Research Projects

Behind the books there are four long-running lines of work. Each took years, and each produced something the field did not have before.

The paremiological experiment

The question sounds simple and had never been answered for any language of the Horn of Africa: which proverbs do Somalis actually know?

Collections tell you what editors have gathered. They do not tell you what is current. The method for finding out was developed by G. L. Permyakov, who established a “paremiological minimum” for Russian of some 300 items — the set genuinely known to most speakers.

The Somali experiment ran from 1999 to 2001, supported by the Deutsche Forschungsgemeinschaft: at the Humboldt University of Berlin in 1999, at the Free University in 2001, and in Aarhus in 2001.

How it worked. The starting pool was the compiler’s own dictionary — over 6,000 paremias drawn from 25 books and more than a hundred oral sources, the earliest of them Reinisch’s Die Somali-Sprache of 1900. Through successive lists this was reduced to 656 proverbs, each cut off after its opening. Forty-five informants — 32 men and 13 women, in three age groups — were asked to complete them.

What came out. Of 29,520 possible completions, 12,342 were filled in correctly: 41.8%. That is far below the equivalent American (88.8%) and Hungarian (82.5%) experiments, and the gap has no confident explanation. 249 paremias were known to more than half of the informants; these form the core.

The results carried a second finding, sociolinguistic rather than paremiological: proverb knowledge rises with age and education, is higher among men, and is markedly higher among speakers who grew up in a nomadic environment — 365 proverbs against 261 for those raised in towns.

Published as Hubsiimo hal baa la siistaa (Moscow, 2002). The title is itself a proverb: to know something for sure, one would even part with a she-camel.

The dictionary of Somali proverbs

The Qaamuuska maahmaahyada Soomaaliyeed (1998) documents over 6,000 proverbs and proverbial phrases with their variants, trilingually — Somali, English, Russian. The majority were taken down directly from Somali informants rather than copied from earlier collections.

The work did not stop at publication. After an interview with WardheerNews in 2006, readers sent in so many proverbs from their own families and regions that the larger dictionary then in preparation had to be reopened. Collection by correspondence has continued since.

Kinship terminology

A comparative project with V. Ya. Porkhomovsky on kinship terms in Somali and Oromo — the two largest Cushitic languages.

Its practical result is a list of 34–38 Somali kinship terms, against the 20 in I. M. Lewis’s list, until then the only reliable source. The work describes the hierarchy xaas → reer → jilib → laf → qabiil, the large proportion of classificatory terms, the absence of simple terms for “grandson”, “nephew” and “niece” — rendered by descriptive formulas — and the vocative system in which parents address their children with the words for “father” and “mother”: aabbo, hooyo.

Published as “Somali kinship terms” in the Semito-Hamitic Festschrift for A. B. Dolgopolsky and H. Jungraithmayr (Berlin: Dietrich Reimer Verlag, 2008, pp. 155–160), with an expanded Russian version prepared under a fundamental research programme of the Presidium of the Russian Academy of Sciences. See Kinship Terms.

Dialects of southern Somalia

Marcello Lamberti’s maps divide Somali into five dialect groups. Field interviews in more than thirty settlements of southern Somalia showed that at least two of those maps — Darod and Benadir — now need serious revision: the civil war set off mass migrations, and dialects mixed and displaced one another.

Two cases document it. In Gedo, the “Marehanisation” of the province, begun as early as 1974, has left the Degodia dialect practically extinct there. In Afgooye, the Habar Gidir variety has expanded into what the maps show as Benadir territory. The conclusion is uncomfortable for anyone relying on a dialect atlas: in the south there are almost no pure dialects left, and speakers use mixed idioms.

This has a practical application in language analysis for the determination of origin — the assessment of asylum seekers’ speech — where a map twenty years out of date can produce a confident and wrong answer.