Sunday, September 04, 2011

Forthcoming talk: the history of Kwarandzyey viewed in areal context

Some readers may be interested in a talk I'll be giving at the end of this month in Paris, at LACITO on 30 September in a colloquium called Journée d'étude : Aires linguistiques. The title is "Du Sahel au Maghreb : essai d'une histoire linguistique du korandjé, langue songhay loin de son aire d'origine". (Yes, I'm going to try to deliver it in French - a foolhardy decision, given that I've only ever studied two years of it, but there you are.)

Basically, Kwarandzyey - the language of Tabelbala in SW Algeria - is a Songhay language, brought originally from at least a thousand kilometres to the south in the Niger valley. The Songhay family typologically fits reasonably well into West Africa - for Güldemann, it is a peripheral member of the Macro-Sudanic area - and shares some features widespread throughout sub-Saharan Africa and rare north of the Sahara (such as Noun-Numeral order). In particular, Songhay shows strikingly close structural similarities to the Mande languages (eg S-Aux-O-V order); these similarities for the most part appear likely to reflect early Mande influence on Songhay, rather than a common genetic origin. The languages of Northwestern Africa - Arabic and Berber varieties alike - share a number of characteristics which contrast sharply with Songhay and with the West African languages around it: some of these reflect common inheritance (eg a two-gender system), others reflect convergence, having been absent from both proto-Berber and early Arabic (eg a vowel system consisting of a i u, plus neutral ə restricted to closed syllables.) Over the past millennium, Kwarandzyey has changed a lot; most of these changes (lexical, grammatical, and phonological) have brought it closer to the Arabic and Berber varieties spoken around its current location. But the changes do not derive from a single language; the lexicon lets us discern influence at least three different branches of Berber (Western, Atlas, and Zenati) and two rather different Arabic dialects (Western Maghrebi and Hassaniya). One way to view this phenomenon is to say that Kwarandzyey, having been isolated from the Macro-Sudanic area to which its ancestor belonged, has been getting integrated into a Northwest African linguistic (and indeed cultural) area. But is this a helpful way of viewing things, or does it misleadingly present an essentially local phenomenon as a product of a wider region? We'll have to see...

5 comments:

John Cowan said...

OT: I've been reading Martin Haspelmath's papers after one of them was featured on Language Hat, and I found one of his papers on the European Sprachbund [PDF scanned sideways] very interesting. This one's on equatives and similatives, but it's the Sprachbund itself, which he calls "Standard Average European" in a hat tip to Whorf, that really gets my attention. There are a couple of other papers, not online, that apparently discuss the idea.

The core languages of the Sprachbund are Romance, Balto-Slavic, West Germanic, and the languages of the Balkan Sprachbund. The periphery includes North Germanic, Hungarian, Finnic, Armenian, and Georgian (perhaps because of Greek influence on the last two?). English and French are on the boundary between core and periphery. The weird languages of Europe — Celtic, Basque, Maltese, Turkish, the other Uralic languages, and the languages of the Caucasus — are definitely excluded.

bulbul said...

(crossposted at LH, expanded here):
John,
have you read Heine and Kuteva's "The Changing Languages of Europe"? They expand on Haspelmath's ideas and argue that even Basque should be included. I would add that even Maltese, weird though it may be, must be counted, even if only for the Italian influence on the verbal system, though that's not the extent of it.

David Marjanović said...

S-Aux-O-V order

Do you mean like German, which has S-finite verb-O-infinite verb in independent clauses?

(A paper I recently read even claims that this "SVOV word order" is such a strong feature that it has contributed to the extinction of the, uh, cognate of the past tense in Upper German -- the cognate of the present perfect is made of two words, so it allows us to fill both V slots. I like that idea a lot.)

Now I'll read the paper on Standard Average European and return to LH at long last. Is the original "SAE as an Exotic Language" available on teh intart00bz somewhere...?

Lameen Souag الأمين سواق said...

Haspelmath's list (http://www.joerg-rhiemeier.de/Conlang/sae.html) looks plausible, but Whorf's original notion of SAE seemed to focus primarily on grammatical semantics (http://sloan.stanford.edu/mousesite/Secondary/Whorfframe2.html), and I'd like to see more work on that...

The philosophical difficulty here is that: languages absorbing certain features of their neighbours, as far as I can tell, is the normal state of affairs. You can draw vague circles encompassing SAE, but other features will align (for example) the Balkans with Turkish, or Iberia with North Africa; without any principled way of counting shared features, are we really justified in picking one bundle of shared features and labelling it a Sprachbund? (My suspicion is yes, but it's hardly obvious.)

David: One of the neater features of West African languages (Eastern Songhay, at least) is that Aux is very clearly a category distinct from V; unlike in German or English, its "slot" cannot be occupied by a verb, even when empty. Aux includes mood, aspect, negation, and (future) tense (all bundled together.)

David Marjanović said...

Haspelmath's list arguably counts a few things 2 or 3 times, and the voice contrast should be renamed "one-way contrast in manner of articulation" or something because the only Germanic language that can be said to have a voice contrast is Dutch, but the rest looks good...

You can draw vague circles encompassing SAE, but other features will align (for example) the Balkans with Turkish, or Iberia with North Africa; without any principled way of counting shared features, are we really justified in picking one bundle of shared features and labelling it a Sprachbund? (My suspicion is yes, but it's hardly obvious.)

What do you mean by "principled"?

And yes, of course it will be blurry at the edges no matter how you count shared features. Even the Balkan Sprachbund is.

One of the neater features of West African languages (Eastern Songhay, at least) is that Aux is very clearly a category distinct from V; unlike in German or English, its "slot" cannot be occupied by a verb, even when empty.

Awesome!