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

Languages included: Koalib, Rere [hei-rer]; Ebang [hei-eba]; Abul [hei-abu]; Laru [hei-lar]; Utoro [hei-uto]; Shirumba [hei-shi]; Tiro [hei-tir]; Moro [hei-mor]; Ko [hei-koo]; Warnang [hei-war]; Logol [hei-log].

DATA SOURCES

I. General

Schadeberg 1981 = Schadeberg, Thilo. A Survey of Kordofanian. Volume One: The Heiban Group. Hamburg: Helmut Buske Verlag. // A brief survey of the Heiban 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-Heiban.

II. Koalib, Rere

Quint 2006 = Quint, Nicolas. Phonologie de la langue koalibe. Dialecte réré (Soudan). Paris: L'Harmattan. // Detailed description of all the aspects of the phonology and phonetics of the Rere dialect of Heiban, illustrated by numerous examples.

III. Ebang

Meinhof 1944 = Meinhof, Carl. Das Heiban in Kordofan. In: Zeitschrift für Eingeborenen-Sprachen, 34, pp. 94-130. // A sketch of Heiban (Ebang) phonology and grammar. Well illustrated with lexical material, but no separate texts or vocabularies included.

IV. Laru

Kuku 2012 = Kuku, Nabil Abdalla. Laru Vowel Harmony. In: Occasional Papers in the Study of Sudanese Languages, 10, pp. 17-34. // A paper on certain aspects of Laru phonology, well illustrated by multiple sets of accurately transcribed lexical examples.

Kuku 2015 = Kuku, Nabil Abdalla. Laru Locatives. In: Occasional Papers in the Study of Sudanese Languages, 11, pp. 1-16. // A paper on Laru grammar, illustrated by multiple lexical examples.

V. Utoro; Tiro

Stevenson 2009 = Stevenson, Roland C. Tira and Otoro: Two Kordofanian Grammars. Ed. by Thilo C. Schadeberg. Köln: Rüdiger Köppe Verlag. // Grammars of two Heiban languages, compiled by R. C. Stevenson based on his own fieldwork in the Nuba Mountains, in 1942 and 1943 respectively. Both grammars include extensive illustrative material, examples of texts, and short glossaries with paradigmatic information on lexemes.

VI. Shirumba

Guest et al. 1998 = Guest, Elizabeth (ed.; w. Mukhtar Wakil Ali, Ahmed Abbas Natilly et al.). Cwaya Phonology. Ms., rev. 28/3/98. // A brief sketch of the phonetics of Shirumba (Cwaya, Shwai), with numerous illustrating lexical examples.

VII. Moro

Black 1971 = Black, Keith; Black, Elizabeth. The Moro Language: Grammar and Dictionary. Khartoum: Sudan Research Unit, University of Khartoum. // Large grammar of the Moro language with an extensive accompanying vocabulary.

NOTES

1. General.

I. Koalib, Rere

As in the case of almost all Kordofanian languages, [Schadeberg 1981] is used as the main source. Comparison with [Quint 2006], a monograph that deals specifically with this variety of Heiban Kordofanian, shows that Schadeberg's semantic glossing is mostly accurate (apart from just a few potential mistakes, e. g. the glossing of 'river' as 'sand'), but the phonetic transcriptions of the two authors vary quite significantly, mostly in regard to vocalism and tonal notation.

Since Quint's study on the whole is (by definition) more thorough than Schadeberg's, more trust should probably be put into Quint's transcription; however, we still choose [Schadeberg 1981] as our primary source, both for the sake of consistency with the following Kordofanian languages and also because [Quint 2006] does not include any wordlists, and its data are insufficient to adequately complete the Swadesh 100-item list.

Items in the primary slot are carried over directly from Schadeberg's records, meaning that nouns are usually given in the singular number (the plural, usually displaying a different class prefix, is adduced in the notes where available), and verbs tend to be given in the imperative form, which often comes close to showing the "pure" stem (but not always). Adjectives are given as stems; prefixes with which they may be joined are indicated in the notes section.

II. Ebang; Abul

Ebang (= Heiban) and Abul are two close varieties of the same language, spoken around the town of Heiban and the adjacent village of Abul. The main source of data for Ebang is [Schadeberg 1981], but where possible, the data have been cross-checked against C. Meinhof's earlier description in [Meinhof 1944]; aside from a few minor phonetic discrepancies, equivalents for most Swadesh meanings are the same between the two sources.

For the Abul variety, Schadeberg's fieldnotes remain the only source of knowledge. They were collected from two 18-year old informants, of which Mr. Ashaya seems to have been the most reliable one and Mr. Fadlalla Usmaan Kunda the less reliable one ("...found it difficult to provide me with the Abul equivalents of the Arabic words I was giving him, he often hesitated or said he could not remember"). Because of this, and also because Ashaya's data are far more complete, we list Ashaya's equivalents in the primary slot and Fadlalla's data in the notes section (referred to as "Fadlalla's idiolect"). Most of the time, Fadlalla's data differs from Ashaya's only in terms of morphology (e.g., different noun classes); occasional lexical discrepancies may be due to errors committed by the informant and should not be trusted.

III. Laru

Schadeberg's data on Laru is limited to the 100-item wordlist, collected at Heiban from a 17-year old native speaker coming from a village in the Laro Hills. No earlier sources on the language are available; as for later publications, the only known author is Nabil Abdalla Kuku, whose two papers on aspects of Laru phonology and grammar [Kuku 2012, 2015] were consulted as control sources, and, in a few cases (i. e. a few items among the 10 additional words), as our primary data source.

IV. Utoro (Otoro)

For this language, Schadeberg relied on two informants from Kacama as well as one informant from Kerindi; data from the latter were found less reliable and were not used as the basis for the grammatical description or the wordlists. All cases where Kerindi data differs from Kacama data have been noted by Schadeberg, and Kerindi variants are included in the comments section.

As a control source, [Stevenson 2009] was used - Schadeberg's edit of a grammar originally compiled in 1943. Stevenson distinguishes between three main dialectal groups of Otoro: Kwara, Kwijur, and Orombe. Apparently, Schadeberg's "Kacama" is closest to Stevenson's "Kwijur", and Schadeberg's "Kerindi" is closest to Stevenson's "Orombe". However, Stevenson himself relies mainly on data from Kwara, so the base forms of Stevenson and Schadeberg often turn out to be significantly different in terms of phonetics or morphological framing (although lexical differences are seemingly minimal in the core basic lexicon at least). On the other hand, in Stevenson's glossary for Otoro, he usually lists variants from all three dialects, and we adduce all of them in the comments.

V. Shirumba

Most of the data on this language still come from [Schadeberg 1981]. Shirumba (Churumba) is the name of the village and the dialectal variety spoken in it; an alternate name is Shwai, or Cwaya, after the local Jebel Shwai ("19 kilometers approximately west of Heiban", acc. to Schadeberg).

Some newer data on Shirumba (Cwaya) is also available in the form of manuscripts on various phonetic and grammatical topics, most of them seemingly prepared by Elizabeth Guest based on her own work with informants. [Guest 1998], in particular, is an overview of the phonetics of that language; we have listed data from that source in the comments section.

VI. Tiro

Spoken in close proximity to the region of the Utoro, but apparently not its closest relative. In [Schadeberg 1981], there is only a 100-item wordlist, collected from a 16-year old informant born in Dengir. The obvious control source is [Stevenson 2009], Schadeberg's edit of a grammar originally compiled in 1942. Unlike Otoro, Tiro has not been described as a dialectally diverse language, although there are enough phonetic differences between Schadeberg's and Stevenson's records to suggest that at least some subdialectal variety does exist.

VII. Moro

The Moro wordlist in [Schadeberg 1981] was recorded from a 29-year old informant living in the small town of Um Dorein, to the west of the Utoro-Tira region. The same source also includes some data from the earlier grammatical and lexical description in [Black 1971], a detailed grammar of the Moro language. Since the dialect of Schadeberg's informant and the dialect described in the Blacks' monograph may be somewhat different, we mark those entries where Schadeberg does not have his own data and relies only on the Blacks as "dubious". Elsewhere, we simply include data from [Black 1971] into the comments section.

VIII. Ko; Warnang

These two idioms are quite distinct from each other, but both are spoken at the exreme south-east corner of the Heiban-speaking region and seem to form a separate subbranch of the group. The only existing records for both languages are the grammatical and lexical data in [Schadeberg 1981], elicited by the author from young informants.

IX. Logol

Spoken in Dugili, at the foot of Jebel Lukha, approximately midway between the Tiro-speaking and the Ko/Warnang-speaking regions. No data on this language are available except for the wordlists and brief grammatical information published in [Schadeberg 1981].

2. Transliteration.

I. General

The principal source for all Heiban 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 > aː, 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).

II. Koalib, Rere

For [Quint 2006], the following should be noted:

1. In his work, Quint uses both an orthographic (phonological) representation (in bold letters) and a more detailed phonetic transcription (in square brackets) where necessary. In those cases where the two differ, we reproduce Quint's phonetic transcription in cursive and his phonological (orthographic) transcription in curly brackets. E. g., our ʈ=íɲɐ́n {ʈínyɐ́n} 'dog' is Quint's ʈínyɐ́n [ʈíɲɐ́n] in the original.

2. Long vowels are transcribed as double aa, ee, ii, etc. both by Schadeberg and Quint and are retranscribed as aː, eː, iː, etc. in the database.

3. Schadeberg's ʌ (the symbol is apparently used as the traditional +ATR correlate to a, employed in Africanistics) more or less consistently corresponds to Quint's mid-central ɐ. Out of caution, we leave both ways of transcription as they are, particularly because there may have been subtle idiolectal differences between informants.

III. Utoro; Tiro

The transcriptional conventions in [Stevenson 2009] are treated as follows:

1. The centralized rounded vowel marked as ö is converted to UTS ɵ.

2. In his phonetic description, Stevenson mentions several +/-ATR and other allophones for both Tiro and Utoro vowels (e. g. i / ı in Utoro, e / ę in Tiro etc.). However, he does not distinguish between them in his regular transcriptions, and we do not believe it necessary to try and mark the phonetics as accurately as possible based on his rules of allophonic distribution.

3. Dental plosives th, dh are converted to UTS t̪, d̪. Stevenson's description of the retroflex articulation in Tiro and Utoro is not very clear: in addition to alveolar plosives t, d, for Tiro he also sets up two retroflex fricatives that he marks as ʂ and ʈ respectively and defines them as allophones (ʂ word-initially, ʈ in word-medial position). It remains unclear if ʈ (retroflex fricative, acc. to Stevenson!) is phonologically distinct from t or not; in basic lexicon, it is very rare anyway, encountered only twice for the entire Swadesh wordlist. For the sake of accuracy, we leave it unchanged.

4. "Retroflex affricates" that Stevenson marks as tr, dr are encountered very rarely in both Tiro and Utoro and most likely represent positional variants of simple alveolar plosives. In those few cases where they are encountered in the basic lexicon, we leave this transcription unchanged.

5. Palatal consonants c, j, ny are converted to UTS ɕ, ʓ, ɲ respectively.

IV. Moro

Moro lexicon as transcribed in [Black 1971], subject to basic typewriter limitations, has some idiosyncratic peculiarities, reflected thus in our transcription:

1. Dental plosives t_, d_ are converted to UTS t̪, d̪.

2. Interdental voiced fricative -d- ("strikethrough d" in the original edition) is rendered as UTS ð.

3. Palatal consonants c, j, n̄ are rendered as UTS ɕ, ʓ, ɲ respectively.

The source contains a very valuable and highly detailed description of the system of Moro allophones (pp. 1-15); however, we do not reflect this in the wordlist, restricting ourselves to simply reflect the base system of phonological oppositions.

Database compiled and annotated by: G. Starostin (latest update: September 2016).