The library of the School of Oriental and African Studies has a manuscript of some 300 folios, accessioned Hausa 98017. On one page it has the annotation in English, 'Histories of Samory and Babatu and others, written in Hausa about 1914 by Mallam Abu (who said he was with them) for Dr. J.F. Corson in the Northern Territories of the Gold Coast.' This was made by F.W. H. Migeod, one of those many colonial officers of a scholarly bent, who acquired the MS from Corson in 1926 and who shortly after added the comment, 'Very little value, 14/9/27.' My God, Migeod! MS 98017 is not only of intrinsic literary interest but is a source of considerable importance for the history of western Africa in the later nineteenth century.
A photocopy of MS 98017 was obtained by the Institute of African Studies, University of Ghana, in 1963, and accessioned IASAR/222. In 1964 I launched a scheme to translate into English various works of Ghana provenance, in Hausa language and Arabic script, which were in the Institute's collection. A.B. Moro and T. Mustapha, both born Hausa speakers, worked with me. In the course of this project MS 98017 was transcribed from Arabic into Roman script, and a provisional translation made of it. Then, in 1966, I left the Institute of African Studies and, sadly, Moro died. Plans for publication of the MS did not come to fruition.
In 1973 Stanislaw Pilaszewicz's Ph.D. from Warsaw University had MS 98017 as its subject. Now, two decades later, he has drawn upon this to give us the book under review. It contains a photographic reproduction of those parts of MS 98017 dealing with the activities of the Zabarma, together with his own transcription and English translation. A second volume on the Samorian sections is promised. Pilaszewicz was a fellow of the Institute of African Studies in 1973-74, but seems not to have located the earlier transcription and translation in its collections. There has thus been considerable duplication of effort.
The part of MS 98017 that concerns Samori and his campaigns consists of 184 pages of text. These are consecutively numbered 1 to 182, one folio having been missed. They comprise over 200 labaru (sing. labari), that is, 'stories' or, perhaps better in the context, 'episodes.' They are arranged chronologically and the last onewhich is reproduced in The Zabarma Conquest (henceforth ZC), p. 205, without explanationreports Samori's death in French captivity. The Samorian material is, then, in good shape and preserves the sequence of Malam Abu's text. The same cannot be said for the remainder of MS 98017.
The Zabarma material is in 76 folios. These comprise 136 pages of Malam Abu's text, and 16 pages that are blank or have nothing other than the first word or words of the next page of that text. The 76 folios contain 97 labaru (and not 87, see ZC, p. 12). All but one of these have to do with the activities of the Zabarma in the Middle Voltaic basin.  The exception concerns the Lobi, and cannot readily be located in either the Samorian or Zabarma corpus.
Migeod worked, but in a desultory manner, upon the Zabarma sections of MS 98017. This is testified by various comments, and attempts at translation, which he penned on a number of the blank pages and on additional sheets. He divided the text into five parts, and paginated each of these independently. In doing so he unfortunately destroyed the sequence of Malam Abu's folios. I find it astonishing that this is not recognized by Pilaszewicz, who preserves the disorder to which Migeod reduced the text, assumes that Malam Abu wrote on the Zabarma in (Migeod's) five 'chapters,' and assigns names to each. Pilaszewicz notes that these 'chapters' are not in chronological order, but suggests that Malam Abu (contrary to his practice with respect to Samori) presented the Zabarma material 'in a retrospective order' (ZC, 12). Making a virtue out of this, Pilaszewicz argues that to disturb the (dis)order of the MS as Migeod bequeathed it to SOAS would 'injure its literary continuity,' and 'harm the literary integrity of the narrative.' This is, in my opinion, to abandon the role of a serious editor.
Pilaszewicz reproduces the (disordered) folios on the Zabarma in ZC, pp. 204-125. I offer one preliminary example of the problems this creates. It does not require any great feat of textual reconstruction to see that the labari in ZC,168 left, which Pilaszewicz assigns to the beginning of 'chapter' III, follows directly on from that in ZC, 125 left, which ends 'chapter' V. Both belong to a cycle of five episodes having to do with relations between the Zabarma and Andani, ruler of the Savelugu division of Dagomba. The reader may wish to consult the relevant transcription, ZC, 67-9 and 55-6, and translation, ZC, 110-11 and 99-100. It will be readily apparent that to assign these episodes to different 'chapters,' and to have Malam Abu report the beginning of the conflict between the Zabarma and Andani many folios after he has narrated the end of that conflict, is nothing short of nonsense.
I believe that the importance of Malam Abu's work is such as to warrant a lengthy consideration of Pilaszewicz's edition. Ideally, Malam Abu's 'literary integrity' might best be rescued by presenting the reader with a restored text and translation. Space forbids any such exercise. Faute de mieux, I shall briefly outline the structure of Malam Abu's account of the Zabarma, showing that he did indeed present a sequential account of their activities in the Middle Voltaic basin over the last four decades of the nineteenth century.
I refer to Malam Abu's text, in what I believe to be its correct order, as MA/folios 1-76: recto (r) or verso (v). I establish a concordance between these and Pilaszewicz's photographic reproductions of the text by referring to the latter as ZC/pages 204-125: right (r) or left (l). Because of the way in which the text was photocopied, a right page in ZC/ will always be verso in MA/. For present purposes it is unnecessary to attempt to replicate, in Roman, the highly idiosyncratic system by means of which the Hausa is rendered in Arabic script. I shall, for convenience, follow the orthographical principles laid out in G.P. Bargery's A Hausa-English Dictionary and English-Hausa Vocabulary (London 1934, xxi).
MA/1:r [ZC/138:l] is a blank page. MA/1:v [ZC/137:r] commences with a descriptive title of the whole corpus: Labari Zabaramawa, 'The Story of the Zabarma.' It also has what seems to be a title for the following 28 labaru or episodes, which form the first opuscule in the corpus: Usulisu, 'Their Origin.' These are found in MA/1:v to 42:v, which correspond to ZC/137:r to 125:l, 168:l to 150:r, 148:l to 140:l, and 149:l. The narrative begins with the arrival of Alfa Hano, the scholar, and Gazari, the warrior, in the eastern Gonja trading town of Salaga, and details their campaigns in Dagomba and the Grunshi country. The death of Alfa Hano is reported on MA/10:r [ZC/129:l]. It occurred in or about 1870.  The death of Gazari in battle is noted on MA/38:v and 39:r [ZC/142:r and l]. We can be reasonably sure that this was in early 1878.  The narrative closes with the succession of Babatu to the Zabarma leadership (MA/42:v). 'The story of Gazari is ended,' Malam Abu commented; 'I greet the whiteman much. The story of Gazari is ended. The Sarkin Zongo [Zaghu, Hausa Zango] Musa, I greet much. The end [Arabic, tammat]. The matter is ended.'
In a second opuscule Malam Abu, logically and chronologically, proceeded to narrate the fortunes of the Zabarma under the leadership of Babatu. The account is entitled Ga wani labari Amiru Babatu daIssa, 'Here is the story of Babatu son of Issa.' It is contained in MA/43:r to 64:v [ZC/204:l to 182:r], and presents 51 episodes, the last of which reports Babatu's death in poverty in Dagomba. 'Amir Babatu, king of the world,' Malam Abu remarked; 'may God forgive him. This story of Babatu is also ended.'
The third opuscule in the corpus, MA/65:r to 75:r [ZC/180:l to 170:l], differs in character from the preceding. It contains 17 labaru describing 17 quarrels which arose between the Zabarma leaders in the times of both Gazari and Babatu. There is one remaining folio, MA/76. The lack of vocalization makes it difficult to read the recto [ZC/139:l], but we appear to have: Wanan labari ta tsaya. Labari Zabaramawa. Sariki. iara. The sense is, perhaps, 'This story is stopped. The Story of the Zabarma. The Rulers [reading sarakai for sariki]. The End.' We may assume that Malam Abu concluded the corpus of stories about the Zabarma with this comment, in which we may perhaps detect a certain sense of relief on his part. On the reverse of this last folio, MA/76:v [ZC/138:r] there appears the 'floating' story to do with the Lobi, which cannot be placed in any context in MS 98017.
I offer the following key for the reader wishing to follow Pilaszewicz's English translation of the Zabarma material in its corrected sequence. The page of ZC is given first, followed by the marginal cross-references to Malam Abu's text supplied by Pilaszewicz. The sequence is: ZC 108/3 to 111/17; 99/1 to 104-105/36; 105/2 to 107/18; 105/1; 72/3 to 93-94/46; 94/2 to 98/22; and 107/1. The fragment on the Lobi is on 108/2.
Having commented so unfavourably about the quality of Pilaszewicz's general editing, it affords no pleasure to have to say that the quality of the translation also leaves much to be desired. It is impossible in the space available systematically to draw attention to the many infelicities in the English text. There are some errors which are a result of sheer carelessness. The reader may look, for example, at ZC, 103 where talatin da bakwai is (correctly) translated as 'thirty-seven' and a few lines later as 'twenty-seven.' There are other errors which are quite inexplicable. For example, the text of Malam Abu, ZC/162:r, has labari Adani da Gazari wani ya iari kuma. This is translated, ZC, 101, as, 'this story of Andani, son of Gazari, is also ended,' thus making the ruler of the Dagomba division of Savelugu a son of the Zabarma leader! The correct reading is, 'this story of Andani and Gazari is also ended.' There is, however, a much graver problem than these examples indicate.
To work from Malam Abu's text to an English translation that satisfactorily conveys the sense of the original is a formidable task. The meaning of his writing is often obscure. It is sometimes unfathomable and frequently open to several different interpretations. Pilaszewicz seldom indicates this, and gives the reader a totally false sense of security in the precision of the translation. In illustration of this I shall arbitrarily take one passage from the opening of the first opuscule and one from the ending of the second. These are, in the reconstructed sequence of the text, on MA/1:v and MA/64:v.
On MA/1:v Malam Abu introduces Gazari into the narrative by means of a series of phrases in his praise. Pilaszewicz's transcription [ZC, 65] reads,
Gazari Takura miji shikafa, ruwa dari gama gari, Gazari mazaja mazaja. Gazari, maskika guda ya fi mashi dubu,and he translates this [ZC, 108],
Gazari Takura, man who owns rice, the wanderer, man of men. Gazari, one your bow is more [efficient] than a thousand of bows.Pilaszewicz takes takura as a proper noun, but none of the many informants with whom I talked about Gazari knew him as such. It means, they said, 'the old one,' and Hausa takura does indeed have the sense, 'bowed with age.' Mijin shinkafa means, literally, 'husband of rice.' Pilaszewicz's treatment of ruwan dare gama gari follows Abraham, who glosses the phrase, 'what a wanderer!'  But I have heard it used proverbially with something like the sense, 'rain at night moves over the sky.' Mazaja mazaja, though corrupt, probably does have the sense of 'man among men.' Mashika guda ya fi mashi dubu, however, surely has nothing to do with bows. Its meaning is clear: 'Your single spear surpasses a thousand spears.'
On MA/64:v Malam Abu refers to the decline in Babatu's fortunes. Pilaszewicz's transcription [ZC, 49] reads,
Turuwa Ikilshi suka yi kisa, suka raba su. Da sawura yakisa Babatu ya tafi Dakoba....This he translates [ZC, 94], 'The Europeans of English origin defeated [them] and separated them. With the rest of his troops Babatu went to Dagomba....'
But suka yi kissa surely means, 'they played a trick,' and probably refers to the independently documented intrigues of Campbell, commander of the British forces at the time.  Raba da is 'to separate from,' and hence I would read suka raba su'they [the English] cut off them [the Zabarma]'da saura yakisa, 'from the remainder of his [Babatu's] war.' By trickery, in other words, the British induced the Zabarma to stop fighting. Hence (as is well known), Babatu ya tafi Dakoba: 'Babatu went to Dagomba.'
I am not concerned here with matters of mistranslation as such; perhaps Pilaszewicz's renderings are preferable to mine. I am, however, concerned with his failure even to indicate that alternative readings of Malam Abu's text are often possible. He has given us what may best be described as a first stab at a translation, but he presents it as definitive.
Who was Malam Abu? In my study of Wa which appeared in 1989 I was cautious, remarking that his 'precise identity remains a mystery.'  Pilaszewicz writes that 'we have no information concerning the author of the manuscript' (ZC, 11), but rashly assumes that Malam Abu was a Hausa. He was not, and this seems an appropriate place further to discuss his identify.
J.F. Corson, born 1878, belonged to the Gold Coast Medical Service. He was on leave in the early part of 1914, but on 1 October 1914 took up an appointment as Medical Officer in Wa. He held this post until 10 May 1916, when he went on leave once again after which he was posted to Akuse in the Gold Coast Colony.  We may be sure, then, that it was in Wa, in the period late-1914 to early 1916, that Malam Abu worked on his history of Samori and the Zabarma.  Corson, we may note, had passed his examination in Lower Standard Hausa.
In encouraging Malam Abu to commit his historical reminiscences to writing, Corson appears to have established something of a tradition. Only a few years later, in 1922, the Wa District Commissioner P.J. Whittall presented Malam Isaka, then Wa Friday Imam, with a ledger 'on condition that he writes the History of the Walas in it in Hausa.' Malam Isaka brought together a series of texts, a number of which had to do with the activities of Samori and the Zaberma. On 3 August 1964 I had a long conversation with Malam Isaka's son, al-hajj Malik, and several other scholars who had known the Friday Imam. I asked about Malam Isaka's procedure. He collected his materials, they said, 'from other malams then alive, especially from 'Uthman Daleri of Yeri Nayiri and Malam Abu, also of Yeri Nayiri.'  In many other conversations, elderly Muslims in Wa insisted that they had heard of no other Malam Abu.
Malam Abu was, then, from the Wa yerihi, the people of Yeri Nayiri (of which ward, incidentally, 'Uthman Daleri was head). This old Muslim community originated with the movement of warbands eastwards at the time of the disintegration of imperial Mali.  It represents the Mande 'warrior' component in Wa society, as opposed to the scholars who constitute Wa Limamyiri. Wa yerihi participated in the various militant Islamic movements of the mid- to late-nineteenth century, including the jihad of Mahmud Karantaw,  and there is no reason to doubt Malam Abu's claim, that he had himself been on campaign.
Pilaszewicz attempts to locate Malam Abu within the tradition of Muslim scholarship which developed in the Voltaic basin in the 18th and 19th centuries, but there are problems with this. The reader will note the absence of doxology from Malam Abu's text, and of those many tokens of devoutness which usually grace the writings of the well-schooled writer. Even the death of Alfa Hano, the man of prayer, is simply recorded, ya rasu, 'he died,' with no expression of concern for the future of his soul. Malam Abu's frequent references to Babatu as 'king of the world' seem unlikely to have come from the pen of one versed in Islamic literary conventions. He appears to belong more to the world of the griot and the praise singer than to that of the 'alim. Famous warriors, and their knavery and cunning no less than their courage and ferocity, are his themes. And finally, in this context, should we assume with Pilaszewicz that Malam Abu not only narrated the labaru, the stories, but also himself committed them to writing? Hausa has come into wide use in Wa, as a vehicle of literary expression, in this century, but Malam Abu may well have availed himself of the services of a scribe or scribes. We simply do not know.
Attention may briefly be drawn to the fact, to which Pilaszewicz makes no reference, that there are five episodes in the Samorian section of MS 98017 which have to do with the relations between Samori's son, Sarankye Mori, and Babatu. These are on the manuscript pages 169-172. They are closely related to four episodes in the Zabarma text, ZC/185:l to 185:l. A comparison of these, which cover much the same ground, is illuminating. In both parts of MS 98017 alike, Malam Abu commends Babatu for his cunning and pokes fun at Sarankye Mori for his foolishness. This may tell us something about Malam Abu's allegiances, but I shall refrain from considering the implications here.
The Zabarma Conquest of north-west Ghana and Upper Volta (why 'Ghana' but not 'Burkina Faso'?) is, then, a highly unsatisfactory work. The scholar will have to await a reliable version of Malam Abu's narrative. Unfortunately it may be a long wait. The appearance of Pilaszewicz's seriously flawed study will none the less deter even the most committed publisher from hastening to sponsor the definitive edition so greatly needed.
2. Holden, 'The Zabarima Conquest', 66. [*]
3. Wilks, Wa and the Wala, 105-6. [*]
4. R.C. Abraham and Malam Mai Kano, Dictionary of the Hausa Language, Crown Agents 1949, 749. [*]
5. Holden, 'The Zabarima Conquest', 84. [*]
6. Wilks, Wa and the Wala, 106. [*]
7. The Gold Coast Civil Service List, Accra 1917, 169. [*]
8. Wilks, Wa and the Wala, 36, 103-8, 122-4. [*]
9. Wilks, FN/112. Copies of these fieldnotes are deposited in the archives of the Africana Library, Northwestern University. [*]
10. Wilks, Wa and the Wala, 53-9. [*]
11. Wilks, Wa and the Wala, 100-5. [*]
12. Compare Wilks, Wa and the Wala, 32. [*]
© The author and Sudanic Africa. Archived 8.4.95