Table of Contents

Kursav

Timothy Usher, Santa Fe Institute

Situation

Kursav (Kulsab,) also known as Faita, is spoken by perhaps 10 people (2015) living on the south bank of the upper Ramu river, just upriver from the confluence of the Peka and Ramu rivers, in the Usino Bundi district of Papua New Guinea's Madang province. Faita is the name of the village in which speakers used to live before disbanding in response to sorcery. Kursav /kuɾsaβ/ is the name that speakers call their language; its meaning is unknown. Kursav is moribund, with middle-aged Kursav people speaking Gende of the Goroka family to the south, Gende speakers having settled along the Ramu both upriver and downriver of the Kursav, Younger Kursav people speak only Tok Pisin (Z'graggen 1975: 32, Daniels 2015: 883-884.)

Sources

Z'graggen (1980: 1-80) 305 comparative terms and (p. 84) pronominal paradigms for Faita

Daniels (2010: 182-183) sketch phonology and (pp. 186-192) 34 reflexes of proto-Sogeram for Kulbab

Daniels (2015 356-392) 136 reflexes of proto-Sogeram and (pp. 883-986) grammar of Kursav

Daniels (2020) (unobtained)

Hisotry of classification

[under construction]

Faita (Kursav) is not mentioned in Z'gaggen (1971.)     Z'graggen (1975: 31-32) places Faita alongside Isabi, Biyom and Tauya in a Brahman subgroup of Madang.

Pawley …

Daniels (2010) places Kursav alongside Aisi in his East Sogeram subgroup of Sogeram. Daniels (2015, 2017, 2020) expands this subgroup to include Gants which is held to be Km ost immediately related to Kursav within East Sogeram, although confusingly he states that the reverse is not necessarily the case, seeing Aisi, Kursav and Gants as forming a chain relationsship despite Gants' current position in the highlands far to the west.

Phonology

Daniels (2015: 887-892) gives 15 to 17 consonants and 6 vowels for Kursav as follows:

\m n ŋ
p t s k [kʷ]
mb nd ŋg [ŋgʷ]
β
w ɾ j ɣ
i ɨ u
e o
a

In addition to the simple vowels given above, five diphthongs are found as follows:

eⁱ oⁱ oᵘ
aⁱ aᵘ

Prenasalized voiced stops /mb nd ŋg/ are realized as plain voiced [b d g] initially.

Daniels characterizes /β/ as a voiced fricative but states that its initial allophone is voiceless [ɸ], suggesting /ɸ/ to be a more appropriate phonemic representation.

Apical non-stop / ɾ/ is realized as lateral [l] initally.

Pronouns

Daniels (2015: 902-904, 936-938) gives free pronouns in four case forms and prefixed verbal objects for Kursav as follows:

subjectpossessiveemphatic emph. poss. verbal object
1 sg.ja ja-ku ja-mba ja-mba ja-ku ja-
2 sg.na na-ku na-mba na-mba na-ku na-
3 sg.nɨ ~ nunu-ku nɨ-mba nɨ-mba nu-ku Ø- ~ u-
1 pl.a-n a-nu-ku a-nɨ-mba a-nɨ-mba nu-ku a-n-
2 pl.na-n na-nu-ku na-nɨ-mbana-nɨ-mba nu-k.na-n-
3 pl.nɨ-n nu-nu-ku nɨ-nɨ-mbanɨ-nɨ-mba nu-kunɨ-n-

Inalienable possessors on kin terms and some body parts are indicated by prefixes. Excepting the first person singular, which can be either /a-/ or /ja-/, the prefixed possessors are clearly derived from the free forms of the pronouns (above,) with the vowel of the third person varying according to the following root (Daniels 2015: 896-899):

root 1 sg. 2 sg. 3 sg. 1 pl. 2 pl. 3 pl.
a- na- nɨ- a-nɨ- na-nɨ- nɨ-nɨ-
aunt mes a-mes na-mes nɨ-mes a-nɨ-mes na-nɨ-mes nɨ-nɨ-mes
mother nia/mɨŋgea-nia na-mɨŋgenɨ-mɨŋge a-nɨ-mɨŋgena-nɨ-mɨŋgenɨ-nɨ-mɨŋge
ja- na- ni- ~ nu-a-nɨ- na-nɨ- ni-nɨ- ~ nu-nu-
wife namba [nduβ ja-ku]na-nambani-namba a-nɨ-nambana-nɨ-nambani-nɨ-namba
arm/handkoma ja-koma na-koma nu-koma a-nɨ-koma na-nɨ-koma nu-nu-koma
friend kʷaⁱ ja-kʷaⁱ na-kʷaⁱ nu-kʷaⁱ a-nu-kʷaⁱ na-nu-kʷaⁱ nu-nu-kʷaⁱ

(The first person singular form of “wife” /nduβ ja-ku/ means merely “woman my”; q.v. /ja-ku/ 1 sg. possessive above, Faita [nduwa] “woman” Z'graggen 1980: 1.)

Historically, these patterns arise from a mixed system, with original first person possessor /a-/ on kin terms and second and third person possessors undifferentiated by number mixed with a newer system in which the free forms (above) are prefixed to the inaliebaly possessed terms, which now include not just kin terms but also body parts. The closely related Gants also shows alternation between first persons /a-/ and /ja-/, though without any differentiation by number (Daniels 2015: 999-1000.)

Verbal morphology

Daniels (105: 920-936) gives subject desinences for Kursav final verbs in six tense/mood paraigms as follows, with he selection between the two non-future paradigms being determined by the root:

non-future Inon-future IIhabitual future irrealisuncertain future
1 sg.-Ø-Ø -ua-Ø -nd-ua-Ø -mnd-ua-Ø -ɨt-Ø ?
2 sg.-Ø-na -ua-na -nd-ua-na-mnd-ua-na-ɨt-a -ma-na-ᵘ
3 sg.-Ø-e -Ø-e ~ -Ø-i -nd-Ø-e -mnd-Ø-e -ɨt-e ?
1 pl.-Ø-ɾ -ua-ɾ -nd-ua-ɾ -mnd-ua-ɾ -ɨtɨ-ɾ ?
2 pl.-Ø-ɾa -ua-ɾa -nd-ua-ɾa-mnd-ua-ɾa-ɨtɨ-ɾa -ma-ɾa-ᵘ
3 pl.-Ø-o -Ø-o ~ -Ø-u -nd-Ø-o -mnd-Ø-o -ɨt-o ?

Imperative verbs take an entirely different set of subject suffixes:

imperative
1 sg. -n
2 sg. -ka
3 sg. -koɾo
1 pl.
-2 pl.-ku-ɾa
3 pl. -konoᵘ

Medial verbs are distinguished according to whether their subjects are the same as or different from the subject of the final verb:

different subjectdifferent subjectsame subjectsame subject
sequential simultaneous unmarked desiderative
1 sg. -ku-Ø -tɨ-Ø-tɨ-Ø -nda -misi ~ -mis
2 sg. -ku-na -t-a-t-a -nda -misi ~ -mis
3 sg. -ek-e ~ -ik-e -t-e-t-e -nda -misi ~ -mis
1 pl. -ku-ɾu -tɨ-Ø-tɨ-ɾ -nda -misi ~ -mis
-2 pl.-ku-ɾa -tɨ-Ø-tɨ-ɾa -nda -misi ~ -mis
3 pl. -ok-o ~ -uk-o -t-o-t-o -nda -misi ~ -mis

While the desiderative is usually medial, it can also function as a final verb in some contexts.

Loans from neighboring languages

Kursav is spoken significantly to the southeast of the other Sogeram River languages, just upriver of the confluence of the Ramu and Peka rivers, and thus nearly surrounded by languages of the Brahman-Peka River subgroup of the South Madang family. Just across the Ramu to the north Usino (Sop) is spoken, with Tauya immediately upriver to the southeast and Biyom just south of Tauya. Despite the closer proximity of Sop, Tauya appears to be the main lender into Kursav. The numerous loans between these languages moved Z'graggen (1975: 31-32) to place them along with the Goroka language Isabi in an invalid Brahan subgroup of Madang (note that the term Brahman is still used here to refer specifically to the subgroup of Brahman-Peka River which consists of Biyom nd Tauya.) Tauya and Biyom attestations below are drawn from Z'graggen (1980: 1-80) and Usino from Z'graggen (1980: 1-159.) Brahman-Peka protoforms which are found only in Brahman are given in brackets:

Faita Tauya Biyom Usino Brahman-Peka
breast amᵊna amɛna ami *amen
younger s.s. sibanow ano -ano [*ano]
sleep akun- akin- kin *kinV-
arrow sp. aŋgrᵊ aʔɛli ʌgir *aŋgiri
sand usuwa osu *…
star ben pɛnɛ
crayfish papas papas
root nikininikini
many sakum saʔumo
snake sʌr sai sʌːri [*sari]
&nbspfence giβu gib
mosquito kasi ʌsi kaːsin*kasin
no kʷe wɛa kʷɛa *kʷV
betel pepper jai jai jar *jaj

Note that “breast” is a doublet, the Brahman-Peka River form replacing its Sogeram River equivalent /*aman/ (Daniels 2015: 360.) In addition ot the above, there is a small number of roots, presumably dating to proto-Madang, which are common to Sogeram River and Brahman-Peka River such that matches between Kursave and Biym-Tauya are probably legitimate retentions in each, e.g. Faita [kum-], Tauya [umu-], Boyim [kumu-]' “die” as well as the singular personl pronouns. “Sleep” is cognate with Sogeram River “stay” (Daniels 2015: 371.)

Some loans have proceeded in the other direction, from Kursav into Brahman-Peka River languages:

Sogeram RiverFaita Tauya Biyom UsinoSumau
butterfly *apapVrV apapureapapare
kunai grass*mɨnda mɨnda mita
loincloth *pɨsa wusa ɸuso wuzᵋ
moon *takun taku tʌʔo
blood/red *jaŋgum jaŋgum inaumo jaʔoi
mosquito *jaŋgur naŋguru

The case of “mosquito” is especially interesting as Kursav appears to have lent the Sogeram River term to Biyom before replacing it with a loan from Brahman-Peka River.

At least two words are found also in Gende and Isabi of the only very distantly related Goroka family, Gende being spoken immediately to the south of Kursav and Isabi just east of Tauya and Biyom. Neither has a known Goroka etymology; “crocodile” is probably from Brahman-Peka River in light of the geography of the region. Gende attestations are drawn from Aufenanger (1952) and Isabi from Z'graggen (1980: 1-80) :

FaitaGendeIsabiTauyaBiyomUsinoBrahman-Peka
crocodileakᵊnaakanaʌgʌnaaʔɛnaakɛnaagəna*aŋgena
bean fuŋgapuŋgapuŋgafuŋga