Akasoy Paganism

Anna AKASOY
PAGANISM AND ISLAM:

MEDIEVAL ARABIC LITERATURE
ON RELIGIONS IN WEST AFRICA
Introduction
Paganism and Islam is a subject which can be approached from
a great variety of perspectives. Before dealing in more detail with one
specific case, I would like to illustrate the breadth of the question with
two examples, a linguistic and a historical one. The earliest preserved
Arabic text on falconry, compiled in eighth-century Baghdad, claims that
among four stages in the life of a bird of prey, the nestling, the young
bird which is still in the nest, should be preferred by falconers. The sec-
ond best age is when the bird has already left the nest but is not yet
a fledgling. It lives on branches and still needs to be fed by its parents.
In Arabic it is called the ungrateful a typical human teenager. When
this Arabic text was translated into Latin some five hundred years later,
the translator did not render the Arabic term for this bird, kafir, accord-
ing to its meaning, ungrateful, or as a technical term, but rather accord-
ing to its usage in inter-religious encounters. He writes of a bird qui
dicitur paganus
1
the bird which is called a pagan.
My second example: the Chronicle of the Discovery and Conquest of
Guinea, a Portuguese text of the fifteenth century, tells of a cargo of
West African slaves shipped to Prince Henry in Portugal. Among these
men were Muslims as well as pagans. The Muslims among them
offered their pagan servants as ransom. The author of the chronicle,
Gomes Eannes de Azurara, states:
* I would like to thank Charles Burnett, Astrid Meier, Patricia Crone and James Mont-
gomery for their helpful comments.
1. For the Arabic text see Al Girif ibn Qudama al-Gassani, Die Beizvgel (Kitab
awari a-ayr). Ein arabisches Falknereibuch des 8. Jahrhunderts, trans. Detlef Mller
and Franois Vir (Hildesheim, 1988), p. 80. A critical edition of the Latin translation is
included in Stefan Georges, Das zweite Falkenbuch Friedrichs II. Quellen, Entstehung,
berlieferung und Rezeption des Moamin (Berlin, 2008), p. 125 for the quotation.
208 A. AKASOY
And here you must note that these Blacks [the ones being offered as ran-
som] being Moors like the others [the captives seeking their freedom], are
nevertheless servants of the former [the cavaleiros], in accordance with
ancient custom, which I believe to have been because of the curse which
after the Deluge, Noah laid upon his son Cain.
2
What these two examples show is that the definition of who is a pagan,
what is paganism, and what it means to be a pagan lies very much in the
eye of the beholder. In the first example, being a pagan seems more
a question of attitude. It is, of course, the Latin translator, who estab-
lishes explicitly the connection between ungratefulness and paganism, a
connection that is implicit in the koranic use of kafir, but we can imagine
how a young bird might display an attitude similar to that of an unbe-
liever or why, to put it in other words, both are referred to with the
same term in Arabic. The bird is not entirely helpless, but happily enjoys
the supplies it receives from its parents. The pagan, in a similar manner,
enjoys the gifts bestowed upon him by God, but does not appreciate
them.
In the second example, paganism is important for the identity of indi-
viduals as members of a specific group which implies also a certain
social standing and even certain rights or the absence of them. In the
Portuguese chronicle, the pagan is the other, the non-Christian, but
there is a hierarchy among the pagans. Gomes Eannes de Azurara
explains it by referring to the Biblical legend of the curse of Ham (con-
fusing Ham and Cain as happened often in the Middle Ages), which
served in the West as a justification for enslaving Africans. From a Mus-
lim point of view, this hierarchy had also a different background: for the
Muslim cavaleiros, their slaves were pagans. As is also evident from
the second example, addressing the problem of Islam and paganism is
in a certain sense paradoxical, since the term pagan is prefigured by the
Christian tradition in which pagan often denoted a Muslim.
3
Christians
are not accustomed to think of themselves as pagans; it is for both that
the pagan is always the other.
2. Quotation from Benjamin Braude, The Sons of Noah and the Construction of Eth-
nic and Geographical Identities in the Medieval and Early Modern Periods, The William
and Mary Quarterly, 3
rd
ser., 54 (1997), p. 103142, at p. 1278. For the Biblical curse
see David M. Goldenberg, The Curse of Ham: Race and Slavery in Early Judaism, Chris-
tianity and Islam (Princeton, 2005).
3. John V. Tolan, Saracens: Islam in the Medieval European Imagination (New York,
2002), chapter 5: Saracens as Pagans (p. 105134).
PAGANISM AND ISLAM 209
In the common use of the modern Western world, the word pagan is
not applied to members of the worlds large and well known religious
communities, i.e. people who can be clearly identified as Muslims, Jews,
Hindus or Buddhists, but rather to followers of religious traditions that
are not easily recognisable through a revelation, a prophet or a literary
corpus (unless the word is avoided altogether because it is considered to
have negative connotations). Usually it is used for historical civilisations,
the marginalised peoples of contemporary Africa and Australasia or,
sometimes and only in recent times as a self-definition of groups in the
Western world that seek a revival of the religious traditions of the ancient
world or pre-modern cultures.
4
This use of the word pagan in the Western world, together with the
hierarchy of Christianity, other Abrahamic religions, other large religions
with a literary corpus, and the miscellaneous has well known roots in the
history of Christianity in a wider sense. These roots are, to a certain
degree, shared by Islam which has, from its very beginnings, a similar
hierarchy of non-Muslims.
5
Privilege is given to the people of the book
(ahl al-kitab), i.e. monotheistic religions or those which could be inter-
preted as such with a written revelation (Jews, Christians, Sabians, Magi-
ans/Zoroastrians),
6
whereas other religious traditions, the pagans, i.e.
usually idolaters and polytheists, are harshly rejected. This hierarchy exists
in theological theory as well as in legal and political practice: pagans
cannot claim the same degree of tolerance granted to non- Muslim mono-
theists who enjoyed the status of protected minorities.
7
Theirs is, at least
4. On the website of the UK-based Pagan Federation, the definition given for a
pagan is a follower of a polytheistic or pantheistic nature-worshipping religion. http://
www.paganfed.org/paganism.php accessed on 14 April 2007.
5. The literature on the subject is abundant. See, for example, the volumes edited by
Jacques Waardenburg (Muslim Perceptions of Other Religions: a Historical Survey [New
York, 1999]) and Robert Hoyland (Muslims and Others in Early Islamic Society [Alder-
shot, 2004]). To name but one prominent example, Shafii distinguished at the beginning
of his famous Risala between those who have written revelations, but distorted them, and
those who did not have any such texts. Both, however, are guilty of kufr (kafaru billah).
6. Zoroastrianism, i.e. the religion of the majus, was used in Arabic literature to
describe traditions of worship and religious affiliation in West Africa (see below). How-
ever, it is not clear how far medieval authors believed these traditions to be connected.
7. However, tolerance granted to non-Muslim monotheists had its limits. They cer-
tainly did not count as believers, as Marietta Stepaniants claims, cf. The Encounter of
Zoroastrianism with Islam, Philosophy East and West, 52 (2002), p. 159172, at p. 163.
The Koranic verse 22:17 God will judge between the believers, those who follow the
Jewish faith, the Sabians, the Christians, the Magians, and the polytheists on the Day of
Resurrection; God witnesses all things. (Translation M.A.S. Abdel Haleem [Oxford,
2004], with a slight modification [Abdel Haleem translates idolaters where the present
210 A. AKASOY
in theory, the choice between conversion, exile, and death. An important
difference was made by legal scholars regarding etnicity. While some
scholars allowed non-Arabs to remain polytheists, there was also a trend
not to allow Arab Christians and Jews to keep their religions.
8
A similar division exists in Islamic studies where privilege is given to
studying the relations between Muslims and monotheists, above all Jews
and Christians. Because of its historical and contemporary significance
and the richness of the available sources researchers have also addressed
the relation between Islam and Indian religions. Studying Islam and
paganism, on the other hand, rarely aims at the relations between Mus-
lims and people classified as pagans, but rather at the detection of ele-
ments of previous religious traditions living on in areas after they came
under Islamic influence. Islam and paganism appear here as two dif-
ferent historical layers.
I. Paganism and early Islam
The two salient features of non-Muslim beliefs in the Koran can be
summarised as kufr and shirk.
9
Kufr, as mentioned before, denotes the
ungrateful rejection or denial of what has been given by God and is con-
nected to error and delusion,
10
whereas shirk means the association of
God with other deities, i.e. polytheism, sometimes expressed in idola-
trous practice. According to wide-spread Islamic views, both failures, in
particular the first one, refer primarily to the beliefs of the pagan Arabs,
but the Koran makes no such identification:
11
there is no separate term
author has chosen polytheists]) makes a clear difference between the believers (alla-
dhina amanu) and the other religious groups. Inna lladhina amanu walladhina hadu
wal-abiina wal-naara wal-majusa walladhina ashraku inna llaha yufailu bayna-
hum yauma l-qiyamati inna llaha ala kulli shayin shahidun.
8. See Yohanan Friedmann, Classification of Unbelievers in Sunni Muslim Law and
Tradition, Jerusalem Studies in Arabic and Islam, 22 (1998), p. 163-195.
9. Marilyn Robinson Waldman, The Development of the Concept of Kufr in the
Quran, Journal of the American Oriental Society, 88 (1968), p. 442455; for a compila-
tion of the relevant verses cf. Muhammad Ibrahim Hafiz Ismail Surty, The Quran and
Al-Shirk (Polytheism): Collection of Relevant Verses (London,
2
1990) and for a thorough
analysis Toshihiko Izutsu, Ethico-religious Concepts in the Quran (Montreal, 2002),
chapters vii and viii. See also Manfred Ullmann, Wrterbuch der klassischen arabischen
Sprache (Wiesbaden, 1970-), s.v. k-f-r.
10. Richard M. Frank, Al-Ghazali on Taqlid. Scholars, Theologians, and Philoso-
phers, Zeitschrift fr Geschichte der Arabisch-Islamischen Wissenschaften, 7 (1991-92),
p. 207-252, at p. 242, note 71.
11. Jacques Waardenburg, Muslim Studies of Other Religions, in idem (ed.), Muslim
Perceptions of Other Religions: a Historical Survey (New York, 1999), p. 3101, at p. 6.
for pagan, which would make it absolutely clear that the kuffar and
mushrikun (i.e. those who practise kufr and shirk) of early Islamic
sources are not the same as Christians and Jews, even though this might
be so in individual cases. The lack of such a separate terminology makes
talking about Islam and paganism even more problematic.
Among revisionist scholars of early Islamic history this unassigned
notion of shirk and kufr has added to a number of speculations regarding
the religious environment in which Islam emerged. Gerald Hawting, for
example, in his book The Idea of Idolatry and the Emergence of Islam
12

has argued that Islam did not emerge as a monotheistic reaction to Arab
paganism, as the traditional Islamic narrative claims, but rather as a truly
monotheistic reaction to other forms of monotheism, i.e. Judaism and
Christianity. The polemics in the Koran against paganism should thus not
be read as a straightforward rejection of local cults of idolatry, but rather
as a denunciation of the earlier monotheistic traditions. Hawting based his
claim on the observation that the nature of the koranic polemic against
the mushrikun does not fit well with the image of pre-Islamic Arab idola-
try and polytheism provided by Muslim tradition.
13
Hawting subscribed
to an earlier claim of John Wansbrough and suggested that the Koran did
not come into existence in a polytheistic Mecca which might have been
an invention of the Islamic tradition of the third century AH in the first
place but rather in a monotheistic milieu outside of western Arabia.
Hawtings theory touches a number of important aspects in the sensi-
tive field of the origins of Islam. In their reactions to this book, historians
of early Islam have indicated some major problems. One of them, as
Fred Donner pointed out, is that the deconstruction of the myth of a
polytheistic Mecca undermines the need for identifying a monotheistic
milieu outside of western Arabia where the Koran was conceived.
14

12. Gerald R. Hawting, The Idea of Idolatry and the Emergence of Islam: from
Polemic to History (Cambridge, 1999).
13. For the following short summary of Hawtings arguments cf. The Idea of Idolatry
and the Emergence of Islam, p. 5. For the view expressed by the founder of this academic
school, John Wansbrough, The Sectarian Milieu: Content and Composition of Islamic
Salvation History (Oxford, 1978). For more recent contributions to this ongoing debate
see Herbert Berg (ed.), Method and Theory in the Study of Islamic Origins (Boston, 2003).
14. Fred Donner, review of Hawting, The Idea of Idolatry in Journal of the American
Oriental Society, 121 (2001), p. 336338. But see also Fred Donner, Narratives of Islamic
Origins. The Beginnings of Islamic Historical Writing (Princeton, 1998), p. 153 for the
need of the early Muslims to distinguish themselves from the Jewish and Christian tradi-
tions. See also the reviews by Francis Robinson (Bulletin of the School of Oriental and
African Studies, 64 [2001], p. 270) Andrew Rippin (Journal of Semitic Studies, 46 [2001],
p. 348351), and Paul Cobb (Journal of Near Eastern Studies, 61 [2002], p. 299300).
212 A. AKASOY
Furthermore, in the Koran there is a clear distinction between monothe-
ists and others,
15
and specific local customs such as infanticide are
targeted which cannot be connected to any of the monotheistic tradi-
tions.
16
Be this as it may, when the prophetical traditions were assembled in
the second and third centuries of the Islamic era and the paganism of
Mecca had become but a distant memory, it continued to appear as a
theme in religious literature. Unlike early Christianity, Islam developed
its classical shape and spread in a milieu much of which had already
adopted monotheism and where it had to defend its original truth. In
an article on Jewish and Muslim rituals connected with death, Fred
Astren pointed out comparable efforts to defend monotheistic customs
against pagan influences. Yet, in the cases under consideration, these
pagan cults were already part of a distant past. Astren therefore sug-
gested that the pagan idolatrous Other of text and memory stood for
other more complex Others of the day,
17
which certainly confirms
for the Muslims (at least for the period in which there were many con-
tacts with Christians and Jews) that references to the mushrikun and kuf-
far were often aimed at these two groups. Which people were originally
targeted as pagans and whether and at what point the attacks against
idolatry and polytheism became primarily a polemical reaction against
other monotheists, shall not concern us here further. What is important,
are the general features of paganism as they appear in Arabic literature
composed during the first three centuries of Islamic history. Paganism is
associated with a certain attitude, even Weltanschauung, certain rituals
and a certain social and political affiliation with non-Muslim groups (as
reflected in the oft-quoted tradition that whoever dies without allegiance
to an imam dies a pagan death),
18
possibly with a hostile stance towards
Islam. At the same time, the confessional identity of these groups is not
entirely clear. They can be Christians, Jews, or proper polytheists. The
boundaries remained flexible in the following centuries. The Brethren of
Purity (Ikhwan al-afa), for example, described the pre-Islamic Arabs as
15. See note 7.
16. For the relevant Koranic passages see Avner Giladi, Some Observations on Infan-
ticide in Medieval Muslim Society, International Journal of Middle East Studies, 22
(1990), p. 185200.
17. Fred Astren, Depaganizing Death: Aspects of Mourning in Rabbinic Judaism and
Early Islam, in J.C. Reeves (ed.), Bible and Quran. Essays in Scriptural Intertextuality
(Atlanta, 2003), p. 183199, at p. 199.
18. Patricia Crone, Medieval Islamic Political Thought (Edinburgh, 2005), p. 137.
being religious (yatadayyanun) by worshipping idols. The difference
between the idols of the polytheists and the prophets was that the latter
were speakers. Both, however, approach God.
19
Difficulties to draw
a clear line between Muslims or monotheists and pagans can also be
observed in connection with explanations for the origins of political
orders. While the political order in areas under Islamic rule was believed
to be legitimised ultimately by prophecy, pagans could not claim such
a foundation for their political orders. The origins could only have been
among the people themselves.
20
Even though the key developments of classical Islam took place
mostly in a Middle Eastern milieu, the geographical area in which Mus-
lims acted during those years soon became much larger. In April 711
(i.e. the year 92 of the Islamic era), ariq ibn Ziyad crossed the Strait of
Gibraltar and landed on the Iberian Peninsula. Forty years later, the sol-
diers of the young Abbasid Empire fought successfully the Chinese
army in the battle of Talas in present-day Kazakhstan.
As Islam expanded into Africa and Asia, Muslims encountered reli-
gious traditions which presented themselves in ways not unlike the idol-
atrous polytheism of pre-Islamic Arabia, whether literary or authentic.
To be sure, many Muslims had been born into these traditions and then
converted to the religion of the conquerors. They were familiar with the
beliefs and concepts of their native regions which were now considered
pagan. But how did those Muslims who came from the central lands of
Islam look at the religious traditions of the newly conquered territories?
Which are the approaches and attitudes that can be reconstructed from
the texts they wrote? Islamic religion, law and ethics follow the para-
digm of Muhammads first community of believers in the Hijaz. Did this
historical situation also provide a matrix for dividing non-Muslims which
remained valid over the following centuries and in the different regions
of the vast and growing Islamic world? Did the polytheistic idolatry of
pre-Islamic Arabia remain a specific pagan culture, or did it turn into
an archetypical paganism? Did the distinction between Arabs and non-
Arabs in law also apply to other fields of knowledge? What did the
notion of paganism imply for medieval Muslims?
19. Hermann Landolt, Ghazali and Religionswissenschaft. Some Notes on the
Mishkat al-Anwar, Asiatische Studien, 45 (1991), p. 1972, at p. 30. The translation being
religious is Landolts. For a discussion of his terminology see Anna Akasoy, Al-Ghazali,
Ramon Llull and Religionswissenschaft, The Muslim World, 102 (2012), p. 33-59.
20. Crone, Medieval Islamic Political Thought, p. 266.
214 A. AKASOY
II. The religions of ancient Greece and India
Among the attitudes of medieval Middle Eastern Muslims to other,
non-Abrahamic cultural and religious traditions, Western scholars have
shown a particular interest in ancient Greece. During the age of the
Graeco-Arabic translation movement in Baghdad from the eighth until
the tenth century, a great number of Greek texts on philosophy and
science were translated into Arabic and became thus part of the intel-
lectual heritage of medieval Islam. When many of these texts were trans-
lated into Latin several centuries later, medieval Christians faced great
problems connected with the religious identity of the philosophical
authorities.
21
It was partly the awareness of these problems which made
it so interesting for modern Western scholars to see how Muslims and,
in fact, earlier Christians (since many individuals involved in the transla-
tions were Christians) had behaved in this case.
The protagonists of this translation movement and their patrons found
different ways of accommodating the ancient Greek pagans in their new
environment. They legitimised them and used them for legitimising
themselves. The Abbasid Caliph al-Mamun, for example, had his
famous dream of Aristotle, the Greeks were identified with the Sabians,
22

Socrates was presented as a martyr of monotheism,
23
and other Greek
philosophers as the pupils of prophets.
24
Greek texts contained explicit
references to ancient Greek religion, and the way these were rendered in
Arabic is crucial for reconstructing early Muslim attitudes to this kind of
paganism always bearing in mind that these attitudes were probably
shared by Arab Christians. Given the nature of the texts, the problems
here seem to have been polytheism and anthropomorphism rather than
idolatry. The translators developed a variety of strategies to render the
ancient Greek gods into Arabic and for a religious environment that had
little tolerance for polytheism. They appear as angels, as those who are
21. See the article by John Marenbon in this volume.
22. Dimitri Gutas, Greek Thought, Arabic Culture: the Graeco-Arabic Translation
Movement in Baghdad and Early Abbasid Society (2
nd
4
th
/8
th
10
th
centuries) (London,
1998), p. 95100 for al-Mamuns dream. John Walbridge, Explaining Away the Greek
Gods in Islam, Journal of the History of Ideas, 59 (1998), p. 389403, at p. 398 for the
Sabian connection. For similar strategies adopted by the Jews see Norman Roth, The
Theft of Philosophy by the Greeks from the Jews, Classical Folia, 32 (1978), p. 5367.
23. Ilai Alon, Socrates in Medieval Arabic Literature (Leiden, 1991), p. 6167 on the
trial.
24. Walbridge, Explaining Away the Greek Gods, p. 400.
deified or who deify themselves (mutaalla/ih) or simply as aliha, gods.
25

In the Hippocratic oath, the gods appear as awliya allah min al-rijal
wal-nisa, the holy men and women, or male and female friends, of
God, a common expression in Sufism.
26
Another case which offers interesting material is India. Even though
their influence did not reach the level of the Greek legacy, Indian imports
in mathematics, medicine or literature were of great importance in
Islamic intellectual history. Indian religions also attracted the interest of
Muslim writers, their religious iconography striking a particularly sensi-
tive chord. The best known medieval Muslim author on India, al-Biruni
(d. 1048), recognised the same pattern in Indian and Greek civilisation
and compared both with Sufism, and, occasionally, with Christianity.
27

These comparisons served al-Biruni, amongst other things, to point out
the esoteric dimension of Indian religions which was less repulsive to a
Muslim reader than its iconography and polytheism. He specifically
observed that in addition to a lofty philosophical side, Greeks and Indi-
ans shared a popular side of anthropomorphism. Al-Biruni explained the
development of this anthropomorphism as a gradual deification of objects
which were originally meant as a device for venerating a prophet or
god.
28
According to the same author, further confusion was caused in
ancient Greece by the vague use of the word god.
29
25. See, for example, the entry qev in Aristotle, The Arabic Version of the Nicoma-
chean Ethics, trans. Douglas Dunlop, ed. Anna Akasoy and Alexander Fidora (Leiden,
2005) and the entry lh in Gerhard Endress and Dimitri Gutas, A Greek and Arabic Lexi-
con (GALex): Materials for a Dictionary of the Medieval Translations from Greek into
Arabic (Leiden, 2002).
26. Manfred Ullmann, Islamic Medicine (Edinburgh, 1978), p. 3031.
27. Walbridge, Explaining Away the Greek Gods, p. 392397; Richard Walzer, Al-
Biruni and Idolatry, in Commmoration Cyrus: Hommage universel (Leiden, 1974), III,
p. 317323, and Mario Kozah, The Epilogue of al-Birunis Kitab Batanjal, in James
E. Montgomery (ed.), Abbasid Studies. Occasional Papers of the School of Abbasid
Studies, Cambridge 6-10 July 2002 (Leuven, 2004), p. 263273. An interesting case of a
reverse comparison can be found in an anti-Christian polemic written by Amad ibn Abd
al-amad al-Khazraji al-Anari al-Qurubi (519/1125582/1187), a Muslim writer in
Christian Toledo. He targeted, among other things, Christian miracles and exposed a fly-
ing cross as held by a magnet. Al-Qurubi perhaps borrowed this from an account of an
Indian sanctuary by al-Qazwini. Cf. Maribel Fierro, Christian Success and Muslim Fear
in Andalusi Writings during the Almoravid and Almohad Periods, Israel Oriental Stud-
ies, 17 (1997), p. 155178, at p. 170.
28. For a discussion of al-Birunis account see Yohanan Friedmann, Medieval Mus-
lim Views of Indian Religions, Journal of the American Oriental Society, 95 (1975),
p. 214221.
29. Walbridge, Explaining Away the Greek Gods, p. 396.
216 A. AKASOY
In other Islamic texts, analogies between Indian religions and Sufism
appear as well, for example in the Life of the Buddha in the Compen-
dium of Chronicles (Jami al-tawarikh) of the Ilkhanid vizier Rashid al-
Din (12471318). Rashid al-Din used terms and concepts associated with
Sufi asceticism such as riyaa (exercise) for describing ascetic prac-
tices in India. He also used expressions belonging to a more mainstream
Islamic vocabulary and referred, for instance, to the Buddha as a prophet.
Similarly open-minded attitudes are rare, another example is Jan-i Janan
in the eighteenth century. He referred to the Koranic claim that a warner
had been sent to every community and claimed that therefore, prophets
must have also been sent to India.
30
This open-mindedness, however,
was not unlimited. Rashid al-Din mentioned idolatrous practices in India
without a critical remark, but he rejected with clear words some of the
metaphysical beliefs of the Hindus.
31
To the claim of Maheshvara (i.e.
Shiva) I am the Lord, the Creator (inni ana al-rabb al-khaliq) Rashid
al-Din adds the words: naudhu billah minhu, We take refuge with God
from this.
32
In his book on India, al-Biruni categorically stated: that
which is not the truth is a deviation, and unbelief (al-kufr) is a single
religion, for it is turning away from the truth.
33
Medieval Muslim writers also established a connection between Indian
idolatry and pre-Islamic Arab idol-worship. A tradition popular since the
ninth century claims that idols were first created in India and washed
from there by the deluge to the Arabian Peninsula. At the same time,
some of those who wrote on Indian religions distinguished clearly
between Indian and Arabian idolatry. Jan-i Janan, for example, claimed
that only the pre-Islamic Arabs treated their idols as independent agents.
30. Friedmann, Medieval Muslim Views of Indian Religions, p. 219.
31. Karl Jahn, Kamalashri Rashid al-Dins Life and Teaching of the Buddha.
A Source for the Buddhism of the Mongol Period, Central Asiatic Journal, 2 (1956),
p. 81128 and idem, Rashid al-Dins History of India: Collected Essays (The Hague,
1965). See also Anna Akasoy, The Buddha and the Straight Path. Rashid al-Dins Life of
the Buddha: Islamic Perspectives, in Anna Akasoy, Charles Burnett, and Ronit Yoeli-
Tlalim (eds), Rashid al-Din, Agent and Mediator of Cultural Exchanges in Ilkhanid Iran
(forthcoming volume of conference proceedings, to be published in the series Warburg
Institute Colloquia).
32. Fol. 2072r in the manuscript formerly preserved in the library of the Royal Asiatic
Society (A 27), now in the collection of Islamic art of Nasser D. Khalili. Reproductions
of the manuscript in Karl Jahn, Die Indiengeschichte des Rasid ad-Din (Vienna, 1980) and
Sheila S. Blair, A Compendium of Chronicles: Rashid al-Dins Illustrated History of the
World (London, 1995).
33. Fa-inna ma ada al-aqq zaigh wal-kufr milla waida min ajl al-iniraf anhu.
Al-Biruni, Kitab fi taqiq ma lil-Hind (Hyderabad, 1958), p. 18. Translation (modified)
from Walbridge, Explaining Away the Greek Gods, p. 395.
He also made a distinction between Hindus before Islam who should be
given the benefit of doubt and not light-heartedly dismissed as infidels,
and those who had the chance to accept Muhammads message a mes-
sage which had not only abrogated Judaism and Christianity, but also the
Indian religions. This, of course, is not relevant for ancient Greece.
There is a great variety of Muslim attitudes towards the religions of
India not only in literature and theory, but also in practice. Because of
the religious zeal of some of its leaders, above all Mahmud of Ghazna in
the tenth and eleventh centuries, Islam has a bad reputation for its role in
Indian history. Yet, other leaders were more accommodating. They
accepted the Hindus as a protected minority (ahl al-dhimma) who had to
pay a special tax (jizya).
34
Another example which deserves consideration in any account of
Islamic attitudes to paganism is ancient Egypt, above all Pharaoh in the
story of Moses, a biblical legacy which appears prominently in the Koran
where Pharaoh is an almost archetypical unbeliever. As in other cases
mentioned so far, his bad reputation stems from an erroneous attitude,
which becomes evident when he asks Moses: What is this Lord of the
Worlds? (26:23).
35
Sometimes Pharaoh and his people as a Koranic
archetype of the infidel are applied to other groups, for example in a
comparison with the Qara Khitai made by the scholar al-Juwayni (1028
1085), who also described this Central Asian people as fire-worship-
pers.
36
Many polemical writers in medieval Islam accused radical Sufis
in the tradition of Ibn Arabi of considering Pharaoh a believer, an opin-
ion which amounted to heresy.
37
34. Awrangzeb accepted jizya from Hindus (David Cook, Understanding Jihad
[Berkeley, 2005], p. 70); Satish Chandra, Jizya and the State of India during the Seven-
teenth Century, in Richard M. Eaton (ed.), Indias Islamic Traditions, 711-1750 (New
Delhi, 2004), p. 133149. See also Yohanan Friedmann, The Temple of Multan. A Note
on Early Muslim Attitudes to Idolatry, Israel Oriental Studies, 2 (1971), p. 176-182.
35. Translation Abdel Haleem, emphasis mine. Other passages which deal with
Pharaoh are suras 20 and 26 and 7:103124, 10:7483, 29:39.
36. Michal Biran, The Empire of the Qara Khitai in Eurasian History: Between China
and the Islamic World (Cambridge, 2005), p. 172176, at p. 172173. According to Biran,
this description might very well reflect reality, since the god of Fire had an important
place in the pantheon of the Liao Khitans, and lighting great fires was a part of their
important rites such as investitures or royal funerals (p. 173).
37. Alexander D. Knysh, Ibn Arabi in the Later Islamic Tradition: the Making of a
Polemical Image in Medieval Islam (New York, 1999), p. 158161; Eric Ormsby, The
Faith of Pharaoh. A Disputed Question in Islamic Theology, in Todd Lawson (ed.), Rea-
son and Inspiration in Islam: Theology, Philosophy and Mysticism in Muslim Thought,
Essays in Honour of Hermann Landolt (London, 2005), p. 471489. For the interest of the
218 A. AKASOY
Yet, there was also a sympathetic attitude to ancient Egypt among
Muslims interested in alchemy and Hermes who was thought to be an
Egyptian prophet. Furthermore, Egyptian Muslims maintained a strong
sense of local pride which was at least partly based on their grand past.
They found ways to rescue individuals in the entourage of the Phar-
aohs, for example by presenting them as secret monotheists. The Pharaoh
in the story of Joseph is another such positive example.
38
Others adopted
a rigorous attitude and tried to destroy the relics of Egypts pagan past.
39

The ambiguity of these attitudes to ancient Egypt becomes obvious for
instance in the case of the green chapel in Memphis which was
destroyed in 1350. Its spoils were re-used in a Khanqah (a Sufi centre),
which, according to Ulrich Haarmann, symbolised the victory of Islam
over paganism, but also reveals a certain belief in the lasting power of
these relics.
40
To conclude this section, when it comes to the great ancient or pre-
Islamic civilisations, Muslims developed a variety of strategies to incor-
porate previous achievements into their own tradition. When confronted
with the pride of the Persians in their cultural legacy, the Muslim Arab
conquerors of Iraq did not hesitate to celebrate their own pre-Islamic
heritage.
41
The challenge they faced might not have been as substantial
as in the case of medieval Christian authors, since Islam is regarded as
the final revelation of a religion, wisdom or truth which has always been
there. Undeniably polytheistic features were explained away, to borrow
an expression of John Walbridge, or they were simply rejected or con-
sidered abrogated. The philosopher and jurist Ibn Rushd (11261198) in
Spain, for example, declared in his Decisive Treatise (Fal al-maqal),
that Muslims should follow ancient authors in the rational sciences, but
wherever they were wrong their views should be corrected.
Abbasid caliph al-Mamun in the Pyramids see Michael Cooperson, Al-Mamun (Oxford,
2005), p. 14.
38. See the articles by Ulrich Haarmann, Regional Sentiment in Medieval Islamic
Egypt, Bulletin of the School of Oriental and African Studies, 43 (1980), p. 5566, and
Medieval Muslim Perceptions of Pharaonic Egypt, in Antonio Loprieno (ed.), Ancient
Egyptian Literature: History and Forms (Leiden, 1996), p. 605627, and Okasha El-Daly,
Egyptology: the Missing Millennium: Ancient Egypt in Medieval Arabic Writings
( London, 2005).
39. Haarmann, Regional Sentiment in Medieval Islamic Egypt, p. 62.
40. Haarmann, Regional Sentiment in Medieval Islamic Egypt, p. 63.
41. See, for example Rina Drory, The Abbasid Construction of the Jahiliyya: Cultural
Authority in the Making, Studia Islamica, 83 (1996), p. 3349.
III. Religious traditions in sub-Saharan West Africa
Encounters with other religious traditions took place not only in the
Eastern and central parts of the Islamic world, but equally in the farthest
West. Here, the inter-faith dimension of Islam has been very well studied
for a certain area: the coexistence and confrontation of Christians, Mus-
lims and Jews in al-Andalus. The impressiveness of this example has
turned it into a paradigm of multi-religious societies, something of a
historical utopia.
42
Participants in modern inter-faith dialogues of the
Abrahamic religions often stress the privileged position granted to Jews
and Christians as people of the book under Muslim rule in al-Andalus.
Yet, it is difficult to estimate how far this tolerance is based on a reli-
gious ideal or rather on specific historical circumstances: there were no
significant numbers of pagans in al-Andalus who could have served as
an example of the treatment given to non-Muslims who were neither
Christians nor Jews. Sub-Saharan Africa on the other hand offers good
possibilities for such a comparison, especially since it formed part of the
same empires as al-Andalus under the Almoravids and the Almohads
(eleventh until thirteenth centuries). It is also here that we find an aspect
of paganism which at least for the medieval period has remained
largely neglected in modern scholarship.
43
Our knowledge of the early history of Islam in West Africa, i.e. the
Senegal and Niger region, is very limited due to a considerable lack of
sources. Islam spread as early as the mid-seventh century in North Africa
and from there into the Sahara and sub-Saharan Africa. In the eleventh
century the Almoravids dominated the region of the former kingdom of
Ghana, and Songhay and Hausaland were Islamised in the fourteenth
century. Scholars often emphasise the reciprocal character of this devel-
opment and speak of an Islamisation of Africa and an Africanisation of
Islam.
44
The main agents were merchants and rulers who were Muslims
42. Also present in scholarly literature, see Mara Rosa Menocal, The Ornament of the
World: How Muslims, Jews and Christians Created a Culture of Tolerance in Medieval
Spain (Boston, 2002).
43. Most of the primary sources referred to in this article can be found in English
translation in Nehemia Levtzion and J.F.P. Hopkins, Corpus of Early Arabic Sources for
West African History (Cambridge, 1981; Princeton,
2
2000).
44. Nehemia Levtzion, Patterns of Islamization in West Africa, in idem (ed.), Con-
version to Islam (New York, 1979), p. 207216 (reprint in idem, Islam in West Africa:
Religion, Society, and Politics to 1800 [Aldershot, 1994]); for Africanisation/Islamisation
cf. also David Robinson, Muslim Societies in African History: New Approaches to African
History (Cambridge, 2004), p. 2759.
220 A. AKASOY
long before the majority of the population adopted this religion.
45
In later
times, the zwaya, clerical tribes, would play a decisive role in spreading
revivalist ideas from the Middle East and North Africa southwards.
46
In
some regions Muslims remained a minority under non-Muslim rulers,
and the Suwarian tradition is an excellent example of the pragmatic atti-
tude Muslims developed in such situations.
47
Because Islamic and non-
Islamic traditions have coexisted in large areas for a long time, Islam in
Africa, not unlike Islam in South Asia, has the reputation of having a
pronounced syncretistic element. We find statements to this effect
already in medieval Arabic sources.
48
Unlike India, West Africa was not part of the Hellenised world. Mus-
lim authors sometimes adapted the cultural sphere created by Alexander
to the geography of the Islamic world and had the great ruler also travel-
ling to Africa and Spain,
49
but West Africa never enjoyed the same rep-
utation as a land of learning as India did. There were no translations of
local (oral or written) texts, for example, and it is no coincidence that the
famous expression of the Prophet runs: seek knowledge, even as far as
China, and not as far as Ghana or, for that matter, as far as Oxford
which, for the medieval Arabs, lay in an equally uncivilised part of the
world.
Compared to Greece, India and Egypt, West Africa seems to have
been of little interest as a land of ancient civilisation to medieval Muslim
45. Levtzion, Patterns of Islamization in West Africa, p. 209210. Islam did not
spread among stateless peoples, even where geographical conditions were the same
(p. 210). Chiefs found themselves in a difficult position between an influential Muslim
minority, living close to the center, monopolizing the trade, and having extensive outside
relations, and the majority of the pagan subjects. The way out of this dilemma was for
them to maintain a middle position between Islam and paganism, to be neither real Mus-
lims nor complete pagans (p. 211). Cf. the observation in the anonymous twelfth-century
Kitab al-istibar that the inhabitants of the town of Kugha are Muslims, but those of the
country are polytheists (Corpus, p. 148; ed. Sad Zaghlul Abd al-amid [Alexandria,
1958], p. 222). The role of chiefs and traders in the spread of Islam has also been observed
by scholars of modern West Africa, cf. Adeline Masquelier, Lightning, Death and the
Avenging Spirits: Bori Values in a Muslim World, Journal of Religion in Africa, 24
(1994), p. 251, at p. 811. For general problems of conversion cf. also Richard Eatons
introduction in idem (ed.), Indias Islamic Traditions, 711-1750 (New Delhi, 2004).
46. Philip D. Curtin, Jihad in West Africa: Early Phases and Inter-Relations in
Mauretania and Senegal, Journal of African History, 12 (1971), p. 1124, at p. 11.
47. See Robinson, Muslim Societies in African History, chapter 9: Asante and Kumasi:
A Muslim Minority in a Sea of Paganism (p. 124138).
48. See note 76 below for Yaquts statement.
49. For such legends cf. Julia Hernndez Juberas, La Pennsula imaginaria: Mitos y
leyendas sobre al-Andalus (Madrid, 1996). H.T. Norris, Saharan Myth and Saga (Oxford,
1972), p. 3133 for Dhu l-Qarnayn/Alexander in Africa.
writers. But how about the more contemporary religious traditions Mus-
lims actually encountered there? Unfortunately, there is nothing in Ara-
bic literature on West Africa that would equal al-Birunis famous book
on India. Likewise, authors who wrote on different Muslim and non-
Muslims sects (such as Shahrastani in the twelfth century) did not dis-
play any interest in the southern regions. The most important source for
medieval Arab Muslim attitudes to African religions are geographical
treatises.
50
This genre emerged in the ninth century and combined infor-
mation from earlier literature with travel reports and knowledge stem-
ming from trade, pilgrimage and military campaigns. The authors had
high-ranking patrons, mostly in the central lands of the Arab world, and
wrote for an educated audience eager to obtain information that was curi-
ous, thrilling, sometimes outright bizarre. Widely known topoi were
often enriched with new details. The descriptions of local religions never
seem to merit attention for their own or even primarily a religious sake.
They appear rather as short pieces of information in more general
accounts of places and tribes among other peculiarities of their culture
such as clothes, food, social and legal customs, trading goods and ani-
mals. Often the purpose of the authors might have been to emphasise the
exotic or marginal character of a certain group of people.
Some of the geographers actually travelled to the regions they
described, or at least close to them they would enquire, for example, in
market places about the customs of people living in more distant regions.
An important source, for instance, is the well-known fourteenth-century
traveller Ibn Baua.
51
Others were armchair-geographers who derived
their knowledge from books. Readers or listeners of such texts in the
urban milieu of the medieval Middle East presumably had general ideas
about how people at the fringes of the known world behaved the more
exotic, the better. On the other hand, if the fanciful element appeared
exaggerated, listeners may have become suspicious as to the accuracy
and credibility of the account. An interesting illustration of such a case,
here from the author himself, can be found in the Sicilian geographer
al-Idrisi, who remarked, on account of a castle in Tibet which causes
50. Levtzion and Hopkins, Corpus; Nehemia Levtzion, Ibn-awqal, the Cheque, and
Awdaghost, Journal of African History, 9 (1968), p. 223233 (reprinted in Levtzion,
Islam in West Africa).
51. Ibn Baua has been the subject of one of the small number of publications rele-
vant for our subject: Nol King, Reading between the Lines of Ibn Baua for the History
of Religions in Black Africa, Milla wa-Milla, 19 (1979), p. 2633. See also Said Hamdun
and Nol King, Ibn Battuta in Black Africa (Princeton, 1994).
222 A. AKASOY
deadly laughter: I believe that this story is not true. It is made up. Nev-
ertheless, it is a precise and well known account.
52
The Islamic geographical tradition was significantly influenced by the
Greek division of the world into different climates and the impact of
these climates and the heavenly bodies on the character of peoples.
53
As
a Greek legacy, the Mediterranean zone was regarded as the most favour-
able climate for the formation of civilisations, but those who lived fur-
ther away, were less lucky. The following accounts on Africa in Arabic
literature are mostly borrowed from Ptolemys Tetrabiblos. aid al-
Andalusi in eleventh-century Toledo, included in his Book of the Catego-
ries of the Nations (abaqat al-umam) a chapter on those who do not
practise any sciences. He wrote:
For those peoples who live near and beyond the equinoctial line to the
limit of the inhabited world in the south, the long presence of the sun at the
zenith makes the air hot and the atmosphere thin. Because of this their
temperaments become hot and their humours fiery, their color black and
their hair woolly. Thus they lack self-control and steadiness of mind and
are overcome by fickleness, foolishness, and ignorance. Such are the blacks,
who live at the extremity of the land of Ethiopia, the Nubians, the Zanj and
the like.
54
Al-Hamdani, a Yemeni author of the tenth century, offers in his Des-
cription of the Arabian Peninsula (ifat Jazirat al-Arab) the astronomical
52. For the Arabic text see Roberto Rubinacci, Il Tibet nella Geografia dIdrisi, in
Gururajamaarika. Studi in onore di Giuseppe Tucci (Naples, 1974), I, p. 195220, 17.
See Anna Akasoy, Tibet in Islamic Geography and Cartography: A survey of Arabic and
Persian Sources, in Anna Akasoy, Charles Burnett and Ronit Yoeli-Tlalim (eds), Islam and
Tibet: Cultural Interactions along the Musk Routes (Farnham, 2010), p. 17-41.
53. In general see Andr Miquel, La gographie humaine du monde musulmane
jusquau milieu de 11e sicle (Paris, 1967); J.B. Harley and David Woodward (eds.), The
History of Cartography (Chicago, 19871994); Fuat Sezgin, Geschichte des arabischen
Schrifttums, xxii (Leiden, 2000).
54. Quotation from Bernard Lewis, Race and Slavery in the Middle East: an Histori-
cal Enquiry (New York, 1990), p. 4748. abaqat al-umam, ed. L. Cheikho (Beirut,
1912), p. 9. Ptolemys wording is: the people who live under the more southern paral-
lels, that is, those from the equator to the summer tropic, since they have the sun over their
heads and are burned by it, have black skins and thick, woolly hair, are contracted in form
and shrunken in stature, are sanguine of nature, and in habits are for the most part savage
because their homes are continually oppressed by heat; we call them by the general name
Ethiopians. Ptolemy, Tetrabiblos, ed. and trans. F.E. Robbins (London, 1980), II.2
(p. 12123). See Koen Vanhaegendoren, Das afrikanische Volk der Ataranten: Zur eth-
nographischen Tradition der Antike (Hamburg, 1998), p. 1415 for the term Ethiopians,
who are rendered as zanj here. For matters of ethnic prejudice see also Akbar Muhammad,
The Image of Africans in Arabic Literature: Some Unpublished Manuscripts, in Slaves
and Slavery in Muslim Africa, I, ed. John Ralph Willis (London, 1985), p. 47-74.
explanation for the character of the inhabitants of Africa as we can find
it in the Tetrabiblos. The naked blacks,
55
he says, are under the influ-
ence of Venus and Mars. Venus shows her effect in the female attitude
of the male population, whereas Mars gives them a manly inner nature.
Evil and wicked they are, liars and cheats, cunning and murderous with
a secret hate.
56
In the Tetrabiblos, the influence of Mars on these people
makes them virile of spirit, rascally, magicians, impostors, deceivers,
and reckless.
57
According to al-Hamdani, in another area of Africa
where people are influenced by Scorpio and Mars, their moral conduct
has become more like that of savage beasts than that of men. They have
become people who squabble, who harbour enmities and disputes and
things which are hateful and abhorrent. They deem life to be of little
worth and they are not compassionate among one another. They do not
show care and affection for each other and sometimes they are cruel to
themselves.
58
We find such characteristics frequently in descriptions of
Africa in Arabic literature. By their very nature, it seems, these people
do not possess the intellectual and moral disposition for accepting Islam.
This geographical framework was also employed by authors who
wrote on other subjects (as is already evident in aid al-Andalusis pas-
sage). The Fatimid propagandist amid al-Din al-Kirmani (d. 1021), for
instance, in his The Repose of the Intellect (Raat al-aql), an abstract
exposition on Ismaili philosophy, dismissed the Turks, Zanj, Berbers,
and their like as by their nature uninterested in acquiring intellectual
knowledge or understanding religious truth.
59
One aspect of the notion of
the pagan is thus the barbarian who lacks any sense of propriety or
understanding the truth. The notion might also reveal a certain sense of
ethnic superiority on the side of Arab Muslims, even if the nature
referred to above had more to do with the climate in which a person was
born than with his or her genes.
55. Nakedness is also mentioned by Masudi in his Meadows of Gold (Muruj al-dha-
hab) (Corpus, p. 31; ed. A.C. Barbier de Meynard and Pavet de Courteille [Paris, 1865],
II, p. 383) and many others. See below.
56. H.T. Norris, The Berbers in Arabic Literature (London, 1982), p. 7. Al-Hamdani,
Geographie der Arabischen Halbinsel, ed. David Heinrich Mller (Leiden, 1884),
p. 40-41.
57. Ptolemy, Tetrabiblos, ed. and trans. Robbins, II.3 (p. 153).
58. Norris, Berbers in Arabic Literature, 7. Al-Hamdani, Geographie der Arabischen
Halbinsel, ed. Mller, 40-41. Compare Ptolemy, Tetrabiblos, ed. and trans. Robbins, II.3
(p. 153).
59. Kirmani, Raat al-aql, ed. M. Kamil Hussein and M. Mustafa Hilmy (Cairo,
1953), p. 241. Translation from Lewis, Race and Slavery in the Middle East, p. 55.
224 A. AKASOY
Islam is often described as colour-blind and the global community of
believers as one where no distinction is made between believers of dif-
ferent ethnic groups, yet and despite the egalitarian achievements of the
Abbasid revolution, it is difficult to deny altogether that Arabs maintain
a privileged position. Muhammad was one of them, the Koran revealed
in clear Arabic, and according to most medieval Muslim authors the
Caliph had to be a member of the prophets own tribe. Many non-Arab
Muslims have to make an extra effort to learn the language which allows
them to read the word of God. Some, who defend very strict positions,
even claim that being a Muslim without knowledge of Arabic is impos-
sible. Such views are often referred to as takfir al-awamm (declaring
simple people unbelievers). Even though there is little clear evidence, we
cannot altogether exclude the possibility that there was sometimes an
ethnic bias involved when Black Africans were declared pagans.
In the texts under consideration here, such an ethnic bias is not spelled
out, although it might have been tacitly present. But the accounts of
African cultures were not always negative. In many Arabic geographical
texts we find neutral or positive descriptions. Al-Hamdani, for example,
even though he depicted the inhabitants of Africa as reversing the ideal
of Islamic manners, mentioned their religious attitudes without critical
comments. Instead, they share the qualities of noble savages. Of the
Berbers in the lands of the Umayyads who are under the influence of
Jupiter and Pisces, he writes: They are free and independent and they
show loyal support and affection for one another. () They love to toil
and work. They are neither servile nor submissive. They are grateful and
they revere Jupiter. They adore and exalt him. They bow down before
him and they name him Ammun.
60
This passage is again literally taken
from Ptolemys Tetrabiblos, but the impartial tone of the passage is not
a unique case. We can find equally matter-of-fact accounts in Arabic
literature which are not borrowed from Greek sources. Without any crit-
icism attached, al-Muahhar ibn ahir al-Maqdisi in his Book on Crea-
tion and History (Kitab al-bad wal-tarikh; written 966), indicates
briefly the names given to superior deities by Africans and Turks:
The Zanj say that their expressions m.l.k.w.y and j.l.w.y mean the greatest
lord, and when the Turks say bir tengri, they mean the lord is one.
61
60. Norris, Berbers in Arabic Literature, p. 7. Al-Hamdani, Geographie der Ara-
bischen Halbinsel, ed. Mller, p. 40-41; ed. al-Akwa, p. 7677. Ptolemy, Tetrabiblos, ed.
and trans. Robbins, II.3 (p. 153).
61. Maqdisi, Le Livre de la creation et de lhistoire, ed. and trans. Clment Huart
(Paris, 18991919), I, p. 63 (Arabic text). I am grateful to James Montgomery for having
The anonymous History of the Ages and Those whom Events have
Annihilated (Akhbar al-zaman wa-man abadahu l-idthan), written
between the end of the tenth and the beginning of the eleventh century,
contains passages in which the religious traditions of several tribes from
among the Sudan are described in a neutral way. One example is a
description of a rain-making ritual:
They often suffer from drought. When they wish to make rain, they assem-
ble bones and pile them up like a mound, which they set on fire, and walk
round, raising their hands to the sky, and utter some words. Then it rains,
and they have water.
62
A more pious story on a similar subject is told by al-Bakri, an Anda-
lusian geographer who compiled his best-known work, The Book of the
Highways and Kingdoms (Kitab al-masalik wal-mamalik), in 1068. Al-
Bakri was a typical armchair-geographer who collected his information
while in Spain, either from written sources (i.e. earlier geographical trea-
tises or travel reports) or from contemporary oral sources. In one of the
sections on sub-Saharan Africa in his book, he deals with a country
called Malal whose ruler has the title al-musulmani. According to al-
Bakri, he received the title under the following circumstances: his people
had been suffering for several years under a severe drought. Their sacri-
fices of cattle had no effect, and their misery became only worse. At that
time the king had a guest who was a Muslim, knew the Koran and was
familiar with the sunna. He told the king that if he became a Muslim
himself, he would pray for him and his people. The king converted, ful-
filled purifying rituals and started praying with the man. The dawn had
just started to break, al-Bakri continues his account, when God caused
abundant rain to descend upon them. So the king ordered the idols (daka-
kir) to be broken and expelled the sorcerers (saara) from his country.
He and his descendants after him as well as his nobles were sincerely
attached to Islam, while the common people of his kingdom remained
polytheists (ahl mamlakatihi mushrikun). Since then their rulers have
been given the title of al-musulmani.
63
drawn my attention to the relevant passage in Ibn Falans text, see Montgomerys trans-
lation Ibn Falan and the Caliphal Mission through Inner Asia to the North: Voyaging
the Volga (available on http://wonka.hampshire.edu/abbasidstudies/pdf/Ibn-Fadlan.pdf;
accessed on 3 Dezember 2007), p. 200 a.1.
62. Corpus, 35; ed. Abdallah al-awi (Beirut, 1966), p. 8788.
63. Corpus, p. 8283; Al-Masalik wal-mamalik, ed. Baron MacGuckin de Slane
(Algiers, 1911), p. 178 and ed. A.P. van Leeuwen and A. Ferre, 2 vols (Tunis, 1992),
no. 1464 = II, p. 875876.
226 A. AKASOY
The texts we have referred to so far already give a good impression of
the kind of material at our disposal. Some authors might have been closer
to the area we are dealing with here, but this did not prevent them from
using earlier written sources. There are also differences in the purposes
they wrote for and the audiences they addressed. Their aim was not sim-
ply to give a truthful account of an area, but also to entertain the listener
by playing with information, images, motifs, narrative elements, etc. that
were already known. The two aims are of course not mutually exclusive.
Some authors such as al-Hamdani had very specific interests when they
wrote their texts. It is thus important to bear in mind that geographical
texts or references should not simply be taken at face value as a straight-
forward reflection of relations between the Arab world and West Africa,
but rather have to be considered within the larger literary tradition.
As already mentioned, some of the information the geographers dis-
played was based on earlier treatises, often translations of Greek texts or
later Arabic adaptations, supplemented by the reports of travellers and
traders. But the authors might have been influenced by other literary and
cultural models. An obvious model could have been pre-Islamic Arabia
with its nomadic society, its adventurous heroes, good camel-riders with
their endurance, pride and generosity, and the hardships of desert life,
but also its polytheistic and idolatrous religions, its brutality and sexual
immorality al-Hamdani, for example, in the text quoted above, points
out the many cases of adultery (zina) among the Africans (a subject of
interest too for Ibn Baua). An example which suggests that Muslims
transferred elements of Arabia into the Sahara is the name of Tadmakka,
the Berber expression for there is Mecca, the settlement Essouk in
present-day Mali.
64
Whether this has primarily geographical, or also cul-
tural and historical implications, is not clear to me. Usually, the Jahiliyya
appears to be more of a subtext than an explicit point of reference. Both
are characterised by their Otherness from an Islamic point of view.
64. John O. Hunwick, Sharia in Songhay: the Replies of al-Maghili to the Questions
of Askia al-ajj Muammad (New York, 1985), p. 6 n. 2: al-Marrakushi, al-Ilam bi-man
alla Marrakush wa-Aghmat min al-alam (Fez, 1936-9), III, p. 222, refuting the explana-
tion that the name means resembling Mecca, cf. al-Bakri in Corpus, p. 85 (ed. de Slane,
p. 181) and Kitab al-istibar, Corpus, p. 149 (ed. Abd al-amid, p. 223). For a compari-
son of Awdaghost and Mecca see Ibn awqal, Kitab urat al-ar (Corpus, p. 46; ed.
J.H. Kramers, 2 vols. [Leiden,
2
193839], I, p. 92), al-Idrisi, Nuzhat al-mushtaq fi ikhtiraq
al-afaq (Corpus, p. 118; ed. R. Dozy and M.J. de Goeje [Leiden, 1866], p. 32) and Yaqut,
Mujam al-buldan (Corpus, p. 168; ed. F. Wstenfeld, 6 vols. [Leipzig, 186673], I,
p. 399400).
Another option was to apply earlier Islamic divisions of religions.
Thus, the inhabitants of West Africa are often called Majus Zoroastri-
ans.
65
At first sight, this association of religious traditions from medieval
West Africa and ancient Iran might look rather odd. However, it was not
the only case in which the Arab Muslims applied the term Majus detached
from its regional origin. They identified the Vikings as Majus, in particu-
lar perhaps because of the prominence of the fire cults, but also because
of their funeral rites.
66
As far as I can see, none of the medieval Arab geographers made the
connection between African Majus and fire. Only the geographer and
traveller Ibn awqal (who wrote in 988) explained that among the north-
ern people of the Rus (who were categorised as Majus) wealthy men
were buried with their slave girls and that the people of Ghana and
Kugha as well as the Indians did likewise.
67
It has been suggested that
Ibn awqal collected most of his information about the Sahara and the
Sudan during his stay in Sijilmasa without ever travelling farther than
that.
68
His account of the customs in Ghana and Kugha was therefore
probably not based on first-hand knowledge, but might have even been
inspired by what he knew about the Rus from geographical literature.
This is even more likely to have been the case with a later geographer,
al-Dimashqi (12561327), who brought different and sometimes contra-
dictory (naked vs. clothes) threads together in his description of the
Zaghawa:
They are naked pagans whose religion (din) is majusiyya. They worship
idols (awthan) which they call dakakir. One of their customs which they
follow and upon which they rely in their government is that when one dies
65. Cf. the entry Majus in Levtzion and Hopkins (eds.), Corpus.
66. For Arabic literature on the Vikings see W.E.D. Allen, The Poet and the Spae-
Wife: an Attempt to Reconstruct al-Ghazas Embassy to the Vikings (Dublin, 1960) and
the articles by James Montgomery Ibn Falan and the Rusiyyah, Journal of Arabic and
Islamic Studies, 3 (2000), p. 125; Ibn Rustas Lack of Eloquence, the Rus, and Sama-
nid Cosmography, Edebiyat, 12 (2001), p. 7393; Serendipity, Resistance, and Multiva-
lency: Ibn Khurradadhbih and his Kitab al-Masalik wa-l-mamalik, in Philip F. Kennedy
(ed.), On Fiction and Adab in Medieval Arabic Literature (Wiesbaden, 2005), p. 177232,
and Spectral Armies, Snakes, and a Giant from Gog and Magog: Ibn Falan as Eyewit-
ness Among the Volga Bulghars, The Medieval History Journal, 9 (2006), p. 6387. For
Zoroastrianism cf. Shaul Shaked, Some Islamic Reports Concerning Zoroastrianism,
Jerusalem Studies of Arabic and Islam, 17 (1994), p. 4384; J. Christoph Brgel,
Zoroastrians as Viewed in Medieval Islamic Sources, in Waardenburg (ed.), Muslim
Perceptions of Other Religions, p. 202212; Stepaniants, The Encounter of Zoroastrian-
ism with Islam.
67. Corpus, p. 52; ed. Kramers, I, p. 397.
68. Levtzion, Ibn awqal, the Cheque and Awdaghost.
228 A. AKASOY
they bury with him those who are most closely related to him and most
beloved of him, as well as his clothes and weapons, just as we have said of
the Slavs.
69
The discussion of the Slavs might thus have influenced the account of
the Zaghawa. Another similarity between the Zoroastrians and some
Africans is the veneration of trees mentioned in the Akhbar al-zaman,
although Zoroastrian tree worship does not seem to be important in Ara-
bic literature.
70
It seems unlikely that the Arabs explained the Zoroas-
trian presence in West Africa as a legacy of the Viking raids in Spain,
assuming, for example, that the men from the north had travelled even
further south and spread their religious customs there. Geographers may
simply have given the label Majus to all those they could not identify
otherwise.
Identifying the followers and practitioners of the religious cults in
West Africa as Majus leaves us again with the question: who are the
pagans? What about those who could neither be identified as followers
of an Abrahamic religion or as Majus? In Arabic geographical literature
there is no term for these people. They were described as unbelievers,
polytheists and idolaters, but so were others (Christians and Majus). It is
only for later centuries and specific regions that we find other terms.
David Robinson describes how in a case around 1500, the older Hausa
religious practices (bori) were considered inferior, but people in the
countryside followed this tradition and consulted a whole range of spir-
its. Apparently, some of the spirits had names which revealed a Muslim
background such as Mallam Alhaji (Mallam = Arabic muallim, i.e.
teacher; Alhaji = Arabic al-ajji, i.e. pilgrim) and Mallam Alkali (Alkali
= Arabic al-qai, i.e. judge). Their presence Robinson concluded,
showed how deeply Islam had taken root in the local society on Hausa
terms. Muslim authorities in the cities often called the practitioners of
Bori by the name for the old priestly caste of the Zoroastrians of Persia:
Maguzawa, Magians. By constructing an analogy to this people of
the book, the scholars transformed these pagans into a dhimmi com-
munity within the Dar al-Islam.
71
This seems to suggest that with bori there was at least an alternative
term or concept that was more specific than non-Muslim, but I have not
69. Nukhbat al-dahr fi ajaib al-barr wal-bar (Corpus, p. 213; ed. M.A.F. Mehren
[St Petersburg, 1866; reprint Leipzig, 1923, p. 268).
70. Corpus, p. 35; ed. al-awi, p. 88.
71. Robinson, Muslim Societies in African History, p. 141.
been able to find out when this term was actually first used, and in any
case the evidence seems altogether rather late. A modern anthropologist
has pointed out that Muslim Hausa call non-Muslim locals Maguzawa
or Arna, but that the latter term is rejected by some groups because of
its negative connotation as paganism.
72
Again, when the term Arna
has been first used, is unknown to me. The origins of this development
of terminologies and categories seem to lie in much earlier centuries.
This might suggest that local religious traditions were first wholeheart-
edly accepted as Majusiyya, but I am unable to tell whether and at what
point in history and under which circumstances a possible backlash
might have taken place against the identification of followers of local
religions as ahl al-dhimma. At any rate, it appears we are dealing here
with two notions of paganism: an accepted, labelled one, and another
one that is neither identified nor accepted. In either case, the term is used
as a description from the Muslim point of view, and it involves religious
identity as well as a position in society.
Then again, it seems, this African Majusiyya was not necessarily con-
nected with specific beliefs. Al-Bakri described the inhabitants of ang-
hana (situated on the Nil al-Sudan) as Majus, who worshipped dakakir
(dakkur being their word for idol), until they were forced to convert to
Islam.
73
The twelfth-century geographer al-Idrisi, on the other hand,
speaks of the Tajuwa people, who are Majus, and do not believe in
anything (wa-hum majus la yataqiduna shayan).
74
Zoroastrianism
appears thus more as a certain set of rituals than a set of beliefs. Some-
thing else that could be implied here, is an ethnic or group identity.
Another piece of evidence for the vagueness which permeates this dis-
course of African paganism is an occasional confusion of Majusiyya and
Christianity in the sources.
75
72. Matthias Krings, Geister des Feuers: Zur Imagination des Fremden im Bori-Kult
der Hausa (Hamburg, 1997), p. 22.
73. Corpus, p. 77 (ed. de Slane, p. 172). The same passage appears in the Kitab al-
istibar (Corpus, p. 144; ed. Abd al-amid, p. 217). The text also describes, based on
al-Bakri, the surroundings of the royal palace in al-Ghaba, where the sorcerers (saara)
and men of the kings religion (ahl diyanatihi) live and were the idols (dakakir) and tombs
of the kings are found. There is, however, also a mosque for the Muslim merchants who
come to the kings court (Corpus, p. 147; ed. Zaghlul, p. 220). Further references see the
entries dakakir and anam and wathan (idol) in the index of the Corpus.
74. Corpus, p. 114 (translation modified); ed. Dozy and de Goeje, p. 13.
75. Nehemia Levtzion, The Almoravids in the Sahara and Bilad al-Sudan: a Study in
Arab Historiography, Jerusalem Studies in Arabic and Islam, 25 (2001), p. 133152, at
p. 144.
230 A. AKASOY
Some authors mention sun worship without labelling it in any way,
perhaps a Ptolemaic legacy.
76
Al-Bakri and the anonymous Kitab al-ist-
ibar (completed in its present form in 1191) describe the case of a
female idol in the land of Damdam, where cannibals live.
77
This passage
is not accompanied by any critical comments. Occasionally we find
descriptions of rituals which must have appeared bizarre and exciting to
a reader or listener in the main lands of the Arab world. Al-Bakri, for
example, offers the following account about the Zafqu:
They are a nation of Sudan who worship a certain snake, a monstrous ser-
pent with a mane and a tail and a head shaped like that of a Bactrian camel.
It lives in a cave in the desert. At the mouth of the cave stands a trellis and
stones and the habitation of the adepts of the cult of that snake. They hang
up precious garments and costly objects on the trellis and place plates of
food and cups of milk and intoxicating drink there. When they want the
serpent to come out to the trellis they pronounce certain formulas and whis-
tle in a particular way and the snake emerges. When one of their rulers dies
they assemble all those whom they regard as worthy of kingship, bring
them near the cave, and pronounce known formulas. Then the snake
approaches them and smells one man after another until it prods one with
its nose. As soon as it has done this it turns away towards the cave. The one
prodded follows as fast as he can and pulls from its tail or its mane as many
hairs as he is able. His kingship will last as many years as he has hairs, one
hair per year. This, they assert, is an infallible prediction.
78
The presence of snakes in West Africa caught the attention of many a
geographer from the main lands of the Islamic world. In the anonymous
76. Yaqut quotes al-Muhallabi (d. 990) who says about the inhabitants of Awdaghost:
They have been converted to Islam by the Mahdi Ubayd Allah, for previously they were
infidels, and worshipped the sun, and used to eat carrion and blood (Corpus, p. 168; ed.
Wstenfeld, I, p. 400). He makes a similar comment about the Madasa, a anhaja people:
Some of them are infidels and some Muslims. Those who are infidels are in a state of
ignorance (jahiliyya). They eat carrion and venerate the sun and yet fear [to commit]
injustice. They intermarry with the Muslims. They and the majority of the Muslims are
uncouth savages. (Corpus, p. 174; ed. Wstenfeld, IV, p. 919). In this case, Islam is
mixed into pure paganism and thereby blemished. Muslims are forbidden to eat carrion or
blood.
77. Al-Bakri (Corpus, p. 8687; ed. de Slane, p. 183) and Kitab al-istibar (Corpus,
p. 151; ed. Abd al-amid, p. 225), also earlier in the Akhbar al-zaman (Corpus, p. 36;
ed. al-awi, p. 89). See also Ibn Said who mentions the Lamlam, abandoned unbelievers
who eat men The Lamlam are rustics and in books are said to have a settlement more
like a village than a town where they have the house of their dakakir, that is idols
(awthan). Kitab bas al-ar fi l-ul wal-ar (Corpus, p. 184; ed. J. Vernet Gins
[ Tetouan, 1958], p. 24).
78. Corpus, p. 7879; ed. de Slane, p. 173174.
Kitab al-istibar, we find the following additional remark concerning
this episode:
I say: This superstition exists amongst them simply because the serpent
may live for more than a thousand years. Their ancestors were brought up
according to this custom and do not know the origin of it. The one who
established this superstition wished simply to rule them thereby. Their
intellects are exceedingly feeble, so he could do with them what he liked.
God preserve us from superstition!
79
This comment resembles the disclosure of similar tricks by Christians
and Indians who make simple people believe they are witnessing mira-
cles. Another trick is described by Yaqut who describes the worship
of kings in the kingdom of the Zaghawa. The author employs a rather
matter-of-fact tone:
Their houses are all reed huts as is also the palace of their king, whom they
exalt and worship instead of Allah. They imagine that he does not eat any
food. There are persons who have charge of this food secretly and bring it
to his house. It is not known where it is brought from and if it happens that
one of his subjects meets the camels carrying his provisions, he is killed
instantly on the spot Their religion is the worship of their kings, for they
believe that they bring life and death, sickness and health.
80
In most cases, however, as far as I can see, the followers of local reli-
gions are not classified in any specific manner. They are often kuffar, but
this does not mean much more than non-Muslims. In several cases Mus-
lims even emphasised the absence of any religious tradition, perhaps
because their terminologies and concepts did not cover the phenomena
observed in and reported from West Africa.
Medieval Muslim geographers made an interesting observation when
they pointed out the connection between statelessness and paganism. The
tenth-century geographer Iakhri, for example, explains in his Kitab al-
masalik wal-mamalik:
We have not mentioned the land of the Sudan in the west, nor the Buja nor
the Zanj, nor other peoples with the same characteristics, because the
orderly government of kingdoms is based upon religious beliefs, good
79. Corpus, p. 146; ed. Abd al-amid, p. 219.
80. Mujam al-buldan (Corpus, p. 171; ed. Wstenfeld, II, p. 932). For dogs eating
sacrificial meat which the Rus then believe to have been eaten by their gods see the
account of Ibn Falan, James Montgomery, trans., Ibn Falan and the Caliphal Mission
through Inner Asia to the North: Voyaging the Volga, p. 210 b.3.
232 A. AKASOY
manners, law and order. () These people lack all these qualities and have
no share in them.
81
For them, the lack of government, just like the eating of carrion and
the nakedness, might have been part and parcel of being the other, but
modern historians of West Africa have observed that the spread of Islam
depended indeed on the degree of political organisation. People without
obvious political organisation seem to have counted automatically as
pagans. Likewise, according to the fifteenth-century scholar al-Maghili,
Askia Muammad had the right to lead a jihad against people without an
amir, who should not be considered Muslims.
82
Being a Muslim could have sometimes simply meant being a subject
to a Muslim ruler. We should probably distinguish here, as Yohanan
Friedmann has done for India, between the submission to a Muslim state
or army and to the Islamic faith.
83
As Nehemia Levtzion pointed out, the
submission to an Islamic ruler might have reduced the level of some-
ones paganism: in Dagbane, the term chefera (from the Arabic kafir)
is reserved for the stateless Konkomba and Bassari, who are completely
untouched by Islam.
84
The identification with paganism was not
reserved for stateless people. Pagans could have been those who were
subject to a pagan king. Al-Bakri, for example, distinguishes between
the Muslims and the followers of the kings religion (ahl din al-malik)
in Ghana.
85
Political loyalty offered thus another option for describing
religious affiliations if the terminology was found wanting by medieval
Arab geographers.
The close connection between the religion of a ruler, the status of
his people, and the form of political organisation is most evident in
legal debates around slavery. According to Islamic law, free Muslims
cannot be enslaved. Over the centuries, more and more people in West
Africa declared themselves Muslims, but the slave raids continued and
the lack of a proper Islamic political organisation often served as an
excuse.
81. Corpus, p. 40; ed. M.J. de Goeje (Leiden, 1870), p. 4.
82. Hunwick, Sharia in Songhay, p. 27 and p. 81.
83. Yohanan Friedmann, A Contribution to the Early History of Islam in India, in
Myriam Rosen-Ayalon (ed.), Studies in Memory of Gaston Wiet (Jerusalem, 1977), p. 322.
84. Levtzion, Patterns of Islamization in West Africa, p. 214; idem, Muslims and
Chiefs in West Africa: a Study of Islam in the Middle Volta Basin in the Pre-Colonial
Period (Oxford, 1968), p. 108109.
85. Corpus, p. 80 (ed. de Slane, p. 175) and Kitab al-istibar, the people who follow
his (i.e. the kings) religion (Corpus, p. 147; ed. Abd al-amid, p. 220). See also
Levtzion, Patterns of Islamization in West Africa, p. 209, n. 7.
The connections between slavery and paganism are manifold. Slaves
and pagans lack personal dignity and are closer to animals than to civi-
lised men. Paganism is the condition of the enslavable who needs to be
fought, conquered and educated. Attitudes to slavery and to paganism in
medieval Islam reveal a similar sense of ethnic and cultural superiority
among the Arabs, even though this was by no means exclusively directed
at black people. The curse of Ham, the legend according to which God
had cursed black people with their colour and with slavery and which
served as a justification for the enslavement of black people, was known
in the Arab world, albeit not universally accepted.
86
In Muslim protests
against the practice of slavery, the insistence of Ahmad Baba in seven-
teenth-century Songhay that the cause of slavery is unbelief went along
with a plea for establishing the definition of paganism along the lines of
religious affiliation.
87
IV. Muslims as pagans
If we return from paganism to the more general term kufr or unbelief,
we find that its use in polemics was not limited to members of other
religions, it was also regularly used against fellow Muslims.
88
By going
back to its etymological roots, the Andalusian Sufi Ibn Arabi used it to
describe an attitude where one veils or hides something.
89
In the Muslim
West a number of examples illustrate that the reproach of kufr against
Muslims often emerged in an intercultural milieu. One of the most prom-
inent defenders of Almoravid Malikism, the Qai Iya (d. 1149), pre-
sented a whole set of non-Muslim habits the adoption of which consti-
tutes kufr billah for a Muslim, including worshipping an idol, the sun,
the moon, the cross or fire, going to church or to a monastery with
those who attend them or wearing the clothes of monks. It is the
86. Goldenberg, The Curse of Ham, p. 171 and n. 15 for Ibn Khaldun who rejected the
legend.
87. Miraj al-Suud: Ahmad Babas Replies on Slavery, ed. and trans. John Hunwick
and Fatima Harrak (Rabat, 2000), p. 11; Bernard Barbour and Michelle Jacobs, The
Miraj: a Legal Treatise on Slavery by Ahmad Baba, in John Ralph Willis (ed.), p. Slaves
and Slavery in Muslim Africa (London, 1985), p. 125159.
88. Curtin, Jihad in West Africa. About groups in partly Islamised regions: While
their leadership sometimes urged holy war against the unconverted, their first aim was
political revolution within a state that was incompletely Islamic (p. 14). See also Crone,
Medieval Islamic Political Thought, p. 385392.
89. Michel Chodkiewicz, Seal of the Saints. Prophethood and Sainthood in the Doc-
trine of Ibn Arabi (Cambridge, 1993), p. 46.
234 A. AKASOY
consensus of Muslims, says the Qai Iya that such acts are performed
only by an unbeliever, and that they indicate unbelief even if the doer
were to declare himself to be a Muslim.
90
The worship of idols and
heavenly bodies might be understood as a Koranic reference (6:7482),
but these are also features of West African religions in Arabic geograph-
ical literature. As for the Iberian Peninsula, conservative Muslims were
obviously most concerned about the adoption of Christian habits.
In West Africa it was particularly the cultivation of pagan habits
among Muslims which evoked the criticism of revival-purification move-
ments.
91
The North African scholar al-Maghili, for example, criticised in
the late fifteenth century the influence of polytheistic culture on Sunni
Ali, nominally a Muslim and the first ruler of Songhay in the Middle
Niger area:
Sunni Alis father was the sultan of its people and his mother was from the
land of Far, and they are an unbelieving people (kuffar) who worship idols
(yabuduna al-anam) of trees and stones; they make sacrifices to them and
ask their needs at them. If good befalls them they claim that it is those idols
who gave it to them and if it does not befall them they believe that those
idols withheld it from them. () These idols have custodians who serve
them and act as intermediaries between the people and them. Among these
people are soothsayers (kuhhan) and magicians (saara).
92
This passage is part of the answers al-Maghili sent to Askia Muam-
mad who had overthrown Sunni Alis government and was hailed for his
jihad against an infidel ruler. Askia Muammad was only half-Songhay
and needed the support of the more western provinces. Islam served as
a unifying tool for the poly-ethnic empire and was used by Askia
Muammad as an instrument of state policy.
93
The soothsayers and
magicians in the quotation are typical elements of Jahiliyya culture.
Paganism served here as a tool for drawing a line between them and
us, a recurrent feature of revival-purification movements in Africa.
94

90. Kitab al-shifa bi-tarif uquq al-Muafa, ed. al-Bajjawi (Beirut, 1984), II,
p. 1072-1073 and p. 1080; from Hanna Kassis, Muslim Revival in Spain in the Fifth/
Eleventh Century, Der Islam, 67 (1990), p. 78110, at p. 87. See also M.J. Kister, Do
Not Assimilate Yourselves La tashabbahu, Jerusalem Studies in Arabic and Islam,
12 (1989), p. 321371.
91. On the closeness of liberation and revival-purification movements see Cook,
Understanding Jihad, p. 83.
92. Hunwick, Sharia in Songhay, p. 14 (Arabic text) and p. 6970 (English transla-
tion).
93. Hunwick, Sharia in Songhay, p. 25.
94. For Uthman dan Fodio cf. Cook, Understanding Jihad, p. 76.
A similar line is drawn in the geographical texts, but their immediate
political impact must have been virtually non-existent.
Jahiliyya was used not only as a historical, but also as a spiritual and
political concept. At the beginning of the Almoravid movement, its lead-
ers claimed that their people, the anhaja Berber, were still living in the
Jahiliyya.
95
This was an expression of the state of ignorance of the Mus-
lims, not a religious affiliation or a historical layer in a certain region
before the spread of Islam.
96
The dichotomy of Jahiliyya and Islam was
also used in later political conflicts, such as the reformist movement of
Uthman dan Fodio around 1800 who targeted the government of Gobir
for its lenience towards bori cults. Uthman and his followers performed
a hijra, regarded Hausaland as being in the state of Jahiliyya and
declared jihad against Gobir.
97
In his On the Difference between the Governments of the Muslims and
the Governments of the Unbelievers (al-Farq bayna wilayat ahl al-islam
wa-bayna wilayat ahl al-kufr), Uthman listed in detail the misdoings of
non-Muslims governments which do not comply with the Sharia and
offer no safety and reliability to their subjects. They pay no regard what-
soever to rules of religion and decency. They wear, eat and drink what
they want, and instead of implementing what Islamic law stipulates, they
change these rules in order to increase their own material benefit. Thus,
in lieu of punishing a delinquent, they take away his money and keep it
for themselves without offering anything to the poor and needy. If they
are subjects to such rulers, Muslims are not allowed to follow their reli-
gious obligations. They are, for example, barred from putting on the
kinds of clothes the Sharia requires them to wear. Their rule is the exact
opposite of any civilised or even human government. Indeed the inten-
tion of the unbelievers in their governments is only, Uthman states, the
95. Fritz Meier, Almoraviden und Marabute, Die Welt des Islams, N.S., 21 (1981),
p. 80163, at p. 116. Tartib al-madarik wa-taqrib al-masalik (Beirut), IV, p. 781. Kana
al-din indahum qalilan wa-aktharuhum jahiliyya laysa inda aktharihim ghayr al-sha-
hadatayn.
96. Ibn Taymiyya compared the military situation at the time of the Prophet with the
presence of semi-Islamised Mongols (Cook, Understanding Jihad, p. 6566).
97. Robinson, Muslim Societies in African History, p. 144. Yet, there were also differ-
ent degrees: for Ibn Waa al-Qurubi (d. 900), his own age corresponded to a fitna of
error (fitnat alala) but not of infidelity (fitnat kufr), and therefore the sword could not be
used. See Maribel Fierro, Spiritual Alienation and Political Activism, Arabica, 47
(2000), p. 230260, at p. 237.
236 A. AKASOY
fulfilling of their lusts, for they are like the beasts.
98
Muslims are thus
obliged to keep themselves away from such rulers and should not even
use the same titles for their own authorities.
99
Uthmans short treatise
brings out how deeply intertwined religious and cultural identity and
membership in a political community often are a dichotomy on all
levels. It also displays another feature of paganism, the reversal of the
rules of justice, both as a divine decree and a right of the believers. This
is not the basic absence of the state as seen above, but rather the total
reversal of its purpose.
Conclusion
When they wrote about the Western fringes of the Islamic world,
medieval Arabic authors certainly made a distinction between Christians
and Jews on the one hand and others on the other hand. After all, Chris-
tians and Jews were well known to the Muslims. They had scriptural
traditions, institutions of learning and religious law, and a basis in urban
centres. The Muslims were familiar with their rituals and religious beliefs
and had established formalised ways of dealing with them. Muslims had
far more difficulties describing and identifying those who belonged to
other religious communities, as is evident from the way they dealt with
religious traditions in areas remote from the Middle East. Yet, it is almost
impossible to determine how far Muslims in fact perceived local people
as they described them.
Identifying a pagan requires a comparison. If we want to understand
why Muslims treated a certain religious community in a certain way,
comparisons will be of great importance. In the Muslim West this is dif-
ficult on the one hand insofar as the structures of the northern and the
southern regions are completely different and for most of the time did
not form part of the same political unit. However, for some 150 years
they did, under the Almoravids and the Almohads. Both displayed a
98. Mervyn Hiskett, Kitab al-Farq: a Work on the Habe Kingdoms Attributed to
Uthman dan Fodio, Bulletin of the School of Oriental and African Studies, 23 (1960),
p. 558579, at p. 567.
99. Hiskett, Kitab al-Farq, p. 569. For the problem of Muslims living under non-
Muslim rule in general see Khaled Abou El Fadl, Islamic Law and Muslim Minorities:
the Juristic Discourse on Muslim Minorities from the Second/Eighth to the Eleventh/
Seventeenth Centuries, Islamic Law and Society, 1 (1994), p. 141187 and Ibn Taymiyya,
Muslims under Non-Muslim Rule, trans. Yahya Michot (Oxford, 2006).
harsh attitude towards all groups of non-Muslims or nominal Muslims,
pagans as well as people of the book.
Al-Maghili, a later figure, is a case in point. He came originally from
North Africa where he incited the population against the local Jews
which resulted in violent persecutions.
100
He justified this with the situ-
ation of the Jews who did not show the abasement demanded of the ahl
al-dhimma. The jizya they paid should not count as such, but rather as a
bribe. Muslims who befriended Jews or defended them against such
attacks were declared unbelievers by him, but other scholars opposed his
opinion, both with regard to the Jews and with regard to the Muslims
cooperating with them.
101
Since he did not succeed in extending his
influence after these pogroms and might have even faced some opposi-
tion, he left North Africa for Gao (east of Timbuktu) where he continued
to spread his anti-Jewish views, but turned, as we have seen, also against
the polytheistic influence on those who claimed to be Muslims. In Askia
Muammad he found someone who shared his hatred of the Jews.
102
The pagan, it seems, is the famous Other, the negative stereotype in
terms of religion, civilisation or culture, political and social order. He is,
however, not necessarily a member of a specific group. There are differ-
ent forms in which paganism becomes apparent: beliefs, forms of wor-
ship, customs (food, clothing), social and political identity. There are
also different degrees of paganism as determined by the elements which
make up paganism. Apart from Otherness it is hardly possible to
describe a concept of paganism in the Arabic texts. Beyond the well-
established religious groups of the Koran, Muslim authors seem to have
found it difficult to find and use categories into which followers of other
religious traditions fit.
Part of the complexity of this issue arises from the pragmatism in
practical approaches. Muslim attitudes to religions in West Africa,
in geographical texts as well as in historical practice, suggest that Mus-
lims employed more flexible schemes than merely perpetuating the
model of the Jahiliyya and Zoroastrianism and their historical character-
istics. The borders between the peoples of the book and the pagans
were blurred, when the former were accused of shirk and kufr, the typical
sins of the latter. Those whom we would classify as typical pagans are
100. John O. Hunwick, Al-Mahl and the Jews of Tuwt: the Demise of a Commu-
nity, Studia Islamica, 61 (1985), p. 155183.
101. Hunwick, Al-Mahl and the Jews of Tuwt, p. 176178.
102. Hunwick, Al-Mahl and the Jews of Tuwt, p. 180.
238 A. AKASOY
often identified as ahl al-dhimma; the privileges granted to Christians
and Jews thus appear less exclusive.
In addition to a religious, ethnic or political affiliation, a habit or a
particular conviction, paganism can also imply the historical state of
a region before the arrival of Islam, more in the sense of Jahiliyya.
Finally, Muslims accuse each other of shirk and kufr, they criticise each
other for being too close to non-Muslims and adopting their habits, and
they lament the condition of the Islamic world as a state of ignorance.
Depending on the exclusive- or inclusiveness of attitudes, the sympa-
thetic integration of alien elements of regional cultures or the critical
expulsion of impure elements, paganism can refer to very different
things in medieval Islam.
One of the reasons why Islam was so successful is its universal doc-
trinal basis and its ability to accommodate very different cultural tradi-
tions. In order to hold this multicultural community of believers together
it is necessary to defend certain boundaries such as essential metaphysi-
cal beliefs, but also the awareness of belonging to the community of
believers and making this public through certain rituals and customs.
This is where, for revivalist movements and purists, the threat of pagan-
ism became virulent.
To sum up, in medieval Islam, the pagan is the enemy in battle, the
noble savage someone to be fought and to be educated, he is the bar-
barian with an inferior nature who engages in his primitive rituals, he is
the enemy on the border who threatens Muslim identity and the enemy
from within who is ungrateful to God.

Akasoy Paganism

Uploaded by

Copyright:

Available Formats

Akasoy Paganism

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Akasoy Paganism

Uploaded by

Copyright:

Available Formats

Anna AKASOY

PAGANISM AND ISLAM:

You might also like