regional loan:

Initial velar prenasalized voiced stop /*ŋg/ is found only in the first person singular pronouns. It is retained as such in all three languages:

… subinitial consonants …

Subinitial bilabial nasal /*m/ …

Subinitial apical nasal /*n/ …

When subinitial /*n/ is followed by high front vowel /*i/ …

Subinitial velar nasal /*ŋ/ …:

Subiniital bilabial voiceless stop /*p/ …

Subinitial apical voiceless stop /*t/ …:

Subinitial laminal voiceless /*s/ …:

Subinitial velar voiceless stop /*k/ …

Subinitial bilabial prenasalized voiced stop /*mb/ …:

Subinitial … /*mv/ …:

Subinitial apical prenasalized voiced stop /*nd/ …:

Subinitial laminal prenasalized voiced /*ndz/ …:

Subinitial velar prenasalized voiced stop /*ŋg/ …:

Subinitial bilabial non-stop /*w/ …:

… subinitial /*w/ …:

Subinitial apical non-stop /*r/ …:

Subinitial palatal non-stop /*j/ …:

Subinitial velar non-stop /*ɣ/ …:

… medial consonants …

Medial bilabial nasal /*m/ …:

Medial apical nasal /*n/… :

When medial /*n/ is followed by high front vowel /*i/ … :

Medial velar nasal /*ŋ/ …:

Medial bilabial stop /*p/ …

… /*f/ …

Medial apical voiceless stop /*t/ …:

Medial laminal voiceless /*s/ …:

Medial velar voiceless stop /*k/ …:

Medial bilabial prenasalized voiced stop /*mb/ …:

… /*mv/ …:

Medial apical prenasalized voiced stop /*nd/ …:

Medial laminal prenasalized voiced /*ndz/ …:

Medial velar prenasalized voiced stop /*ŋg/ …:

Medial bilabial non-stop /*w/ …:

Medial apical non-stop /*r/ …:

Medial palatal non-stop /*j/ …:

Medial velar non-stop /*ɣ/ …:

… final consonants ….

Final bilabial nasal /*m/ is dropped in Rao:

One unexplained exception may be a loan into Rao gtom Anor or Aiome:

Final apical nasal /*n/ …:

… /*n/ … :

Final velar nasal /*ŋ/ is supported only by outcomparison:

Neither voiceless stops /*p *t *s *k/ nor prenasalized voiced stops /*mb *nd *ndz *ŋg/ occur finally.

Final bilabial non-stop /*w/ …:

Final apical non-stop /*r/… :

…:

…:

…vowels ….

Rao reduces the first syllable of disyllabic roots. There are two ways in which this occurs, depending upon whether the root begins with a consonant or a vowel. In disyllabic roots with initial consonants, the vowel of the first syllable is dropped and carried over to the second, leaving an initial cluster which is in some instances simplified:

…

…

[under construction]

South Ramu nominative pronouns are as follows, with attestations drawn from Capell (1951-1952: 183, 186-187,) Z'graggen (1972) and Stanhope (1980: 9) as indicated:

….

[under construction]

Both Aram and Aren have borrowed basic terms from the Sogeram River subgroup of Madang, specifically from Mand and Nend immediately to the east which like South Ramu River languages drop most initial consonants (Daniels 2010.) Anor (Aram) and Aren (Aiome) attestations are drawn from Z'graggen (1972,) Paynamar (Manat,) Angaua (Nend) and Atemple (Mand) from Z'graggen (1980: 1-80) and proto-Sogeram and proto-West Sogeram from Daniels (2010: 186-192, 2015: 356-392):

A few of these etymologies deserve comment: 1) /*iŋkɨn/ “ground possum” is found also not only in Rao (above) but in Moam River /*iŋgʉan/ and Ruboni Range /*iŋgan/, an atymology which might be considered stronger than Daniels' Sogeram ,but developments wihin the latter may explain why the Aram-Aren medial is /*k/ rather than /*ŋg/ (cf. Daniels' Mend /(bor)ikɨn/, not in Z'graggen 1972,) . 2) /*akʷasa/ “areca nut” is found only in Atemple, Mand and Manat (a.k.a. Paynamar,) and is considered proto-Sogeram only because Daniels places Manat with Central Sogeram rather than with West Sogeram, as we suggest to be more accurate. 3) /*βaŋan/ “net bag” (/*paŋan/ 2010) is most likely related to Ramu River [*waɣan] (presumably really /*paɣan/) but the change /*ɣ/ to /*ŋ/ would be irregular. 4) Daniels glosses /*kinaŋ/ as “axe,” which is no doubt the same root as [*ɣinaŋ] found in other subgroups of Ramu River; however the Mand reflex is given as“bamboo knife” while Z'graggen (1972. 1980: 33) gives both Atemple and Aiome words as “bamboo”. 5) /*kura/ “forest” is appended with a question mark hy Daniels (2010: 188-189) himself for reasons which are not immediately obvious, but a similar and possibly related root /*uraŋ/ is found also in Ramu River; however the loss of final /*ŋ/ here would be irregular.

Some loans may have propagated in the other direction as they have etymologies in South Ramu River but not in Sogeramn River:

Again, some comments are in order: 2) “dog” is found also in Emerum (Apalɨ) as [iŋam] (Z'graggen 1972) though Daniels doesn't reconstruct it for proto-Sogeram River (cf. 2015: 390 /*upri/ “dog”.) 2) Daniels (2015: 363) reconstructs Sogeram River /*iau/ “fish” on the basis of the Mand (Atemple) word and outcomprison to Tauya /jau/. 3) Daniels (2015: 362) reconstructs Sogeram River /*arum/ “good” based upon Mand /arom/ “good”, Manat /arum/ “big, old”; and Mum /aru/ “good”; however Ramu River /*ɣarum/ “big/old” is better-supported, the initial consonant cannot be explained as a loan from Sogeram River and the meaning “good” is a development peculiar to Aram-Aren. Z'graggen (1972) gives Paynamar (Manat) [arumandiɛ] as “big”.

Still others have no known etymologies in either family: