History and Theory 42 (February 2003), 61-81

Wesleyan University 2003 ISSN: 0018-2656

FORUM ON TRANSLATION AND HISTORIOGRAPHY

2.

TRANSLATION AND THE COLONIAL IMAGINARY:

IBN KHALDN ORIENTALIST

ABDELMAJID HANNOUM
No translation would be possible if its ultimate essence strove for likeness to the original.
Walter Benjamin
Imagining is rst and foremost restructuring semantic elds.
Paul Ricoeur
ABSTRACT

Despite the increasing interest in translation in the last two decades, there has been no investigation of the translation of historiography and its transformation from one language to
another. This article takes as a case study the translation into French of Ibn Khaldn, the
fourteenth-century North African historian. It considers specically the translation done by
William de Slane in the context of the colonization of Algeria. The Histoire des Berbres,
the French narrative of Ibn Khaldn that relates to the history of Arabs and Berbers in the
Maghreb, has become since then the source of French knowledge of North Africa. It is upon
that French narrative that colonial and post-colonial historians have constructed their
knowledge of North Africa, of Arabs, and of Berbers. The article shows how a portion of
the writing of Ibn Khaldn was translated and transformed in the process in such a way as
to become a French narrative with colonial categories specic to the nineteenth century.
Using a semiotic approach and analyzing both the French text and its original, the article
shows how colonialism introduced what Castoriadis calls an imaginary by transforming
local knowledge and converting it into colonial knowledge. In showing this the essay
reveals that not only is translation not the transmission of a message from one language to
another, it is indeed the production of a new text. For translation is itself the product of an
imaginary, a creationin Ricoeurs words, a restructuring of semantic elds.
I. TRANSLATION AND THE IMAGINARY

Translation was a part of the whole enterprise that the early colonial administration in Algeria set in place, an enterprise that made knowledge indispensable for
colonization. In order to know the natives, one had not only to observe them,
study them, and understand their culture and their society, but also to know their
past. The present was believed to be out there, to be apprehended by observation; the past was assumed to be recorded in documents, to be grasped only by a
work of translation, either direct or indirect.
The ofcers of the Arab Bureau, the military institution that assured both the
production of knowledge and the establishment of order, were mainly ethnographers, with little concern for history.1 The later interest in history among the
scholars of the civilian regime was an attempt to go beyond the Arab Bureaus
1. A. Hannoum, Colonialism and Knowledge in Algeria: The Archives of the Arab Bureau,
History and Anthropology 12, no. 4 (2001), 343-379.

62

ABDELMAJID HANNOUM

politics, which were seen as too sympathetic to the Arabs. In fact, now that the
opposition between Arabs and Berbers had been established and had gained the
self-evidence of truth, one had to explain it. To explain it meant mainly to seek
its origin and its development, and above all to establish the extent of what was
Arab and what was Berber in colonial Algeria. Only historical research would be
able to respond to the challenge. To do this, the scholars of the civilian regime
were faced rst with a search for documents. Usable documents already existed
at that time. The Exploration Scientique de lAlgrie (the Scientic Exploration
of Algeria) translated Arab historiography on a small scale in the rst decade
after the conquest of Algiers. Of the thirty-nine published volumes dedicated to
the study of Algeria, three volumes were translations of Arabic works.2 However,
Ibn Khaldn, the fourteenth-century Maghrebi historian, was not translated by
the members of the Exploration Scientique, but he was extensively used by one
of its scholars, Ernest Carette, who was the rst to trace the history of the tribes
and to reveal the locations of Arabs and Berbers in colonial Algeria. This is to
say that Ibn Khaldn was already a subject of indirect translation in the context
of the Exploration Scientique.3
However, it is the translation by William de Slane of a fragment of Ibn
Khaldn related to the history of the Berbers that might well be considered the
greatest textual event in the history of French Orientalism.4 De Slanes translation of Ibn Khaldn, under the title Histoire des Berbres, formed the foundation
of French historical knowledge of North Africa. It is this translation and its
effects that I wish to explore in this article. In particular, I wish to reveal the
mechanisms and the politics by which a fragment of the `Ibar of Ibn Khaldn
was discovered by French Orientalism and converted into a colonial text with
colonial categories. It is this text that became central in French colonial historiography of North Africa, a text upon which all colonial historians built their
work. This same colonial historiography became, from 1930 onwards, the foundation of national historiography in North Africa.
The article also has a theoretical dimension, for it attempts to show not only
how knowledge is regulated by power, but also how colonialism introduced and
established a specic imaginary by transforming local knowledge and converting it into colonial knowledge. Let me now dene what I mean by the imaginary.
I use the concept in the rich sense given to it by Cornelius Castoriadis. The
imaginary about which I talk, he says, is not an image of [something]. It is an
2. Those are related to history, geography, and law: Qayrawani, Histoire de lAfrique, transl. E.
Pellissier and Ch. Remusat (Paris: Imprimerie Impriale, 1845); Al-Aiachi-Moula-Ahmed, Voyages
dans le sud de lAlgrie et des Etats barbaresques de lOuest et de lEst, transl. Louis Andrien
Berbrugger (Paris: Imprimerie Impriale, 1846); Khalil B. Ishaq, Prcis de jurisprudence musulmane:
ou, Principes de lgislation musulmane civile et religieuse selon le rite mlkite, transl. Nicolas
Perron (Paris: Imprimerie Impriale, 18481852).
3. E. Carette, Recherches sur lorigine et les migrations des principales tribus de lAfrique septentrionale et particulirement de lAlgrie (Paris: Imprimerie Impriale, 1853); E. Carette, Etudes sur
la Kabilie, 2 vols. (Paris: Imprimerie Impriale, 1848).
4. Ibn Khaldn, Histoire des Berbres, transl. William de Slane, 4 vols. (Algiers: Imprimerie du
Gouvernement, 18521856).

TRANSLATION AND THE COLONIAL IMAGINARY

63

incessant and essentially determined creation (sociohistorical and psychic) of gures, forms/images, from which it must be only a question of something. What
we call reality and rationality are its products.5 The imaginary refers both to
the product of imagination (our worlds made up of systems of meanings) and to
activity, the ability by which we create a system of meaning that we identify as
our world. Anthropologists have already explored the domain of the imaginary
in a certain way. They consider knowledge as a part of that complex whole that
Tylor calls culture, and culture is a system of meaning, the product of the function that Clifford Geertz calls a program. As Geertz puts it: Culture is best
seen not as complexes of concrete behavior patternscustoms, usages, traditions, habits, clustersas has, by and large, been the case up to now, but as a set
of control mechanismsplans, recipes, rules, instructions (what computer engineers call programs)for governing behavior.6
The imaginary is a function of producing meanings, and it is the product of
this function. However, unlike Castoriadis, I maintain that the imaginary does
not create ex nihilo, but rather operates within systems of meanings, transforming them in such a way as to create new meanings out of old ones. Knowledge,
in all its forms is in fact a product of imagination. It is widely argued today that
knowledge is regulated by power; however, its analysis cannot be reduced to that
level. Knowledge was not only the means by and through which colonialism
governed. Knowledge is also regulated by the power of the mental structure that
produces it. Its function went beyond knowing the natives in order to govern
them; it assured colonial domination even long after the collapse of the colonial
enterprise. Colonial knowledge shaped postcolonial identities; it introduced
colonial categories and institutions that outlived colonialism. Indeed, colonialism
produced the knowledge by and through which it governed. It also transformed
the product of imagination and, in fact more importantly, the domain, the function of imagination. The local culture, the webs of signicance, has not only
been transformed (the natives have not only been caught in webs of signicance
that they have not spun), but their native function (program) has been transformed in such a way as to continue producing similar webs of signicance in
the absence of colonial scholars or their collaboration. In short, colonialism not
only produced a body of knowledge, but it also implemented the colonial function of the production of knowledge, and, by extension, of that culture that
assured the production of colonial webs of signicance. This function was implemented not only by the existence of a powerful discursive formation, but also by
colonial institutions that were themselves the product of the imaginary. For society as a whole is the product of the imaginary insofar as it is a system of interpretation. Castoriadis writes:
It would be supercial and insufcient to say society contains a system of interpretation
of the world. Each society is a system of interpretation of the world, and again, the term
interpretation is at and inappropriate here. Each society is a construction, a creation of
5. Cornelius Castoriadis, Linstitution et limaginaire (Paris: Seuil, 1975), 8.
6. Clifford Geertz, The Interpretation of Cultures (New York: Basic Books, 1973), 44.

64

ABDELMAJID HANNOUM

a world, of its own world. Its own identity is nothing but this system of interpretation,
this world it creates. And that is why (like every individual) it perceives any attack as a
mortal threat, as an attack upon its identity, upon itself.7

Society is a system of meaning; it constitutes who we are today. In order to

answer the Kantian question of who are we today? we in fact should rst
attempt to answer the implicit question of who were we yesterday? Or to be
more precise, one has to nd out how we have become who we are today. The
present is made up of the past, and the past itself is our present. Hence the fundamental importance of historical knowledge in human societies. Thus, a system
of meaning is always historical; it always draws from the past, or as Ernst
Cassirer put it: Historical knowledge is the answer to denite questions, an
answer which must be given by the past; but the questions themselves are put and
dictated by the presentby our present intellectual interests and our present
moral and social needs.8
However, the past is available to us mainly through documents, that is, through
narratives that, however close to the event they may be, are themselves representations of the real, and not the real itself. Therefore, historians do not nd
answers in the past, but they make them from documents of the past, that is, from
answers given by others. In other words, the past does not provide answers to
questions of the present, but historians do, using other representations of the
past.9 Further, historians are the ofcial producers of meaning in a society; their
knowledge is not only informative about things past, but it constitutes and shapes
identities and determines relations in the present. Identities, whether collective or
individual, are by denition historical. History, Merleau-Ponty says, is a
strange object, an object which is ourselves.10 However, historians are not solitary people busy looking at documents of the past and making meaning out of
them; their function takes various forms in different social institutions: schools,
the family, the media, the press, the church. The social is, then, historical, and the
historical is everywhere; it is not only a realm, but also the whole of reality.11
To say with Croce that all history is contemporary history is in fact to say that
above and beyond the realm of history there is no other realm of being.12
All of this means that history involves the translation of (documents from) the
past into the language of the present. But what is translation? A central aim of
this essay is to answer this question. A number of theorists have already shown
that translation is not the reproduction of a foreign text in ones own text; it is not
7. Cornelius Castoriadis, The Imaginary: Creation in the Social-Historical Domain, in Disorder
and Order: Proceedings of the Stanford International Symposium (Sept. 1416, 1981), ed. Paisley
Livingston (Saratoga, Calif.: Anma Libri, 1984), 152. For the French text, see Limaginaire: La cration dans le domaine social-historique in Domaines de lhomme (Paris: Seuil, 1977), 226-227.
8. Ernst Cassirer, An Essay on Man (New Haven: Yale University Press, 1944).
9. See also R. G. Collingwood, for whom history is re-enactment of past experience in the mind
of the historian, The Idea of History (Oxford: Clarendon Press: 1946), 282.
10. Maurice Merleau-Ponty, The Crisis of Understanding, in Adventures of the Dialectic, transl.
Joseph Bien (Evanston: Northwestern University Press: 1973), 11.
11. Cassirer, An Essay on Man, 178.
12. Ibid.

TRANSLATION AND THE COLONIAL IMAGINARY

65

a transmission, faithful or not, of a message from one language to another; it is

not even a betrayal of the initial message. Translation is a domestication of a
foreign text. Lawrence Venutis remark about literary translation is also valid in
the case of historiography: Not only does a translation constitute an interpretation of the foreign text, varying with different cultural situations at different historical moments, but canons of accuracy are articulated and applied in the
domestic culture and therefore are basically ethnocentric, no matter how seemingly faithful, no matter how linguistically correct.13 However, I go beyond this
to argue that a translated text is not only an interpretation, it is also the production of a new text. For translation is itself the product of an imaginary, a creationin Ricoeurs words, a restructuring of semantic elds.14 The foreign
text serves as the raw material, almost an excuse, for the production of another
text, a text that emerges within a specic discursive formation, with its own rules
and constraints. For while the foreign text is the expression of an imaginary
structure, which is the product of a historical moment, the translated text is the
expression of an imaginary structure that is also the product of a historical
moment. Maybe this is why Berman notes that a translator without a historical
consciousness remains a prisoner of his representation of translating and to those
representations that convey the social discourses of the moment.15
This article, through its investigation of translation and the imaginary, serves
to explain the relation of a colonial present to another pastspecically, exploring what questions a colonial present produced and how it answered them using
the past of another culture. For, as will become clear, the colonial questions and
answers are regulated by a European epistemologya way of knowing made up
of certain categoriesspecic to the nineteenth century, and the past as written
by the natives contains another epistemology, another way of knowing relative
to a specic time and culturethe Arabic culture of the fourteenth century. By
examining an Arabic text, its translation, and its subsequent interpretations, the
article sheds light on what Michel Foucault would call the conditions of emergence of the colonial discourse in the form of a translation.
Let us now see how colonial scholars discovered Ibn Khaldn, and discuss the
way in which that discovery has been the object of a whole discourse, inseparable from the French translated version of Ibn Khaldn.
II. DISCOVERING IBN KHALDN

Ibn Khaldns history was linked to the discovery of Islamic manuscripts that
Silvestre de Sacy undertook with an uncanny vitality and vigor. One of de Sacys
achievements was that he put before the profession [Orientalism] an entire sys13. Lawrence Venuti, The Scandals of Translation: Towards an Ethic of Difference (New York:
Routledge, 1998), 81-82.
14. Paul Ricoeur, Imagination in Discourse and in Action, in Rethinking Imagination, ed. Gillian
Robinson and John Rundell (New York: Routledge, 1994).
15. A. Berman, Pour une critique des traductions: John Donne (Paris: Gallimard, 1995). Quoted
in Venuti, The Scandals of Translation, 84.

66

ABDELMAJID HANNOUM

tematic body of texts, a pedagogic practice, a scholarly tradition, and an important link between Oriental scholarship and public policy.16 Ibn Khaldns work
or fragments of his work known in 180817was an important part of that body
of texts. In 1818, de Sacy wrote again, this time to give an impression of the
whole book, the `Ibar; he straightforwardly stated that among the historical
books written in Arabic there is none that deserves more the honor of publication
than this one.18 Soon afterward, frequent attempts were made by de Sacys students to make Ibn Khaldn known. Thus in 1822 Joseph de Hammer published in
the Journal Asiatique a summary of the book and insisted on the importance of
translating it.19 Two years later, Garcin de Tassy presented another review.20 F. E.
Schulz gave a review of the whole book the following year, in the same journal.21
What sort of book is this history by Ibn Khaldn?22 He undertook his project
about the history of the Maghreb soon after he became dissatised with politics.
He then retreated to the castle of Salama in what is today Algeria, a place that has
since vanished from sight. It was there that he wrote the rst part of his project,
known as the Muqaddima, and most of the second part related to the history of the
Berbers. Having found that he needed some materials to verify or complete his
manuscript, Ibn Khaldn then moved to Tunis, where he ultimately offered the
rst copy to the Hafsid Sultan, Abu al-`Abbs. Thus it seems that the initial project was complete, for Ibn Khaldn deliberately chose not to write about the
Machrek for lack of materials. This, however, changed once he settled in Egypt. It
is there that the part concerning the history of the Persians and the Romans was
written, and it is also there that the last part, concerning his own life, more political and intellectual, was composed. Thus, the book is comprised of three parts: the
Muqaddima; the history of the Berbers and the Arabs, the Persians, the Romans
and other peoples; and nally the part usually considered a biography.23
Ibn Khaldn is thus a syncretism of authors. As regards the Muqaddima, he is
an epistemic subject, who was not only able to evaluate the eld of historiography in his time, but also, especially, to elaborate a rigorous way of writing history, a way that was deemed rational, modern, and in full conformity with the
European practice of history in the age of positivism. However, Ibn Khaldn is
another author when it comes to his historiography proper, in which he failed to
follow his own preaching. He is a researcher, a practitioner of historiography, who
16. Edward Said, Orientalism (New York, Vintage Books, 1979), 124.
17. Silvestre de Sacy, Chrestomathie Arabe, 3 vols. (Paris: Imprimerie Royale, 18261827), II,
387, 393-401. Also de Sacy, Relation de lEgypte par Abd-Allatif, (Paris: Treuttel et Wrtz, 1810),
509.
18. De Sacy, Ibn Khaldn in Biographie universelle (Paris: Michaud, 1818), t. 21, 154.
19. Joseph de Hammer, Sur lintroduction la connaissance de lhistoire, clbre ouvrage dIbn
Khaldn, Journal Asiatique 1 (1822), 267-278.
20. Garcin de Tassy, Supplment la notice de M. de Hammer, Journal Asiatique 4 (1824), 158161.
21. F. E. Schulz, Sur le grand ouvrage historique et critique dIbn Khaldn, Journal Asiatique 7
(1825), 213-300.
22. Ibn Khaldn himself narrates the history of the composition of his book in the last part, entitled al-ta`rif bi ibn Khaldn.
23. A. Cheddadi, Ibn Khaldn revisit (Casablanca: Les Editions Toubkal, 1999).

TRANSLATION AND THE COLONIAL IMAGINARY

67

is no different from the previous historians whom he criticizes in the Muqaddima

for their lack of a critical approach (tamhs, nazar).24 Later, even Ibn Khaldn the
practitioner of historiography was divided into two: one is the author of the portion related to the history of North Africa, who not only brought new knowledge,
an ethnographic history as the colonial expression goes, but who also succeeded in applying, up to a point, the epistemological remarks elaborated in the
Muqaddima.25 As for the author of the history of the Orient, he is usually considered less competent, less original than the authors of the two other parts. The autobiography has not drawn any attention despite the recent interest in the genre.26
When de Sacy drew attention to the need to translate the Muqaddima, he also
suggested its French title: Prolgomnes historiques, a title that later would
become the denitive title of the French translation by William de Slane, a student of de Sacy. De Hammer, on the other hand, insisted on the urgency of translating the part related to the history of the Berbers, that is, the second book. In
1847, de Slane, the principal interpreter of the Arme dAfrique, published,
under the auspices of the Imprimerie du Gouvernement (the Government Press),
the Arabic text related to the history of the Berbers.27 In 1854, he published the
translation under the title Histoire des Berbres. In 1858, Quatremre published
the Arabic text of the Muqaddima.28 De Slane published his translation of the
Muqaddima in 1863 under the title Prolgomnes. The two texts have had different trajectories: the Muqaddima was used fruitfully by sociologists such as
Emile Durkheim29 and Robert Montagne,30 whereas the Histoire des Berbres
served as the framework for the history of North Africa. In fact, it was published,
signicantly enough, by the Ministry of War, and its importance for the colonial
enterprise was immediately noted. Indeed, Joseph Reinaud wrote: The History
of the Berbers of Ibn Khaldn could not fail to draw the attention of the French
government. With the establishment of the French in Algeria daily relations, contacts of friendships and war have been established between them and the tribes
that occupy the interior of the country.31
24. This view was rst expressed by de Sacy and reproduced in the Arab discourse of Ibn Khaldn;
see, for instance, Taha Hussein, Etude analytique de la philosophie sociale dIbn Khaldn (Paris:
Vrin, 1919). It was conrmed by most of the subsequent interpreters of Ibn Khaldn.
25. Such is the view of Robert Brunschvig, La Bebrie orientale sous les Hafsides (Paris: AdrienMaisonneuve, 19401947), vol. 1.
26. An example of this is the collective work Interpreting the Self: Autobiography in the Arabic
Literary Tradition, ed. Dwight Reynolds (Berkeley: University of California Press, 2001). It contains
no treatment of the autobiography of Ibn Khaldn.
27. Ibn Khaldn, Histoire des Berbres, Arab text (Paris: Imprimerie du Gouvernement, 1847).
28. Ibn Khaldn, Muqaddima, in Notices et extraits des mss. de la Bibliothque impriale, vols.
XVI-XVIII (Paris: Bibliothque impriale, 1858). Translated by de Slane, Prolgomnes, in ibid.,
vols. XIX-XXI.
29. See Ernest Gellner, Muslim Society (Cambridge, Eng.: Cambridge University Press, 1981).
However, Gellner does not investigate the relation between Ibn Khaldns thought and Durkheims.
In addition to the fact that the work of Ibn Khaldn was available by mid-nineteenth century,
Durkheim also had an Egyptian doctoral candidate, Taha Hussein, working on the Muqaddima.
30. See A. Hannoum, Lauteur comme authorit en ethnographie coloniale, in La sociologie
musulmane de Robert Montagne (Paris: Maisonneuve et Larose, 2000), 249-264.
31. Joseph Reinaud, Ibn Khaldn, in Nouvelle biographie gnrale (Paris: Firmin Didot, 1854),
vol. 25, 746.

68

ABDELMAJID HANNOUM

Ibn Khaldn as a colonial author was, therefore, born in fragmentsArabic

texts for the purpose of pedagogy; summaries of the entire book addressed to
draw attention to his importance to the colonial enterprise; but also, equally
importantly, reviews of the book, entries in bibliographies, and statements about
the man, his age, and his culture. In short, a discourse on him, not of him,
emerged that cannot be separated from the translation. In other words, the analysis of Ibn Khaldns translated text, what Ibn Khaldn says in French, should go
hand in hand with what is said about Ibn Khaldn. He thus appears in this discourse as a genius, skillful, and also solitary, a man who contrasts sharply
with his own milieu. Further, the discourse on Ibn Khaldn is made up of two
diametrically opposed statements: on the one hand he is a modern and, therefore,
represents Europe with its rationality and its historicity; on the other hand, being
what he is, he contrasts with the civilization from which he came. While he is
enlightened, it is dark; while he is rational, it is obscurantist; while he is excellent, a genius, it is poor and mediocre. (How is this possible? He may not be from
that civilization. He may be European!32) And nally, while he is ignored in his
own country by his own people, he is discovered and given his due respect in
Europe.33 The colonial discovery of Ibn Khaldn has thus become a discovery
not only for Orientalism, but even more importantly for his own peoplehe is
the genius who has long been ignored by his people, wrote Jacques Berque.34
To discover Ibn Khaldn means above all that the French scholar is a competent epistemic subject, able to evaluate and sanction an intellectual activity. This
is a competence that is lacking in Arab and Islamic traditions for they have failed
to recognize the genius of Ibn Khaldn. The supposed Orientalist discovery of
Ibn Khaldn thus legitimizes the French interpretation of Ibn Khaldn. Indeed,
despite the fact that in social science it has become evident that even a genius
is the product of a historical, social, and intellectual moment,35 this discourse
continued to be produced not only by French scholars in the postcolonial period,36 but also by natives and post-natives. For instance, when a French
Orientalist repeats the ide reue and says that Ibn Khaldn is a solitary mind
that arose all of a sudden to explain the profound causes of history, and that it
is no wonder that this mind did not have an echo in a world that was not prepared
to receive him,37 a native scholar echoes the judgment even more bluntly, using
the language of the Enlightenment to mark his afliation to it and his own distance from the civilization of Ibn Khaldn: Ibn Khaldns star shines the more
brightly by contrast with the foil darkness against which it ashes out. . . . He is
32. Emile-Felix Gautier, Les sicles obscurs (Paris: Payot, 1927 ); F. Rosenthal, The Muqaddima
(Princeton: Princeton University Press, 1966).
33. Jacques Berque, Lintrieur du Maghreb (Paris: Gallimard, 1978), 13.
34. Ibid.
35. Especially in the work of Michel Foucault, The Archeology of Knowledge, transl. A. M.
Sheridan Smith (New York: Pantheon Books, 1972).
36. Regis Blachre, Place dIbn Khaldun dans lhumanisme Arabo-islamique, Revue des Deux
Mondes (October 1972), 70-79; Berque, Lintrieur du Maghreb; Yves Lacoste, Ibn Khaldun: The
Birth of History and the Past of the Third World (London: Verso, 1984).
37. Blachre, Place dIbn Khaldun dans lhumanisme Arabo-islamique, 79.

TRANSLATION AND THE COLONIAL IMAGINARY

69

indeed the one outstanding personality in the history of a civilization whose

social life on the whole was solitary, poor, nasty, brutish, and short.38
Colonizing the imaginary also means that colonial judgments, whose birth and
formulation are linked to a specic context, continue to be reproduced, passed on
uncritically, even in the postcolonial period, carrying with them their original
myths, and perpetuating colonial relations.
III. THE TRICKERY OF THE INTRODUCTION

De Slane precedes his translation of Ibn Khaldns text related to the history of
the Berbers by an Introduction in which he states that
The task of the translator is not limited to the exact reproduction of ideas uttered in the
text that is the subject of his translation. There are other obligations as well. He should
rectify the errors of the author, clarify the passages that offer some obscurities, provide
ideas that lead to the perfect understanding of the narrative and give the necessary assistance to make the book better understood.39

The translator is not only required to render ideas in one language in another,
but also to add information necessary to make the translated work comprehensible. In other words, the translator is required, by the act of translation, to add his
own voice to that of the author. His is in part a rectifying voice, without which
not only could errors not be corrected, but the text itself could not be understood.
For the translator contests dates, events, and interpretations, he reveals deciencies, and provides information. Thus, the translator not only transmits a
message from one language to another, but he also recties the errors of the message. In the colonial context this means that the translator converts the original
text into a colonial one, for correction implies an inadequacy in the cognitive universe of the author; the correction creates an adequacy, but at the expense of the
author, whose voice, whose cognitive universe, has been changed to conform to
the cognitive universe of the translator.40
In the Introduction, de Slane maintains that despite its richness, the Arabic text
of Ibn Khaldn is obscure, and when it is not, it is verbose and repetitive. In
fact, he continues, it is a simple draft, very badly written.41 Hence, the task
of the translator is both to clarify what is obscure and eliminate what is repetitive. While the translator sees his or her task as the reproduction of the same text
in another language, leaving aside whether this is indeed possible, what he or she
in fact does is construct another text that is alien to the original text, but is
nonetheless part of it.
What function does the Introduction fulll? An introduction usually takes the
form of a preface which is a metadiscourse; it is written a posteriori, and is usu38. Charles Issawi, An Arab Philosophy of History: Selections (Princeton: Darwin Press, 1987), ix.
39. William de Slane, Histoire des Berbres et des dynasties musulmanes de lAfrique septentrionale (Algiers: Imprimerie du Gouvernement, 18521856), i.
40. A.-J. Greimas, Le savoir et le croire: un seul univers cognitif, in Du Sens II (Paris: Seuil,
1983), 115-133.
41. De Slane, Histoire des Berbres, lxix.

70

ABDELMAJID HANNOUM

ally a reection of the author on his own discourse. As Algirdas-Julien Greimas

wrote:
The preface is not a part of the corpus. Along the temporal axis, it is a post-face and follows on the discourse of the research and its being put into writing. Its status is that of a
metadiscursive thought concerning a discourse that has already been produced. This
metadiscourse is supposed to reveal what the author himself thinks of his own discourse,
its goal, and its organization.42

The Introduction to the translation of Ibn Khaldn is not only a reection on

the translated text, its composition, and its theses. It is not only a reection on
the task of the translator, his cognitive endeavor, and his translation. The
Introduction to the translated text of Ibn Khaldn is also a discursive strategy to
determine the reading of the translated text. For de Slane suggests an interpretation that in fact conditions any reading of his translated text. His Introduction is
also composed of a summary of the translated text. This summary is in fact a
means, a trap; it makes the reader perceive the text in a certain way. This is a cognitive manipulation to make the reader understand the translated text as interpreted by the translator.
However, de Slane rst gives a general description of the whole book by Ibn
Khaldn, the `Ibar, whose second part is the object of his translation. Thus, one
nds the familiar discourse on Ibn Khaldn as being a genius and yet attached
to superstition.43 De Slane then proceeds to give a general view of the translated book, the Histoire des Berbres, which, he says, is made up of the fourth
section of the `Ibar, a section that narrates the history of the Arabs in North
Africa; he also says it is made up of the third book of the `Ibar, which contains
a genealogical history of the great indigenous race that dominated for centuries.44 To make the reader understand the chronology of the history of North
Africa, the translator provides a linear narrative of the different empires that succeeded one another in North Africa, from the Arab conquest of the seventh century to the fourteenth century, that is, to the time of Ibn Khaldn.
De Slanes narrative rst describes a situation of lack. According to him, sectarian divisions, indigenous revolts, and a Vandal conquest weakened Roman
North Africa before the Arab conquest. The Latin population had abandoned
their possessions to the Berbers and moved to the coast. The Vandals delivered a
fatal blow to Byzantine rule in North Africa, and despite the fact that Belisarius
and Salomon re-established Byzantine rule, North Africa became a fragile land,
easy prey for a religiously unied and aggressive adversary coming from the
OrientArabia. To make things worse, the Goths of Spain took over Tingitana,
thus creating a more divided North Africa: one part, mainly Carthage, still loyal
to the Empire; the other part, the province of Tingitana, under the Goths yoke.
This is to say that North Africa was weakened and divided as a result of a
42. Algirdas-Julien Greimas, On Chance Occurences in What We Call the Human Sciences, in
On Meaning, transl. Paul J. Perron and Frank H. Collins, foreword by Fredric Jameson (London:
Pinter, 1987).
43. De Slane, Histoire des Berbres, iii.
44. Ibid., viii.

TRANSLATION AND THE COLONIAL IMAGINARY

71

Christian heresy and a European conict. Fragile and fragmented, it was soon
faced with the presence of the Arabs, unied by their new religionIslam. One
can see, then, that the text opposes two entities. One is represented by Arab
Muslims, that is, by the Orient; the other is represented by a Christian Europe
whose territory extends to, and includes, North Africa.
In any case, it is this situation of lack that made the conquest of North Africa
possible. North Africa was taken away from the Romans. But what about the
Berbers? They appear rst as the indigenous population that had revolted, thus
participating in the weakening of the empire, and then as servants of Roman
lords who were liberated by the Arabs. However, their lives did not improve.
Soon afterward, they joined the Romans against the Arabs who had burdened
them by religious duties and heavier taxes, taking half of their harvest. This
alliance between Berbers and Romans resulted not only in a joint victory against
the Arabs, but more importantly in the formation of the rst Berber Empire.
The actors in de Slanes narrative are all distinguished from the outset: on the one
hand there are the Romans, on the other the Arabs, and in between there are the
Berbers. The relation between the rst and the last is perfectly clear, for it is one
of opposition. The Arabs were at war against the Romans; not only did they want
the land, but they also wanted to impose their religion. The Romans, on the other
hand, were the masters of North Africa, weakened yet defending their possessions. However, it is the relation between the Berbers and the Arabs on the one
hand, and the Berbers and the Romans on the other, that seems unclear. The
Berbers revolted against the Romans and joined the Arabs only because they
were burdened by imperial taxes. Likewise, they initially joined the Arabs and
then revolted against them only because of a misunderstanding. For they rst saw
the Arabs as liberators, but opposed them once they realized the Arabs were like
the Romans, if not worse. In other words, once they came to know the Arabs, the
Berbers deliberately chose the Romans. Indeed, they defeated the Arabs. A
Berber kingdom ensued under the leadership of Kusayla. This was only transitory, however, for the Arabs returned and crushed the Berbers. This is the beginning of what de Slane calls the age of Arab domination. It is also the beginning
of an opposition between the Arab nation and the Berber race that has always
been impatient under the foreign yoke.45 To show that his narrative is a summary of the translated work, de Slane asks his readers to consult such and such
chapters for more details about the events he has just summarized.
This narrative became fundamental in French colonial historiography. For not
only does it explain how North Africa was subjugated by the Arabs in the seventh century, but it also shows that the country had been a part of the Roman
Empire. Further, the narrative is an expression of that same imaginary structure
developed long ago in Europe, namely the frequently analyzed opposition
between the Occident and the Orient. Moreover, when it is reproduced again in
this colonial context, it suggests a specic colonial policy, namely, divide and
conquer. For instance, when de Slane mentions the great discord among Arabs
45. Ibid., xx.

72

ABDELMAJID HANNOUM

that arose after the death of the Prophet in 622, he happily notes that the largest
part of the Arab nation was led into this quarrel that, fortunately for Europe, forever broke the unity of the Empire.46 In all of this, Europe, along with its struggle against the Orient, is now a fundamental part of that French text called the
Histoire des Berbres.
IV. THE TRANSLATION ITSELF: COLONIAL SEMANTICS

De Slanes narrative is supposedly a summary of his translation of Ibn Khaldn.

Let us now examine the translation itself. It should be repeated that the translation of the Arabic text into French means not so much the transmission of a message from the rst to the second language, but rather the conversion of local categories into colonial categories, a conversion that is the result of the passage
from one culture to another, from one historical moment to another. A long time
ago Benjamin Whorf suggested that facts are unlike to speakers whose language
background provides for unlike formulation for them.47 But how do these facts
change? Why are the facts of a text endowed with different or even contradictory meaning when translated into another language?
It is important to mention that the type of translation under consideration is
what Roman Jakobson calls interlingual translation, which means that one
interprets a verbal sign as a verbal sign in another language.48 Therefore, one is
in fact removed not only from one cultural space to another, from the Maghreb
to France, but also from one cultural time to another, from fourteenth-century
Islam to nineteenth-century Europe. For after all, language is not only an instrument by which a group communicates, but it is also a tool made up of categories
by which a group orders and signies its world. To quote Whorf again:
Every language is a vast pattern-system, different from others, in which are culturally
ordained the forms and categories by which the personality not only communicates, but
also analyzes nature, notices or neglects types of relationship and phenomena, channels
his reasoning, and builds the house of his consciousness.49

However, a language does not exist in theory only, but it is practiced in real
situations, it is a speech-act (parole, as Saussure put it); it is also a textual practice (langue, in the same Saussurian terminology).50 It is clear that the type of
language to be considered in this instance is nineteenth-century French as actualized by de Slane in his translation. Therefore, I shall examine the components
of the translators language, or rather text. However, I shall assume that the translated text as a whole, including the Introduction, is made up of a number of levels. The text is not only divided into signier (the expression, in the terminology
46. Ibid.
47. Benjamin Whorf, Language, Thought, and Reality: Selected Writings (Cambridge, Mass.: MIT
Press, 1956), 235.
48. Roman Jakobson, On the Linguistics Aspects of Translation, in Language and Literature
(Cambridge, Mass.: Harvard University Press, 1987), 429-430.
49. Whorf, Language, Thought, and Reality, 252.
50. Ferdinand de Saussure, Cours de linguistique gnrale (Paris: Payot, 1985).

TRANSLATION AND THE COLONIAL IMAGINARY

73

of Louis Hjelmslev) and signied (the content), but each has a form and a substance. Thus, the text is made up of the form of the expression (the linguistic system) and the substance of the expression (the phonetic system). Further, the content itself has a form (grammar) and a substance (semantics).51 It is in terms of
these various levels that I will now explain the transformation of the original text
into a colonial narrative.
LHistoire des Berbres, as presented, that is, interpreted by the translator to
the reader, is in two parts: one is about the Orient, including the non-Arab dynasties such as those of the Persians and the Israelites; the other is about the Berbers,
but also the Arabs in North Africa. But whereas the histories of the non-Arab
dynasties are narrated only for their relevance to the history of the Arab races,
the history of the Arabs in North Africa is told for its relevance to the history of
the Berber races. In short, the history of Ibn Khaldn is in the nal analysis a
history of two races: a conqueror race coming from the Orient, and a conquered
one, indigenous to North Africa. Further, both Arabs and Berbers are not only
two races, but two nations as well. How does Ibn Khaldn designate them in the
original? What are the meanings of the terms he uses?
Before answering these questions, let me rst attempt a denition of the terms
race and nation as they were used in nineteenth-century France. It is worth
noting that prior to the translation of de Slane, both Arabs and Berbers were designated as races in the ethnography of the Arab Bureau.52 The terms race and
nation had by then gained specic meanings. In nineteenth-century Europe,
with the rise of nationalism, the concepts of race and nation became fundamental in the understanding of the development of human societies. Three inuential
names dominate in France: Comte de Gobineau, Ernest Renan, and Hyppolite
Taine.53 But it was Gobineau who rst articulated the most widely accepted discourse on race in France. His views became most inuential not so much because
they were novel, but because they were widely accepted before he even articulated them. As Hannah Arendt put it in a comparative study of race-thinking in
Europe:
The Comte de Gobineau developed an opinion already generally accepted within the
French nobility into a full-edged historical doctrine, claiming to have detected the secret
law of the fall of civilization and to have exalted history to the dignity of a natural science. With him race-thinking completed its rst stage, and began its second stage whose
inuences were to be felt until the twenties of our century.54

Gobineaus conception consisted of at least two fundamental points: rst, the

world is inhabited by races that are unequal, as are the various breeds of dogs;
51. See Louis Hjelmslev, Prolgomnes une thorie du langage (Paris: Minuit, 1968), translated from Danish by Una Canger. Also see the signicant contribution of A.-J. Greimas, Smantique
structurale (Paris: Larousse, 1966). For a very brief review see the entry Text in Dictionnaire
raisonn du language (Paris: Hachette, 1979).
52. See A. Hannoum, Colonialism and Knowledge in Algeria.
53. John White, Taine on Race and Genius, Social Research 10 (February 1943), 76-99.
54. Hannah Arendt, Race-Thinking Before Racism, Review of Politics 6 (January 1944), 47.
Also, Theophile Simar, Etude critique sur la formation de la doctrine des races au 18 et son expansion au 19 sicle (Brussels: M. Lamertin, 1922).

74

ABDELMAJID HANNOUM

second, the races are necessarily antagonistic. But what is a race? A race is
marked by inherent, hereditary, permanent qualities, psychological, physical, and
intellectual. No matter how much a race is subjected to change and accidents of
nature, it will always preserve these elements.55 Gobineau states unwaveringly:
The existing races constitute separate branches of one or many primitive stocks. These
stocks have now vanished. They are not known in historical times at all, and we cannot
form even the most general idea of their qualities. They differed from each other in the
shape and proportions of the limbs, the structure of the skull, the internal conformation of
the body, the nature of the capillary system, the color of the skin, and the like; and they
never succeeded in losing their characteristic features except under powerful inuence of
the crossing blood.56

According to Gobineau, the Aryan race has proved its excellence throughout
the ages. Excellence in a race is manifested in its phenotypes, and the Aryan race
is by far the most beautiful, the most well shaped, and especially the most capable of creating civilization. Aryan civilization has always shown proofs of its
excellence.57 Civilization is thus associated with race, and a race is by denition
a people with a common origin, common blood, common identity, and the
specic ability (or lack of ability) to create civilization. Indeed, civilization is the
product of a race, it is specic to this race, and it is the expression of its genius.
If one wants to study a race, one studies the form of its civilization.58 If, then,
civilizations are expressions of races, they are therefore, like races, antagonistic
to each other, and when they combine, this results in degeneration. Gobineau
writes: Civilization is incommunicable, not only to savages, but also to more
enlightened nations. This is shown by the efforts of French goodwill and conciliation in the ancient kingdom of Algiers at the present day, as well as the experience of the English in India, and the Dutch in Java.59
If, then, by the mid-nineteenth century, race-thinking had come to constitute a
subconscious culture, in which human relations in general and civilizations in
particular were dened essentially by inequality and antagonism, it is no wonder
that the translated text of Ibn Khaldn carried with it this colonial semantics.
This can be seen in the Arab concepts that de Slane translated as race and
nation.
The Arab concepts translated as race and nation are these: jl, umma, and
tabaqa. The most often used by Ibn Khaldn is jl. This term appears as a syn55. See A. Gobineau, Essai sur lingalit des races humaines (Paris: Firmin Didot, 18531855).
English translation by Adrian Collins, Inequality of Human Races (New York: H. Fertig, 1967). See
also H. Taine, Histoire de la littrature anglaise (Paris: Hachette, 1863), vol. 1.
56. Gobineau, Inequality of Human Races, 133.
57. Ibid, 92-93.
58. It is these forms that constitute Taines work, which was mainly a study of the genius of races,
interestingly enough European races, especially the Latin and German. Taine was one of the most
dominant gures in nineteenth-century Europe. Among his admirers was Nietzsche. After a solid
background in philosophy, he turned to the study of art and literature. Among his well-known work
is his dissertation La Fontaine et ses Fables (Paris: Hachette, 1861); Histoire de la littrature anglaise
(Paris: Hachette, 1863); De lintelligence (Paris: Hachette, 1861).
59. Gobineau, The Inequality of Human Races, 171.

TRANSLATION AND THE COLONIAL IMAGINARY

75

onym of both umma and tabaqa; however, it has a specic meaning that the two
others may not share. Etymologically, jl means a generation, but Ibn Khaldn
uses it in the sense of a tribe or a group of tribes living in a specic historical
moment. Ibn Khaldn speaks of the jl al-`arab (generation of the Arabs) and
the jl al-barbar (generation of the Berbers). Ibn Khaldn also uses the term jl
to designate a specic stage in human development: jl al-badw (those who are
in the stage of nature, or in a much more literal translation, those with a Bedouin
lifestyle) or jl al-hadar (those who are in the stage of culture, or again in a more
literal sense, those with an urban lifestyle).60
One would conclude, then, that the concept jl is temporal, chronological. Jl
is in fact a human grouping clearly dened in time (for Ibn Khaldn speaks of
the jl al-awwal li-al-`arab [the rst generation of Arabs], al-jl ath-thni li-al`arab [the second generation of Arabs], and so forth). Jl is also a genealogical
term, for jl are also linked to each other by a common ancestor (nasab), real or
often imaginary. In other words, jl as a unied group is differentiated from others mainly by a representation of its origins (today one would say by its history).
However, a jl in the understanding of Ibn Khaldn may also be otherwise
dened. First, a jl can be dened by its phenotypes, which Ibn Khaldn explains
rst by the climate that changes the color of a groups skin and may in fact even
inuence its members moods. Second, a jl can also be dened by its culture as
a result of its economic activity (inna ikhtilaf al-ajyal innama huwa bi ikhtilaf
nihlatahum mina al-ma`ash) (The differences among the ajyl are the result of
their different ways of making their living).61
As for the term umma, it is more global. It refers to either a religious community with its total historical experiences (the umma of Islam is, in fact, not only Arabs,
but all those who have shared the Islamic religious experience), or a historical community with its different stages that does not take into account transformations,
religious or dynastic, to be referred to as one entity (as is the case of the umma of
the Arabs, or the umma of the Persians). Let me quote Ibn Khaldn again:
The genealogists were led into error by their belief that the difference between umma
[people] is in their nasab [genealogy]. This is not so. The difference between a jl or umma
for some is due to genealogy [nasab] as in the case of the Arabs, the Children of Israel,
and the Persians. For others such as the Zinj, the Abyssinians, the Slavs, and the Sudanese
Negroes it is due to geographical location and phenotype [sima]. It may be caused by customs [`awaid] and sentiments [shi`ar] as in the case of the Arabs. It may be caused by
other conditions [ahwal], features [khawas] and specic traits [mumayizt] of umam [pl.
umma]. But to generalize and say that the inhabitants of such and such region in the south
or in the north are the descendants of the well known so and so because they have a common nihla [sect] or color or phenotype of the forehead, is of [the order] of these errors
caused by the ignorance of the nature of beings and places which changes through generations [a`qab] and does not stay the same.62
60. According to the translation of Jacques Berque, Ibn Khaldn et les Bedouins, in Maghreb,
histoire et socits (Algiers, Duculot, 1974), 18. Berque, too, translates jl as race.
61. Muqaddima, 129.
62. Muqaddima, 91. I intentionally, but unavoidably, depart from the English translation of Franz
Rosenthal because it is too inscribed in the same colonial discourse. He too translates jl or umma as
race and nation.

76

ABDELMAJID HANNOUM

It should be noted that history, that is, the trkh of the Arabs, is originally a
religious discipline, and its development was a result of the success of the
Islamic conquests.63 Given the fact that the Arab conquest was ultimately an
experience of the Other, the explanation of this otherness itself had a religious
character. In Islam, too, human unity is presented as an unquestionable fact.
Humanity is not only one, the creature of the same Creator, but it is also equal in
the sense that humans are all able to receive the divine message. The inequality
of humans does not occur on a genetic levelit is not the result of the ability or
lack of ability of a group to create civilizationbut rather on a moral level.
Therefore, humanity is not divided into inferior and superior races, but rather
into believers and nonbelievers. Moreover, this inequality is not given, nor is it
irreversible; a nonbeliever is as susceptible to becoming a believer as a believer
is susceptible to becoming a nonbeliever. Hence, the importance of the concept
of the fath, that is the transition from belief to nonbelief.64
Be that as it may be, Ibn Khaldn is too much of a historian not to notice that
on a practical level tribes but also individuals are unequal as far as their might
(jh) is concerned, and that despite the fact that religion provides cohesion for a
group (this is why he talks about the umma of Islam and the umma of
Muhammad), there is also diversity and inequality within the same umma. In
other words, the umma is also made up of a number of jls that differ in might,
in conquest, in deeds. It is their diversityone may even say their inequality
that Ibn Khaldn explains. How is this inequality engendered?
It is here that Ibn Khaldn uses his concept of `asabiya (tribal solidarity). For
him, `asabiya is general, but uneven in strength among tribes. The bigger the
tribe the stronger its tribal solidarity; the stronger the tribal solidarity, the more
powerful the tribe. However, the power of the tribe, however great it may be, is
not a sufcient condition for it to become a dynasty. Religion is an important factor for `asabiya to become operational; it allows the tribe not only to solidify its
power but also to go beyond itself and become a dynasty. However, given the fact
that one of the keys of Ibn Khaldns explanation of human transformations is
the fact that power is provisional, the tribe-dynasty itself is subject to a process
of disintegration. Its power results in luxury, in unjust practices, which themselves result in weakening the `asabiya that provides the cohesion and therefore
the strength of a group. The dynasty then comes to an end and is replaced by that
of another tribe that goes through the same patterns of rise and fall. The superiority of a tribe is contingent, and that superiority is not inherent in the tribe, as
strength is inherent in superior races according to Gobineau, Taine, and Renan.
For Ibn Khaldn inequality is constant and variant at the same time. It is constant
in that humans are constituted of dominant and dominated. But as the material
world, including humans and human conditions, is subject to dissolution, neither
inequality nor superiority is inherent in a group; they are contingencies
63. See Tarif Khaldidi, Arabic Thought in the Classical Period (Cambridge, Eng.: Cambridge
University Press, 1994).
64. The transition from belief to disbelief is also possible; it is the murtadd, apostate, but this
possibility is to be eliminated.

TRANSLATION AND THE COLONIAL IMAGINARY

77

(`awrid), and as such they are doomed to end. Among the people who dominated and later lost their strength, Ibn Khaldn mentions the Greeks, Romans,
Persians, Arabs, and Jews. For states (duwal) have ages, he asserts, like those of
individuals.65
Moreover, there is also diversity within the same jl. Of course, not all the
members of a jl are equal in terms of might (jh). How can this inequality be
explained? Here Ibn Khaldn uses a number of concepts to account for the difference in power (qudra) among individuals or groups within the same jl. For
Ibn Khaldn the superiority is always maintained by a superiority in number, by
a noble genealogy (hasab sharf), but equally importantly by jh, that is, a power
(qudra) to impose the will in order to establish a religious and political order.
(This should be understood as a necessary domination for the benet of both the
dominator [ghlib] and the dominated [maghlb]. They both benet from the
elimination of what is harmful and the pursuit of what is useful.66) To say the
same thing differently, Ibn Khaldn suggests that people, or rather tribes, are
unequal only insofar as some of them have a hasab (nobility) that is not inherent
in their nature but is generated from their own deeds, associated, one would
assume, with might and conquest. However, hasab does not last more than four
generations (ajyal).67
One can see that there is neither progress, nor degeneration, nor racial
hierarchy in Ibn Khaldn. His historical narrative is regulated by other categories such as jl, `asabiya, nasab, and hasab. In some ways historical events
reproduce themselves in the same way: tribes become dynasties due to `asabiya
and religion; they decline after they reach their peak, only to be replaced by
another tribe that follows the same pattern of rise and fall.
The text of de Slane is marked not only by a racial ideology that sets Arabs
and Berbers apart and in opposition, but also by the theme of domination, which
connotes a colonialism of the bad type because it is Oriental and therefore based
on force and destruction. Even in his introduction, de Slane speaks of Arab domination, and he translates the title of the rst chapter of Ibn Khaldn as Arab
Domination. Further, de Slane translates the term fath as domination or conquest, both meaning forced subordination (of course to other people) imposed
by the force of weaponry, a relation of a loser to a winner, and nally, as a form
of colonialism in which the goal is to impose religion on the vanquished.68 What
does the word fath actually mean? It is a positive term for Ibn Khaldn as for any
Muslim; it designates the opening of an area of belief, of Islam, and the ending
of a period of disbelief, of Jhiliya; therefore, it is a divine favor for both those
who arrive and those who join of their own free will or by force. Submission, in
65. Muqaddima, 180.
66. See Cheddadi, Ibn Khaldn revisit.
67. Muqaddima, 145.
68. In the nineteenth-century French dictionary, domination means: Authorit qui, accept ou
non des subordonns, sexerce pleinement. (Emile Littr, Dictionnaire de la langue franaise [Paris:
Gallimard/Hachette, 1967). As for the word authorit, it means pouvoir de se faire obir (ibid.).
Further, whereas Ibn Khaldn always mentions the fath as Islamic (al-fath al-islami), de Slane mentions it also as Arab conquest. The word conqute means to soumettre par les armes (ibid.).

78

ABDELMAJID HANNOUM

this view, is not of people to other people, but to God. Historically, the early
Muslim community itself went through this transformation, from the Jhiliya to
Islam. Fath is a necessary condition for a Muslim. Once again de Slane translates
a term (here, fath) through the colonialist lens that perceived the world in terms
of domination, and in the process both misrepresents Ibn Khaldn and at the
same time reinforces the French domination of North Africa.
V. THE TRANSLATION ITSELF: COLONIAL SYNTAX

In the colonial text, the term Arab does not mean what it does in the original
text of Ibn Khaldn: the various Arab populations, either originally or more
recently Arabized, with their various generations (ajyl, pl. of jl), each marked
with different levels of cohesiveness (`asabiya) and the power (jh) that results
from levels. Rather, the term Arab refers to the Oriental of Orientalism whose
image is all too familiar to the European reader; they are one and the same. They
do not change, nor do they disappear. They are always dominant, and their domination brings destruction. In addition, the term Berber does not mean in the
colonial discourse what it means in Ibn Khaldn: the diverse Berber tribes,
whose origins are mostly Oriental, and whose histories and deeds are associated
with Arabs. Rather, the term Berber refers to newcomers in the eld of
Orientalism. They did not exist prior to the conquest of Algiers, or rather they
were the Arabs, the Moors, the Saracens. The Berbers now exist in the French
imaginary, but they join a category of people who have also been familiar to the
European since the beginning of the nineteenth century: they are the primitives
who represent the past of Europeans. Let us now see how this new semantics
affects or, better, determines the narrative syntax of the translation of de Slane,
that is, how it determines the actors, the objects, and the relations involved in the
narrative.
Colonial historiography, based on the translated Ibn Khaldn, is marked by
two major narratives, the narrative of the Khina, whose signicance in colonial
scholarship is immense, and the episode of the Banu Hilal. Both narratives
explained the present of North Africa to colonial scholars, and how it had
become Arab and Muslim despite a massive Berber presence and a long Roman
history.69 These two episodes of Arab invasion constitute in French colonial historiography not only narratives of origins but also great discontinuities that have
changed the course of North African history and have shaped it in a certain way
relevant to the French presence in the Maghreb. Let me now show how racethinking and nation-thinking have shaped the colonial narrative. To this end, I
briey discuss the objects found in Ibn Khaldns original text in order to understand their transformations in the colonial translation.
De Slane admits that the Introduction has not attained all its objectives of correction, rectication, and orientation: The translator has proposed to devote a
number of pages in this short introduction to examine the origins of the Berbers.
69. See A. Hannoum, Colonial Histories, Postcolonial Memories (Portsmouth: Heinemann, 2001).

TRANSLATION AND THE COLONIAL IMAGINARY

79

His work was about to be nished when a number of items of information came
to modify his conclusions.70 In other words, despite the fact that one nds in Ibn
Khaldn an explanation of the origins of the Berbers, this seemed to de Slane to
be unsatisfactory and needed to be corrected. De Slane promised his readers that
he would put his text regarding the origins of the Berbers in the third volume. But
he did not do so. Why does de Slane judge Ibn Khaldns views of the origins of
the Berbers as unsatisfactory?
The early ethnographers of the Arab Bureau had already set in place a discourse on the origins of the Berbers that provided them with a European origin.
For, by the color of their skin, by their culture, their architecture, and even their
religious beliefs the Berbers were believed to be descended from both an indigenous population about which little is known and Roman, Greek, and German
populations that inhabited the country at the dawn of historical time. The Arab
invasion had changed little in the racial landscape of North Africa: now as before
the Berbers still speak a distinct language, they still preserve their same skin
color (white), and they are still liberal, unconcerned about religion, that is,
Islam (they still observe Roman law [qnn], not Islamic law [shar`a]). Further
they still live in a primitive republic with democratic principles: in Berber custom, a municipal council (djemaa) is elected. This is to say that by the time the
translation of Ibn Khaldn was made, there was already a widely accepted discourse about the Berbers, a discourse that made them primitive Europeans.71
Thus, when Faidherbe, a general in the army and a vice president of the Socit
danthropologie, summed up the situation of anthropological research in Algeria
in 1874 as regards the Berbers and their origins, he reached a statistical conclusion.72 For him the inhabitants of Algeria by the beginning of the nineteenth century were divided in the following manner:
Indigenous Libyans
} Berbers
Blonds from the North
Phoenicians
Romans, their auxiliaries, and Greeks from the Late Empire
Vandals (in the East)
Arabs (a good number of them remained pure; stronger proportions in the East
and the West)
Negroes of all races and sorts (the majority at the stage of interbreeding in all
degrees) more numerous in the South
Jews (completely analogous to the Arabs)
Turks from all the provinces and European renegades

75%
1%
1%
0.5%
15%
5%
2%
0.5%

North Africa was, then, believed to be inhabited by a predominantly Berber,

that is, European population. They constituted a little more than 76 percent of the
population, as opposed to the Arabs who constituted 15or rather 17percent,
70. De Slane, Histoire des Berbres, lxvi.
71. See Hannoum, Colonialism and Knowledge in Algeria.
72. Louis Leon Cesar Faidherbe, Instructions sur lanthropologie de lAlgrie (Paris: Hennuyer,
1874), 12.

80

ABDELMAJID HANNOUM

given that at this date Arabs and Jews were still part of the same discourse (they
were the Semites of Renan and Gobineaua degenerate race, a mixture of white
and black). Algeria then, in French colonial ethnography, was numerically largely Berber and historically European. Such a discourse contradicts the narrative of
Ibn Khaldn on the origins of the Berbers, but it does t nicely with the translated text of de Slane that makes the history of North Africa the history of a struggle between Arabs and Berbers.
What was Ibn Khaldns account of the origin of the Berbers? The `Ibar of Ibn
Khaldn was polemical in its time. It was a defense of the Berbers against the
negative image that was created soon after the Arab conquest of North Africa in
the late seventh century. Ibn Khaldn states the goal of his book clearly. For, he
says, he wants to show the strength [the Berbers have shown] through time, that
they inspired fear; they are brave and powerful comparable to other nations and
peoples of the world such as the Arabs, the Persians, the Greeks and the
Romans.73 It is with this intention that Ibn Khaldn describes the genealogy of
the Berbers:
The truth that should be accepted regarding the origin of the Berbers is that they are
descendants of Canaan, son of Ham, son of Noah, as indicated in the presentation of the
genealogies of men since the Creation. The name of their ancestors is Mazigh; the
Kerethetes [Arqish] and the Philistines were their brothers; their brother Casluhim, son of
Mizrain, son of Ham. Their king Goliath is well-known as one of their kings. There were
well-known wars between the Philistines and the Children of Israel in Syria. The
Canaanites and the Kerethites were allies of the Philistines. This is what one must believe,
because it is the truth that must not be ignored.74

The project of Arabizing the Berber was the culmination of a long-term effort
undertaken from the ninth century onward to fuse two populations that seemed
to be opposed as conqueror and conquered.75 In other words, the conquest of
North Africa by the Arabs had to be interpreted in such a way that it would not
seem a conquest, that is the domination of one group by another. After the establishment of Berber dynasties, the conquest of North Africa came to be seen rather
as a reunion of two groups of the same origins. They were both Arabs; one of
them came at the dawn of time, and the other came in the late seventh century to
transmit the prophetic message.76
This discourse, which I have elsewhere characterized as inclusive and integrative,77 questions the whole colonial discourse that was built on the division of
Arab versus Berber. Obliterating the opposition between Arabs and Berbers
would lead to one of two conclusions: either that both groups are Algerians, of
mixed origins, and that therefore Algeria belongs to them, and to them only; or,
if the view of Ibn Khaldn is adopted, that both groups are Arabs, and therefore
here too, not only is the French presence illegitimate, but so also was the Roman
73. `Ibar, 121.
74. Ibid., 6, 113.
75. For more details see Hannoum, Colonial Histories, Postcolonial Memories, 1-28.
76. Ibid.
77. Ibid.

TRANSLATION AND THE COLONIAL IMAGINARY

81

presence. Small wonder that, when de Slane could not convert the view of Ibn
Khaldn on the issue by an act of translation, he simply deemed it incorrect and
declared it as such by his authority as translator, an authority based not only on
a bilingual competence, but also on a wide reading in two traditions.
The Arabic text of Ibn Khaldn reveals, as I have attempted to show, a different
imaginary structure, one that is marked neither by racialism nor by a colonial ideology, nor does it contain a plot in which a hero and anti-hero pursue opposite
quests. Nor does it point to a major lack that is at the heart of North African history. Instead, Ibn Khaldns long and complex narrative reveals a complex set of
relations between tribes; it is a recording (trkh) of various states of successive
generations; it is an ordering of the Maghreb of the fourteenth century according to a number of categories specic to his time: tribal, genealogical, religious,
`asabiya, jh, and nasab. Above all, history happens in a cyclic way; it reproduces the same patterns, with actors grouped in different polities, tribal and
dynastic. It is a history that undoubtedly implies change and transformation,
but excludes the notion of progress and a shared human future towards which
all peoples head with different and unequal speed.
Yet, Ibn Khaldns history became the Histoire des Berbres, a colonial text
that employed colonialist categories to perceive, to think, and to re-present North
African history. This occurred not only in the colonial period by colonial historians, but also in the post-colonial period, by national and post (?) orientalist historians alike.78 In all of this de Slanes translation too is a product of the imaginary.
Simons Rock College of Bard

78. The most known of this is Abdallah Laroui who, despite his criticism, was caught in colonial
categories. The most illustrative example of a post (?) orientalist historiography is Maya Shatzmiller,
The Berbers and the Islamic State (Princeton: Markus Wiener, 2000) in which she writes the history
of the Berbers using racial categories found in de Slane.