Navigation – Plan du site

The mutakallimūn and Non-Muslim Religions

Prof. David Thomas
p. 39-61


Les succès militaires des viie et viiie siècles qui ont conduit de nombreuses populations non musulmanes sous la loi musulmane ont constitué un défi pour les musulmans: que faire de ces populations et comment les comprendre? Comme pour presque toutes les autres questions, les musulmans se sont tournés vers le Coran pour y trouver un éclairage, et, s'appuyant sur les passages mentionnant les juifs, les chrétiens et les autres croyants, ils ont construit un cadre légal leur permettant de décider du statut des non musulmans dans la société musulmane, ainsi que du cadre théologique élucidant leur relation à la vérité musulmane.
D'après le Coran, une succession de messagers ont été envoyés avant l'islam, apportant chacun une version successive de la révélation musulmane. Depuis les premiers temps de l'islam, les différentes histoires du salut ont toujours placé l'islam comme point culminant des révélations divine antécédentes. La conséquence en est que les historiens et les théologiens musulmans ont eu tendance à parler de groupe historiques de croyants (les « juifs », les « chrétiens ») plutôt que de traditions religieuses distinctes (le « judaïsme », le « christianisme »). Pour eux, il n'existe qu'une seule religion se déployant à travers l'histoire, et toute divergence par rapport à la forme finale de ses enseignements et de ses dogmes est une erreur.
Les théologiens musulmans des ixe et xe siècles ont formalisé cette vision en démontrant que les doctrines des groupes non musulmans sont incohérentes, y compris dans leurs erreurs. Leur argument selon lequel tout écart par rapport à un monothéisme simple manque de logique prouve que pour eux seul l'islam répond aux exigences de la raison. De même, le fait qu’ils isolent dans la discussion les dogmes des non musulmans, sans égard aux systèmes théologiques qui les portent, montre qu’ils dénient aux non musulmans la possibilité même qu’ils aient des corps de dogmes intégrés et cohérents.
En général, la vision des théologiens musulmans des croyances des non musulmans dans les débuts de l’islam est qu’elles sont des formes aberrantes de la vérité incarnée dans sa totalité et dans sa forme définitive par l’islam lui-même. Ceci est en réaltié une amplification de l’enseignement coranique, qui ne prend pas en considération la cohérence interne des croyances des juifs, des chrétiens ou des autres non musulmans.

Haut de page

Texte intégral

Muslims and Others in Early Islam

  • 1 Ibn Bābawayh, al-Tawḥīd, ch. 37, trans. David Thomas, “Two Muslim-Christian Debates”, p. 53‒65.

1The 10th century Shīʿī scholar Abū Ǧaʿfar Muḥammad ibn ʿAlī ibn Bābawayh (d. 381/991‒992) includes in his Kitāb al-tawīd the story of an encounter between the early Shīʿī scholar Hišām ibn al-Ḥakam (d. 180/796) and a Christian bishop named Barīha. The story is told by Hišām himself, who recounts how one day as he was sitting in his shop in the Karḫ market in Baghdad with students reading out the Qurʾān for him, ‘the chief patriarch’ Barīha, with a hundred or so cloaked followers and attendants, came up and started a theological discussion. The two argued over issues related to the doctrine of the Trinity and Hišām eventually won, causing Barīha to return home in a distressed state. He was so upset that his serving-woman persuaded him to return to Hišām in secret to try to find out where he had obtained arguments that had proved so effective. The two met, and Hišām explained that he had learned them from the Imam Ǧaʿfar al-Ṣādiq. Barīha asked to meet him, so the two travelled together to Medina where Barīha converted to Islam at the Imam’s hands.1

2Engaging as it is, this story as it stands is very unlikely to reflect historical events. Written by a Shīʿī, its express purpose is evidently to demonstrate the inspired knowledge of the line of Imams and to show their pre-eminence as holy individuals. To this end, the arguments that Hišām relays from his spiritual leader are made to excel in contrast to the rather lame points made by Barīha, who, despite the long experience in debate with which he is credited, is reduced to frustrated silence. And these arguments closely resemble arguments known from 9th and 10th century Muslim polemicists, and hence probably derive from them. However, what could have a ring of truth is the general circumstance in which a known Muslim religious thinker was virtually accosted by a Christian leader and his retinue, and compelled to debate abstract theology. Hišām was a famous Shīʿī expert from the latter 8th century, and he is known to have been a merchant who moved from his native Kufa to the newly-founded Abbasid capital Baghdad. So it is not impossible that the story preserves a slim basis of fact, and maybe reflects the kind of meeting that would not have been uncommon between Muslims and Christians (as well as others) in the early centuries of Islam.

3The speedy expansion of the Islamic state in the century or so following the death of Muḥammad meant that Muslims were confronted with the pressing problem of how to deal with the considerable populations of mainly Christians and Jews who now lived within their borders. In legal terms, at an early stage, probably before 800, they were able to place those who bore a relationship to themselves as Ahl al-Kitāb (‘People of the Book’, possessors of inspired scripture) in the category of ahl al-imma (client communities), and to set requirements for and limitations on their conduct in public, thus imposing considerable control on their behaviour, at least in principle. These ahl al-imma frequently played prominent parts in the public life of the early Islamic state as bureaucrats, physicians and translators, and for most of the time the majority about whose lives there is any information appeared to have suffered little from the legal impositions under which in principle they lived.

4From a religious point of view, matters were rather different. Like Hišām ibn al-Ḥakam, Islamic religious thinkers in the early centuries were faced with the problem of how to relate non-Muslims to themselves as possessors of the truth and of putting it into practice, or quite simply in terms of their perception of God and their relationship to him. On this there were ample hints in the Qurʾān, which has strong claims to contain more explicit remarks than any other scripture about communities of outside believers, and their relationship with God and his chosen community.

The Qurʾān and Non-Muslims

  • 2 There was an ongoing debate among Muslim scholars in the medieval period as to whether Jews, Chris (...)

5In the first place, the Qurʾān makes it abundantly clear that the Muslim community has indeed been chosen by God, and with a particular duty to perform: “We have appointed you a middle nation (umma wasaā) that you may be witnesses against humankind” (Q II, 143, cf. e.g. Q III, 102‒104, 110). It is also a community that is distinguished from unbelievers (kāfirūn) on the one hand, (Q CIX), and from the People of the Book (Ahl al-Kitāb) on the other (e.g. Q XXIX, 46).2 It has obviously emerged after other communities and has been given the vocation of setting an example and of calling others to the right way. This is vividly depicted at the very opening of the Qurʾān, where in the first chapter God is asked to “show us the straight path, the path of those who you have favoured, not of those who earn your anger nor of those who go astray (ġayr al-maġūb ʿalayhim wa-lā al-āllīn)” (Q I, 5‒7). The implication is that the ‘we’ here are to follow the path set by God both in order to avoid the errors into which others have fallen and as an example in contrast to them (commentators on the Qurʾān tended to identify these others as the Jews—“those who earn your anger”—and the Christians—“those who go astray”).

6In addition to this, the Qurʾān identifies itself as a revealed scripture sent like others before, but also confirming (muaddiq) them (Q VI, 92) and giving an exposition of what God decrees (wa-tafīl al-kitāb lā rayb fīhi min Rabb al-ʿālamīn) (Q X, 38), evidently a fuller revelation than its predecessors, and it names itself the Criterion (al-furqān), revealed after the Torah and Gospel (Q III, 3‒4, XXV, 1). Similarly, Muḥammad is identified as a messenger like others before him (“our prophet”, rasūlunā), but also as “the messenger of God and the seal of the prophets” (Q XXXIII, 40) as though bringing the line of God’s apostles to an end and embodying its climax in himself.

7The Qurʾān mentions in a number of places the various revelations that God has previously sent, often implying a temporal order (the Torah and Gospel came after Abraham, Q III, 65; Muḥammad is referred to in the Torah and Gospel, Q VII, 157; the Torah and Gospel are always referred to in that order), and it also says that in each period there is a book (li-kull aǧal kitāb) (Q XIII, 38). There are thus clear indications that while the Qurʾān does not lay out a set chronology, it contains enough hints to suggest some form of historical progression in which a succession of prophetic messengers brought revelations to various communities. Most importantly, it portrays the Muslim community as a model to others, and strongly suggests that the Qurʾān has been sent as a final revelation from God and that Muḥammad is the last messenger—Q VII, 157 and LXI, 6 affirm that earlier scriptures and messengers foretold his coming.

Early Qurʾānic Chronologies

8It appears that from an early time Muslim scholars made use of these and similar teachings to construct a chronology of prophetic proclamations in which each of the messengers named in the Qurʾān (including the non-biblical ones) was given a biography and his activities sketched out. The purpose was to show how history was moved by God towards its climax at which the Prophet Muḥammad would appear with the last and complete revelation that was to form the basis of true worship and service of God, and of the perfect society. In this, they were doubtless employing previous models constructed by Christians to explain how Christ was sent as the fulfilment of prophetic predictions contained in earlier scripture, and how the church was established as the perfect model of society. They would have been helped by the emerging doctrine that Muḥammad and the Qurʾān had been sent not only to endorse and fulfil scriptural teachings but also to correct them.

9The origins of the Muslim doctrine of tarīf, the corruption of earlier scripture, can no longer be known in any detail, though by the time of its first known mention, the middle of the 8th century, it had matured into an accusation that Jews and Christians had either intentionally misinterpreted their scriptures or consciously or otherwise altered the text. This was based on hints in the Qurʾān (e.g. II, 75, II, 140, V, 41), and it was given impetus by the failure to find the references to Muḥammad that Muslims were led by the Qurʾān to expect in the Torah and Gospel.

10So, something was amiss among the main religious communities that early Muslims encountered among them, and their own community and its revealed teachings were not only a further stage to what had appeared earlier but also a corrective to the mistakes that were now endemic within earlier beliefs and practices.

  • 3 Khalidi, Arabic Historical Thought, p. 35.

11One of the earliest known Muslim prophetic chronologies is the Kitāb al-mubtadaʾ that originally formed the first part of the great history by Muḥammad ibn Isḥāq (d. c.150/767) in the 8th century. This is now known almost entirely from the biography of Muḥammad that survives as the earliest account of his life (referred to simply as the Sīra), though this biography was originally only one part of a universal history that moved forward from Adam, showing how God’s will was executed through the ages and culminated in the decisive sending of the last messenger.3

  • 4 Newby, The Making of the Last Prophet, p. 8‒25.
  • 5 Newby, The Making of the Last Prophet, p. 175.

12The Kitāb al-mubtadaʾ has disappeared in its original form and can only be reconstructed in part from later works in which it is quoted, mainly the Tārī al-rusul wa-l-mulūk of the 10th century historian Abū Ǧaʿfar al-Ṭabarī (d. 311/923).4 Its surviving portions, which focus on prophetic figures from Adam up to Jesus and St George, though not in any familiar historical order, are sufficient to show that Ibn Isḥāq’s intention was to depict a line of messengers from God stretching through the eras that preceded the coming of Muḥammad. There is at least one hint that Ibn Isḥāq perceived this succession as forming a crescendo that climaxed in the appearance of Muḥammad. In his account of the prophet Isaiah he remarks: “When this king [Hezekiah] ruled, God sent Isaiah, the son of Amoz, along with him. This was before the coming of Zechariah, John and Jesus. Isaiah was the one who announced the glad tidings of Jesus and Muḥammad.”5 Here Ibn Isḥāq both recognises an elementary chronological sequence in the appearance of these prophetic figures, and makes a link between what is evidently the prophecy in Isaiah 21: 7, “If it is a column of horsemen in pairs, a column of donkeys, a column of camels…”, and the appearance of Jesus and Muḥammad. (This prophecy, in which the columns of donkeys and camels were identified in the singular as Jesus, the rider on the donkey, and after him Muḥammad, the rider on the camel, became a familiar element in Muslim polemical interpretations of the Bible.) In this instance of prophetic anticipation, he hints at the same kind of rising expectation for Muḥammad among earlier prophets that Christian apologists recognised among the Hebrew prophets for Jesus. It prompts one to wonder whether the original unedited and un-excerpted Kitāb al-mubtadaʾ contained many more links of this kind that left no doubt about the appearance of Muḥammad as an event that was universally foretold.

  • 6 Rosenthal, The History of al-Ṭabarī, p. 327‒331, giving many variant versions.

13Later historians in the early years of Islam repeat this pattern of a progression from one prophet to another with some frequency. The best-known among them, Abū Ǧaʿfar al-Ṭabarī, even includes the succession of prophets, together with kings, in the title of his work, “The history of messengers and kings”, indicating their central importance in the divinely planned structure of events that preceded the coming of Muḥammad and Islam. Towards the beginning, he even tells a story that encapsulates the intimate connection between Adam and the prophets who came after him. He relates that God gave Adam a picture of himself and all his descendants. “There were people there crowned by light. When Adam asked his Lord who they were, God replied: They are the prophets and messengers whom I shall send to my servants. Among them, there is a man who is the most luminous of them. However, a life of only forty years was written down for him.” And the story goes on to specify this “most luminous” individual as David.6 This feature must mark out its Jewish origins, but its significance in al-Ṭabarī’s narrative is in its binding together the prophetic messengers who appeared at points in history into an intentional sequence and a rising fulfilment of the divine plan.

  • 7 ʿAlī al-Ṭabarī, al-Radd ʿalā al-Naṣārā, in Ebied, Thomas, The Polemical Works of ʿAlī al-Ṭabarī, p (...)

14The significance implied in this approach towards history, in which everything leads up to the dramatic moment of the coming of Islam, is not difficult to catch. It is that all the earlier communities to whom messengers were sent with revelations were preliminary forms of the Islamic community, and all the teachings they were given in their scriptures were, at least in their initial form, anticipations of Islam and in agreement with it. This suggests firstly that the Jews, Christians and other possessors of scripture were each divinely confirmed communities, and secondly that their religious beliefs were not in essence different from those of Islam. It is why, for example, the 9th century convert from Christianity, ʿAlī ibn Rabban al-Ṭabarī (the historian’s elder countryman and his teacher) could specify in the refutation of the Christians he wrote as a sort of repudiation of his former faith that he was not arguing against “Christ (peace be upon him) or the people of his truth, but those Christian sects (unūf al-Naārā) that oppose Christ and the Gospels and corrupt the words (wa-arrafa al-kalimāt)”.7 Christ taught and the true Christians believed the same as other messengers and their communities had done beforehand, and the same as Muḥammad and the Muslims were to do afterwards. There only ever has been one divine religion, and history is essentially a sequence of restatements and re-affirmations of this until the revelation of its final and unimpeachable articulation in the Qurʾān.

15This view is not so very different from that of Christianity, of course, which sees Hebrew religion as a preparation for itself, and, according to some authors mainly from the Eastern Churches, sees Islam not as an innovation in belief that came to supersede earlier religions but as a localised version of monotheism that is not incompatible with Christianity and was intended specifically to bring the pagans of Arabia to an elementary form of belief in the one God before progressing to the full Trinitarian truth. It is reflected in the terms used by Muslims in the early centuries for the religious groups among whom they lived. In the first place, the Qurʾān itself tends only to use concrete names that denote the members of the groups—Yahūd, Naārā, Ahl al-Kitāb, Maǧūs, ābiʾūn—rather than abstract names—Yahūdiyya or Narāniyya—which would suggest a set of defining religious principles or beliefs and an institutional form of faith that could be placed alongside Islam as an alternative way of believing and acting. This would contradict the understanding that Jews, Christians and others were the historical descendants of the communities to whom God had sent the same revelation in earlier times as he had now sent through Muḥammad, making them “People of the Book”.

Heresiographies and Non-Muslim Faiths

16Writers on non-Muslims in the early centuries followed this lead. Thus, in the 9th century many works entitled Radd ʿalā al-Naārā or Radd ʿalā al-Yahūd were written, the main surviving refutation from the century being Radd ʿalā al-alāth firaq min al-Naārā (on which, there is more below). Its author, the ingenious Abū ʿĪsā Muḥammad ibn Hārūn al-Warrāq (d. after c.250/864), knew intimately the differences between the Christian traditions he called Malkiyya, Nasūriyya and Yaʿqūbiyya (Melkites, Nestorians and Jacobites) and their differences over divinity from the Muslims, but he still appeared to acknowledge them as part of the same historical continuum as Islam, or part of the sequence of religious dispensations as that was given to the Muslims later on. In principle, all the earlier communities were Muslim, because while their revealed teachings and the doctrines that were derived from them may have contained elements unique to themselves they all derived from the one divine source and were identical in essence. Thus, there might be communities of Jews, Christians and others, but there could not be any religions other than the single religion of Islam.

  • 8 Thomas, “Abū ʿĪsā al-Warrāq and the History of Religions”.

17In like fashion, authors of works on the range of different beliefs they encountered tended to refer to them in terms that indicate their communal identity rather than any real distinctiveness. Abū ʿĪsā al-Warrāq himself entitled a work that according to later references and brief summaries (it itself is now lost) contained accounts of the beliefs of the pagan Arabs, Jews, Christians and Persian dualists, Kitāb maqālāt al-nās wa-itilāfihim (“The opinions of people and the differences between them”). Here (as far as can be told) he sets out the diverse explanations about the nature of deity held by religious communities that could be encountered in or around the Islamic world, but he does not appear to see these as indicative of perceptions that were so radically divergent that they could not be fitted together into a single religious scheme.8

  • 9 Gimaret, “al-Milal wa’l-niḥal”, comments on the vagueness of these terms in the early centuries of (...)
  • 10 Ibn Ḥazm, Fiṣal, vol. 1, p. 48‒63, and vol. 2, p. 2‒81 respectively.

18The same applies to extant large-scale works on the relationships between the versions of religious traditions that Muslims could encounter in the early Islamic centuries. The 11th-century exposure of the shortcomings in non-Islamic traditions by the Andalusian theologian Abū Muḥammad ʿAlī ibn Ḥazm (d. 456/1064) is descriptively entitled Kitāb al-fial fī al-milal wa-l-ahwāʾ wa-l-nial, which can be translated as “Judgement regarding the confessions, inclinations and sects”. Ibn Ḥazm does not say clearly what he means by these distinctions (milal could be taken to refer to religious beliefs and doctrines external to institutional Islam, while al-ahwāʾ wa-l-nial could refer to religious variants within Islam),9 though the criticisms he levels against them indicate clearly enough that he sees them all without much distinction as divergent forms of the one true religion and products of the errors that have been introduced or allowed to creep in by their adherents. Thus, for example, his attack against Christianity focuses on the doctrines of the Trinity and Incarnation as perversions of the doctrine of divine oneness that is preserved in Islam, and also on the divergences between Gospel accounts as indications that the text of the original Gospel has suffered extensive corruption at the hands of the Christians.10 The former attack is along the lines of a traditional Muslim refutation that originates from the presupposition (as set out above) that the original Christian teachings as delivered by the divinely sent messenger Jesus emphasised the oneness of God in the same way that Islam was to, while the latter much more original detailed examination of the various Gospels springs from the belief that the differences between the historical Gospels and the Qurʾān were the result of neglect or intentional tampering on the part of Christians. Both forms of attack show that Ibn Ḥazm shared the common Muslim belief that, strictly speaking, there was no ‘Christianity’ as a distinct religion, but a revelation that preceded the final revelation of Islam and originally concurred with it in all essentials, though it had now degenerated into eccentric doctrines based on an irretrievably corrupt scripture.

19The net result of this approach that judges the various religious groups according to their closeness to and divergence from the norm that finds its embodiment in a form of Islam (depending on the particular Muslim author) is that diversity in religious expression is interpreted as the product of human error within a single succession of divine dispensations that have been meted out at points throughout history. Ibn Ḥazm would presumably give the narrowest interpretation to verses in the Qurʾān that suggest latitude and expansiveness, such as Q V, 48, “To each among you have we prescribed a law and an open way [širʿa wa-minhāǧ]. If God had so willed he would have made you a single people [umma wāida], but (his plan is) to test you in what he has given you; so strive as in a race with all virtues”; or even more Q XXII, 67, “To every people (umma) have we appointed rites and ceremonies which they must follow”.

  • 11 Al-Šahrastānī, Kitāb al-milal wa-l-niḥal, trans. Gimaret, Monnot, Livre des religions, p. 17‒20.

20The best-known heresiographical (or doxological) work from the early Islamic period is the Kitāb al-milal wa-l-nial of the 12th century theologian Abū al-Fatḥ Muḥammad al-Šahrastānī (d. 548/1153). This has a very clear structure that makes it possible to see how its author understood the range of what he regarded as religious and intellectual traditions and how he ranked them with respect to one another. He identifies two major divisions: the possessors of divine religion and the confessions (arbāb al-diyānāt wa-l-milal), and the people of inclinations and sects (ahl al-ahwāʾ wa-l-nial).11 Within these he details many more distinct groups, from the Muslims with their sub-divisions into the ahl al-uūl and ahl al-furūʿ, each with many further sub-divisions, to the Sabians, philosophers, pre-Islamic Arabs and Indians, each again sub-divided into many smaller groups.

  • 12 Al-Šahrastānī, Kitāb al-milal wa-l-niḥal, p. 45; Gimaret, Monnot, Livre des religions, p. 159‒160.

21Al-Šahrastānī clearly sees a substantial distinction between the people he terms “religions and confessions” and “inclinations and sects”. As he explains at the beginning of his exposition of the first of his two major divisions, “The inhabitants of the world are divided as regards their teachings into the people of religions and the people of inclinations. In actuality, when a man believes or professes something, in this he is either a recipient from elsewhere (mustafīd min ġayrihi), or he is an independent exerciser of his own opinion (mustabidd bi-raʾyihi). The recipient from elsewhere is submissive and obedient, since religion (dīn) is obedience and submission: the obedient man is the man of religion. The free-thinker is an inventor and innovator”.12 In effect, religions and confessions are derived from revelation, of which there can only be one, while inclinations and sects are the product of conjecture and surmise, of which there can be many. Al-Šahrastānī leaves little doubt about which groups are likely to be right and which wrong, and he implies strongly that what he calls religions and confessions are all, at least in their original form, versions of revealed teachings from the one source. In his mind, there is only revealed truth and there are also human attempts to attain to an approximation of this.

  • 13 Al-Šahrastānī, Kitāb al-milal wa-l-niḥal, p. 7; Gimaret, Monnot, Livre des religions, p. 108.

22He goes further in the first of his five prefaces to the work, where he says that all the divisions he will be describing are each split up, the people of inclinations and sects into diverse opinions, and people of religions into determined numbers, the Mazdeans into 70 sects, the Jews into 71 sects, the Christians into 72 sects, and the Muslims into 73 sects. “And of the sects, only a single one will be saved.”13 Here he makes use of a prophetic hadith that is attested by Abū Dāʾūd, al-Tirmiḏī and Ibn Māǧa, and reveals the force behind his multiple divisions and sub-divisions into the long sequences of groupings that follow one another in the work.

23The ominous last sentence in this explanation of the structure of the Kitāb al-milal wa-l-nial shows clearly al-Šahrastānī’s understanding of the relationship between the religious traditions, though this would anyway not be too difficult to work out. But it is important not to lose sight of the distinction he makes between religions founded on revealed scripture and sects based on human inventions. If salvation is the goal, then the latter are excluded by definition because they follow beliefs and practices that have no transcendent authority and therefore cannot be acceptable to God. It is only the religions with a foundation in revelation that have even a theoretical possibility of winning eternal salvation. From the viewpoint of his understanding of the relationship between Islam and other religious traditions, “the possessors of divine religion and the confessions” are essentially all descendants of the recipients of the same revelation that was given at various times through the line of prophetic messengers, and they thus only differ from one another in their individual subjective interpretations of the one truth, which differ insofar as they reflect subjectivity, while “the people of inclinations and sects” are beyond serious consideration because they cannot claim to possess any authority in their teachings apart from their own opinions. For al-Šahrastānī, like other Muslims writing about the variety of religious expressions in the early Islamic period, there has only been one revelation; it has been reiterated over the centuries, and through human frailty it has been corrupted and misinterpreted. Thus, the relationship between Islam and other traditions based on revelation is, as the Qurʾān says, one of confirmation and correction.

The mutakallimūn and Other Faiths

24It is against this general background witnessed in the histories and the heresiographies written in the early centuries that the attitude shown in what can be called works of systematic theology, may be largely understood. These works show graphically how their authors did not look on any individual scriptural tradition as an integral whole complete in itself, but rather as an earlier version of Islam that could be treated according to the same norms and be expected to conform to the same principles as Islam itself. In consequence, individual teachings from the earlier traditions could be separated out from the context of the tradition and subjected to the same rational tests that were applied to Islamic doctrines.

25Knowledge about the earliest forms of Islamic theological writings, mainly from the 9th century, is hampered by the absence of more than a small handful of works by theologians from before c.900 (it is symptomatic, maybe, of the likelihood of works from this period being preserved that the surviving writings of even such a theological giant as Abū al-Ḥasan al-Ašʿarī, who was active in the late 9th and early 10th centuries (d. 323/935), number only about five relatively minor compositions out of a known corpus of more than a hundred, including at least one apparently multi-volume treatise that has disappeared more or less without a trace). The titles that are listed by later authors, particularly the bibliophile al-Nadīm (d. c.380/990; he is often known as Ibn al-Nadīm) in the later 10th century, suggest that in the main they tended to write on specific topics, often rejoinders to particular points raised by colleagues or opponents. Titles such as Kitāb al-tawīd (“Divine unity”), Al-ǧawāhir wa-l-aʿ (“Substances and accidents”), Kitāb ʿalā al-Naẓẓām fī alq al-šayʾ wa-ǧawābuhu ʿanhu (“Against al-Naẓẓām concerning the creation of a thing and his reply about it”), and Kitāb ʿalā al-Maǧūs (“Against the Zoroastrians”) are typical.

  • 14 Di Matteo, “Confutazione contro i Cristiani dello zaydita al-Qāsim b. Ibrāhīm”.

26Many leading theologians from the 9th century wrote works entitled al-Radd ʿalā al-Naārā, suggesting focused arguments against the beliefs of the Christian community, and the few surviving examples from this early time confirm this. In fact, they show that Muslim theologians in general were extremely focused in their interest in Christianity. To take the earliest surviving example from the period, the truncated Radd ʿalā al-Naārā of the Zaydī Imam al-Qāsim ibn Ibrāhīm al-Rassī (d. 246/860), this essentially comprises compact though very thorough arguments directed against the doctrine of the Trinity as an accurate account of the true being of God, and against the divinity of Christ as a logical impossibility and a violation of the contents of the Gospels (the Radd contains one of the earliest surviving Arabic translations of the first chapters of the Gospel of Matthew, which is rendered in saǧʿ).14 While al-Qāsim shows that he has much more than a minimal knowledge of Christian accounts of the atonement, his interest is clearly in the two doctrines that are related to the being of God, and whether they stand up to logical scrutiny. Evidently, he was concerned to test these alternatives to the Islamic principles of divine unity and uniqueness and to demonstrate that they were not viable. He was not interested in Christianity as such, but only in those aspects of the Christian tradition that were most relevant to the Islamic tradition and its insistence on the absolute unity and transcendence of God.

27The same can be said of the much longer and vastly more detailed Radd ʿalā al-alā firaq min al-Naārā of Abū ʿĪsā al-Warrāq, an intriguing figure whose true allegiance within Islam was never properly decided, but who probably espoused a form of Šīʿism after abandoning the Muʿtazila. He died sometime after c.250/864, and was remembered for his detailed and objective (some even thought them sympathetic) accounts of non-Islamic faith traditions, particularly Persian dualist sects, to which al-Šahrastānī actually accused him of belonging.

28With such interests, it is not surprising that Abū ʿĪsā should show a more than thorough knowledge of Christianity (based, probably together with his knowledge of other faiths, on the accounts in his lost Maqālāt al-nās), not only of its basic doctrines, together with something of its history, but also of the variations in the doctrines of the three major sects within the Islamic Empire, Nestorians, Jacobites and Melkites. Few people showed more knowledge about this, as is attested by the appearance of quotations, nearly always unacknowledged, in later works for well over a century.

  • 15 Thomas, Anti-Christian Polemic in Early Islam, Abū ʽĪsā al-Warrāq’s ‘Against the Trinity’; Early M (...)

29Despite all this, when Abū ʿĪsā came to make his refutation of Christianity he concentrated on the same two doctrines as al-Qāsim ibn Ibrāhīm, the Trinity and the Incarnation. His refutation constitutes two extremely long and detailed sequences of arguments against these doctrines, in which he demonstrates with the help of unadorned common-sense logic that the explanations offered by Arabic-speaking Christians either contravene reason, or lead to internal confusion, or are simply incoherent.15

30Abū ʿĪsā gives no explanation of what he is doing in his refutation. (It is possible that he in fact did this originally, because his refutation has only been preserved in quotations in the response made by the 10th-century Christian Yaḥyā ibn ʿAdī, though if there were any they would presumably have left some trace in the extant parts, which are very extensive, including the formal introduction and the conclusion.) The reader is left to surmise about what this could exactly have been, though there cannot be much doubt that he was both intending to expose the numerous logical flaws he saw in the doctrinal accounts offered by his Christian opponents, and to show that any alternative to simple monotheism such as that promoted in Islam was prone to failure. In other words, his refutation is at the same time a polemical exercise against the two major Christian doctrines and also an apologetic exercise in defence of tawīd. His Maqālāt may well have been a preparation for this in bringing together descriptions of what may be called the different versions of deity that would be known from the various religious groups active in the Islamic world at the time. His refutation of Christianity, and also the lost refutations of Judaism and dualist sects he is known to have composed, were based on these descriptions, and formed a body of arguments that led to the inevitable conclusion that only the strictest form of monotheism was tenable. Abū ʿĪsā may even have taken this to its logical conclusion—and here would be a reason why he was routinely excoriated by the later Muslim authors who bothered to mention him by name—because he is known to have questioned and ridiculed beliefs held dear in Islam. So, his ultimate aim could have been to arrive at a form of independent monotheism that conformed to rational dictates alone, rather than being derived from claims to prophetic or any other kind of revelation.

31Abū ʿĪsā’s Radd is such a substantial intellectual achievement that in the absence of firm information about the intention behind it speculation is inevitable. If what is said here reflects any measure of the purpose of the work, the inference to which it leads is that, like so many other Muslim authors, Abū ʿĪsā did not see Christianity or any of the faith traditions he wrote about as distinct in essence from Islam (or the kind of Islamic monotheism he favoured). Rather, he saw them as alternative interpretations of the same body of teaching that could be compared with the sound rational version he saw in Islam and corrected. In addition, he (and contemporaries whose works are now lost) saw that the examples of error provided by these versions could be used as illustrations of the intellectual and spiritual chaos that results from abandonment of the simplicity of strict monotheism. The underlying theological stance is that the historical versions of Christianity and other religious traditions are spurious because they have departed from the original authoritative and rationally sound form, and their adherents should, and can, be argued or educated out of them by properly deployed rational methods.

32That this may be the thinking behind Abū ʿĪsā al-Warrāq’s refutation is supported and to a large extent confirmed by the treatment of Christianity and other faiths in works that are first known from the 10th century.

33Whereas in the 9th century all the signs (such as they are) point to Muslim theological works generally focussing on single topics, from the early 10th century there are reports of works consisting of collections of topics being discussed together: the lost multi-part work by al-Ašʿarī referred to above, known simply as al-Fuūl (“The chapters”), was probably one of the earliest examples. Later in the century, Abū Bakr al-Bāqillānī’s (d. 1013) Kitāb al-tamhīd (“The introduction”) and ʿAbd al-Ǧabbār’s al-Muġnī fī abwāb al-tawīd wa-l-ʿadl (“The summa in sections on divine unity and justice”) are two striking instances that make it possible to see for the first time how Muslims from different theological schools, al-Bāqillānī from the Ashʿariyya and ʿAbd al-Ǧabbār from the Muʿtazila, understood the range of questions raised in theological discussions and the relationship between them. These are early examples of Muslim systematic theological treatises.

  • 16 For a description of this structure, see Peters, God’s Created Speech, p. 25‒35.

34The more elaborate of these two works is ʿAbd al-Ǧabbār’s Muġnī, which he says himself took him twenty years to write. Structured in twenty books according to the five principles of the Muʿtazila, and in particular the principles of divine unity and justice, it seeks to treat a wide range of traditional and current theological topics within a tight structure that starts with epistemological considerations, moves on to questions related to the existence and character of God, then to the relationship between God and the world, and so on to the nature of society and political issues.16 It is a prodigious achievement that demonstrates the coherence and completeness of Islamic theology and also its harmony with reason, and therefore its correctness.

35As would be expected from a work that emerged from the welter of theological controversies of the 9th century and reflected the many that were still seething in the tenth, in amongst its positive presentations of Muʿtazilī theology the Muġnī includes many arguments against opponents. This is only to be expected because defeating opponents is an understandable and necessary contribution to establishing the sound theological points that ʿAbd al-Ǧabbār wished to make. But what is, at first glimpse, somewhat surprising, are the attacks on groups from outside Islam, prominent among them dualists, Christians, Jews and Hindus.

36These attacks do not appear in one section of the work, as though they were being dealt with together as rivals to Islam. Rather, as might be inferred from the treatment of Christians in the works from the previous century discussed above, they appear in conjunction with the topics of doctrinal exposition to which they are most relevant. Thus, the attacks on dualists and Christians come at the conclusion of the long discussion on the existence and character of God that take up the first five volumes of the work (in fact, the first four books are lost, but a general idea of their contents and lay-out can be gathered from later works that summarise the Muġnī).

37To take the section on the Christians, which is a long set of arguments that includes quotations and summaries from works by earlier Muʿtazilī theologians and others (including Abū ʿĪsā al-Warrāq, whose arguments are quoted without attribution), this comprises two major sections plus an introductory summary of Christian doctrines. The first of these is a set of arguments against the Trinity and the second is against the Incarnation, or more precisely the possibility that the divine and human could have united in the person of Jesus Christ. ʿAbd al-Ǧabbār has extracted the two doctrines from their historical, scriptural (there is almost no reference to the Bible or Qurʾān) and doctrinal contexts, and he addresses them as statements standing in isolation to be analysed and deconstructed according to principles of logic that would have been shared by everyone, Muslims and other theological experts in the Islamic world, who participated in the theological discourse of which ʿAbd al-Ǧabbār would be acknowledged as a leading expert.

38Two important points emerge from this form of approach to this other faith in which, incidentally, ʿAbd al-Ǧabbār was followed by a long succession of authors who wrote theological compendiums like the Muġnī. The first is that he sees Christians as corrigible rather than lost in error. As he adds argument to argument in his elaborate analysis of the various components of the two Christian doctrines, he shows that the versions of belief propounded by his opponents can be corrected. While they are not logically sustainable as they are, and must in the case of the Trinity be pared down to a perception of a single, entirely undifferentiated divine essence, and in the case of the Incarnation be revised in order to maintain the radical disjunction between God and creaturely beings, resulting in an effective abandonment of the two major Christian doctrines, they do nevertheless contain traces of belief in the one, transcendent God, even though this has been perverted and confused.

39In adopting this approach to the two Christian doctrines, ʿAbd al-Ǧabbār was engaged in the same exercise as Muslim polemicists before him, as well as many after. For them, it was a matter of restoring other faith traditions to their pristine form, in accordance with the teachings in the Qurʾān that the Jewish, Christian and other communities had been given revelations that were all characterised by the same emphasis on the oneness and transcendence of God as the Qurʾān itself. It was the duty of the polemicist to uncover or recover this original true form by demonstrating that alternative forms had no rational sustainability. This approach gave Muslim polemic as it was conducted by the mutakallimūn in the early centuries a pronounced educative character, and it presupposes an attitude towards these other traditions of a certain degree of acceptance of them (accompanied by frustration, maybe) as the errant fellow travellers they were. They were Ahl al-Kitāb, even though, in the words of Sūrat al-Fātia, they were maġūb ʿalayhim and āllīn.

40The second point is that ʿAbd al-Ǧabbār and fellow polemicists evidently saw no procedural problem or methodological failing in extracting individual doctrinal elements from a whole body of beliefs and addressing them in isolation, in the case of the Jews their teaching about prophethood and of the Christians their teaching about the Trinity and Incarnation. The implicit presupposition on which they were working must have been that the beliefs and doctrines they encountered in other scriptural traditions were not part of an alternative structure of teachings that had no relationship to Islam but were part of the same current as their own and could be treated in the same way. In other words, there were no other separate scriptural religions as such, but only other versions of the one truth that was embodied in Islam, though in the case of these traditions transmitted in defective forms.

41This is emphasised by the fact that ʿAbd al-Ǧabbār (and indeed al-Bāqillānī) makes no differentiation between opponents within Islam, such as the Ašʿariyya, and these external opponents. They can all be treated in the same way because their unsound teachings and their distorted ideas all issue from the one source of divine revelation.

42The similarities detectable between the very abstract approaches adopted in the Muġnī and similar Muslim theological works on the one hand and on the other hand the discursive references to other faiths found in the Qurʾān show how little Muslim attitudes towards other religious traditions moved away from the founding inspiration. Arguments were given admirably sophisticated expression, and buttressed with the most powerful logic. But the underlying attitude hardly changed, if indeed it changed at all.

The Muslim Attitude Towards Other Faiths

43The fundamental thrust of the Qurʾān and of Muslim theology is, of course, that there is only one God and that God is one. The choice before humans is therefore either to accept or reject, and the experience of the Muslim community was that there were numerous individuals and communities along the spectrum between acceptance and rejection. At one extreme were those who called themselves Muslim through and through, although their identity inevitably shifted as different groups among those who called themselves Muslims made the claim to be the bearers of the pure faith, the one sound sect among the 73 in the Prophet’s hadith. Then there were the historical descendants of the first recipients of a previous revelation, who exhibited signs of the original pure faith that had been held by the communities of the prophetic messengers who had been sent to them. These were precursors of institutional Islam, and they had been equally Muslim in their first reception of revelation. In consequence, there remained a close bond of recognition between them and Muslims, and they could be argued or educated out of the mistaken perception they had adopted. Further along the spectrum were those who followed a faith that approximated to the scriptural faiths, for example by believing in one or more transcendent powers, and at the other end were those who rejected the idea of a God who actively revealed knowledge and truth through revealed messages, particularly philosophers who claimed to be able to arrive at transcendental truth through their own unaided reason. According to this model, historical Islam as it had been held by the Prophet Muḥammad and his true followers was the repository of teachings that would be applied as criteria by which all other claims to truth could be judged. Islam was the guarantor and confirmer of other traditions, and the corrector of mistakes and errors within them.

44To a certain extent, this theological attitude provided the underpinning for a degree of toleration for other faiths, most graphically expressed in the position of ahl al-imma accorded to the Ahl al-Kitāb by Muslim jurisprudents, with the privileges and many more impositions that accrued, and by the limited roles allowed them in Islamic public life (aspiring Christian and other non-Muslim politicians were known to convert in order to achieve the highest positions). But if Muslims perceived their religion as a ‘corrector’, ultimately their role was to convert. Hence the intricate, sometimes prolix, arguments constructed against teachings that were seen to undermine or contravene tawīd, and the generally condescending attitude shown towards opponents who had the temerity to challenge the truth of Islam.

  • 17 Winter, “The Last Trump Card”, p. 138.

45In the final analysis, for those scholars who extended the teachings in the Qurʾān about believers who did not profess full Islam into abstract theological principles there was only one faith, of which groups such as Christians and Jews possessed an approximate version in which glimpses of original truth could be detected among the confusion.17 In this respect, the general Muslim approach towards other faiths can loosely be labelled inclusivist because it accords a measure of value to them, though this term can only be applied loosely because, with regard to scriptural traditions, in the final analysis there were no other faiths but rather earlier iterations of the one faith that had been distorted over time. Islam had been revealed to fulfil them and demonstrate their partial characters, and the assumption must be that all right-thinking followers of these traditions would recognize this culmination of the prophetic line and accept it. For those who did not, Muslim individuals and the community as a whole were given the task of exposing their errors and trying to educate them out of these. Despite this, in the implied insistence that these approximations to the truth should adopt more rigour in their beliefs, acknowledge the errors in the beliefs to which they clung, and abandon them for the full embodiment of truth, the Muslim regard towards other faiths is qualified by a considerable degree of intolerance, and it approaches closer to exclusivism. The question remains as to whether this dogmatic model could be interpreted leniently, so that Muslims might recognize a modicum of truth in other religious traditions, or whether it must be interpreted harshly, so that other believers could only ever be reminded of the inadequacies of their position. The history of relations between Muslims and others shows that Muslims scarcely acknowledged common beliefs and morality, and rarely hesitated to emphasise the differences and their own superiority.

Haut de page


Primary Sources

ʿAbd al-Ǧabbār (416/1025), al-Muġnī fī abwāb al-tawīd wa-l-ʿadl, ed. Ṭaha Ḥusayn et al., Cairo, 1960‒1969.

ʿAlī ibn Rabban al-Ṭabarī (c.246/860), Abū Sahl, al-Radd ʿalā al-Naārā, in Rifaat Ebied, David Thomas, (eds. and trans.), The Polemical Works of ʿAlī al-abarī, Leiden, Brill, 2016.

Ibn Bābawayh, al-Tawīd, ed. Ḥasan al-Ḥusaynī al-Ṭihrānī, Tehran, 1967, trans. David Thomas, “Two Muslim-Christian Debates from the Early Shîʿite Tradition”, Journal of Semitic Studies 33, 1988, p. 53‒80.

Ibn Ḥazm (456/1064), Abū Muḥammad ʿAlī, Kitāb al-fial fī al-milal wa-l-ahwāʾ wa-l-nial, Cairo, Būlāq, 1899‒1903.

al-Qāsim ibn Ibrāhīm (246/860), “al-Radd ʽalā al-Naārā”, ed. Ignazio di Matteo, “Confutazione contro i Cristiani dello zaydita al-Qāsim b. Ibrāhīm”, Rivista degli Studi Orientali 9, 1921‒1922, p. 301‒364.

al-Šahrastānī (548/1153), Abū al-Fatḥ Muḥammad, Kitāb al-milal wa-l-nial, ed. Muḥammad Fatḥ Allāh Badrān, Cairo, 1947‒1955; trans. Daniel Gimaret, Guy Monnot, Livre des religions et des sectes, vol. 1, Leuven, Peeters/UNESCO, 1986.

al-Ṭabarī (310/923), Abū Ǧaʿfar Muḥammad ibn Ǧarīr, Tārī al-rusul wa-l-mulūk, trans. Franz Rosenthal, The History of al-abarī. Volume 1, General Introduction and From the Creation to the Flood, Albany NY, State University of New York Press, 1989.

al-Warrāq (c.250/864), Abū ʿĪsā Muḥammad ibn Hārūn, al-Radd ʿalā al-alā firaq min al-Naārā, ed. and trans. David Thomas, Anti-Christian Polemic in Early Islam, Abū ʽĪsā al-Warrāq’s ‘Against the Trinity’, Cambridge, New York, Cambridge University Press, 1992; Early Muslim Polemic Against Christianity, Abū ʿĪsā al-Warrāq’s ‘Against the Incarnation’, Cambridge, New York, Cambridge University Press, 2002.

Secondary Sources

Gimaret, Daniel, art. “al-Milal wa’l-niḥal”, in Encyclopaedia of Islam, Second Edition, Leiden, Brill [accessed 23 April 2017].

Khalidi, Tarif, Arabic Historical Thought in the Classical Period, New York, Cambridge University Press, 1994.

Newby, Gordon, The Making of the Last Prophet, Columbia, South Carolina, University of South Carolina Press, 1989.

Peters, Johannes R.T.M, God’s Created Speech. A Study in the Speculative Theology of the Muʿtazilî Qâdî l-Qûât Abû l-asan ʿAbd al-Jabbâr bn Amad al-Hamaanî, Leiden, Brill, 1976.

Thomas, David, “Abū ʽĪsā al-Warrāq and the History of Religions”, Journal of Semitic Studies 41, 1996, p. 275‒290.

Winter, Tim, “The Last Trump Card. Islam and the Supersession of Other Faiths”, Studies in Interreligious Dialogue 9, 1999, p. 133‒155.

Haut de page


1 Ibn Bābawayh, al-Tawḥīd, ch. 37, trans. David Thomas, “Two Muslim-Christian Debates”, p. 53‒65.

2 There was an ongoing debate among Muslim scholars in the medieval period as to whether Jews, Christians and other scripturalists were kāfirūn or not.

3 Khalidi, Arabic Historical Thought, p. 35.

4 Newby, The Making of the Last Prophet, p. 8‒25.

5 Newby, The Making of the Last Prophet, p. 175.

6 Rosenthal, The History of al-Ṭabarī, p. 327‒331, giving many variant versions.

7 ʿAlī al-Ṭabarī, al-Radd ʿalā al-Naṣārā, in Ebied, Thomas, The Polemical Works of ʿAlī al-Ṭabarī, p. 64‒65.

8 Thomas, “Abū ʿĪsā al-Warrāq and the History of Religions”.

9 Gimaret, “al-Milal wa’l-niḥal”, comments on the vagueness of these terms in the early centuries of Islam.

10 Ibn Ḥazm, Fiṣal, vol. 1, p. 48‒63, and vol. 2, p. 2‒81 respectively.

11 Al-Šahrastānī, Kitāb al-milal wa-l-niḥal, trans. Gimaret, Monnot, Livre des religions, p. 17‒20.

12 Al-Šahrastānī, Kitāb al-milal wa-l-niḥal, p. 45; Gimaret, Monnot, Livre des religions, p. 159‒160.

13 Al-Šahrastānī, Kitāb al-milal wa-l-niḥal, p. 7; Gimaret, Monnot, Livre des religions, p. 108.

14 Di Matteo, “Confutazione contro i Cristiani dello zaydita al-Qāsim b. Ibrāhīm”.

15 Thomas, Anti-Christian Polemic in Early Islam, Abū ʽĪsā al-Warrāq’s ‘Against the Trinity’; Early Muslim Polemic Against Christianity, Abū ʿĪsā al-Warrāq’s ‘Against the Incarnation’.

16 For a description of this structure, see Peters, God’s Created Speech, p. 25‒35.

17 Winter, “The Last Trump Card”, p. 138.

Haut de page

Pour citer cet article

Référence papier

Prof. David Thomas, « The mutakallimūn and Non-Muslim Religions », MIDÉO, 33 | 2018, 39-61.

Référence électronique

Prof. David Thomas, « The mutakallimūn and Non-Muslim Religions », MIDÉO [En ligne], 33 | 2018, mis en ligne le 05 juillet 2018, consulté le 20 juillet 2018. URL :

Haut de page


Prof. David Thomas

University of Birmingham

Haut de page

Droits d’auteur

Institut Dominicain d'Études Orientales

Haut de page
  • Logo Institut dominicain d'études orientales - IDEO
  • Logo Institut français d'archéologie orientale - IFAO
  • OpenEdition Journals