THE LANGUAGES OF THE NORTHERN SUDAN: AN HISTORICAL PERSPECTIVE
BY BRUCE G. TRIGGER
THE JOURNAL OF AFRICAN HISTORY VOLUME VII, Nos. 1, 2 & 3, 1969
For several years I have been investigating various problems of the culture history of the northern part of the Republic of the Sudan. As one approach to solving these problems, I have sought to examine the linguistic history of the region. While doing so, I came across a number of contradictory hypotheses concerning specific historical problems, some of which were advanced by historians who admittedly lacked adequate linguistic data. A solid basis for the re-examination of rival claims has been provided by the recent classifications of African languages put forward by Joseph Greenberg.l My indebtedness both to his published works and to other information with which he has provided me is gratefully acknowledged.
Most of the northern part of the Republic of the Sudan is presently occupied by Arabic speakers who are the descendants either of groups that entered the region in the fourteenth century or of native peoples who have learned the language from them. The Beja peoples, who inhabit the Eastern Desert, speak Cushitic languages that are related to others found throughout Ethiopia and Somalia. According to Greenberg, the Cushitic, Semitic, Berber, Ancient Egyptian and Chadic languages form co-ordinate branches of an Afro-Asiatic language stock. There is reason to believe that the Beja have inhabited the Eastern Desert for a long time and are descendants of the Blemmye mentioned in classical accounts and the Medjay of ancient Egyptian ones. The confinement of the Berber languages to the regions west of the Nile, the Cushitic languages to the regions east of it, and ancient Egyptian to the Nile Valley itself almost certainly antedates the rise of Egyptian civilization. The other indigenous languages of the Sudan mainly belong to a proposed Nilo-Saharan stock which is located between the Afro-Asiatic languages of North Africa and the Niger-Congo languages to the south. In the Sudan, the most widespread subfamily of NiloSaharan is Eastern Sudanic. The group predominates in the southern part of the Sudan and in the Great Lakes region of East Africa. The main branches found in the southern Sudan are Beir-Didinga, Temein and vilotic. Farther north, in the predominantly Arab-speaking part of the Sudan, six more branches are found. These are Barea, Tabi, Merarit, Dagu, Nyima and Nubian. Other languages of the Sudan belong to the Saharan, fur, Maban, Koman and Chari-Nile divisions of the Nilo-Saharan stock. In the south there are also small enclaves of Kordofanian and AdamawaErstern languages which belong to the Congo -Kordofanian stock.
Our main interest in this paper is reconstructing the distributions of the languages of the northern part of the Sudan prior to the spread of Arabic. Certainly the data are not sufficiently detailed to permit us to trace the distributions of all the smaller groups. Most speculation has centred on the distribution of the Eastern Sudanic languages and particularly in their Nubian division. The latter can be subdivided into four groups: Nile Nubian, Kordofan Nubian, Gebel Meidob, and Birged.2 Nile Nubian consists of two languages, Dongola-Kenuzi and Mahas-Fadidja, which are spoken in the Nile Valley; the other three, which are often referred to collectively as Hill Nubian, are isolated groups in Kordofan and Darfur. It is known from historical accounts and local inscriptions that Nubian was more widely distributed in the Nile Valley in medieval times than it is now. It was spoken not only in Dongola, where it is today, but also in the kingdom of Alwa, which had its capital near the present city of Khartoum.
The principal debate concerning the history of the Nubian languages has been whether they spread to the Nile Valley from Kordofan and Darfur or moved in the opposite direction. H. A. MaclliIichael3 was the first to propose that the Nubian languages had been introduced to the south-west from the Nile Valley and had been adopted by the indigenous population of the former region. Herzog,4 who noted that Hill Nubian contained Egyptian and Christian terms, felt that this had happened after A.D. 500. Murdock5 suggests that Nubian was spoken in the Nile Valley as early as the Khartoum Mesolithic, and that the Hill Nubian languages were carried to their present location by Nile Nubians fleeing the Muslim incursions in the fourteenth to sixteenth centuries. Arkells believes that this occurred about A.D. 30o and that the refugees were fleeing the Ethiopians. Local traditions among the tribes of Kordofan and Darfur are claimed as support for such a westward migration ,7 but such traditions may simply reflect the influence of the Meroitic and Nubian Christian kingdoms iq that region.
E. Zyhlarz,8 on the other hand, believes that Nile Nubian was carried to its present location from the Kordofan region. He postulates two migrations, one moving north across the desert to found the kingdom of Nobatia in Lower Nubia, the other eastward to emerge as the Black Noba who settled in the vicinity of Meroe in the fourth century A.D. J. W. Crowfoot,9 P. L. Shinnie,(10) and L. P. Kirwan(11) support at least the latter part of Zyhlarz's theory and have suggested a southern origin for the Tanqasi (Alwa) culture which they associate with the Black Noba.
N. B. Millet,(12) the excavator of Gebel Adda, has recently suggesteda variant of Zyhlarz's hypothesis. According to him, Meroitic may have been only the written language of Lower Nubia when it was resettled in the third century B.c. after it had been for many years a depopulated land. The majority of the new settlers were speakers of a dialect immediately ancestral to Dongola-Kenuzi. In the third century A.D. Mahas-Fadidja-speaking Nubians conquered the central part of Nubia, driving the Dongola-Kenuzi speakers north into the region just above Aswan and south into the Dongola area. This would explain the central position of Mahas-Fadidja between the closely related Kenuzi and Dongola dialects.
Wyatt MacGaffey(13) has proposed yet another theory which combines a late appearance of Nubian in the middle reaches of the Nile Valley with a Nile Valley origin for Hill Nubian. He suggests that there was a rapid northward movement of Eastern-Sudanic-speaking peoples from the present Shilluk country about A.D. 300. These people settled in the Nile Valley and the language they brought with them became the Nubian of the present day. Later the Hill Nubian languages were carried to the southwest by refugees from the Nile Valley.
Each of these hypotheses, implicitly or explicitly, involves assumptions concerning the relationships among the different Nubian languages. Those of MacNIichael, Herzog, Arkell, Murdock and MacGaffey require a close relationship between Nile Nubian and the three Hill Nubian groups. Millet and Zyhlarz allow for more distant relationships. Greenberg's research indicates that these four groups form the primary divisions of Nubian and a glottochronological study dates the more recent split of Nile Nubian into the Dongola-Kenuzi and Mahas-Fadidja languages at A.D. 860 ± 200 vears.(14) While Greenberg cautions against too much reliance on a date achieved by a controversial method, his findings can be used as a reasonable guide. The greater antiquity of the splitting between the major divisions of Nubian, which Greenberg describes as `considerable', suggests that any hypothesis which requires the derivation of the Hill dialects from Nile Nubian in the last 2,500 years is incorrect. Millet's thesis, on the other hand, seems to require a too early separation between Mahas-Fadidja and Dongola-Kenuzi.
The erroneous notion that Nile Nubian and Hill Nubian are closely related may result in part from a failure of travellers to distinguish between the speech of Nile Nubians who have emigrated to Kordofan and Darfur in recent centuries and the language of the Hill Nubians as such. Such migrations seem to have gone on during the period of oppressive Turkish rule.
Greenberg suggests that solely on the basis of linguistic evidence the principle of least moves indicates the Kordofan-Darfur area as the place of origin for the whole group. `Allowing for a period of more restricted geographical occupation and therefore maintenance of linguistic unity, the
glottochronological determination supports a view such as that of Zyhlar2 who sees Nile Nubian as coming in from the south about n.D. 400.'15 This interpretation is supported by ancient historical sources. According to Strabo,ls Eratosthenes is said to have mentioned a people called the Nubai about 200 B.c. He described them as a large tribe which was independent of the Meroitic empire and lived west of the Nile between Meroe and the bends of the river. The latter expression almost certainly refers to the Dongola region.l' According to Pliny the Elder," who wrote during the first century n.D., the Nubai were by then living in the Nile Valley, perhaps in the vicinity of Argo Island, and in A.D. 150 Ptolemyi3 placed them on the west bank and islands of the Nile. In the fourth century the Ethiopians appear to call this same group the Red Noba.2° Bare as these accounts are, they suggest a slow north-eastward migration of the Nubians. I have suggested elsewhere that the Nobatae of Procopius, who are said to have settled near Aswan in the reign of Diocletian, may have been raiding Kharga from the Nile Valley rather than actually living in the vicinity of the oasis. 21 If this is so, Zyhlarz's separate migration across the desert is unnecessary and the arrival of the Nobatae in the vicinity of Aswan is merely another event in their occupation of the Nile Valley. Presumably, these people became acculturated fairly soon after they arrived in the Nile Valley and while they were still in the Dongola region.
It seems to me to be quite incorrect, however, to equate the X-group or Ballana culture of Lower Nubia solely with the Nubians, especially since the culture seems to develop out of the Meroitic one and its new elements are, for the most part, ones which diffused from Egypt. The X-group culture was probably common to all the peoples living in Lower Nubia between the third and sixth centuries A.D.-Meroites and Blemmyes as well as Nubians. During this period, however, the Nubian language appears to have gradually supplanted the others that were spoken there.
Farther south, another Nubian group, called in Ethiopian inscriptions the Black Noba, overran Meroe in the fourth century.22 These people seem to be associated with the primitive Tanqasi culture, and are described in Ethiopian inscriptions as living in towns of reed houses (which seem to be their own) and towns of brick houses which they had taken from the Kasu or V'Ieroites. Presumably, these groups had less opportunity for acculturation than did the Nubian speakers to the nortli, prior to their arrival in an area known archaeologically. Since there is little written material from Alwa, we do not know if their dialect was the same as that spoken farther north or was somewhat different.
Prior to the spread of Nubian into the central reaches of the Nile Valley, the language of that region seems to have been Meroitic, a language known to us from many inscriptions in an alphabet partly derived from Egyptian demotic but as yet untranslated. Before F. L. Griffith23 had established the approximate values for the Meroitic alphabet, R. Lepsius24 had suggested that Meroitic might be either a Cushitic language or an old form of ~,Tubian. For a time, Griffith thought that it might be ancestral to Nubian and recognized a number of words which were, or seemed to be, common to both languages. Little progress was made along these lines, however, and soon Griffith abandoned the Nubian hypothesis.25 A Cushitic or `Hamitic' hypothesis has been pursued by Carl Meinhof26 and Zyhlarz.2' Although this hypothesis has attracted a certain following,28 it has not led to the translation of Meroitic and, in a recent important article, the German scholar, Fritze Hintze,29 has demolished the basic arguments in its favour.
My own study of Meroitic was prompted by the observation that many of the proposed cognates between Nubian and Meroitic are also cognates in many other Eastern Sudanic languages. Following this line of investigation, I found a number of new grammatical and lexical similarities which led me to conclude that Meroitic may be an Eastern Sudanic language and, therefore, related to Nubian, but less closely than Griffith originally expected.3° Secure knowledge of Meroitic is still very limited and I wish to emphasize that I regard this hypothesis as a very tentative one.
Today, the Eastern Sudanic languages occupy a large area in the southern part of the Sudan and in East Africa. Farther north a number of isolated Eastern Sudanic languages form enclaves within a sea of Arabic. These can be interpreted either as isolated languages which have penetrated north in fairly recent times or as remnants of a more solid northward extension of the Eastern Sudanic family. If my reconstruction is correct, the northwestern part of the Nilotic Sudan was largely occupied by Eastern-Sudanicspeaking peoples at the beginning of the Christian era. The Meroitic language occupied the middle reaches of the Nile Valley and apparently had done so for a considerable period of time. In the steppe land to the west was a large block of Nubian speakers. Early in the Christian era, some of the speakers of the Nile Nubian branch of these languages pushed north into modern Nubia, while others pushed eastward into the Khartoum region, replacing Meroitic with Nubian. After the Arab incursions, Nubian disappeared in the latter region and Arabic spread into the steppe lands.
Murdock,(31) Bentley and Crowfoot(32) and others(33) have commented on the extraordinary longevity of many cultural traits in the Sudan. Such continuities can be seen in pottery,34 personal ornaments,35 and body mutilations(36) and seem to be associated with an equal stability in population (physical type) and in language. In the Nile Valley this should not be correlated with the geographical stability of the Nubian languages as such, but with the stability of the broader Eastern Sudanic language group of which it is a member. Whether or not Meroitic is a member of this widely distributed language family, its speakers have played a leading role in the history of this part of Africa. Eastern Sudanic languages have been spoken by pastoral and in earlier times by hunting and gathering peoples and also by the most highly civilized peoples in the Sudan in medieval times. If Meroitic, too, is an Eastern Sudanic language, it was also spoken by the people who built the first civilization in Africa beyond the Mediterranean.
SUMMARY
Various theories have been proposed concerning the homeland of the Nubian languages. Current historical and linguistic evidence suggests that it was in Kordofan or Darfur and that Nubian first entered the middle reaches of the Nile between A.D. zoo and 500. There it replaced Meroitic, which appears to have been spoken in the area for a long time. Meroitic may, but has not yet been proved to be a language of the Eastern Sudanic group. If so, the persistence of many cultural traits in the Sudan may be correlated with the stability of the Eastern Sudanic-speakers in the same area.
BIBLIOGRAPHY
ADAMS, w. Y. An introductory classification of Meroitic pottery. Kush, xII, (I964), I26-73•
ARKELL, A. J. A History of the Sudan from Earliest Times to 1821 (London, I96I).
BENTLEY, o. and CROWFOOT, J. w. Nuba pots in the Gordon College. Sudan Notes Rec. vii (I924), I8-28.
CROWFOOT, J. w. Further notes on pottery. Sudan Notes Rec. VIII (1925), I25-36.
GREENBERG, J. x. Studies in African Linguistic Classification (Branford, I955).
GREENBERG, J. x. The languages of Africa. Int. J. Am. Ling. xxix, no. I (I963).
GRIFFITH, F. L. Meroitic inscriptions. In D. R. MacIver and C. L. Woolley, Areika(Philadelphia, I909).
GRIFFITH, F. L. Karanog, The 1lleroitic Inscriptions of Shablul and Karanog (Philadelphia, I9II).
GRIFFITx, F. L. Meroitic studies. J. Egypt. Archaeol. III (I9I6), 111-24.
HERZOG, R. Die Nubier (Berlin, I957)
HILLELSON, s. Nubian origins. Sudan Notes Rec. XIII (I930), I37-48
HINTZE, F. Die sprachliche Stellung des Meroitischen. Afrikanistische Studien (Berlin, I955).
KIRWAN, L. P. The Oxford University Excavations at Firka (Oxford, I939).
KRONENBERG, A. The thumb-ring: a modern parallel to a Meroittc object. Kush, x (I962), 336-7.
LEPSIUS, R. Briefe aus Agypten, fl`thiopien und Sinai (I852).
LEPSIUS, R. Nubische Grammatik (I880).
MacGAFFEY, w. The history of Negro migrations in the Sudan. SWest.J. Anthrop. XVII, 178--97
MAC MICHAEL, H. A. History of the Arabs in the Sudan, 2 vols. (Cambridge, I922).
Meinhof, c. Die Sprache von Meroe. Z. Eingeborenensprachen, XII, I-I6.
MURDOCk, G. P. Africa: Its Peoples and Their Culture History (New York, I959)
OLDEROGGE, D. A. and POTECHIN, I. I. Narody Afriki (MOSCOW, 1954).
SHINNIE, P. L. The fall of Meroe. Kush, III (I955), 82-5
TRIGGER, B. G. Meroitic and Eastern Sudanic: a linguistic relationship? Kush, XII (I964), 188-94
TRIGGER, B. G. History and Settlement in Lower Nubia (Yale University Publications in Anthropology, no. 69, I965).
ZYHLARZ, E. Zur Stellung des Darfur-Nubischen. Wien. Z. Kde Morgenlandes, XXXV (I928).
ZYHLARZ, E. Das meroitische Sprachproblem. Anthropos, xxv (I93o), 409-63
ZYHLARZ, E. Die Fiktion der `kuschitischen' Volker. Kush, Iv (I956), I9-33•
JJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ
1 Greenberg (iq55, iq63).
13 MacGaffey (tg6i).
14Greenberg, personal communication. Originally cited in Trigger (1965), 135-
1'Greenberg, personal communication. Originally cited in Trigger (1965), 135-
z Greenberg (iq63), 85. 3 MacMichael (igzz).
° Herzog (i957)• 5 Murdock (i959). I58•
fi Arkell (i961), 174. ' Ibid. i74-7.
e Zyhlarz (iq28). Discussed in Hillelson (i93o). ° Crowfoot (iq25).
1° 5hinnie (iqg5). '1 Kirwan (i939), 42•
v3 Griffith (tqoq, tqtt). z' Lepsius (t852, t88o).
~s Griffith (rqt6), 123. zo N'Ieinhof (tqzo).
z'Zyhlarz (1930, rq56). This was based on correspondences in a supposed gender system, personal prefixes to verbs, etc.
ze Olderogge and Potechin (rgg4). za Hintze (iq55).
"° Trigger (tq64). In a recent letter, P. L. Shinnie informs me he has independently ..'.sicned 1Teroitic to the larger Nilo-Saharan stock of which Eastern Sudanic is a member.
Is Ibid. rs Strabo, xv•n, t, 2.
" Arkell (196 t), 177, misinterprets this passage as suggesting that the Nubians were then living in their present location. In the same book, he supports both an east-west (t74) and a west-east (t78) migration.
18 Pliny, Nat. Hist. vi, x_xxv. zo Arkell (tq6t), 173.
zz Arkell (tq6t), 172, t7;.
la Ptolemy, Geography, IV, i48-83. zi Trigger (tq65)> r34-5
z Greenberg (iq63), 85. 3 MacMichael (igzz).
° Herzog (i957)• 5 Murdock (i959). I58•
fi Arkell (i961), 174. ' Ibid. i74-7.
e Zyhlarz (iq28). Discussed in Hillelson (i93o). ° Crowfoot (iq25).
1° 5hinnie (iqg5). '1 Kirwan (i939), 42•
la N. B. Millet, Some Notes on the Linguistic Background of Modern Nubian. Paper _read at a Symposium on Contemporary Nubia, Aswan, Egypt, 1964.
91 Murdock (i959), I5g• 3ZBentley and Crowfoot (i924).
3'Adams (tg64), i6q, yo. In addition some C-group pottery seems to be genetically related to that of the Khartoum Mesolithic even earlier.
34 See note 33. 3bKronenberg (iq62). 16 Murdock (ig59), I58•