Central Ok

Timothy Usher, Santa Fe Institute

The Central Ok family consists of two or three closely-related languages spoken in villages along the Sauw River and its tributaries in the Merauke Regency of Indonesia's Papua province (Hughes 2009: 11-14.) The designations Nakai, Naki and Tangko are the words for “what?” in each language (q.v. Wilbrink 2004: 109f, 110f, Hughes 2009: 37.)

The internal classification of Central Ok is as follows:

Central Ok

Tangko

Nakai-Naki

Nakai

NakI

Jang (2003: 39-47) 239 comparative terms for Nagi of Kawor 2 village

Wilbrink (2004: 109a-110f) 239+ comparative terms for Sait (Nagi) and Nakai after Menanti

Hughes (2009: 27-37) 239 comparative terms for Tangko of Kawemaot village and Nakai of Burunggop

Williamson (2010) grammar of Nagi

The Central Ok family was first recognized by Hughes (2009: 3,) who calculated a 29-32% lexicostatistical resemblance between Tangko of Kawemaot and Nakai of Burunggop.

Proto-Central Ok had perhaps 14 consonants and 6 vowels as follows:

Any consonant except apical non-stop /*r/ can occur initially. Prenasalized apical stop /*nd/ is found in only a small number of examples, in which it is initial. Even though this is in complementary distribution with /*r/, and medial reflexes of Ok /*nd/ have merged those with /*r/, for historical reasons they are not shown here as allophones of a single phoneme, although such an equivalence may be synchronically justified.

Consonant clusters do not occur except across morpheme boundaries in compounds.

Only a restricted set of consonants occurs finally:

Initial consonants correspond as follows:

Medial consonants correspond as follows:

Final consonants correspond as follows:

Vowels generally correspond as follows:

These correspondences are exemplified as follows, with Tangko attestations drawn from Hughes (2009: 27-37,) Nakai from Hughes (2009: 27-37) and Wilbrink (2004: 110a-110f) and Naki from Jang (2003: 39-47) and Williamson (2011 ibid.) Jang's, Wilbrink's and Hughesˈ survey vocabularies are phonetic transcriptions; Williamson's are orthographic. For clarity of presentation, the general dispositions of initial consonants will presented first, then those of medial consonants, then those of final cononants, followed by those of vowels. Some details of the forms which follow are necessarily tentative.

[under contruction – Hughesˈ Nagi to be replaced with Jang]

Any consonant except apical non-stop /*r/ can occur initially.

Initial bilabial nasal /*m/ is retained as such in all descendants:

Initial apical nasal /*n/ is retained as such in all descendants:

Initial bilabial plain stop /*p/ is fricated to [ɸ] in Tangko and dropped in Nakai, a pattern of loss shared with West Ok and Lowland Ok:

Initial apical plain stop /*t/ is retained as such in all descendants:

Initial velar plain stop /*k/ is retained in both as such in all descendants. It is often realized as uvular [q] and occasionally as voiced velar [g]:

Initial laminal /*s/ is retained as such in all descendants:

One apparently irregular example is likely a loan from Dumut:

Initial prenasalized bilabial stop /*mb/ is as in other subgroups of Ok typically realized as plain voiced [b]:

Initial prenasalized apical stop /*nd/ is uncommon:

Several examples known to have initial /*nd/ or /*r/ in other Ok subgroups are here indeterminate between /*nd/ and /*j/ (below):

“Heart” might be thought a loan, as expected vowel /*y/ is reflected as /u/ in Nakai rather than as /i/ (below.)

Initial prenasalized velar stop /*ŋg/ is somewhat more common than it is in other Ok subgroups. As with /*k/, it is often realized as uvular [ɴɢ ɢ], or in Tangko as nasal [ɴ]:

“Cold” and “wrist/joint” can be traced to proto-Ok, or alternately are loans from Mountain Ok. “Ant” may be a loan from Dumut or vice-versa. The examples of “cold” and “garden” suggest that the contrast between initials /*k *ŋg/ is disappearing.

One example is a loan from Indonesian:

Initial bilabial non-stop /*w/ is occluded to [β] in Tangko and Nakai but not Naki, especially when followed by front vowels /*i *e/:

Initial palatal non-stop /*j/ is is occluded in all descendants:

The occlusion of /*j/ is shared with Ninggerum to the southeast, where Lowland Ok /*j/ is realized as [d r].

The quality attested by the following correspondence is not yet known. It remains a problem in Lowland Ok as well, from which it is conceivably a loan:

Any consonant can occur medially except rounded velar /*kʷ/, which didnˈt occur medially in Proto-Ok, and prenasalized apical stop /*nd/, which has been merged with non-stop /*r/. Medial bilabial stop /*p/ was rare in Ok, and has not been found to occur.

Medial bilabial nasal /*m/ is retained as such in all descendants:

Medial apical nasal /*n/ is retained as such in all descendants:

No native examples of root-medial plain bilabial stop /*p/ have been identified.

Medial plain apical stop /*t/ is merged with /*r/ in Tangko:

In one example, /*t/ is fricated to /s/, probably due to preceding /*i/ as is the rule in Ngalum:

Medial laminal /*s/ is generally preserved as such in all descendants:

Medial plain velar stop /*k/ is generally lenited between vowels in all descendants:

The general loss of medial /*k/ in the Lowland Ok languages to the southeast should be viewed as a continuation of ths areal trend.

Medial prenasalized bilabial stop /*mb/ is reflected as plain voiced [b] in Tangko and merged with the reflexes of /*w/ [w β] in Nakai, a development shared with Ninggerum:

As in Mountain Ok, the reflexes of medial apical prenasalized stop /*nd/ have merged with those of /*r/ (below).

Medial prenasalized velr stop /*ŋg/ is often reflected as uvular [ɴɢ], or as merely [ɴ] in Tangko:

Medial apical non-stop /*r/ is typically dropped in Naki, a loss shared with Ninggerum:

There is at least one exception:

Medial palatal non-stop /*j/ is typically occluded, becoming nearly indistinguishable from the reflexes of /*r/ (above):

Because the reflexes of medials /*r *j/ overlap in range, it's possible that some forms shown here with one really had the other.

Medial /*w/ is generally retained as such in all descendants:

Only a restricted set of consonants occurs finally, including nasals /*m n/, plain stops /*p *t *k/, prenasalized velar stop /*ŋg/, which as in other Ok subgroups is invariably realized as [ŋ], and non-stops /*r *j/.

Final bilabial nasal /*m/ is preserved as such in all descendants, excepting some instances when it is assimilated to [ŋ] when followed by a velar stop /*k *ŋg/ in a compound:

Final apical nasal /*n/ is preserved as such in all descendants, excepting some instances when it is assimilated to [ŋ] when followed by a velar stop /*k *ŋg/ in a compound:

…:

Final plain bilabial stop /*p/ is typically retained as such in all descendants:

Final plain apical stop /*t/ is fricated to [s], which is not otherwise known to occur finally, or dropped in Tangko:

It's conceivable that these represent two correspondences reflecting finals /*t *nd/ respectively, as “young” and “breadfruit” are known to continue final /*t/ while “skin” is thought to continue /*nd/; if so these would be distinguished only in Tangko.

One example in which this does not occur may be a loan into either Tangko or Naki:

Final plain velar stop /*k/ is often realized as uvular [q q̚]:

Prenasalized stops /*mb *nd/ do not occur finally because they have been merged with /*p *t/ (above).

Final prenasalized stop /*ŋg/ is realized as nasal [ŋ] in all descendants, as it is throughout the Ok family, or as uvular [ɴ]:

In one example. final /*ŋg/ is lost in Nakai when followed by /su/ “leaf”:

Final apical non-stop /*r/ is distinguishable from /*t/ by its palatalization and loss in Naki:

Nakai's merger of final /*r/ with the reflexes /*t/ is shared with Muyu-Yonggom to the southeast.

The reconstruction of final laminal glide /*j/ in the following two roots is highly tentative, as outcomparison to other Ok subgroups suggests /*r/; however in none of these does /*j/ otherwise occur finally, so it's possible that Central Ok preserves a distinction which has been neutralized in other subgroups. The alternative would seem to be that both Nakai words are loans from Naki:

No examples of bilabial non-stop /*w/ are known to occur finally.

High central vowel /*ʉ/ is retained as [ɯ] in Tangko and merged with high front /i/ in Nakai and Naki:

Sequence /*uwi/ is reduced to /ɯ/ in Tangko:

Central Ok free pronouns are reconstructed as follows, with Tangko forms drawn from Hughes (2009: 28-29,) Nakai from Hughes (2009: 28-29) and Wilbrink (2004: 110a) and Nagi from Jang (2003: 40) and Williamson (2011: 12-14):

[under construction]