This book "asserts that the origin and spread of languages must be examined primarily through the time-tested techniques of linguistic analysis, rather than those of evolutionary biology" and "defends traditional practices in historical linguistics while remaining open to new techniques, including computational methods" and "will appeal to readers interested in world history and world geography."
Review of Un Créole Arabe: Le Kinubi De Mombasa, Kenya
Date: Tue, 1 Nov 2005 12:04:05 -0800 (PST) From: John McWhorter Subject: Un créole arabe: Le kinubi de Mombasa, Kenya & Kinubi Texts
AUTHOR: Luffin, Xavier. TITLE: Un créole arabe SUBTITLE: Le kinubi de Mombasa, Kenya SERIES: Lincom Studies in Pidgin and Creole Linguistics PUBLISHER: Lincom GmbH YEAR: 2005
AUTHOR: Luffin, Xavier. TITLE: Kinubi Texts SERIES: Languages of the World / Text Collections PUBLISHER: Lincom GmbH YEAR: 2004
John McWhorter, Manhattan Institute, New York
What happens when speakers of Nilotic languages serve in a British army under Arabic-speaking commanders battling a revolutionary uprising in Sudan, the revolution is successful enough that the British withdraw the troops to areas where Arabic is not spoken, and the troops stay there permanently, developing a new language based on the Arabic words and constructions they have been using to communicate with one another?
In terms of language acquisition, the situation is similar to the importation of African slaves to New World plantations, and thus we might expect a creolized Arabic in which much of the accreted elaborations of ancient Arabic varieties would be shaved away. Yet two popular schools in creole studies would predict otherwise. The Relexification Hypothesis would predict a hybrid of Arabic and Nilotic languages, susceptible to analysis as Nilotic grammar with Arabic lexicon. Others would predict a combination of features "selected" from the contact "ecology," lightly seasoned by Nilotic influence, but with so much of the basic machinery of Arabic intact as to render the concept of creole unnecessary.
However, Nubi Creole Arabic (Kinubi) is a stark contradiction to both of these premises, as Xavier Luffin's grammar amply demonstrates. Kinubi emerged among soldiers of southern Sudan, serving a Turko- Egyptian government controlled by the British. The Mahdists gained control of southern Sudan in 1881, and at the end of this decade the British transferred these troops to Uganda, Kenya and Tanzania. Meanwhile, a new contact language had developed among them, which they carried with them in their exodus. Their descendants still speak the language today, some having also settled in the Democratic Republic of the Congo. The language that emerged from this situation is neither Nilotic grammar with Arabic words nor just one more colloquial Arabic variety like Moroccan or Egyptian Arabic. It is a creole language -- in the sense that its grammar is a vast subtraction of Arabic's, has only fragmentary inheritances from Nilotic languages like Bari, Dinka, and Mamvu (or a few Central Sudanic ones), and is plainly the result of a people who learned only a rudimentary Arabic and expanded it into a new natural language. Nonconcatenative morphology, richly productive in all colloquial Arabics, is completely extinct in Kinubi, as is grammatical gender, the definite article, phonemic length contrast, and other features.
Kinubi has played little part in discussions of creole genesis, partly because its literature has rarely been channeled to creolist venues, and partly because most creolists are more familiar with European languages than with Arabic. Luffin's grammar, with its accompanying volume of transcribed texts, joins a growing body of literature on Kinubi which contains a valuable lesson for not only creolists but all language contact specialists.
That is, it is often suggested that claims about the nature of creole languages are premature in that most creoles emerged as the result of contact between Romance and Germanic languages and certain Niger-Congo ones. The suggestion is generally made with an implication that creolization may not be as inherently subtractive a process as often assumed and that creoles based on, say, Georgian or Warlpiri might be well-inflected and perhaps even ergative or the like. Many suppose, for example, that creoles are low on inflections simply because Kwa languages like Twi and Fongbe are, rather than because creolization entails incomplete acquisition of a target language. Similarly, many note that the colloquial Frenches that slaves were exposed to are much less inflectional than standard French, such that the isolating typology of French creoles is merely a small step further along a pathway of ordinary grammar-internal evolution.
There are serious problems with both of those claims, but the very nature of Kinubi nicely addresses them by sheer example. Kinubi's lexifier, Sudanese Arabic, is a richly inflected language, displaying the famous nonconcatentative morphology of the Semitic family. Thus Kinubi's analytic structure cannot be treated as a "natural" development. Meanwhile, Nilotic languages are hardly chary of inflection, and yet again, Kinubi is an isolating tongue. The lesson is that creole genesis entails not only mixture, but simplification. All Arabist analysts of Kinubi readily see that Kinubi's grammar renders it, and its sister Juba Arabic spoken by those who stayed in Sudan, in a class apart from colloquial Arabics elsewhere.
Luffin's grammar includes not only the usual sections on phonology, morphology and syntax, but a substantial introductory section on the language's history and sociological position, a brief discussion of schools of thought on creole genesis, one on sources of its lexicon, and even a rather extensive discussion of code-switching between it and its main adstrate languages Swahili, English and Arabic (the latter mostly from the religious literature adhered to by its speakers, who are Muslims). Luffin limns a rich portrait of the language as it varies both in time and, especially, space, based on fieldwork not only on his principal focus, the variety spoken in Mombasa in Kenya, but also in Uganda as well as with speakers of the variety spoken in Sudan.
Luffin gives especial attention to adstratal influences as they vary in these locations, mostly lexical, and to an extent that renders the grammar perhaps excessively listy in places. At times one supposes that it may have been more useful to present this lexical coverage in a separate article or monograph, as those seeking to use the book strictly as a grammar must wade through quite a bit of diligent tabulations of lexical variations.
Another questionable aspect of the coverage is the occasional exhaustive comparisons of Kinubi with Sudanese Arabic. It is certainly not a flaw that the book is as much a comparative presentation as a strict description: this makes the tome useful in assessing Kinubi's implications for tracing the nature of creole genesis (especially since the overlap between creolists and Arabists is very small, meaning that presentations of Kinubi will be most useful with ample examples from Arabic itself). The problem is simply that the comparison so resoundingly reveals again and again that Kinubi has simply flushed away so very much of the machinery of Arabic. For example, when Luffin takes each Arabic triconsonantal verb pattern one by one and lists the handful of Kinubi forms that happen to have been based on it rather than the usual source, the simple imperative, he is showing not living grammar but what today qualifies as fossilizations: these patterns have no grammatical status in Kinubi itself. Or, Luffin carefully tabulates the conjugational patterns in Sudanese Arabic with the unchanging verb in Kinubi -- which leads one to wonder whether just stating that Kinubi has no conjugational patterns would not have sufficed.
But the grammatical coverage also reveals assorted interesting features that demonstrate that, as many creolists consider so urgent, complexity is not alien to creole grammars. Stress alone encodes the passive: kútu "put," kutú "be put" and also deverbalizes to create nouns: kúruju "cultivate, grow," kurúju "agriculture." (Heine 1982 is apparently unique in treating Kinubi as tonal, and thus describing distinctions like this as tonally encoded.) Kinubi pronouns, typically of a creole, do not vary for case, but there is a paradigm of bound pronominal affixes used in certain contexts, and unlike many creoles which have clitic pronominals phonetically still close to the free ones, two of the Kinubi ones are suppletively distinct from their free equivalents (1S and 2S -i and -ki for free ána and íta). Morphophonemic processes include assimilation in the vowel of progressive marker gi- to the initial vowel of the verb: gi- kélem "talking" but gu-úza "buying" and ga-já "coming." Stative verbs are rarely modified by tense or aspect markers, and moreover, as has proven the case with many creoles, the semantic contribution of some of Kinubi's preverbal markers is by no means semantically prototypical, conditioned by complex aspects of pragmatic and discourse phenomena.
Something else that stands out in the grammar, however, is a striking amount of free variation. Words often occur in alternate forms, sometimes conditioned by regular processes such as a tendency for final /aC/ to become [e] in color terms (áswad > áswe "black), but just as often varying randomly (ábyad ~ ábya "white," or alsán ~ alshán "because," káfu ~ kófu "fear"). Preverbal gi- does not assimilate vocalically to the verb in regular fashion, since there are ample exceptions such as gi-só "going." Future marker bi- usually does not assimilate -- but does so with a few roots. Stress only encodes the passive and the deverbal in a subset of cases, and in fact Luffin only finds the latter process marginally in Mombasa: these appear to be changes in progress like the English stress variation distinguishing between nominal and verbal uses of words like survey, record, and permit. It is an especially admirable aspect of Luffin's grammar that he pays assiduous attention to this kind of variation, its distribution and extent, especially since little of it is a matter of gradient influence from the Arabic lexifier; that is, Nubi's variation is not of the well-covered sort on view in creole dialect continua in the Anglophone Caribbean.
Luffin's portrait of Kinubi can be taken as indicating what a "real" language is like, i.e., languages other than the 200 or so that have extensive written traditions and norms: a basic pattern varying in fluid fashion according to surrounding languages as well as the processes of transformation that all linguistic systems undergo as used in human mouths over time. However, the sheer amount of variation Luffin presents is not typical even of unwritten languages that are old -- and is also quite common in creole languages, especially in areas such as morphophonemics in which complication usually has no semantic function and is obviously the result of phonetic transformations over time. Luffin's presentation of Nubi -- although his discussion does not occasion specifying this -- has all of the hallmarks of a language only a century-plus old, having yet to freeze into the degree of obligatory stricture typical of a grammar in which countless speakers over countless centuries have converged upon arbitrary conventionalizations.
Meanwhile, Kinubi Texts collects transcriptions of the speech of twenty speakers, showing the language in all of its reality including ongoing code-switching and borrowings. The volume is compact enough that ideally it would be an appendix to the grammar.
The grammar's format is user-friendly. The text is clear, except the account of Kinubi's hitsory: the political situation of late nineteenth- century Sudan is remote to most today, and Luffin's description is rather difficult to follow without very close reading (and, truthfully, is best engaged with supplementary descriptions from elsewhere). There is also an unsuitable amount of repetition of the sort that word processing allows: certain text blocks were copied into other sections without being deleted from their original place.
Notwithstanding, the two books together stand as the most comprehensive single source on this creole now available. Luffin's presentation presents Kinubi in all of its dynamic variety such that it is alive on the page in the same way as Tok Pisin has been in work by Peter Mühlhäusler, Suzanne Romaine and other scholars. This will hopefully allow Kinubi to play a larger part in scholarship on language contact and creolization in the future.
Heine, Bernd. 1982. The Nubi language of Kibera -- an Arabic creole. Berlin: Riemer.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
John McWhorter was formerly Associate Professor of Linguistics at U.C. Berkeley and is now Senior Fellow in Public Policy at the Manhattan Institute. He is the author of The Word on the Street, The Missing Spanish Creoles, The Power of Babel, Defining Creole, and the forthcoming Language Interrupted: Signs of Non-Native Acquisition in Standard Language Grammars.