There has been a flood on the Web of confusing and confused discussions about the “Arab”/”African” divide in Darfur recently. Perhaps a little historical background will help to clarify the issue.

The population of Darfur is approximately 2/3 to 3/4 “African” and 1/3 to 1/4 “Arab”; here I am basing myself on the 1955-56 Census. All the later censuses are not helpful in this context. All in Darfur are Muslim of the Maliki madhhab and a high proportion are adherents of different branches of the Tijaniyya tariqa. Of course, who is “African” and who “Arab” is ultimately a matter of self-ascription.

The oldest self-description of Darfur’s ethnic identities comes in an undated land-charter issued by the sultan, Muhammad Tayrab (ruled 1752/3-1785/6), where the ruler is described as sultan al-‘Arab wa’l-‘Ajam, which may be translated as “sultan of the Arabs and non-Arabs”. Now what ‘Ajam meant in eighteenth century Darfur (assuming that the phrase is not simply some borrowed rhetoric) it is impossible to know; conventionally the word has something of the meaning of the classical Greek, barbaroi, i.e. people who speak barbar, i.e. not Greek, or in our case not Arabic. Generally in Arabic, ‘Ajam has a pejorative sense, but bear in mind that the ruling dynasty, the Keira from the Fur people, had already begun to construct an Arab pedigree for itself, in other words they may already have identified themselves as Arab. The same sultan paid a 1,000 gold coins for a copy of the greatest Arabic dictionary ever written, the Taj al-‘Arus ‘ala sharh al-Qamus of al-Zabidi (written ironically by an Indian!).

Ethnicity in Darfur (and in pretty well everywhere else) is a very slippery and elusive concept, involving language, real or manufactured pedigrees or identities, colour categories, and ideas of superiority. Also to be factored in are processes of Arabisation and Arabicisation. I shall not attempt to pursue all these themes here. Many are discussed in greater detail in my The Darfur Sultanate. A History (London: Christopher Hurst & New York: Columbia University Press, 2008).

Let us start with slipperiness: in 1969, a student of mine at the University of Khartoum from Darfur, now Dr. Hasan Imam Hasan, went to Jabal Mun (on the Chad/Sudan border north of al-Jinayna) financed by the University’s Vacation Fund to collect material about his people, the Mileri of Jabal Mun. He himself is a Mileri, but from the Mileri community of Kas in southern Darfur. Dr. Hasan did an excellent job and together we published an article in Sudan Notes and Records, 51, 1970, 152-61. The Mileri speak a dialect of Tama, Milerinka, but had begun by the 1960s to assume a quasi-Arab identity as members of the Misiriyya Arab tribe (the article describes how this came about). Now they are known as Jabal Misiriyya or “Mountain” Misiriyya, linking themselves consciously to the Misiriyya Arabs of southeastern Darfur and Kordofan. The point here is that language in itself is not ncessarily an ethnic marker either way. The Birged and Berti of eastern and northern Darfur have lost their languages within the last hundred years, but seemingly do not yet regard themselves as Arab simply because they now speak only Arabic. They may well do so in the future.

The manufacturing of genealogies retroactively has been much studied by historians and anthropologists in the Sudan. If one looks closely at the pedigrees given in MacMichael’s History (see below) and more recent Sudanese works, they can usually be divided into three segments; 1) x number of generations going back to ‘Adnan, reputed ancestor of the northern Arabs of Arabia, i.e. north of the Yemen, copied ultimately from one of the genealogical works in common circulation in the Muslim world (many of these in the Sudan are ascribed to a legendary, probably fictitious, genealogist, al-Samarqandi), 2) a middle section linking Arabia to Africa of usually about ten generations which may be fictitious or not, 3) and a third section (usually seven to eight generations) that reflect realities within the tribe in question, especially in regard to group responsibility in regad to blood-money payments (diya). Crudely, the Arabs in Darfur have quite elaborate genealogies which have a charter-like function within their communities, some of the non-Arabs like the Zaghawa and Berti have less elaborate genealogies but with a similar function, and the Fur do not really care about such matters, since they have very different descent patterns and different ways of dealing with blood-money issues.

The most elaborate written genealogy I ever photographed in Darfur (fifteen pages long, detailing some forty generations) was from a Fulani community in north central Darfur (Khiriban near al-Fashir), tracing their descent to the Arab Muslim conqueror of North Africa in the years just after the Prophet, ‘Uqba ibn Nafi’, but their ancestors came from Futa Toro in Senegal. And I could go on.

Colour as an ethnic marker: “black” is “black”, but there are many shades of “black” in Darfur and indeed in the Sudan generally; colour-coding is built into Sudanese Arabic, as anyone who has overheard Sudanese talking among themselves will know. Nineteenth-century documents concerning slaves usually begin with a wasf or formal physical description, which usually included height, whether having scars (shillukh), ethnic origin, and colour””e.g. safra/asfar “yellow” i.e. sallow skin, ahmar “red” or lightish “Arab” skin (in Fur, solongdungo, the nickname of the firs historical Keira ruler, Sulayman), azraq “blue”, i.e black etc. How racially loaded these terms were to those writing the documents is hard to say. What they signify in present-day language is beyond my competence to comment on. It would be interesting to know if language usage in this respect is changing now in the IDP and refugee camps.

Concepts of superiority in Darfur were historically linked to Islam and not directly or solely to ethnic identity. The Fur, the largest ethnic group in Darfur, are clearly an amalgam of ethnicities that grew by assimilation defined by Islam, but marked by the use of the Fur language (Fur was the court language until 1916) and allegiance to the Keira dynasty. In Saviors and Survivors, Professor Mamdani applies Spaulding’s ideas about Arabisation/Arabicisation under the Funj sultans along the Nile to Darfur. This is misleading; the ethno/political realities of the Funj state and Darfur were quite different.. Spaulding’s ideas of “enclaves” of Arabisation/Islamisation in the towns along the Nile, do not apply in Darfur if only because there were virtually no towns in the latter. Darfur is no way part of the Nilotic Sudan.

Finally, let us consider the British as possible villains in this debate. Professor Mamdani quotes Churchill. In the case of Darfur, the key figure is rather Sir Harold MacMichael (d. 1968); in 1915-6 MacMichael was intelligence officer in the lead-up to the conquest of the Darfur sultanate in 1916. He wrote a report (one of several), Notes on Darfur 1915, much of which I reproduce in my forthcoming Darfur and the British. A Sourcebook (London: Christopher Hurst, 2009). MacMichael’s report is profoundly racist, based on the assumption of Arab superiority and “African” inferiority, especially the Fur. MacMichael consolidated his views on the inherent superiority of the Arabs in his The History of the Arabs in the Sudan, 2 vols. (London, 1922), one of the most enduring and influential books in Sudanese historiography.

The next item in my book is the Western Darfur District Handbook from 1936, largely concerned with the administration of the Fur of the Zalingei District; here the Fur are lauded as hardworking (both men and women) and law-abiding””model citizens, their only failing being a casual attitude to sex. The Arabs appear in the Handbook as shifty characters skulking around the markets looking for what they could steal. An abrupt change of view in just over twenty years.

Darfur and the British documents how the British viewed the Arab/African divide in 1930s and 40s and how exasperated they often were by especially the northern Arab camel nomads and the Zaghawa; the former because they lacked defined tribal homelands and the latter because their tendency to move ever southwards.

These are the comments of an historian; what would be interesting to know is how concepts of ethnicity and identity are changing in the IDP and refugee camps now. My impression from visits in 2005 was that the IDP camps were still organised on community lines, but perhaps this is changing. I would guess, however, that ethnicity is still as slippery and elusive as ever.

R. S. O’Fahey is Professor at the University of Bergen. His website contains the finest collection of online historical documents on Darfur.