Annotated Swadesh wordlists for the Talodi group (Kordofanian family).

Languages included: Talodi (Jomang) [tal-jom]; Nding [tal-ndi]; Dagik (Masakin) [tal-dag]; Ngile (Daloka) [tal-ngi]; Tocho [tal-toc]; Acheron [tal-ach]; Lumun [tal-lum]; Torona [tal-tor].


I. General.

Schadeberg 1981 = Schadeberg, Thilo. A Survey of Kordofanian. Volume Two: The Talodi Group. Hamburg: Helmut Buske Verlag. // A brief survey of the Talodi languages of Kordofan, based on the author's own fieldwork. Includes brief descriptions of the phonetic inventories and nominal grammar (noun class systems, pronouns, etc.) of ten different languages, as well as 200-item wordlists, a lexicostatistical classification, and a first attempt at the lexical reconstruction of Proto-Talodi.

Norton & Alaki 2015 = Norton, Russell; Alaki, Thomas Kuku. The Talodi Languages: A Comparative-Historical Analysis. In: Occasional Papers in the Study of Sudanese Languages, 11, pp. 47-161. // A large paper that outlines in the detail a classification model for the Talodi languages (including superficial lexicostatistical analysis) and a basic reconstruction of the phonology and morphology of Proto-Talodi. The etymological appendix to the paper contains a wealth of lexical data on all Talodi languages, compiled from various sources and also including original data on such previously unattested Talodi varieties as Acheron, Lumun, and Torona.

II. Dagik (Masakin).

Vanderelst 2016 = Vanderelst, John. A Grammar of Dagik, a Kordofanian Language of Sudan. Köln: Rüdiger Köppe Verlag. // Detailed grammatical description of one dialectal variety of Dagik (Masakin), with a few examples of glossed texts and a short accompanying vocabulary.


1. General.

I. Talodi (Jomang); Nding (Liri); Ngile (Daloka).

Systematic data on these languages are currently only available from the survey lists in [Schadeberg 1981].

The Jomang (ʓoːmʷaŋ) list was elicited from the 27-year old Jima Umar Kafi from Talodi (there is no single name used for this language, according to Schadeberg).

The main list for Nding (Liri) was acquired from the 35-year old Joga Cweda Tutu at Tungaru (Liri); the name for the language is either Nding (ndiŋ) or Liri (El Liri, Eliri). Additional items are also reproduced by Schadeberg from his fieldwork with an 8-year old boy by the name of Jarinebi; sometimes they demonstrate significant deviations from the full list, but their degree of reliability is also questionable. We include Jarinebi's equivalents in the notes section.

The main list for Ngile (Daloka) was provided from the 40-year old Ding Koa of Daloka, a village about 18 km south-south-east of Kadugli. The proper name for the language is t̪ɛ̀=ŋgìlɛ́, derived from the local name of the village (ŋgìlɛ́). Schadeberg's notes also include alternate forms (usually phonetic or morphological variants) elicited from several other younger informants; since they are not particularly relevant, we just list them all together in the notes section as "idiolectal variants".

II. Dagik (Masakin).

According to [Vanderelst 2016], Dagik is currently spoken in the Buram district by inhabitants of approximately 5-6 villages. Suggested names for the language are Masakin (an Arabic ethnonym < masaakiin 'poor'), Dagik (probably also from Arabic daqiiq 'small' ), and, in [Schadeberg 1981: 14], Dengebu [ðɛ́ŋgɛ́bu].

For the sake of consistency, we use [Schadeberg 1981] (a list recorded from the 45-year old Mr. Gallo Kuku Giti Kori of Reikha) as our default source, although [Vanderelst 2016] is a much more modern and detailed description and would, in itself, be sufficient for the compilation of an acceptable Swadesh wordlist. We do, however, systematically list all the correlates in the notes section. A few notable discrepancies between Schadeberg's and Vanderelst's materials may be due either to elements of dialectal variety or semantic inaccuracy; available data are insufficient to prove anything with certainty.

III. Acheron; Lumun; Torona; Tocho.

These varieties of Talodi probably constitute separate languages, but are traditionally classified into a single "Moro Hills" dialect cluster. Data on Tocho was included in [Schadeberg 1981], where it is attributed to a 40-year old informant, Mr. Mahmur, who was recorded by the author at Doleibaya, a new settlement about 20 km from Talodi; however, Mahmur's birthplace is listed as the village of Tocho. For the three other varieties, the only major usable data source is the comparative study [Norton & Alaki 2015], which allows to construct near-complete Swadesh wordlists for all of them, based on the original fieldwork conducted by the authors.

2. Transliteration.


The principal source for all Talodi data in the lists is [Schadeberg 1981]. Since his transcription is largely based on the IPA, we have preserved it almost intact, with cosmetic differences concerning IPA > UTS transliteration (ʃ > š, c > ɕ, etc.). Doubled vowels in Schadeberg's transcription, marking length, have been converted to single vowels plus the length sign (aa > , etc.). Tonal systems in the languages usually involve two registers, and the low register in Schadeberg's transcriptions usually remains unmarked; we consistently mark it with the low tone diacritic (V̀).

The transcriptional system in [Norton & Alaki 2015] is also IPA-based, but seems to be slightly tipped towards more accurate phonetic rather than phonological description. We preserve most of the details intact, with the usual cosmetic changes from IPA to UTS. It should be noted that prosodic notation in this source is somewhat inconsistent: some languages seem to be given without any prosodic information whatsoever (Acheron), whereas for Lumun and Torona high tone seems to be explicitly marked and low tone left unmarked (as in Schadeberg's work).

Our morphological segmentation of the data (a very important matter for Kordofanian data, where words frequently begin with nominal class prefixes or agreement markers) largely follows the segmentation in the original works, but sometimes we make additional decisions on segmentation based on external data (e. g. Schadeberg and Norton / Alaki usually do not separate prefixes and suffixes if they behave like fossilized markers in a certain language; but in most cases, external data from closely related languages makes it crystal clear that the morpheme is a historical prefix / suffix, so we segment it).

Database compiled and annotated by: G. Starostin (last update: May 2017).