Families‎ > ‎Papuan Gulf‎ > ‎

Tua River

    Timothy Usher, Santa Fe Institute


    [under construction]

    The Tua River family, also called Teberan-Pawaian, is comprised of at least four very different widely-spoken languages, Dadibi, Folopa and Pawaia, and Witu…Tua River and Pio River…… tributaries of the Purari around the intersection of Papua New Guinea's Gulf, Southern Highlands and Simbu provinces.…MacDonald (1973: CITE.)


    The internal classification of Tua River is as follows:

    Tua River

    MacDonald (1973: 122-124) gives lexicostatistics for Tua River as follows:

  Witu Pawaia Dadibi Polopa
 Witu --- 2 8 8
 Pawaia 2 --- 10 16
 Dadibi 8 10 --- 35
 Polopa 8 16 35 ---

    [under construction]

    Franklin (1973: 264-268) lexicostatistics…


    Murray (1918-1919) 60 comparative terms for Pawaia (Huaruha)
    Brown (1919-1920) 113 comparative terms for Bara variety of Folopa
    Brown (1919-1920) 117 comparative terms for Pawaia (Aurama)
    Johnston (1919-1920) 126 comparative terms for Ibukairi variety of Folopa
    Lind (1921-1922) 94 comparative terms for Foraba variety of Folopa
    Lind (1921-1922) 102 comparative terms for Ro, Keai and Worugi variety of Folopa
    Chance (1923-1924) 112 comparative terms for Sesa variety of Folopa
    Kerr (1967) grammar of Witu
    Kerr (1975) comparison of Wiru to proposed East New Guinea Highlands stock
    Wurm (1975: 488-490) recapitulates Wiru phonemic inventory, pronouns and verbal desinences drawn from Kerr (1967)
    Franklin (1968) Gulf District …
    Franklin ed. (1973) 100 comparative terms for Wiru, Uraru Pawaia, Dadibi and for Sopese, Boro, Suri and Tebera varieties of Folopa
    Franklin (1975: 264-268) 84 comparative terms for Wiru
    Trefry (1969) comparative study of Kuman and North Pawaia, 200 comparative terms, miscellaneous examples and grammar notes
    Trefry (1972) phonology of North Pawaia
    MacDonald (1973) comparative notes on Dadibi-Folopa and Tua River
    MacDonald and MacDonald (1974) phonology of Dadibi
    MacDonald (1976) grammar of Dadibi
    MacDonald (1992) sketch phonology of Dadibi
    MacDonald survey vocabulary of Dadibi provided by Paul Whitehouse via Summer Institute of Linguistics in Ukarumpa
    Anderson and Wade (1988) Folopa ergativity
    Anderson (1989) Folopa existential verbs
    Anderson (1994) sketch phonology of Folopa
    Anderson and Anderson (1995) phonology of Folopa
    Anderson (2009) Folopa serial verbs
    Anderson (2010) Folopa grammar and (pp. 97-107) vocabulary
    Anderson survey vocabulary of Folopa provided by Paul Whitehouse via Summer Institute of Linguistics in Ukarumpa
    Whitby and Bai (1990, 2008) dictionary of Dadibi

    Additionally, the Summer Institute of Linguistics provides sketch phonologies of Witu and North Pawaia, but they're unattributed.

    History of classification

    [under construction]

    Dadibi-Folopa Teberan …
    Teberan-Pawaian …
Wurm (1962: 118) Mikaru placed with East New Guinea Highlands Phylum. Franklin (1968: 25) Mikaruan family. MacDonald (1973) renames Teberan family     Franklin (1975: 860) Trans-Murray stock, Dadibi-Folopa( Teberan,) Pawaia and Foe. … Ross (1995: CITE) … to which Franklin (2001: 153) concedes “Despite these relationships, there is not sufficient evidence to suggest any proven genetic relationship between Kutubuan and the Teberan Family, nor to suggest that Kutubuan and Engan should be considered as one genetic related group or as a subgroup of Trans New Guinea phylum.”

    Several previous studies have assumed that Witu is either a member of or coordinate to Enga-Southern Highlands family. Wurm (1964: CITE) placed Witu with the West Central family (i.e. Enga-Southern Highlands) of his East New Guinea Highlands Stock, a classification which Kerr (1967: 1-2) characterized as “reasonably secure, [but] so far based on relatively slender evidence.” Kerr (p. 4) asserts that “Kewa is probabkly both geographically and linguistically closer to Witu than any other languages, but preliminary studies require it to be classed in a separate subfamily.” Kerr (1975: 277, ibid.) …. Franklin (1975: 264-268, CITE) included Wiru in his Proto-Engan, …

    Franklin (2001) comparison of Lake Kutubu (Kutubuan) and Enga-Southern Highlands (Engan,) represented by Huli, Kewa and Pole. Conclusion (p.153): "A number of common kinship terms also occur between Kutubuan and Engan, as well as with Dadibi, again much more frequently than those of other word classes or semantic groups. Notice, however, that MacDonald (1973: 122) assigns Dadibi only a 5% relationship with Foe and calculates a 14% relationship between Polopa (the westernmost Teberan language) and Foe.

    The hypothesis advanced here is that the (q.v. Franklin 1975: 264-168) is that most Witu words shared with Enga-Southern Highlands are loans from the latter, with perhaps a small number of legitimate retentions which would presumably be ancestral to both Enga-Southern Highlands and Tua River.
    We agree with Franklin that Tua River is related to Lake Kutubu and hence to Foe; however, Lake Kutubu and Kikori River seem to be at least similar to one another than either is to Tua River.
    This is somewhat in accordance with Greenberg's (1971: 829) Barika subgroup of his proposed South New Guinea, as was sharply cricitized by Franklin (1973: 382): “In fact the inclusion of the Teberan Family with those of the Turama-Omatian Family is probably the most misleading of the SNG subgroup.” However, Greenberg didn't second-guess Wurm's placement of Lake Kutubu with … Highlands, and his treatment seems a careless conflation rather than a taxonomic insight, as these two families are sharply distinct where Greenberg's lowest-level subgroups were intended to be (and generally are) unambitious and obvious.

    Franklin (1968: CITE)
    MacDonald (1973) …

    Historical phonology

    [under construction]

    Consonants correspond as follows:

Tua River Witu Pawaia Dad.-Fol.
 *m- m m *m
*-m2- m       ø (Ṽ)
 *n- n n n
*-n1- n n *n
*-n2- n       ø (Ṽ)       ø (Ṽ)
 *p- p p *p
*-p- p *b
 *t- t s *t
*-t- t     *t [r]
 *s- t s *s
*-s- t s *s
 *k- k h          *k [k h]
 *mb- p p *b
*-mb- mb p *b
 *nd- t t *d
*-nd- nd
 *ŋg- k      k [k] *g
*-ŋg- ŋg      k [g] *g
 *j- t j *j
 *w- w w *w
*-w- w *w

    Vowels correspond as follows:

Tua River Witu Pawaia Dad.-Fol.
*i i i *i
*e e *e
*a a a *a
      *a/_Ci ɛ *a

    These correspondences are exemplified as follows:

    Witu words are drawn from Kerr (1967, 1975) and from Franklin (ed. 1973: 592, 1975: 264-268.) All Pawaia words below are drawn from Trefry (1969, 1972) unless otherwise indicated. Abbreviated sources are as follows: kh = Kerr, kf = Frankiln (ed. 1973, 1975) gm = Murray (1918-1919,) lb = Brown (1919-1920.)
    Pawaia tone is only indicated in Trefry (1972) and in limited portions of (1969.)


   Tua River  Witu  Pawaia  Dadibi-Folopa
   *m    m  *m
 meat/flesh  *mi    mi  *mi
 not  *menV  mená [kf]    *meni
 give  *mV  me-te- [hk]  ima-  *ma
 breast  *am[i/e]    ɛmi  *ame
 tongue  *kamina    hɛmina  *kamina
   *n    n  *n
 eat/drink  *nV-  nV-    *nV-
 louse  *n[ɔ/o]mV  nomò [kf]    *n[ɔ/o]
 tongue  *kamina    hɛmina  *kamina


   Tua River  Witu  Pawaia  Dadibi-Folopa


   Tua River  Witu  Pawaia  Dadibi-Folopa
 tree  *in[i] (?)    in [dt]; ĩ [kf]  *ni


   Tua River  Witu  Pawaia  Dadibi-Folopa
   *p-  p  p  *p
 lie down  *pi(-ndi)-  *pi-tí- [kf]    *pi(-di)-.
 all  *pe[j]a  pea    *pe[j]a
 shake  *pese-  pete- [sil]    *pese-
 root      pɛɾadiɛ  *...
 go  *pV-  pV- [hk]    *pV-
 valley (?)  *pao      *pao
 blow  *p[ɔ]-  ?popo-ka- [hk]    *pɔ -
 hot/sharp/angry  *poso  poto [kf]    *poso
 mud/dirty  *pũ[Nu]    pũ  *pũ
   *-p-  p    b
 belly  *ndepe  tepe [kf]    *debe
 ear/temple (?)  *ŋgapindi   kapindi [hk]    *gabidi

    Aspirated apical stop /*t/ is fricated to /s/ in Pawaia:

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *t-  t  s-  t-
 earth/ground  *[i]tono  itonó [kf]
 ?tóno 'mtn.'
 sòó  *tõ
 body  *tiŋgi[ni]  tiŋgini [hk]    *tigi
 old/ripe  *tãũ    sãu  *tãũ
 head  *tombo[u]  tombou    *tobo
 path  *tũ    sũ̂  *tũ

    Aspirated velar stop /*k/ is fricated and deoccluded to /h/ in Pawaia:

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *k-    h  *k [k h]
 dog  *kã[m]õ    hã  *kãõ
 many  *k…  kãĩa [kf]  howoti [dt]
 hoitejɛ [kf]
 tongue  *kamina    hɛmina  *kamina
 see  *kadi    hɛti-  


   Tua River  Witu  Pawaia  Dadibi
   *-k-    -h-  -h-
 yellow  *ma-k[anu] (?)    máhànù ~


   Tua River  Witu  Pawaia  Dadibi-Folopa

    Initial prenasalized stops /*mb *nd *ŋg/ are merged with plain voiceless /*p *t *k/ in Witu and Pawaia:

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *mb-  p  p  *b
 knee/wrist  *mboŋg[i]  poŋgi    *b[o]g[o]
   Tua River  Witu  Pawaia  Dadibi-Folopa
   *nd-  t  t  *d
 who?  *nde  te [hk]    *de
 belly  *ndepe-  tepe [kf]    *debe
 do  *ndV-  tV- [hk]    *dV-
 tail      tul  
 elbow  *nd[u]k[u]  tuku [sil]    
   Tua River  Witu  Pawaia  Dadibi-Folopa
   *ŋg-  k  k [k g]  *g
 garden (?)  *ŋgi      *gi
 claw      kɛ́pít  
 ear/temple (?)  *ŋgapindi   kapindi [hk]    *gabidi
 stone axe  *ŋgVmbe    kopɛ  

    Medial prenasalized stops /*mb *nd *ŋg/ are preserved as such in Witu … :

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *-mb-  mb  p [p b]  *b
 head  *tombo[u]  tombou    *tobo
 navel  *sombVtV  tombotoi [sil]    *s[o]bVtV
 stone axe  *ŋgVmbe    kopɛ  
   *-ŋg-  ŋg  k [k g]  *g
 pandanus nut  *aŋgia    ɛgia  *agia
 claw  *[o/ɔ]ŋg[o/u]  oŋgo  ogu  *[o/ɔ]g[o]
 knee/wrist  *mboŋg[i]  poŋgi    *b[o]g[o]
 body  *tiŋgi[ni]  tiŋgini [hk]    *tigi

    Due to these mergers, it's not possible to directly establish the quality of the medial where the Witu reflex is absent or ambiguous:

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *-[p/mb]-  ?  p [p b]  *b
 father  *a[p/mb]a      *aba
 father/f.'s bro.  *a[p/mb]bu    àpú  *abu
 claw      kɛ́pít  


   Tua River  Witu  Pawaia  Dadibi-Folopa
   *-nd-    -t-  
 see  *kadi    hɛti-  
 ear/temple (?)  *ŋgapindi   kapindi [hk]    *gabidi

    Fricative /*s/ is occluded to /t/ in Witu:

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *s  t    *s
 navel  *sombVtV  tombotoi [sil]    *s[o]bVtV
 die  *su[ku]  tu-    *sugu-
 firewood  *isi  iti [hk]    *isi
 woman/female  *[a]so[a]  atoà [kf]    *so
 shake  *pese-  pete- [sil]    *pese-
 hot/sharp/angry  *poso  poto [kf]    *poso

    The validity of this correspondence is supported by the following loan (q.v. Kerr 1975: 279

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *s  t    *s
 tobacco  [*soko]  toko  sogo  *sogo

   Tua River  Witu  Pawaia  Dadibi-Folopa
       s  *s
 fire/sun      síà  *sia
 yellow      sɛwai  *sewa


   Tua River  Witu  Pawaia  Dadibi-Folopa

    Witu occludes initial palatal glide /*j/ to /t/:

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *j  t  j  *j
 rain  *ja[ma]  ta  jama [dt]
 dʒa [kf]
 dog  *j[o]w[e]  tue    *j[o]wi

    This change is partially shared with the Kewa languages immediately to the west, where Southern Highlands initial /*j/ is occluded to /*s/ when followed by high back vowel /*u/. It seems probable that the Witu change was preceded by a merger with /*s/, which was then merged with /t/ (above.)


   Tua River  Witu  Pawaia  Dadibi-Folopa
   *w  w ø  *w  *w
 water  *wẽi  uè [kf]    *wẽi
 speech/speak  *wa-  wa o-    *wa-
 seed  *w[ae]      *wae
 net bag  *w[a/o]    wó  *w[a/o]
 yellow      sɛwai  *sewa[e]

   Tua River  Witu  Pawaia  Dadibi-Folopa

   Tua River  Witu  Pawaia  Dadibi-Folopa

   Tua River  Witu  Pawaia  Dadibi-Folopa

    Pawaia raises /*a/ to /ɛ/ when followed by /*i/ in the next syllable:

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *a/_i    ɛ  *a
 breast  *am[i/e]    ɛmi  *ame
 pandanus nut  *agia    ɛgia  *agia
 tongue  *kamina    hɛmina  *kamina
 see  *kadi    hɛti-  

   Tua River  Witu  Pawaia  Dadibi-Folopa

   Tua River  Witu  Pawaia  Dadibi-Folopa

   Tua River  Witu  Pawaia  Dadibi-Folopa
   *u    *u  
 tail  *    tul  

   Tua River  Witu  Pawaia  Dadibi-Folopa

Subpages (3): Dadibi-Folopa Pawaia Witu