Bantu’s Swahili.

Kamau Muiga makes a point in his essay Bantu’s Swahili, or How to Steal a Language from Africa that I wouldn’t have thought needed to be made in this day and age: that Swahili is “a Bantu language, an African language that wholly emerged in Africa among Africans.” But apparently it’s still widely said to be “a lingua franca, a pidgin, between African languages and Arabic on the East African coast,” so I’m linking to his demolition of that idea. He starts out citing a passage from Henry Louis Gates, Jr.’s documentary series Wonders of the African World in which Sheikh Badawi, “one of Lamu’s most venerable Islamic scholars,” said “he carries little if any African ancestry, and that he can trace his own lineage back to Arabs who emigrated to the East African coast over the last few centuries and even back to the Prophet Muhammad himself”; I remember being struck by that myself when I saw the show — it was so evidently false. Muiga also quotes the great book that gave me my own first understanding of African history, W.E.B. DuBois’s The World and Africa. It’s a good, eloquent essay, and I thank Trevor for the link.

Comments

  1. David Eddyshaw says:

    Kamau Muiga is quite right about the historic European tendency to attribute anything magnificent in Africa to outside influence; in West Africa the Fulani have been the subject of extensive racial fantasy not unconnected with the fact that they were the rulers of large areas at the time of the European invasions, for example.

    In this particular case, though, I think he’s he’s fingering the wrong culprit. The wish-fulfilment of people like Ali Mazrui in this regard has everything to do with Islam, and nothing to do with a general desire to downplay African excellence as such; quite the contrary.

    I recall an (on the whole rather good) television series about African history presented by Mazrui which was somewhat marred by this. At one point he quite seriously suggested that the digging of the Suez Canal was a European plot to separate Africa from the Arabia that it evidently culturally belonged with (in his view.)

    This is also all to do with the fact that the actual Swahili people really can trace their ancestry in the male line back to Arabs, and really are thoroughly Muslim. Moreover, Mazrui’s thesis (though wrong) is by no means as daft as Mulga makes it sound: Swahili plainly is a Bantu language; but the idea is that although it originated as a pidgin it’s now highly decreolised. The idea is not self-evidently stupid by any means.

    Derek Nurse and Thomas Hinnebusch took the idea seriously enough to publish a detailed refutation (“Swahili and Sabaki: A Linguistic History”) at least in part as a response.

  2. Patrick Ingram says:

    Reminds me of all the people who say that English isn’t a “true” Germanic language, but rather a Germanic-French creole.

  3. I wonder what can be said about Persian then. After all, if you get a sufficiently large Persian dictionary, you will find there every Arabic word in existence.

  4. In principle, if you obtain a thoroughly comprehensive Garo dictionary (which does not exist) it will contain not only the whole of the OED but the whole of an unabridged Bengali dictionary too, because (as Robbins Burling pointed out years ago) Garo speakers will freely borrow words from either lexicon when and as needed, just as anglophones feel free to use any word in the OED (well, perhaps not the ones marked Obs.) when and as needed.

    Presumably many of the phonological features that distinguish Kiswahili from its relatives, like the loss of tone and nasal vowels, reflect the inability of ignernt Arabs to pronounce them. In any case, though the People of the Coast are now Africans speaking an African language, the People of the Highlands speak an unimpeachably Asian language, despite constituting the very flower of African-ness for Asians and Europeans over the last two and a half millennia, a distinction they share with the leopard.

  5. : At one point he quite seriously suggested that the digging of the Suez Canal was a European plot to separate Africa from the Arabia that it evidently culturally belonged with (in his view.)

    In terms of language families, and indeed language areas, it would be more accurate to say Arabia evidently belongs with Africa.

  6. You’re on very thin ice when you start arguing with people that their own definition of themselves is obviously wrong “because just look at you”. Anyone who’s been to the Arabian peninsula will know that there are plenty of people who were born in the region, speak Arabic and describe themselves as Arabs, but who could easily be described as “black African” in appearance by an outsider.

  7. David Eddyshaw says:

    it would be more accurate to say Arabia evidently belongs with Africa.

    Indeed yes; the reverse would be a clear case of the tail wagging the dog. Still, it’s not that Mazrui is wholly wrong; for example, the vast majority of Semitic languages are African, and the majority of speakers of Semitic languages are African too. Mazrui’s general point (I’ve been a bit unfair to him, in that he probably was actually joking about the Suez canal, and I was too dim to realise it) was that the European idea of precolonial Africa as separate from the rest of civilisation is resoundingly wrong. One can’t argue with him there. He’s right; and in this regard, Mulga (although you can see where he’s coming from) is actually wrong. And it doesn’t in any way detract from African creativity and cultural vigour to point out that Africa has always been part of world civilisation. The opposite, if anything.

    I came across a Kusaasi folk tale which is evidently the same story as Chaucer’s Pardoner’s Tale. And it is the same story; it’s an Indian tale that got to Chaucer via successive translations into Pahlavi, Syriac, Arabic and Latin, and got to the Kusaasi (who are not a Muslim group themselves, and have no tradition of writing in Arabic) via Arabic.

  8. J.W. Brewer says:

    Muiga is fighting over the provenance of Swahili because the stakes are high, because apparently everyone agrees it is the “crown jewel,” objectively more beautiful and expressive than the other (uncontroversially purely African in origin) languages of the region? Really? What would Mr. Ngugi from the other thread (who inexplicably chose to write in Kikuyu rather than Swahili after turning his back on English) think of that? (Separately, the analogy to Spanish is interesting, but he seems to have overlooked that Spanish did not acquire its stratum of Arabic loanwords from what you might call a position of strength. Rather, it acquired them during a historical period lasting a number of centuries when Romance speakers were politically subjugated to Arabic speakers, which he seems invested in claiming never ever happened along the East African coast.)

  9. Anyone who’s been to the Arabian peninsula will know that there are plenty of people who were born in the region, speak Arabic and describe themselves as Arabs, but who could easily be described as “black African” in appearance by an outsider.

    Yes, and those people have black African ancestors. Surely you’re aware of the massive importation of black African slaves into the Arabian peninsula that went on for many centuries.

    Really? What would Mr. Ngugi from the other thread (who inexplicably chose to write in Kikuyu rather than Swahili after turning his back on English) think of that?

    If you’d been paying attention, you’d have learned that Muiga’s mother tongue is also Gikuyu. He’s not making a claim that Swahili is “objectively more beautiful and expressive” (which would be not only absurd but contrary to the thrust of the essay), he’s saying that that’s how it’s commonly thought of. I suspect he has extensive evidence for that, if you want to argue with him.

  10. David Eddyshaw says:

    Presumably many of the phonological features that distinguish Kiswahili from its relatives, like the loss of tone and nasal vowels, reflect the inability of ignernt Arabs to pronounce them.

    This is a large part of Mazrui’s thesis; Nurse and Hinnebusch make a good case for it being a language-internal development, given that the closely-related languages have pretty stripped-down tone systems etc. The other main plank (and the one which seems most conclusive to non-linguists, i.e. most people, who think of languages as basically big bags of words) is the large Arabic component in the vocabulary, but it’s relatively easy to demonstrate that this must reflect borrowing rather than survivals from some Arabic-based creole.

  11. It’s certainly not as large as the French component in English, and nobody takes seriously any more that English is decreolized.

    ObOffTopic: While people in the Appalachians don’t speak Elizabethan English, it seems that people on Jabal Rāziḥ in the northwest of Yemen actually do speak a descendant of Old South Arabian, a Semitic language group thought to be long extinct. There is a lot of Arabic influence, of course, but the language is entirely unintelligible to other Yemenis. Rāziḥis delight in this, and apparently maintain their language because they enjoy baffling the lowlanders. (Isolation helps too, of course.)

  12. “use any word in the OED (well, perhaps not the ones marked Obs.) ”

    Some of those are in Shakespeare, whose plays are not Obs., which ought to impel an exception to the OED’s policy of forgoing pronunciation information for Obs. words.

  13. J.W. Brewer says:

    I took sentences like “it [Swahili] happens to be exceptionally beautiful to the African ear” to reflect the author’s own opinion, not merely dubious conventional wisdom from which he was subtly dissociating himself in the interests of scholarly objectivity. And really, “THE African ear”? It’s a mighty big and varied continent, and I don’t imagine folks on the Atlantic coast think much of the Hollywood tendency to treat Swahili as The Generic African Language.* The “beautiful” in the earlier reference to Swahili as “a beautiful language spoken as the lingua franca in seven East African countries” might be more of a pro forma tic (pushing back against historical prejudices against African things by throwing in a gratuitous laudatory adjective in front of anything and everything African), but “exceptionally beautiful” is hard to understand other than as a comparative claim made at the expense of other local languages. Note also how he follows his own description of Swahili as a “lingua franca” with vigorous criticism of other erroneous sources also use “lingua franca” as a description. Obviously it’s an important point that a language that functions as a lingua franca need not originate as a pidgin or contact language, and can simply happen to be the pre-existing local language of some ethnic group well-situated for purposes of the larger polyglot set of folks engaged in trade and thus in need of a lingua franca. Maybe better editing would have made it clearer that that was his point.

    Is there any reliable historical data on how early v. how late Swahili spread inland from the coast and became a common trade-facilitating L2 among speakers of disparate L1’s?

    *I may have previously told here the anecdote about a very brief bit of multiculturalism at the American School in Japan c. 1975, when I would have been in 4th or 5th grade. One of the teachers, without any particular context I can recall, decided we ought to be taught that “Jambo” was how one said “Hello” in Swahili. We didn’t even get taught a second lexeme — it was sort of jambo means hello; here endeth the lesson.

  14. Finally read it, and my reactions are not that positive. It sounds like his teachers taught him a lot of outdated, half-digested ideologically motivated nonsense, as is unfortunately not uncommon all over the world, and I sympathise with his efforts to push back on that. But he’s conflating such absurd claims with ones that are basically correct:

    “Swahili clearly borrows a very small proportion of its vocabulary from Arabic, but these few loanwords were deemed sufficient basis for declaring that Swahili is not a conventional African language but a hybrid, a lingua franca, a creole, birthed by the arrival of Arabs on the East African coast. And so a beautiful language was stolen from Africa.”

    The 100-word Swadesh list is not intended to be a representative sample of the lexicon – “very small proportion of its core vocabulary” would be accurate, or even “of the words most frequently used in daily speech”, but a very large proportion of its total vocabulary (far larger than Arabic influence on Spanish). Swahili is no creole, nor by any reasonable definition a hybrid, but it certainly is and was a lingua franca, and the “arrival of Arabs on the East African coast” played as important a role in its history as the arrival of Normans in that of English. And if it were a creole, it would still be entirely reasonable to call it an African language; are we to dismiss Krio or Pichi (let alone Nubi) as insufficiently African?

    It’s great that he’s gotten past feeling like “an African looking at a foreign culture” when he goes to the coast. But that feeling of being “an outsider looking in” doesn’t just come from this somewhat artificial continental divide. As he points out, the Swahili have their own culture and their own identity distinct from the highlands, and Africanness doesn’t erase that.

  15. Interesting about Jabal Rāziḥ. The seminal paper is Watson et al., The language of Jabal Rāziḥ: Arabic or something else? Proceedings of the Seminar for Arabian Studies, Vol. 36, Papers from the thirty-ninth meeting of the Seminar for Arabian Studies held in London (2006), pp. 35-41. One of the authors is a native speaker.

    Although Watson et al. consider it a unique Arabized South Arabian language, they also note that many Yemeni dialects have a large percentage of words without obvious Arabic cognates. I wonder how advanced the study of Yemeni Arabic dialects is. I would guess not very, but right now that is not my main worry about Yemen and its people.

  16. Here’s a link to a long document (93 pages, in the English version) in English, French, Portuguese, and Swahili. The subject matter is United Methodist Church doctrine and bureaucracy, so the content will probably be of interest only to UMC policy wonks, but the four versions might be good reading practice for language students.

    http://www.umc.org/who-we-are/way-forward-report-released-in-all-four-official-languages-of-general-confe

    (The URL ends after “confe”; this is not a typo.)

Speak Your Mind

*