Dadibi-Folopa

Timothy Usher, Santa Fe Institute

The Dadibi-Folopa family is comprised of at least two languages spoken around the intersection of Papua New Guinea's Southern Highlands, Gulf and Simbu provinces (MacDonald 1973: CITE.) Dadibi and Folopa are different enough from one another to be considered separate families, while Folopa dialects are probably diverse enough to be considered a family of several closely-related languages.

Brown (1919-1920) 113 comparative terms for Bara lleriety of Folopa

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

Franklin (1968) Gulf District

Franklin ed. (1973) 100 comparative terms for Dadibi and for Sopese, Boro, Suri and Tebera varieties of Folopa

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

Proto-Dadibi-Folopa had 11 or 12 consonants and 6 or 7 vowels:

The originality of low mid front vowel /*ɛ/ is debatable.

Vowels are further distinguished by the presence vs. absence of nasalization:

Both Dadibi and Folopa contrast aspirated and plain stops initially, but only one of the series is analyzed as occurring medially, with the Mac Donalds interpreting medial stops as allophones of plain /b d g/ and the Andersons as allophones of /p t k/. We proceed under the assumption that the first interpretation is diachronically correct. In Anderson and Anderson's description, Folopa lacks /g/; this is because their /k/ is historically /*g/ and corresponds to Dadibi /g/. Hence /k/ is unaspirated, while /p t/ are initially [ɸ tʰ] (1995: 1, 2-3.) Folopa /h/ occurs only initially, suggesting it to be the aspirated reflex of /*k/, i.e. /*p *t *k/ [ɸ tʰ h].

Neither final consonants nor consonant clusters occur.

[under construction]

Consonants correspond as follows:

Vowels correspond as follows:

All Dadibi words below are drawn from Whitby and Bai (1990, 2008) and all Folopa words from Anderson and Anderson (1995,) Anderson (1989, 1994, 2009,) Anderson and Wade (1988,) and Anderson (2010: 97-107, ibid.) unless otherwise indicated. Abbreviated sources are as follows: lb = Brown (1919-1920), sc = Chance (1923-1924,) kf = Franklin ed. (1973,) gmd = MacDonald (1973,) MacDonald and MacDonald (1974.) Several of the comparisons below are found in MacDonald (1973.)

Initial nasals /*m *n/ are retained as such in both languages:

Medial bilabial nasal /*m/ is retained as such unless following a nasal in the previous syllable (below):

Medial apical nasal /*n/ when not following a nasal in the previous syllable (below) is merged with /l/ in Folopa:

Medial nasals /*m *n/ are dropped in Folopa when following a nasal in the previous syllable:

Initial voiceless bilabial stop /*p/ is realized as aspirated [pʰ] in Dadibi and as bilabial fricative [ɸ] in Folopa:

…

Initial apical stop /*t/ is realized as aspirated [tʰ] in both languages:

Initial velar aspirated stop /*k/ is aspirated and deocluded to /h/ in Folopa. In Dadibi, /*k/ is retained as such when followed by /*i *e/ and deoccluded to [h] when followed by /*a *ɔ *o/:

This irregular correspondence indicates a loan:

Initial voiced stops /*b *d/ are retained as such in both languages:

Initial velar voiced stop /*g/ is given as /k/ in Folopa:

Medial peripheral stops /*b *d/ are not distinguished from voiceless /*p *k/. They are given as voiced in /b g/ Dadibi and as voiceless /p k/ in Folopa:

The choice of voiced /*b *g/ rather than voiceless /*p *k/ to represent these segments isn't entirely arbitrary, as outcomparison to Witu shows at least /*b/ to correspond to prenasalized /mb/; e.g. /*tobo/, Witu /tombou/ “head,” /*gabidi/, Witu /*kambindi/ “ear.” What's not immediately clear is whether medials /*p *k/, which we assume must have once existed, were merged with the reflexes of the voiced stops or were lenited (as they are in Witu) and dropped.

…

…

…

The following correspondence is tentatively identified as voiced affricate /*dz/:

…

Fricative /*s/ is retained as such in both langugaes. Medially, it can also be realized as voiced [z]:

…

…

…

High front vowel /*i/ is retained as such in both languages:

High back vowel /*u/ is retained as such in both languages:

High-mid front vowel /*e/ is retained as such in both languages:

Words of the form /*CeCi/ are typically leveled to /CiCi/ in Anderson and Anderson's Folopa:

High-mid back vowel /*o/ is retained as such in both languages:

Low-mid back vowel /ɔ/ is merged with /o/ in Dadibi. It is significantly less common than high-mid /*o/ (above):

Low central vowel /*a/ is retained as such in both languages:

…