Skip to main content Accessibility help
Hostname: page-component-55597f9d44-n4bck Total loading time: 0.517 Render date: 2022-08-18T19:28:46.317Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "useRatesEcommerce": false, "useNewApi": true } hasContentIssue true

The Science of Kalām

Published online by Cambridge University Press:  24 October 2008

Richard M. Frank
The Catholic University of America, Department of Semitic and Egyptian Languages, Washington DC 20064, USA


Our intention here is to present the essential character of classical, sunnī kalām within a strictly formal perspective and to set out its basic aspects. It was conceived by the mutakallimīn as a rational, conceptual, and critical science and, although kalām differed in a number of basic concepts and constructs and in its analytic system, the topical organisation of the major compendia parallels that of metaphysics as understood in the contemporary Aristotelian tradition. The debates between kalām and falsafa need to be examined within this context. Kalām, however, is theological in the strict sense of the term and it is as such that its problematic and its procedures are primarily to be understood. Thus seen, the object of kalām is to rationalise the cognitive content presented to Believers in the symbolic language of the koranic revelation. It has, then, four principal tasks, sc, to conceptualise, to order, to explain, and where possible to justify the primary doctrines of the community whose belief is held to be normative. Within this framework the differences that characterise the major schools as such and the various tendencies of individual masters within each school may readily be brought to light.

On se propose ici de présenter, d'un point de vue strictement formel, la nature du kālam classique sunnite et d'identifier ses caractéristiques principales. II avait été conçu par les mutakallimin comme une science rationelle, conceptuelle et critique. L'organisation des matières dans ses traités reprend celle de la métaphysique dans la tradition aristotélicienne de l'époque, bien que le kalām s'en distingue par plusieurs de ses structures et concepts fondamentaux, ainsi que par son système analytique. C'est dans ce contexte qu'il faut considérer les debats qui s'instaurèrent entre kalām et falsafa. Le kalām, cependant, est d'ordre strictement théologique et c'est principalement dans ce cadre qu'il faut comprendre sa problématique et ses procédures. Le kalām a pour fonction de rationaliser le contenu cognitif offert aux croyants dans le langage symbolique de la révélation coranique. Il en résulte quatre tâches principales; il s'agit de conceptualiser, ordonner, expliquer et, dans le mesure du possible, justifier les doctrines principales reconnues par la communauté faisant référence en matière de croyance. Dans ce cadre, il sera possible de mettre en lumière les différences entre les principales écoles, ainsi que les tendances qui distinguent certains de leurs grands maîtres respectifs.

Research Article
Copyright © Cambridge University Press 1992

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


1 As such they not really pertinent to our present problem and may, indeed, be distracting if one would understand the nature and the formal content of classical kalām. The historical origins, remote or medieval, of European dramatic art in religious ritual and ceremony may be interesting, but they will not reveal the nature and form of the literary art of Sean O'Casey or Samuel Beckett.

2Al-fiqh’, as the name for the science of jurisprudence, is clearly a distinct lexeme. Explanations of the lexical basis of its meaning in this use generally take it to mean understanding and formal knowledge in a rather broad sense, i.e., not in one that is specifically descriptive of jurisprudence as such and as distinct from other realms of insight and understanding. ‘Al-fiqh’ is often seen as a direct parallel to the ‘prudentia’ of Latin ‘jurisprudentia’, but those who remark on this do not cite specific discussions of the word that mean to explain its sense explicitly as the name for the science of the revealed law and formally to distinguish the lexical basis of this use from others. Ibn Fāris (Maqāyīs al-luġa, s.v.) takes ‘faqiha, yafqahu’ as a general verb for apperception and knowing that “came to be associated particularly with the knowledge of the šarī'a (‘ilm al-šarī'a’), so that any one who knows the licit and the prohibited is called a faqīh.” B. Weiss, on the basis of the analogy with ‘jurisprudentia’, offers an interpretation of the word as a formally conceived description of the activity of the Muslim faqīh in interpreting of the šarī'a (“The primacy of revelation in classical islamic legal theory,” Studia Islamica, 59 [1984]: 79109, p. 99), but offers no evidence that the word was explicitly so interpreted as a lexically distinct descriptive name for jurisprudence.Google Scholar

3 Ibn Fāris, Maqāyīs al-luġa, s. ✓KLM.

4 E.g., al-Bāqillānī, Tamhīd, ed. McCarthy, R. (Beyrouth, 1957),Google Scholar §11 and Fūrak, Ibn, Muğarrad maqālāt al-Aš'arī, ed. Gimaret, D. (Paris, 1987), p. 31, 10, and p. 285,Google Scholar 1. I do not mean to suggest in what follows that the word ‘al-kalām’ has, in its every occurrence in the kalām texts, the formal meaning we are presenting or that, even when it is used formally, the immediate context may not occasionally cause other meanings to surface. What I will insist, however, is that as the name for the discipline or science of kalām it is understood by the leading masters of the 4th and 5th centuries A.H. in this formal meaning as properly descriptive of their science. Concerning the origin of the term, cf.van Ess, Joseph, Theologie und Gesellschaft im 2. und 3. Jahrhundert Hidschra, Bd. 1 (Berlin and New York, 1991), §A 5.1.1, pp. 48ff.Google Scholar and for a brief examination of the wider semantic scope of the verb as it is related to the formal meaning we are discussing and their deliberate exploitation by al-Aš'arī for rhetorical purposes, cf. al-Aš'arī's Kitāb al-Ḥathth ‘alā l-baḥth,” Mélanges de l'Institue Dominicain d'Études Orientales (MIDEO), 18 (1988): 83152, pp. 115ff.Google Scholar

5 For the formal sense of the term, cf., e.g., al-Isfarā'īnī in Frank, R., “Al-Ustādh Abū Isḥāq,” MIDEO, 19 (1989): 129202, pp. 143f., Frr. 3.6, Ibn Fūrak, Muğarrad, p. 32, If. and p. 292, 13ff.Google Scholar (where, e.g., “mā lā ilā ma'rifatihi illā bi-l-nazari mitlu l-kalāmi fī hadati l-'ālam”) and generally ibid., pp. 31, 20ff., where “al-nazaru huwa l-fikru wal-ta'ammulu wal-i'tibāru wal-muqāyasatu wa-raddu mā ġāba ‘ani l-hissi ilā mā wuğida l-'ilmu bihi …” as well as al-Bāqillānī, Tamhīd, §11. Thus, for example, al-Aš'arī, speaks of “al-kalāmu fī l-ğismi wal-harakati …” (al-Hatt 'alā l-baḥlt, ed. Frank, R., MIDEO, 18 [1988], §1.12) in the sense of the speculative inquiry into and the resultant knowledge concerning material bodies,Google Scholar etc. Cf. also, e.g., Fūrak, Ibn, Muğarrad, p. 285.Google Scholar A number of distinct lexical equivalents of ‘al-nazar’ (among them ‘al-tafakkuru wal-i'tibār’) which involve formal inference (al-istidlāl) are offered also by al-Bayhaqī, al-I'tiqād wal-hidāya, ed. al-Hūt, K. Y. (Beyrouth, 1403/1983), p. 74.Google Scholar Generally, thus, ‘ ‘ilm al-nazar’ is equivalent to ’ ‘ilm al-kalām’ and ‘al-nuzzār’ to ‘ahl al-kalām’. That al-nazar is consistently taken to be formal reasoning that leads to “knowledge,” i.e., to an indubitably justified conclusion or inference, is obvious enough in the frequent discussion of the question of how knowledge comes to exist following the nazar (e.g., al-Ğuwaynī, , al-Šāmil, ed. al-Naššār, A. S. [Alexandria, 1969], pp. 112ff. and 624f.Google Scholar and al-Šāmil, K., Some Additional Fragments of the Text, ed. Frank, R. [Tehran, 1981], pp. 10ff., 85ff. and 94f.).Google Scholar That kalām must, by definition, be formally conceptual is taken for granted. It is for this reason that al-Ğuwaynī says (al-Kāmil fī ihtisār al-Šmil, MS Ahmet III 1322, fol. 98) that kalām has to employ a formal terminology (istilāh) and cannot be carried on in ordinary language (alfāzu l-'awāmm). Al-Ġazālī (Mi'yār al-'ilm [Cairo, 1329], pp. 55,7 and 57,5) speaks of the formal terminology of those who practice rational thinking or speculation (al-nuzzār) with regard to logical terminology.

6 ‘Speculative’ (one might substitute ‘theoretical’) is perhaps a more appropriate equivalent, in that there are other modes of rational discourse than that held to be represented in kalām. (That lexically the base meaning of the common verb ‘nazara, yanzuru’ is “to look at,” “to consider” and so also, in the proper context, “to gaze at” - θεωρεĩν, speculari — is essentially irrelevant to the question of what is for us the appropriate English equivalent of the word in the formal use we are considering. It may be semantically interesting and even historically significant, that is, but it is not immediately pertinent.) Again here, in trying to present the formal meaning of the term, one has to keep in mind the different meanings and connotations that ‘speculative’ and ‘rational’ can evoke. In the way we most commonly use these words today they are not equivalent to ‘rationalis’ and ‘speculativus’ as these were employed in the middle ages. ‘Speculative’ today most often describes a judgement or conclusion as one that is based on insufficient evidence or is otherwise not adeuately justified, and so also often with ‘theoretical’. Likewise, ‘rational’ may evoke the connotations of a rationalism of the 18th and 19th century variety or one of positivist leanings. Our present concern, though, is with medieval texts and thus ‘speculative’ is probably the more exact rendering of ‘al-nazar’ in this expression, if one hears the word in the sense it had in ‘scientia speculative’ for medieval philosophy and theology. However they may have disagreed each with many of the other's primary assumptions and conclusions, the mutakallimūn and the falāsifa shared a common conception of the nature of speculative (theoretical) science, as a science that, like mathematics, begins in absolutely certain axioms and principles and proceeds deductively so as to achieve true “knowledge” ('ilm, έπισιήμη). It is within the context of this assumption that one is to understand the basis of al-Ġazālī's feeling that by reasoning he ought to be able to attain a knowledge of metaphysical truths, that is, as removed from rational doubt as are mathematical conclusions.

7 In many cases the equivalence is clear enough. It is indicated already by al-Aš'arī, e.g., in al-Hatt where the investigation of the basic doctrines of Islam, condemned by his opponent (§1.12), is finally identified explicitly with “the speculative science” (§2.331). Note that ‘al-dīn” is not so broad an expression as English ‘religion’, but rather refers to religion strictly as a something common and public, in its embodying beliefs and obligations that are, in principle, incumbent on every Believer. Note also that ‘‘ilm al-kalām’ may be synonymous with’ ‘ilm al-tawhīd’, where ‘al-tawhīd’ is taken to denote the essentials of Muslim belief as a whole.

8 In brief, if ‘al-ta'ammul’ ≠ ‘al-ğadal’, then ‘al-kalām1’ (= ‘al-at'ammul’) ≠ ‘al-kalām2’ (= ‘al-ğadal’); and so too ‘al-nazar1’ (= ‘al-ta'ammul’) ≠ ‘al-nazar 2’ (= ‘al-ğadal’).

9 For the use of ‘takallama, yatakallamu’ with reference to a formal disputation as such, cf., e.g., al-Qiftī, Ibn, Anbāh al-ruwāh, ed. Ibrāhīm, M., 4 vols. (Cairo, 1973), vol. I, p. 250,Google Scholar cited MIDEO, 18, p. 116f. and Hağar, Ibn, Dayl Tabaqāt al-hanābila (Beyrouth, n.d.) 1, p. 9, 4 (§5) and p. 13, 11f. (§9).Google Scholar Thus al-tarīqatu l-'ilmiyya is distinguished from la-tarīqatu l-ğadaliyya (Ibn Fūnakdīm, Šarh al-hamsa, ed. ‘Utmān, A. [Cairo, 1965], p. 307).Google Scholar It is reported (Ibn Fūrak, Muğarrad, p. 293) that al-Aš‘arū says, distinguishing and explaining ‘al-nazar’ in the sense of ğadal‘, that the latter (the activity of dialectical disputation is used for instruction and the refutation of false theses. ‘Al-nazar’ in the sense of disputation is fairly common and is used, moreover, even by the Hanbalites, who certainly were no devotees of speculative reasoning or of rational theology (e.g., Ibn Hağar, Dayl 1, p. 9, 2 and 6f. and p. 13, 11f.).

10 Variations occur in where various sub-sections or minor topics are placed and often in what may be included or omitted, depending on the length of the tract, etc. The abridged, catechetical manuals generally omit §1 of our table. Note too that under some constructions topics subsequent to the discussion of prophecy as such, sc., those dealing with the status of Muhammad and questions that depend directly and as such on the prophetic revelation, are not considered by some to be usūl but rather furū' since they cannot be rationally demonstrated.

11 Here, thus, “the creation of human actions” (ḫalq al-af'¯l) for the Aš'arites and its denial for the Mu'tazilites.

12 In some works, particularly where contemplative vision of the divine economy is envisaged, that there exists no true agent other than God will become the primary focus of attention, uniqueness and ontological unity being taken as already and more easily known.

13 Cf. Gutas, D., Avicenna and the Aristotelian Tradition (Leiden, 1988), pp. 243–45.Google Scholar

14 Cf. Gutas, Avicenna, p. 359 and generally pp. 248ff. With this order, note too that “political science,” for al-Fār¯bī, follows metaphysics' treatment of the governance of the world by the immaterial beings of the celestial world.

15 Al-Mustasfā (Bulaq, 1325) 1, pp. 5f. We should note that because of naïve or incomplete readings of the texts, al-Gazālī's conception of the nature of kalām and of its place within the religious sciences and its role within the Muslim community is consistently misunderstood and misrepresented. The matter is too complex to take up here and will have to be made the subject of a separate study.Google Scholar

16 Cf., e.g., Ibn Fūrak's report of al-Aš‘arī’s conception of “this rational speculation” (Muğarrad, p. 250, 4ff.), viz., that it is a special reasoning (naẓarun maḫṣūṣ), viz., one that is proper to an adult of sound mind who “should not believe one doctrine rather than another by taqlīd and should not prefer one position to another because he finds the one congenial and the other difficult, and should not prefer a particular doctirine becausee it has some worldly authority or status or because this is the doctrine of his forebears or his compatriots… On the contrary, he should (1) hold himself in every respect as one who enquires and seeks understanding and looks for right guidance and (2) he should see diverse theses and contrary doctrines as equivalent and equal in truth and falsity, so that he make a true beginning of thought and reflection (fikratan wa-ta'ammulan) on each matter he examines … when he is unprejudiced and his thoughts are free of those barriers and impediments that come between the inquirer and knowledge of that into which he inquires, then he inevitably achieves the knowledge of the object of his study in the way in which he sought.” For al-Gazālīs's position, see our, “Al-Gazāli on Taqlīd,” forthcoming in Zeitschrift für Geschichte der arabisch-islamischen Wissenschaften.

17 What one sees here is the set of basic problems that are characteristic of metaphysics as such. It is important to note that, although the topics are traditional and are, moreover, arranged following the order that was usual in the hellenistic tradition, absent are the assumptions, the analyses, and the conclusions that belonged to the received tradition. Kalām tried, as it were, to make a new beginning, rejecting the logical, physical, psuychological, cosmological, and metaphysical doctrines of the Neo-platonised Aristotelianism that was the common philosophical tender of the age. The analytic procedures of classical kalām are philosophically sound. Some of the assumptions and conclusions are basically naïve; others, however, manifest genuine, occasionally profound, philosophical or theological insight.

18 It should not, that is, if, accepting the claim, one follows the definition of Bochenski, J. (The Logic of Religion [New York, 1965], p. 14), sc., that theology is “a study in which, along with other axioms, at least one sentence isassumed which belongs to a given creed and which is not sustained by persons other than believers in a givern religion.” The distinction bewteen philosophy and theology is fundamental and has to be maintained. Note that this distinction, as made here, does not rest upon any assumption or judgement concerning the cognitive value of religious belief. Rather, it is founded simply on the recognition of religious experience and language as distinct from ordinary experience and languavge.Google Scholar

19 I. e., their understanding of what beings are fundamental and what things are essentially assemblages of basic entities.

20 By the time there was a generally felt need for this science, the Aristotelian tradition was rapidly gaining ground in some of the milieux that were served by classical kalām.

21 These works are later, but the same things is evident in the long fragment of the section concerning prophecy from al-Bāqillānī's Hidāya that is preserved in MS al-Azhar, al-tawḥīd, no. (21) 242.

22 It should be noted that one likely reason for which some of the proofs of theses in the major compendia regress through so many sub-theses in order to demonstrate one or another element in the primary syllogism is that, unlike the scholastic theologies of the Christian middle ages, classical kalām does not have a tradition of philosophical disciplines (physics, purely philosophical metaphysics, etc.) available as commonly recognised scientific background to which to appeal. On the other hand, propositions and terms that they consider to be established, are often not argued, save in sub-sections of fundamentally extraneous theses in connection with some elements of which they may customarily have been called into question.

23 R. fī l-ağrām al-‘ (in Tis' rasā’ilCairo, 1326/1908]), p. 48 (the numbers rise exponentially when subsequently he comes to talk about living beings);Google Scholar cf. also The Life of Ibn Sina, ed. Gohlman, W. E. (Albany, 1974), pp. 26ff.Google Scholar and see Gutas, Avicenna, pp. 26ff., where corrections to the Arabic text of this section of the autobiography are presented.

24 This inversion of things is more apparent in the case of those mutakallimīn who hold that the revelation cannot be totally rationalised and becomes particularly conspicuous when one considers that, in practice as well as in principle, the systematic theology of the schools and the sufi theology are not opposed to one another, as is often assumed, but that the latter is complementary to the former.

25 There is a terminological problem here that should be noted, Given their conception of the nature of their own speculative science, the mutakallimūn in effect considered themselves to be philosophers in the proper sence of the term. They would not, however, refer to themselves as falāsifa or to their science as falsafa, for as they as well as those who were pleased to call themselves falāsifa heard the word, ‘falsafa’ did not mean philosophy or theoretical science simply as such, but denoted the largely neoplatonised Aristotelianism (including its core of physical, psychological, cosmological, and metaphysical teachings) that was the received tradition and doctrine of the falāsifa. In order to avoid lexical ambivalences we have chosen here to use ‘falsafa’ to refer to this tradition and ‘philosophy’ to refer to philosophy as such in the currently usual sense.

26 For their part, the falāsifa, in a way, did recognise kalām as a distinct kind of discourse and as a theology insofar as some of its basic assumptions derive from common Muslim belief. If, however, we will say that they recognised it as specifically a kind or mode of religious discourse, we shall have to examine and to make clear exactly what we mean here by ‘religious’. Is what formally understand by the term ‘religious’ in this and in analogous contexts really the same as what the falāsifa intended and expressed when they said that the discipline of kalām belonged to al-milla? Because they acknowledge the supramundane source source of prophetic inspiration (al-wahy) in a separated intellignence and recognise genuine prophetic revelation, be particular instances however perfect or imperfect, as presenting metaphysical truths according to a certain modality and having such validity as properly belongs to that modality, they do not hold it to be a phenomenon claims of whose divine origin belong peculiarly to a realm of religious experience, language and belief that is distinct from the universal realm of common experience and profane language and whose rationalisation, therefore, belongs properly to theological rather than to philosophic discourse in our contemporary sense of these terms. It is for this reason, and for this reason alone, that falsafa may, without inconsistency, claim to arbitrate and rectify the truth claims of religious speech and to warrant the validity of sound religious discourse as such. Kalām, according to the falāsifa, is characteristically associated with religion, its proper function being that of a particular kind of rational exposition and vindication of the prophetic texts that are authoritative for the community. As a kind of rational discourse, however, kalām is not essentially distinguished from philosophical discourse (i.e., from the metaphysics of falsafa) merely because it is in some way bound to the symbolic language of prophetic revelation; it is essentially distinguished from the science of metaphysics because its rationalisation of prophetic speech reaches only to the level of dialectic and does not achieve the demonstrative knowledge that belongs to philosophy. The distinction we make between religious speech and profane speech simply had not been drawn. Kalām does not claim that in religious speech a realm or dimension of being that as such is not accessible to profane rationality is symbolically disclosed, nor did the falāsifa see it as making such a claim. To pertain to religion in the sense of belonging properly to a milla, thus, is not to pertain to the domain of religion in the sense in which we nowadays normally conceive it. In distinguishing falsafa from kalām, thus, the falāsifa did not distinguish philosophy from theology, not, at any rate, in the same way we normally do when we draw a formal distinction between them. In sum, though they did in their own way manage to isolate religion as a distinct theme for discussion — and this was a major achievement – it remains that the falāsifa did not, as we do, conceive religion as representing (or at least least claiming to represent) a domain of experience and of speech that is altogether distinct from that of common, secular experience and profane speech (concerning this matter, cf., e.g., Collins, J., The Emergence of the Philosophy of Religion (New Haven, 1967), pp. 357ff.Google Scholaret alibi and Schmitz, Kenneth, “Philosophy of religion and the redefiniton of philosophy,” Man and World, 3 [1970], pp. 54–18).CrossRefGoogle Scholar Nor, therefore, did they envisage the possibility of theology's being a theoretical science sui generis. In regard to this question, again, one should keep in mind the historical distance that separates theworld of thefalasifa and the mutakallimīn from that of our own day, in which it is generally questionable whether there can be, other than mathematics, any science ([έπιστήμη, 'ilm 'aqlī) as it was conceived by the philosohers and theologians of the middle ages. Concerning the theology's relation to philosophy see, e.g., Welte, B., “Die Wesensstruktur der Theologie als Wissenschaft,” in Auf der Spur des Ewigen (Freiburg, 1965), pp. 351–65Google Scholar and Hünermann, P., Theologie als Wissenschaft, Quæstones disputatæ 45, (Freiburg, 1970), pp. 73123.Google Scholar

27 Cf, e.g., the citation of Ibn Fūrak, Muğarrad, p. 250, translated in n. 16 above. Thus al-Qušayrī speaks frequently of the “just” and the proper use of reason, e.g. Latā'if al-išārāt, (Cairo, 19691971) 1, p. 170, 3–10 et alibi pass. The same is true for al-Gazālī; see our “Al-Gazālī on Taqlīd,” n. 76.Google Scholar

28 Like al-Aš‘arī and al-Gazālī, Gilson makes explicit his philosophy's ultimate subordination to religious doctrine, whence his “Christian philosophy” is really a theology; see Wippel, J., Metaphysical Themes in Thomas Aquinas (Washington, 1984), pp. 1ff. For a more nuanced conception of what a Christian philosophy might be, cf., e.g., MGoogle Scholararticle, Metz's, “Christliche Philosophie,” in Lexikon für Theologie und Kirche 2, pp. 1141–47Google Scholar and Welte, “Wesensstruktur,” pp. 362ff. Kalām is often referred to as “dialectical theology,” though exactly what is meant by those who use the expression is not always clear. (They plainly do not mean to class it with the so-called dialectical theology of the 19th century.) The use would seem likely to have its primary origin in the description of kalām and its classification as a discipline or science given by the falāsifa and would seem to have found a kind of evidential support in the convention according to which, as in the Latin West, the structure of the scholastic disputation supplied the literary form in which theses and argumentation are usually presented in the compendia. The evidence of the literary form is essentially secondary, however, and remains materially irrelevant as such to the basic question of the nature of kal¯m unless subtantiated by more cogent considerations. By contrast, the assertion by al-Fārābī, followed by later fal¯sifa, that kal¯m is dialectical, howbeit self-serving, has a cogent theoretical foundation. Falsafa, they hold, is a “demonstrative science,” i.e., one that begins from what they hold to be unquestionably justified principles and axioms in order to deduce demonstratively verified conclusions; kal¯m, on the other hand, begins with generally accepted assumptions and axioms (i.e., the endoxa of ordinary people, sc., of those who have not the intellectual gifts required to grasp the scientific truths of falsafa) and proceeds therefore by dialectical reasoning to explain the publicly authorised religious beliefs of the society and to defend them for those whose knowledge and understanding of the world is to be formed by them. By this criterion all systematic theology, of whatever form, however “philosophical” (lege “Aristotelian”) in its content and its procedures, will be essentially dialectical. In considering the historical meaning and significance of this thesis, several things have to be kept in mind, among them that, as the falāsifa conceived it, the “demonstrative science” that is falsafa embodies a received core of authoritative doctrine, including first principles and axioms as well as metaphysical and cosmological dogmas. In the historical context, thus, kalām represented a competing body of conceptually articulated doctrine that claimed the scientific foundation of rational demonstration and the corroboration of divine revelation ti boot. Besides being theoretically both consistent and satisfying, the assertion that kalām is essentially dialectical was convenient in that it permitted the falāsifa to assent to the truth of the revelation as such while dismissing the cognitive claims of kalām and without reexamining either the starting points of their own arguments or their doctrinal conclusions.

29 In a number of passages the Koran appeals to the evidence of natural phenomena in order to argue its point about the universal dominion of the One God. These are generally taken to be calls for a properly rational consideration of the evidence (e.g., al-Qušayrī, Latā'if 5, p. 216) and for some authorities, the “proofs” contained in the scripture (al-dalā'il al-sam 'iyya) may, even without formal elaboration, furnish an adequate basis for the rational justification of religious assent. Al -Aš'arī, in an early work, went so far as to claim that the scriptural proofs for the existence of God are in fact better than the the philosophical demonstrations based on the ontology of “accidents” (R. ilā ahl al-taġr, ed. Gulaynid, M. S. [Cariro, 1987], pp. 54f.)Google Scholar and in al-Luma' (ed. McCarthy, R. [Beyrouth, 1953], § 3) he attempts to present one of the Koranic arguments directly as a formally rational proof (cf. also. al-Hatt, §2.2).Google Scholar Here again, in the failure of this proof, we witness a failure to make an appropriate distinction, for these “Scriptural proofs,” as they are found in the Koran, do not make an appeal to the empirical evidence of ordinary experience but rather to that of religious experience. They call attention to — they are išārātun ilā — something that is hidden to ordinary perception, though immediately transparent to religious vision as such, viz., the overwhelming, effective presence of the transcendent God. (Here this does not imply that the world is mythologized; for the Muslim, as for the Christian, God is not in any sense a part of the world or an aspect of the world. It is because of this that kalām is able, in its own way, to share the theoretical attitude that is characteristic of philosophy.) Kalām is the rational exposition of the cognitive content of the religious vision articulated in the Koran and the theololgian may choose to focus on one or another facet or aspect of the vision or of whatr is viewed and to describe it from one or another formal perspective. The theolgy of al-Aš'ar/11th century may be seen as seeking to maintain thealmost mystical vision of the prophetic revelation. All being depends upon God; as creations, contingent beings and events are attributes of God, His “action attributes” (sifātu af'ālihi) and all phenomena are thus tokens (āyāt) of His operative presence and power. It is upon this aspect of being that the Aš'arites principally focused, and accordingly, the facets and moments of things and events that are brought forward in the theological rationalisation are those which are most prominent to religious insight, not the manifest antecedents, conditions, and concomitants of their occurrence. In its analysis of the coming to be of voluntary human actions, for example, the classical Aš'arite theology is not concerned to give an account of how they come to take place. The physical state of the agent and the role of antecedent belief, motivation, interntion, and volition, albeit explicitly recognised, are scarcely mentioned. The aim is, rather, to conceptualise those ontological aspects and relations of their being that ground their reality insofar as they appear as events and states of affairs for religious intuition and to describe them in such a way as to disclose the coherence of this intuition and the logic of trye rekuguiys hydgements. The Aš'arite analysis has to be seen from this perspective, for othewise it can only appear as a kind of fanciful abstraction of speculative imagination and its acceptance in the schools and by many of the sufis will be inexplicable. (Concerning the Aš‘arites’ focus on the logic of theological predicates cf. our “ Two islamic views of human agency,” in Makdisi, G., Sourdel, D., et Sourdel-Thomine, J. (ed.), La Notion de liberté au moyen âge [Paris, 1985], pp. 3749, pp. 43ff.Google Scholar In an earlier work, which appeared in Studia Islamica, 25 (1966): 1375, written without having understood the system and its intentionality, we misinterpreted al-Aš‘arī’s doctrine as an attempt to give a causal account of the occurrence.) The Mu'tazila, though insisting, for their own theological reasons, on the nigh absolute autonomy of the human agent, hold that God is the agent of all events that do not occur directly or indirectly as the actions of living, created agents. Their primary intention, however, is simply to account for the material occurrences, not to see God's immediate activity in the world.Google Scholar

30 Al-A'arī, al-Tagr, p. 76, §10.

31 According to the mutakallimīn, assent is required to the truth of every theological assertion that is made in the Koran and (under given conditions) in authenticated traditions. This primary assent, however, is prior to the thematic understanding of the proposition intended by the revealed text. When the authoritative sentences are formally conceptualised, they will be ranked in order of importance. In some cases, however, a unique and cgnitively certain interpretation may not be attainable for a given sentence or may not be agreed upon, in which assent to the truth of the sentence is required but not to any given conceptualisation or interpretation of its content. Something analogous to a process of formalisation will be required at some level even for the Hanbalites, since in order to set forth an 'aqīda, it is necessary to know which verses of the Koran and which authenticated prophetic logia are substantially equivalent to one another in content and implication and which most clearly express each given primary dogma.

32 Although the Mu'tazila and the Aš'ariyya alike insist that most of the primary dogmas may be discovered by pure reason and must be rationally justified, both hold that all are presented in the revelation and that, in the real order of things, the revelation is prior in the order of discovery. Moreover, as we have seen, the revelation is formally prior to their rationalisation. This needs to be nuanced, however, if one is to give anadequate account of the way the mutakallimūn conceive the order of things here. So again one encounters the problem of their conceiving fundamental theology as an autonomously rational metaphysics.

33 As was noted in note 31 above, compromise is sometimes achieved concerning important formulations (e.g., “mounts the Throne” as a description of God), by asserting that the sentence is true as intended by God, but we, not being privy to its meaning, are unable to determine which, if any, of several theologically coherent and logically consistent interpretations is intended.

34 See Bochenski, Logic, pp. 62ff. and pp. 119f. and note the comparisons with the procedures of the natural sociences inour own day, pp. 64f. and 82.

35 I.e. some such hypothesis would seem plainly to be needed, for it is hard to imagine how it should have come to pass that out of so many philosophers, all of them in principle dedicated to impartial, critical rationality, not one thought to question the need to appeal to the operation of a separated, celestial agent if the only thing required was to explain the cognitive intuition of essential forms and of the basic principles of reasoning.

36 Their assent to the “fact” of prophetic inspiration or revelation has to be examined and understood within the context of the problem indicated in note 26 above. It is to be noted also that whether the mutakallimūn and the falāsifa held or did not hold similar conceptions of the nature and autority of prophetic revelation is a matter of indifference to the validity of the distinction we are making. Note also that we do not mean in any way to suggest that the falāsifa were wholly critical in their thinking, for they were, in many respects, as bound to the basic doctrines of the inherited philosophical tradition as were the mutakallimūn to those of the revelation.

37 According to al-Aš'arī in al-Ṯaġr this was the order followed by the Prophet in his teaching of the community. Bochenski presents this as one among several basic forms that are employed for the justification of religious discourse in Logic, §45, where he notes that our (4) cannot be deductively demonstrated. I would suggest that it is precisely because the truth of a proposition of this kind cannot be demonstrated deductively that al-Fārābī and others of the falᾱsifa generally hesitated to assert formally the relative superiority or perfection of the revelation given to Muhammad (or to say that Muhammad is or may be, in some sense, “the Seal of the Prophets”). There is no evidence in the texts to support the assertion, strongly advocated in some circles, that some of the falāsifa, among them al-Fārābī, did not assent, modo suo, to the fundamental claims of the Islamic revelation. Quite to the contrary, it would seem incontestable that al-Fārābī considered Islam at least to be “a virtuous religion” and in Iṯbāt al-nubuwwāt Avicenna states explicitly that Muḥammad isa prophet (Tis‘rasā’il, p. 134; I am indebted to my colleague, Prof. Thérèse-Anne Druart for this reference).

38 Cf., e.g, al-Ḡuwaynī, Iḫtisār, fols., 248v and 234v. Al-Ġazālī has his own problems with the probative value of miracles (e.g., Faysal al-tafriqa, ed. al-Dimaŝqī, M.Q. [Cairo, 1319/1901], pp. 8f.Google Scholar and al-Qisṭās al-mustaqīm [Cairo, 1381?/1962], pp. 66f.), but the question of whether it is possible for God to lie can have no major significance within the overall context of his theology, since (following the doctrine of Avicenna) he holds that it is impossible that God do other than what He does and that the celestial mechanism is programmed so that a prophet's mind will receive celestial enlightenment according to what is optimal for the universal system. Concerning this see our forthcoming “Creation and the cosmic system: Al-Ġazālī and Avicenna.”Google Scholar

39 This is not to say that a similar situation does not sometimes arise in philosophy too. Such is the case, for example, with the Platonic dyad (on which see Robert Sokolowski, Presence and Absence, a Philosophical Investigation of Language and Being [Bloomington and London, 1978], pp. 172ff)Google Scholar and with what Husserl terms “inner time consciousness” (das innere Zeitbewußtsein, on which see Sokolowski, R., Husserlian Meditations [Evanston, 1974], §§52ff., pp. 132ff.).Google Scholar

40 The utilisation by the mutakallimīn of the philological sciences in determining in what meanings particular words may be employed as predicates of God and in analysing those that are validly said of God is remarkably perceptive. This, however, is formally secondary to the primary problem of abstraction, since it comes into play chiefly in the task of sorting out the descriptions of God that occur in the Koran and the Sunna and ordering them within an already established theological framework.

41 Many primary differences between the schools are manifested as differences in the classification of predicates within a basic three- or four-fold system. In the beginning of al-Maqsad al-asnā al-Ġazālī goes to some length in order to reject, in the guise of a rhetorically constructed caricature, his predecessors' conception and method of analysis in favor ofthosee of Avicenna's Aristotelianism. On examination of the evidence it becomes doubtful, however, that al-Ġazālī's embrace of an Aristotelian logic with its categories and the associated metaphysics allowed him to make any substantial advance in logical rigor, intentional clarity, or depth of philosophical and theological insight over what had been achieved by his predecessors.

42 The Mu'azilite treatment of this problem, for rather obvious reasons, proved unsatisfactory to the deeper religious intuitions of the orthodox community. One major difficulty here is that they located the origin of the meaningfulness of sentences in the will while holding God's will to have its being as a multitude of distrinct and temporally contingent volitions. Their overall treatment of the problem would seem to leave several questions hanging, but this is not altogether clear, as the teaching of the Mu'tazila on this subject has not been adequately studied.

43 The only theological work by a direct student of al-Aš‘arī, the Ta’wīl muškil al-āyāt of Abū al-Hasan al-Tabarī preserved in Ms Tal'at, mağ. 419 of the Egyptian National Library remains unpublished and there is as yet no adequate edition of Ibn Fūrak's Bayān ta'wškil al-aḥādīṯ even though a considerable number of manuscripts are readily available.

44 The main question here does not so much concern the adequacy of concepts or the degree to which the divine essence may be known as it does the completeness of the inventory of the basic elements or facts that may be known. Related to this is the question of what, if any, level of rational justification of belief is essential to the validity of religious assent. A study of the correlation of the positions taken by individual masters with regard to these questions might furnish significant criteria for classifying the theological tendencies of the school and their margins.

45 Again, adequacy of knowledge is not in question; for some the question may involve the completeness or perfection of the explicit assent which constitutes Belief and for others the completeness of the guide to what knowledge may be sought, e.g., in contemplative vision.

46 The study of the law as such, viz., of the particular rules and their application to individual cases, belongs to the science of jurisprudence (al-fiqh), while what we should term ethics in a more traditional sense, sc., the study of the virtues as such and of their acquisition, was not pursued by the fuqahā', but by the sufi theologians. The basic theology of kalām supplies the theoretical foundation for both.

47 Systems — and Avicenna's is certainly one of the most remarkable – tend, as such and in themselves, to be seductive the more all-inclusive and tightly constructed they are, as each thing has its own cozy place and everything is explained.

48 Superficial readings have consistently overlooked this fact and thereby posited, explicitly or implicitly, inconsistencies that are not in the text. There is, for example, no place (including al-Iqtisād) in which al-Ġazālī formally asserts any kind or degree of occasionalism.

49 This is clear already in the formally conceptual analyses that occur, for example, in the work of al-Ğunayd and al-Hall¯ğ and which, although often carried beyond what one finds in the school manuals (e.g., al-Hallāğ's discussion in K. al-Tawᾱīn of the difference between our tawḥīd and God's tawḥīd), remain formally and materially within the boundaries of “the science of the fundamental doctrines of Islam.”

Cited by

Save article to Kindle

To save this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the or variations. ‘’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

The Science of Kalām
Available formats

Save article to Dropbox

To save this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about saving content to Dropbox.

The Science of Kalām
Available formats

Save article to Google Drive

To save this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about saving content to Google Drive.

The Science of Kalām
Available formats

Reply to: Submit a response

Please enter your response.

Your details

Please enter a valid email address.

Conflicting interests

Do you have any conflicting interests? *