Introduction
Abū l-Qāsim Maḥmūd b. ʿUmar al-Zamakhsharī (467/1075–538/1144) was a Muʿtazilite man of letters and grammarian from Khwārazm. The author of about fifty works, two-thirds of which have survived (many in print), his main fields of interest were adab, grammar and lexicography, but he also composed works in theology and law, as well as works on the Quran and the Tradition.Footnote 1 He is best known, however, for his Quran commentary, al-Kashshāf ʿan ḥaqā’iq ghawāmiḍ al-tanzīl wa-ʿuyūn al-aqāwīl fī wujūh al-ta'wīl (The Discoverer of the Truths of the Hidden Things of Revelation and the Choicest Statements concerning the Aspects of Interpretation) which he completed in Mecca in 528/1134.Footnote 2 Throughout its history, the Kashshāf has been described as a “Muʿtazilite interpretation of the Quran”. By the seventh/thirteenth century, Nāṣir al-Dīn ʿAbd Allāh ibn ʿUmar al-Bayḍāwī (d. c. 685/1286) had already composed his Anwār al-tanzīl wa-asrār al-ta'wīl, a Quran commentary which Robson describes as “largely a condensed and amended edition of al-Zamakhsharī's Kashshāf”, which sometimes refuted the latter's Muʿtazilite views and sometimes simply omitted them.Footnote 3 Al-Bayḍāwī was not the only one preoccupied with the Muʿtazilism of the Kashshāf: his contemporary, Alexandrian qāḍī Nāṣir al-Dīn Aḥmad b. Muḥammad Ibn al-Munayyir (d. 683/1284), also wrote a counterblast to the commentary at this time.Footnote 4 Half a century later, Shams al-Dīn Muḥammad b. Aḥmad al-Dhahabī (d. 748/1348) warned readers to be wary of the Kashshāf and, even a century after that, Aḥmad b. ʿAlī Ibn Ḥajar al-ʿAsqalānī (d. 846/1442) repeated the warning, saying that the Kashshāf was off limits to all who wished to study it unless they were aware of its dangers.Footnote 5
In modern times a number of scholars have made a connection between the Kashshāf and its supposed Muʿtazilism. Nöldeke spoke of “[al-Zamakhsharī's] most clever and over-subtle investigations of philosophical and theological matters” in the Kashshāf; Nassau Lees referred to the Muʿtazilite doctrines that “pervade the whole Preface”; Goldziher said that in the Kashshāf al-Zamakhsharī “produced a concise fundamental work for Muʿtazilite Qur’ān interpretation”; and Brockelmann spoke of “Muʿtazila bias”.Footnote 6 More recently, Smith noted that “[al-Zamakhsharī's] interpretation of and commentary on the Qur’ān were strongly influenced by his theological viewpoints”; McAuliffe referred to the Kashshāf as a “mouthpiece for the dogmas of the [Muʿtazilites]”; and Madelung stated that al-Zamakhsharī's “rationalist Muʿtazilī interpretations … provoked criticism among traditionalist Sunnīs”, mentioning al-Bayḍāwī and Ibn al-Munayyir as representatives of this criticism.Footnote 7
Other authors, however, hold a position opposed to the one above. Jansen states: “[I]n spite of traces of Mutazilite dogmatical attitudes, [the Kashshāf] is not a dogmatical commentary”; Gimaret writes: “The Ḳur’ānic commentaries of Abū Djaʿfar al-Ṭūsī (d. 459/1067) and of al-Ṭabarsī (d. c. 548/1155) are overtly Muʿtazilī commentaries, even more so than the Kashshāf of al-Zamakhsharī”; Rippin argues: “The famous work of al-Zamakhsharī (d. 538/1144), renowned for its Muʿtazilī perspective, is distinctive primarily for its special outlook and not for the presence of an overall theological argument per se, nor for the quantity of such argumentation” (though what this “special outlook” is, if it refers to Muʿtazilism, is unclear); and Saleh notes: “His Muʿtazilī opinions are few and far between to be significant (sic) and too buried in a normative Sunnī approach to allow them a distinctive voice” (though the term “normative Sunnī approach” requires some clarification). Later, though, in his remarks on al-Zamakhsharī's commentary on Q93:7, Saleh seems to lean towards the first position.Footnote 8
A closer look at the preceding statements raises more questions than it answers. One might ask how the content of the Kashshāf exercised such medieval scholars as al-Bayḍāwī, Ibn al-Munayyir, al-Dhahabī and Ibn Ḥajar al-ʿAsqalānī so much that they were still condemning the commentary three-hundred years after it first appeared. Why had it not simply been destroyed, banned or replaced by al-Bayḍāwī's version? Here again, though, there is a problem, for al-Bayḍāwī would not have removed everything that was objectionable in al-Zamakhsharī's Kashshāf, so it was left to two later scholars, Shams al-Dīn Muḥammad ibn ʿAlī al-Dā’ūdī al-Mālikī (d. 945/1538) and Aḥmad al-Nūbī (d. c. 1037/1627), to complete his task for him. In their respective works, al-Itḥāf bi-tamyīz mā tabiʿa fīhi l-Bayḍāwī ṣāḥib al-Kashshāf and Kashf al-aqwāl al-mubtadhala fī sabq qalam al-Bayḍāwī li-madhhab al-Muʿtazila, these scholars drew attention to the places in the Kashshāf where al-Bayḍāwī would have “failed to remove al-Zamakhsharī's heresies”.Footnote 9 These later efforts to “de-Muʿtazilize” the Kashshāf, several centuries after al-Bayḍāwī, make one wonder, not only if there was ever agreement as to what was specifically Muʿtazilite and, therefore, hereretical in the commentary, but also if there was not more afoot than saving al-Zamakhsharī's readers from exposure to his heresy.
More recent statements on the Kashshāf are not particularly helpful either. While not wishing to criticize the scholars cited here, it is worth noting that their statements on the Kashshāf are broad and more or less assume that readers are well informed. The result of this is that readers may assume they have learned more than they have. Goldziher, for example, referred to the Kashshāf as “a concise fundamental work for Muʿtazilite Qur’ān interpretation” and Gimaret argued that al-Ṭūsī and al-Ṭabarsī wrote “overtly Muʿtazilī commentaries, even more so than the Kashshāf”, but neither told readers what makes a commentary a Muʿtazilī interpretation of the Quran or, in the case of Gimaret, why the Kashshāf was less of a commentary of this kind than those of the others.Footnote 10 Would this have something to do with the “special outlook” to which Rippin refers? Even if it does, we are no further along the path to clarity as to what makes the Kashshāf, or any commentary, “Muʿtazilite”. Would a commentary be Muʿtazilite, then, not because of its outlook but simply because it contains specifically Muʿtazilite theology? As noted, Rippin has stated that there is little in the way of Muʿtazilite theological argumentation in the Kashshāf; yet Smith argued that al-Zamakhsharī's commentary was “strongly influenced by his theological viewpoints”; and McAuliffe referred to the Kashshāf as a “mouthpiece” for the dogmas of the Muʿtazilites. How far is the Kashshāf specifically Muʿtazilite dogma? No-one seems to know. For a variety of undoubtedly good reasons, no one has studied al-Zamakhsharī's Kashshāf to any great extent (and the same could probably be said for other Quran commentaries). The views presented above therefore tend to repeat, to some degree, broad statements that have their roots in primary sources that have been around for centuries.
In the light of these various opinions, this article sets out to study al-Zamakhsharī's Kashshāf with an eye to discovering what can be said about its Muʿtazilite contents. It may be possible, afterwards, to agree with Goldziher's statement that the Kashshāf is “a concise fundamental work for Muʿtazilite Qur’ān interpretation”, or with any of the statements above, and for these statements to have a more precise meaning than they do now. On the other hand, these statements may be unwarranted, not because they contradict the evidence but because there is insufficient evidence to support them. Once the matter of the Kashshāf's Muʿtazilism has been studied, it will be possible to see if its condemnation was in any way justified; and if not, to see what motives might have led to its censure.
While a detailed, line-by-line reading of the commentary might be the best way to answer questions concerning the content of the Kashshāf, I adopt here a shorter but no less sound approach, looking at the text from a number of angles. The first involves a detailed study of a limited part of the Kashshāf, the commentary on Q44 (Sūrat al-Dukhān/Smoke) and Q54 (Sūrat al-Qamar/The Moon), to establish the extent to which Muʿtazilite theological ideas can be found therein. The continuous reading of these two sūras avoids the trap of focusing on verses that might seem to fit a Muʿtazilite agenda more easily and thus retains the possibility that Muʿtazilite views might emerge in unexpected places.Footnote 11 The second approach has two parts, each involving comparing the Kashshāf with a theological treatise that may have influenced it. The first is al-Zamakhsharī's theological tractate, al-Minhāj fī uṣūl al-dīn; the second al-Muʿtamad fī uṣūl al-dīn, the major theological treatise of al-Zamakhsharī's teacher Rukn al-Dīn (Ibn) al-Malāḥimī (d. 536/1141). The third approach focuses on manuscripts of the Kashshāf in order to test the veracity of the well-known anecdote according to which al-Zamakhsharī would have begun his commentary with the words al-ḥamdu li-llāh alladhī khalaqa l-Qur’ān (Praise be to God who created the Quran), a clear declaration of his Muʿtazilite position. By approaching the Kashshāf from these various angles, it will be possible to gain a clearer idea of its Muʿtazilite contents and the extent to which it may be considered a Muʿtazilite commentary on the Quran.
1. The first approach: a study of Q44 (Sūrat al-Dukhān) and Q54 (Sūrat al-Qamar)
As noted above, al-Zamakhsharī has been said to have given a Muʿtazilite interpretation of the Quran. A study of his commentary on Q44 (Sūrat al-Dukhān/Smoke) and Q54 (Sūrat al-Qamar/The Moon) shows, however, that the Kashshāf hardly speaks constantly of Muʿtazilism; nor does it seek occasion to do so. Within the framework of the traditional tafsīr musalsal, al-Zamakhsharī offers some Muʿtazilism when the opportunity arises, but does not present a detailed development.Footnote 12 It might have been possible for him to have used many passages as a starting point for a presentation of his Muʿtazilite opinions but he does not. What distinguishes this commentary from others, then, is not an excess of Muʿtazilism.Footnote 13 This restrained use of the tafsīr for expressing Muʿtazilite views is shown by the fact that, in his commentary on Q44 and Q54, al-Zamakhsharī makes only one reference to a Muʿtazilite tenet. In his commentary on Q54:17 we read:Footnote 14
Now We have made the Quran easy for Remembrance. Is there any that will remember?
wa-la-qad yassarnā l-Qur’āna li-l-dhikri fa-hal min muddakirin.
Al-Zamakhsharī explains the passage as follows:Footnote 15
that is, we have made it easy to remember and to take a warning from, in that we have filled it with healing exhortations and by means of it have turned [people] away from the promise and the threat [but] “is there anyone” to heed the warning?
ay sahhalnāhu li-l-iddikār wa-l-ittiʿāẓ bi-an shaḥannāhu bi-l-mawāʿiẓ al-shāfiya wa-ṣarrafnā fīhi min al-waʿd wa-l-waʿīd «fa-hal min» muttaʿiẓ.
He begins his commentary by using synonyms to explain the Quranic passage:
1. sahhala (to facilitate, to make easy, to ease) means the same as yassara in the Quranic verse;
2. iddikār (maṣdar of iddakara [idhtakara]: to remember, to think about, to bear in mind) has the same meaning as dhikr in the verse.
At the same time, however, al-Zamakhsharī's explanation goes beyond simple synomyms; he adds: and to take a warning from (ittiʿāẓ).
He then continues:
in that we have filled it with healing exhortations and through it have turned [people] away from the promise and the threat (al-waʿd wa-l-waʿīd) [but] “is there anyone” to heed the warning?
This last expression, translated as “the promise and the threat”, is one of the five principles of the Muʿtazilite school of theology.Footnote 16
Having interpreted the text in line with his own theological views, al-Zamakhsharī then gives two other interpretations. First:
It was said: We have made it easy to memorize (sahhalnāhu li-l-ḥifẓ) and those who wish to memorize it are helped to do so but is there anyone seeking to memorize [the Quran] in order to be helped?
Here the emphasis is on the understanding of the word dhikr – it means ḥifẓ. Second, al-Zamakhsharī says:
It is possible that the meaning is: We have made it easy to remember (hayya'nāhu li-l-dhikr) as one prepares (yassara) his camel for a trip when he saddles it; and one prepares (yassara) his horse for a raid when he puts a saddle and bridle on it.
Here the accent returns to the meaning of yassara; the idea of making remembering the Quran easy by specifically preparing it for this. It is interesting to note, however, that al-Zamakhsharī explains yassara as hayya'a but then proceeds to give an example in which the original verb, yassara, is used to clarify the meaning of the verb hayya'a. This may be because he is thinking of the lines of poetry he wishes to use next, which include:
wa-qumtu ilayhi bi-l-lijām muyassiran
I stood up and went to [my horse], with the bridle making him ready to ride.Footnote 17
The verb yassara, then, means: making performance easy through prior preparation. This nuance is clearly explained with the help of the example and the verse; passing via the verb hayya'a, then, appears superfluous.Footnote 18
Before ending his comments, al-Zamakhsharī stresses the validity of the second of his three interpretations (that the Quran has been made easy to memorize (ḥifẓ)). He says:
It is related that the followers of [other] religions recite their books, like the Torah and the Gospel, only by looking at them, and that they do not memorize them as [the Muslims do with] the Quran.
Having presented what could be said about this passage, both Muʿtazilite in meaning and other, al-Zamakhsharī moves on to the next verse. The only indication that he might prefer the Muʿtazilite interpretation is that it was given first.
In line with the previous analysis is al-Zamakhsharī's rather imaginative use of one of his sources in his commentary on Q35:8/9: “And what of him, the evil of whose deeds has been decked out fair to him, so that he thinks it is good?” Here we see him ostensibly quoting this source, al-Zajjāj's Maʿānī l-Qur’ā n, but closer analysis of the original text indicates that al-Zamakhsharī is giving a Muʿtazilite bent to al-Zajjāj's text.Footnote 19 Q35:8/9 is obviously an important verse concerning free will and divine predestination, since it speaks of God's leading astray and guiding whom he will. In order to understand the explanations of the text, the full verse is presented here schematically:
This verse begins with a question to which no answer is given.Footnote 20 Al-Zamakhsharī, and al-Zajjāj before him, realized that an answer was apparently missing. Al-Zamakhsharī creates an elaborate dialogue between God and “his prophet (nabiyyihi)”, into which he injects and then explains parts of this passage. The point is that those who think their evil deeds are good are the same as those whom God leads astray (wa-maʿnā tazyīn al-ʿamal wa-l-iḍlāl wāḥid); and the Prophet is not to waste away regretting any of them.Footnote 21 Al-Zamakhsharī then gives the two explanations he says al-Zajjāj mentioned (dhakara). The latter proposes his explanations by supplying an answer (A, below) to the question (Q, below); this answer is, each time, based on what follows in the Quranic verse, what could be called the “Quranic answer” (Aq, below). Since the contents of the answer are, therefore, indicated by the later passages (dalāla ʿalayhi), that is, the “Quranic answer” (Aq), al-Zajjāj says that the answer (A) to the question (Q) itself has been dropped (ḥudhifa). The two question-and-answer explanations that al-Zamakhsharī says al-Zajjāj gives are as follows:Footnote 22
Q: “And what him, the evil of whose deeds has been decked out fair to him?”
A: Thy soul was wasted in regrets for them (dhahabat nafsuka ʿalayhim ḥasratan).
Aq: “so let not thy soul be wasted in regrets for him (sic)” (ʿalayhi in the Kashshāf).
Here the answer (A) has been dropped because, as shown above, it is given later in the verse (Aq). This is in accord with what al-Zamakhsharī has already stated: the Prophet is not to waste time regretting such people. As for al-Zajjāj's second explanation, as quoted by al-Zamakhsharī, the answer does not fit the question quite so snugly; nor does al-Zamakhsharī give any additional explanations:
Q: “And what of him, the evil of whose deeds has been decked out fair to him?”
A: [He is] like the one whom God has guided (ka-man hadāhu llāh).
Aq: “God leads astray whomsoever He will, and whomsoever He will He guides”.
*
When the Maʿānī l-Qur’ān itself is examined, it becomes clear that al-Zamakhsharī is practically quoting al-Zajjāj, with some slight, but not insignificant, changes. In the first explanation, al-Zajjāj's original question and answer reads:Footnote 23
Q: “And what of him, the evil of whose deeds has been decked out fair to him” and whom God has led astray (fa-aḍallahu llāh)?
A: Thy soul was wasted in regrets for him (dhahabat nafsuka ʿalayhi ḥasratan).
Aq: “so let not they soul be wasted in regrets for them”.
Al-Zamakhsharī's version does not mention the phrase about God leading the evildoers astray. While this could be an oversight on the part of the author or copyist, the context is far too tantalizing to allow easy acceptance of such an explanation. It looks, rather, like the elimination of a statement contrary to the Muʿtazilite explanation that al-Zamakhsharī has just given concerning the true meaning of God's leading someone astray.
As for the second explanation, we see al-Zamakhsharī's Muʿtazilite hand at work even more clearly. In the Maʿānī l-Qur’ān, al-Zajjāj's original sequence reads:
Q: “And what of him, the evil of whose deeds has been decked out fair to him?”
A: [He is] like the one from whom God has turned away (ka-man taʿāddāhu llāh)
Aq: God leads astray whomsoever He will, and whomsoever He will He guides.
In this sequence we see that al-Zamakhsharī has done the opposite of al-Zajjāj. Since the Quranic answer (Aq) has two parts, one answer (A) is as acceptable as the other. However, as we have seen, al-Zamakhsharī's answer does not fit as easily as one would expect; it appears, again, to be an attempt to avoid saying that God leads people astray. More seriously, perhaps, he appears to attribute this interpretation to al-Zajjāj. The whole passage is a clear example of al-Zamakhsharī altering a passage in the Maʿānī l-Qur’ān to make it fit his own theological views more closely.
The aforementioned study of al-Zamakhsharī's commentary on Q44 and Q54 showed that the Kashshāf had all the elements of a traditional Quran commentary (tafsīr musalsal) and in no way distinguished itself within the genre. Its author employed all the traditional techniques (tafsīr al-Qur’ān bi-l-Qur’ān, grammar, questions-and-answers (masā’il wa-ajwiba)) and information (variant readings (qirā’āt), occasions of revelation (asbāb al-nuzūl), traditions (aḥādīth), poetry and other transmitted knowledge) at his disposal to explain the meaning of the text on a word-by-word or phrase-by-phrase basis. This indicates that the Kashshāf retained all the elements of a traditional work of exegesis, no matter how it was viewed or defined with respect to the Muʿtazilism of its author. Furthermore, with respect to this theology and despite the fact that al-Zamakhsharī was accused by some of using the Kashshāf as a means for propagating it, an attentive reading of his commentary on the forty-fourth and fifty-fourth sūras reveals nothing by way of polemics, proseletyzing or even theological discussion; one sees only a passing reference to a Muʿtazilite principle. Likewise, a comparative study of the Kashshāf with one of its sources, al-Zajjāj's Maʿānī l-Qur’ān, revelaed only one instance reflecting al-Zamakhsharī's theological position, in his “quotation” from the Maʿānī. The Kashshāf received strong criticism in some circles, and this suggests that it was caused more by al-Zamakhsharī's being a Muʿtazilite than by the content of the Kashshāf.
2. The second approach: possible theological influences on the Kashshāf
From the previous section it would seem that, despite the general accusations of heresy, al-Zamakhsharī's commentary was on the whole well within the accepted exegetical tradition. In this section, I will compare the Kashshāf with two theological treatises to see if any of the theological ideas from the latter can be detected. The treatises are al-Zamakhsharī's own al-Minhāj fī uṣūl al-dīn, and al-Mu ʿtamad fī uṣūl al-dīn, the major theological treatise of his teacher Rukn al-Dīn (Ibn) al-Malāḥimī (d. 536/1141).
2.1 Al-Zamakhsharī's al-Minhāj fī uṣūl al-dīn
Al-Minhāj fī uṣūl al-dīn is al-Zamakhsharī's short, and only, truly theological treatise. However, its theological content is not the main interest here; rather, it is used to see to what extent the “Muʿtazilite content” of Quranic verses used in the Minhāj can be found repeated or reflected in the Kashshāf. There is no assumption that the Minhāj was written before the Kashshāf or that it was a source. We do not know when the Minhāj was completed, although the Kashshāf was completed in 528/1134. The goal, then, is to see to what extent, if at all, the theological ideas al-Zamakhsharī associated with certain Quranic passages or verses in the Minhāj were also associated with them in the Kashshāf. It is this association of theological ideas and Quranic texts that is our focus here.
The Minhāj fī uṣūl al-dīn is described by Madelung as “a brief summary of his theological creed”, and by Schmidtke as “a short creedal tract on theology”. It is divided into nine chapters, each laid out in the usual question-and-answer (masā’il wa-ajwiba) format. A tenth chapter on the imamate is lacking from surviving Yemeni manuscripts, since it did not agree with the doctrine of the Kufan Zaydīs who brought the Minhāj to Yemen.Footnote 24 The chapter titles given by Schmidtke indicate that al-Zamakhsharī deals with themes that were dear to the hearts of the Muʿtazilites.Footnote 25 In particular, the titles of Chapter 7 (Chapter on the Promise and the Threat) and Chapter 8 (Chapter on the Command of what is Proper and the Interdiction of what is Reprehensible) state two of the five principles of the Muʿtazilite School. Their position at the end of the treatise probably indicates their relative importance. The long second chapter (Chapter on the Knowledge of God and His Attributes) deals with the unity of God (tawḥīd), a fundamental Muʿtazilite fundamental principle, while the next three chapters (Chapter on the Imposition of Moral Obligations, Chapter on the Facilitating Favours and Chapter on Pains) deal with themes associated with the other fundamental principle, divine justice (ʿadl). The sixth (Chapter on Sustenance, Prices and Terms of Death) brushes with the question of divine predestination (qadar). These nine chapters are of similar length, two to three pages in Schmidtke's edition, although the seventh, on ‘the promise and the threat’ is four full pages. The only exception is the second chapter which, at ten pages, amounts to nearly a third of the entire treatise (even if a few of these pages bear some ample footnotes).
In the Minhāj, Madelung argues, al-Zamakhsharī avoided entering into the controversies that separated the two branches of the Muʿtazila that existed in Khwārazm during his time: those who followed ʿAbd al-Jabbār (d. 415/1025), known as the Bahshamiyya, and those who followed the latter's student Abū l-Ḥusayn al-Baṣrī (d. 436/1044). While al-Zamakhsharī's method was masā’il wa-ajwiba, his approach was “catholic”: if he mentioned two conflicting views, he did not take sides or even directly indicate his preference; nor did he ever refute a Muʿtazilite view. He was influenced by the views of Abū l-Ḥusayn al-Baṣrī and Rukn al-Dīn (Ibn) al-Malāḥimī (d. 536/1141), a disciple of the former and one of al-Zamakhsharī's contemporaries, as a closer reading of the Minhāj reveals – nowhere does he uphold views of the Bahshamiyya that the latter had rejected.Footnote 26 While being influenced by the theology of Abū l-Ḥusayn al-Baṣrī and Ibn al-Malāḥimī, Madelung notes that al-Zamakhsharī did not identify completely with it, and so “lent support to a broadly based, catholic Muʿtazilism”.Footnote 27 The question now is: are these influences stated or at least reflected in the Kashshāf?
2.1.1 Overall view on the use of the Quran in the Minhāj
In the Minhāj, al-Zamakhsharī makes specific reference to twenty-eight Quranic passages, either short verses or parts of longer ones. These are not divided equally among the nine chapters (which, with the exception of the second, are of a similar length). It is not surprising that a significant number of the Quranic references are found in chapter 2 (12 of the 28). Chapter 7, however, also has a good number of references (9); the rest are divided between Chapter 8 (two) and Chapter 9 (five). The remaining chapters do not contain Quranic references. Furthermore, it should be noted that Quranic references are not spread evenly through these chapters. Al-Zamakhsharī frequently supports a position with three or four passages from the Quran, so that, even in the relatively short Chapter 7, the nine references are found in just three places: a single verse on one occasion and two groups of four on the other.
One might ask about the role of these Quranic passages in the Minhāj. A study of the text reveals that this is twofold: an illustrative role, for example at the end of Chapter 9 on prophecy, where al-Zamakhsharī deals with Muḥammad's miracle and the miraculous character of the Quran. First he states what a miracle is and gives the conditions for its occurrence; he then asks: “What is the miracle (muʿjiza) of Muḥammad?” He begins his answer by mentioning the Quran, but then adds others, such as the splitting of the moon (inshiqāq al-qamar), the glorification of God by the pebbles (tasbīḥ al-ḥaṣā), the moaning of the tree trunk (ḥanīn al-jidhʿ), the feeding of the crowds with little food and the quenching of their thirst with little water (iṭʿām al-jamāʿāt min ṭaʿām yasīr wa-saqyuhum min mā’ qalīl) and, finally, Muḥammad's communication of hidden matters (ikhbāruhu bi-l-ghuyūb). Al-Zamakhsharī then returns to the Quran to describe in more detail its miraculous character and gives two aspects of this: the Quranic passages are used to illustrate the second characteristic of the Quran's miraculous character, its “reports on hidden matters”.Footnote 28
The miraculous character of the Quran is two-fold: (first,) the inability of the people of eloquence [(ahl al-fasāḥa)] to match it and (secondly,) the reports on hidden matters in it like the saying of God the Exalted “and you will not do”,Footnote 29 “Certainly the host shall be routed”,Footnote 30 “that He may uplift it above every religion”,Footnote 31 “God will protect thee from men”.Footnote 32
Al-Zamakhsharī's answer and his demonstration of the Quran's two-fold miraculous character ends rather abruptly here. No explanations are given as to the context of these Quranic passages; as with the list of miracles, he accepts that his reader is familiar with them.Footnote 33 The context of these verses is as follows:
1. Q2:24/22 has to do with the challenge to the polytheists to produce a sūra like one in the Quran; but the Quran says that they will not do this (lan taf ʿalū);
2. Q54:45 is taken as a reference to the victory at Badr;
3. Q9:33 says the religion brought by Muḥammad will be raised above other religions;
4. Q5: 67/71 is an order to the Prophet to deliver his message and is accompanied by a promise of divine protection.
Three of these verses were later understood to refer to events that came to pass afterwards. What is important for al-Zamakhsharī, though, is that they are reports of hidden matters known only to God and revealed in the Quran.
Besides this more illustrative role in the Minhāj, Quranic passages can sometimes be raised to the level of a “proof” (dalīl). An example is found in Chapter 7 on ‘The promise and the threat’, where the question is asked as to the proof (dalīl) that the persistent offender will stay eternally in hell; here al-Zamakhsharī gives three Quranic passages followed by a saying of the Prophet: “Whosoever throws himself off a mountain will fall into hell-fire for eternity; and whosoever kills himself with a knife will be in hell-fire eternally, sticking his knife into his belly with his [own] hand”. Here al-Zamakhsharī also mentions that his opponents, whom he calls the Murji'ites, are wrong to cling to traditions which are countered by others which agree with the Quran, although he does not elaborate. On such occasions, though, there does not appear to be a distinction between a rational proof (dalīl) and a more traditional argument from authority where the revealed text is simply brought forward to support a position; in each case either a series of passages from the Quran or sometimes even a single passage is presented.Footnote 34 It may be that al-Zamakhsharī takes the word dalīl to mean “indication” or “sign” rather than “decisive proof” or “clear demonstration”.Footnote 35 Either way, in the preceding section of the Minhāj, the question is asked as to the scriptural indications (sing. māniʿ al-samʿ) prohibiting prophetic intercession for the grave sinner, and here again we find Quranic passages. Al-Zamakhsharī's “proof” (dalīl), then, that the persistent sinner remains in hell does not involve carefully crafted rational arguments with respect to the correct understanding of the traditional material he has presented. In fact, the only developed discussion of this kind is found in the latter section where, after giving the scriptural indications prohibiting prophetic intercession, al-Zamakhsharī has to explain the tradition where the Prophet says: “[My] intercession is for the grave offenders among my nation”. Al-Zamakhsharī concludes this section with another appeal to authority: “We have held to what is agreed upon (fa-akhadhnā bi-mā ttufiqa ʿalayhi)”.Footnote 36
When al-Zamakhsharī's demonstrations go beyond a simple list of Quranic passages, one might question their relevance; perhaps there is a history with which his readers are familiar. In his use of one cluster of four Quranic passages in Chapter 2, for example, the Quranic passages are incorporated into a proof that God can be neither perceived by the senses nor seen.Footnote 37 The section begins with the question: “What is the proof (dalīl) that He cannot be perceived by any sense (lā yudraku bi-ḥāssa) and that He is not seen (laysa bi-mar’ī)?” The answer has three related parts: (i) God is neither a body (jism) nor an accident (ʿaraḍ) and so cannot be perceived by any of the senses; (ii) God remains unseen even if one has good eyesight and all impediments (al-mawāniʿ min al-ru'ya) have been removed;Footnote 38 and (iii) God cannot be faced (muqābala) since he is not a body and does not subsist (ḥāl) in one. Al-Zamakhsharī then abruptly quotes four Quranic passages without introduction: “The eyes attain Him not” (Q6:103), “Thou shalt not see Me” (Q7:143/138), “(those) … shall never create a fly” (Q22:73/72) and “for they said, ‘Show us God openly’. And the thunderbolt took them for their evil doing” (Q4:153/152).Footnote 39 When we look closely at these passages, it becomes clear that the main thrust is the impossibility of seeing God, of looking at him. They do not deal with God's not being a body (jism) or subsisting in one (ḥāl fī jism), nor do they refer to either of the two preceding points. They “prove” that God can be neither perceived nor seen, in much the same way as the verses in the preceding example proved that the persistent sinner will remain in hell for ever: by making a broad statement that is not at odds with the theological position. It should be added here that, as in the previous example and on other occasions where al-Zamakhsharī offers more than Quranic proof for the position he is defending, references to the Quran are not always given first. Sometimes Quranic verses are used merely to support secondary points that arise in the course of the argument. Perhaps this is Gardet's thinking when he writes: “The fact that these ‘traditional’ arguments are in some manuals listed after the “rational” arguments indicates that the former are to be regarded as a confirmatur to the results of dialectical reasoning”.Footnote 40
The overall impression left by the Minhāj is that it is “a brief summary of [al-Zamakhsharī's] theological creed”, a short work in which the main tenets of Muʿtazilism were exposed and “defended” by set answers and Quranic references. Within this framework, Quranic verses and passages have a role as an authority next to rational arguments, although in what we have seen there were few of these arguments. Unlike tafsīr, where he would comment on individual words or phrases, in the Minhāj he uses them to support or illustrate the theological position he is upholding. In these cases al-Zamakhsharī either presumes that the meaning of the passage is clear – and in line with the position he is taking – or he explains it in terms of this position. This was not new to al-Zamakhsharī: he is merely reflecting the approach of his predecessors in the field of Kalām, even if he makes relatively little use of direct scriptural quotes for his theological purposes. What is interesting, though, is how he blurs the lines between rational and traditional arguments in his presentation, since he often quotes scripture or tradtions as part of what he calls a dalīl.
2.1.2 Reflections of the Minhāj in the Kashshāf
Whatever the final evaluation of the Minhāj as a theological work is, it is important in that it incorporates Quranic references. In the Minhāj we are presented with 28 Quranic passages which had a Muʿtazilite connection, if not content, for al-Zamakhsharī. While the use of these passages in Muʿtazilite treatises probably did not start with al-Zamakhsharī, what is of interest is how far their Muʿtazilite connection or content is to be found in the Kashshāf. Studying the Minhāj and the Kashshāf allows us to see al-Zamakhsharī's interpretative process at work from both ends. In the Minhāj we saw him illustrating or defending his theological positions with reference to the Quran; in the Kashshāf we can see the same process starting from the Quran. What, then, do we see when we study al-Zamakhsharī's commentary on these 28 passages in the Kashshāf? The statistics are shown in Table 1.
Table 1. Muʿtazilite “content” of the 28 quranic verses according to the Kashshāf
A careful examination of the exegesis of these verses or passages reveals that in only ten cases (35.7%) does al-Zamakhsharī raise the theological points for which he used them in the Minhāj. This does not mean that in his commentary on some of the remaining eighteen passages (64.3%) he does not express ideas that could be attributed to a Muʿtazilite position. Seven of these eighteen verses (38.9%) have some Muʿtazilite content while the other 11 (61.1%) do not. However, in these seven cases, al-Zamakhsharī does not mention the ideas he defended in the Minhāj. This means that in approximately 60% of the 28 verses Muʿtazilite content can be found, and that more than half of these (58.8%) are in the Minhāj. In 40% of the cases nothing Muʿtazilite was raised.
An example of the intermediary situation can be found in al-Zamakhsharī's commentary, which he brings forth in the seventh chapter of the Minhāj on the Promise and the Threat as scriptural indications (sing. māniʿ al-samʿ) prohibiting prophetic intercession for the grave sinner (see pp. 38–9 (E); p. 75 (A)). He here presents passages from four verses (in italics below) without further comment: Q24:23; 40:18; 2:48/45; 39:19/20.Footnote 41 The first passage might also indicate what, for al-Zamakhsharī, amounted to a grave sin, although the main theme of these passages is definitely elsewhere.
Q24:23: Surely those who cast it up on women in wedlock that are heedless but believing shall be accursed in the present world and the world to come; and there awaits them a mighty chastisement.Footnote 42
Q40:18: And warn them against the Day of the Imminent when, choking with anguish, the hearts are in the throats and the evil doers have not one loyal friend, no intercessor to be heeded.
Q2: 48/45: And beware of a day when no soul for another shall give satisfaction, and no intercession shall be accepted from it, nor any counterpoise be taken, neither shall they be helped.
Q39: 19/20: He against whom the word of chastisement is realized – shalt thou deliver him out of the Fire? Footnote 43
When the commentary on these verses in the Kashshāf is examined, we note that at Q24:23 al-Zamakhsharī does not raise the matter of Quranic impediments to prophetic intercession for the grave sinner. In his commentary on Q40:18, he devotes a few lines to the topic of intercession (shafāʿa), saying that it is an additional favour (ziyādat al-tafaḍḍul) and only those who repent receive it, a position similar to that expressed in the Minhāj in his explanation of the tradition on prophetic intercession: “[My] intercession is for the grave offenders among my nation”.Footnote 44 His commentary in the Kashshāf would, of course, cover prophetic intercession, but al-Zamakhsharī does not raise this topic, nor is there reason to assume he was thinking solely of prophetic intercession, since the verse speaks of “loyal friend” (ḥamīm) and “intercessor” (shafīʿ). At Q2:48/45, the topic of intercession for the “disobedient” (al-ʿuṣāt) is dealt with in the same way, but there is no specific reference to the intercession of a prophet.Footnote 45 Finally, although in his commentary on Q39:19/20, al-Zamakhsharī makes it clear that God alone can release a soul from the Fire and, as a consequence, the role of the prophet is non-existent, he does not raise the topic of intercession (shafāʿa). The focus is the prophet's actions in this world and his inability to bring about faith, not his intercessory powers in the next world.Footnote 46
In the Kashshāf, then, the interdiction of prophetic intercession is not mentioned explicitly in the commentary on any of these verses; nor are they presented as being “revealed prohibitions”. In the first verse nothing is said on the topic. In the second, intercession is mentioned as an additional favour for those who have repented; in the third, there is a question about intercession for the disobedient not being accepted; and in the fourth, the stress is on God's omnipotence, that is, that Muḥammad can neither create faith in this world nor save those who merit chastisement in the next.
While, in the previous example, al-Zamakhsharī's commentary reflects Muʿtazilite ideas not specifically mentioned in the Minhāj, elsewhere in the Kashshāf we find him discussing, with respect to certain Quranic passages, the ideas he raised in the Minhāj when these passages were quoted. For example, the first cluster of Quranic references in the Minhāj, alluded to above, contains four verses: Q6:103; 7:143/139; 22:73/72; 4:153/152. They are incorporated into a discussion in Chapter 2 on the impossibility of seeing God. In the commentary on some of these verses in the Kashshāf we see some of the same ideas.
In his commentary on Q6:103: lā tudrikuhu l-abṣāru (“The eyes attain him not”) in the Kashshāf, al-Zamakhsharī is evidently in the same zone as when composing the Minhāj. There he uses only a part of Q6:103; the full verse is as follows:
The eyes attain Him not, but he attains the eyes; He is the All-subtle, the All-aware.
lā tudrikuhu l-abṣāru wa-huwa yudriku l-abṣāra wa-huwa l-laṭīfu l-khabīru.
In the Kashshāf he first defines “vision (al-baṣar)” as: “the subtle substance that God has set in the sense of sight, by which what can be seen is attained”.Footnote 47
He then explains the passage:
The meaning is that the eyes do not connect with Him and do not attain Him, because in His essence He is above being seen. The eyes connect only with what is in some position, either primarily or secondarily, like bodies (ajsām) and forms (hay’āt).Footnote 48
In the Kashshāf he tends to repeat himself and even gives one part of the verse as an explanation of another. Nevertheless, it is clear that he is expressing the same ideas as those in Chapter 2 of the Minhā j, where he used the verse as a proof-text, though here the development is greater. The philosophical vocabulary is present, and he states clearly that only bodies and forms can be perceived by the faculty of sight, but that God is in his essence beyond such visible things, too subtle to be seen, although he sees all and nothing is too subtle for his perception. It could be noted, however, that al-Zamakhsharī seems to be grappling with either the ideas or the vocabulary, for in his commentary on the last three short phrases he uses words derived from the fourth form of the root d-r-k a total of eight times.Footnote 49
The next relevant verse is Q7:143/139: lan tarān ī (“Thou shalt not see Me”). This is a long verse, containing a dialogue between God and Moses, and al-Zamakhsharī devotes a lot of ink to it.Footnote 50 Of greatest interest here is the connection between lan tarānī or other parts of the verse, and the Kalām topic for which it was used in Chapter 2 of the Minhāj. When al-Zamakhsharī comes to the statement lan tarānī, it is evident that he has this Kalām topic clearly in mind. When he asks his theoretical question as to why Moses should have asked to see God, since as a prophet he should have known better, his question is six lines long and covers a number of theological ideas that are only touched upon in the Minhāj.Footnote 51 He shows that God cannot be seen because of what he is and then establishes that Moses' request was made only to silence those who had asked – a point already made in the Minhāj.Footnote 52 He returns to this second point, stressing that such a request would imply a crass anthropomorphism unworthy of a prophet.Footnote 53 Throughout the commentary on this verse, the theological ideas of the Muʿtazilites are clear. If any greater indication of where al-Zamakhsharī stands were needed, there is the list of the early members of the school he gives, from the founding fathers to its most famous theologians of the fourth/tenth century: Wāṣil ibn ʿAṭā’ (d. 748), ʿAmr ibn ʿUbayd (d. 761), al-Naẓẓām (d. 836), Abū l-Hudhayl (d. 841?), Abū ʿAlī al-Jubbā’ī (d. 915) and his son Abū Hāshim al-Jubbā’ī (d. 933). Both the ideas and the vocabulary used to express them are identical to those found in the Minhāj. In the Kashshāf, however, the topic is developed far more than in the theological treatise. The extra time and space allow al-Zamakhsharī to take a swipe at his opponents (whom he calls the “Mujbira”); he argues that what they say does not hold up since they are sinners!
The third verse used in the Minhāj to illustrate the point is Q22:73/72: lan yakhluqū dhubāban (“ (those) … shall never create a fly)”. In the Minhāj, the phrase was used merely to indicate impossibility, the impossibility of seeing God; no context was given for it. In fact, the verse deals with the gods of the polytheists, which cannot create even so much as a fly. In the Kashshāf, al-Zamakhsharī does not make any reference to seeing God or to any point from the theological debate that surrounds the phrase in the Minhāj.Footnote 54 In the Kashshāf it illustrates the impotence of the polytheists’ idols; in the Minhāj it illustrates the impossibility of seeing God.
The fourth and final verse used in the first cluster of Minhāj is Q4:153/152: fa-qālū arinā llāha jahratan fa-akhadhathumu l-ṣāʿiqatu bi-ẓulmihim (“for they said, ‘Show us God openly’. And the thunderbolt took them for their evil doing”). In the Minhāj, al-Zamakhsharī states clearly that the punishment was due to their demanding something impossible, that is, seeing God, and gives an example of Abraham's demanding something that was possible:
If they had asked for something that [was] possible, they would not have done evil and would not have been taken by the thunderbolt, as when [Abraham] demanded the revivification of [the] dead, [for] he did no evil and was not taken by the thunderbolt.Footnote 55
In the Kashshāf, at this passage, there is no reference to the theological questions that were the context for its use in the Minhāj. Al-Zamakhsharī merely states that the Quranic “for their evil doing” (bi-ẓulmihim) means “because of their asking to see” (bi-sabab su’ālihim al-ru'ya). He then gives the same example about Abraham in practically the same words, ending with a shot at his opponents (whom he calls the “Mushabbiha”): “May evil and thunderbolts befall the Mushabbiha”.Footnote 56
With Q4:153/152, the first cluster of verses from the Minhāj has been examined. While the commentary on the first two verses in this cluster reflects Muʿtazilite theological thinking, the other two do not; Q22:73/72 does not even deal with the theme of seeing God.
One might question the meaning of these statistics. What do they say about the Muʿtazilite content of the Kashshāf in general, assuming that one can extrapolate from them? If it is recalled that in five of the Minhāj's nine chapters, al-Zamakhsharī made no direct reference to the Quran, the first idea that springs to mind is that in al-Zamakhsharī's Kalām, Quranic passages played a secondary role. Furthermore, considering that al-Zamakhsharī was not primarily a theologian, there is no need to assume that his knowledge of the minutiae of Kalām debate and the Muʿtazilite positions was excessive. Not only was the Minhāj a brief summary of al-Zamakhsharī's theological position, as Madelung says, but it could also have been all that he was capable of writing on the topic. As for the Kashshāf, al-Zamakhsharī appears to manifest an overall lack of concern there for the specific “Muʿtazilite content” of the verses that he used in the Minhāj. As shown above, there was a lack of reciprocity between the Kashshāf and the Minhāj at the 28 verses used in the latter; for in about 65% of cases, what was to be found in the Minhāj did not reappear in the Kashshāf. Al-Zamakhsharī was clearly not constantly thinking about Kalām as he composed his Quran commentary. This is in line with the results of the study of Q44 and Q54. Consequently, a comparison with the Minhāj fī uṣūl al-dīn supports the view that the Kashshāf was not written primarily as a vehicle for Muʿtazilite theology.
2.2 Rukn al-Dīn (Ibn) al-Malāḥimī's (d. 536/1141) al-Muʿtamad fī uṣūl al-dīn
In Rukn al-Dīn ibn al-Malāḥimī, “the main representative of the school of Abū l-Ḥusayn al-Baṣrī in the first half of the 6th/12th century”, we find a scholar whose lifespan corresponds with that of al-Zamakhsharī.Footnote 57 Ibn al-Malāḥimī taught al-Zamakhsharī theology, while the latter taught the former exegesis.Footnote 58 Furthermore, the extant parts of Ibn al-Malāḥimī's Kalām work, al-Muʿtamad fī uṣūl al-dīn, have been published and are available for scrutiny.Footnote 59 This work could offer us a direct and contemporary source for the Muʿtazilism of the Kashshāf.Footnote 60 However, the fact that material is found in both the Muʿtamad and the Kashshāf is hardly a guarantee that the former was a source for the latter. It may be that al-Zamakhsharī had completed the Kashshāf before Ibn al-Malāḥimī even began the Muʿtamad, and that they both drew on earlier, common sources; no dates indicate when the latter was finished so it is necessary to keep this point in mind.
2.2.1 Overall view on the use of the Quran in the Muʿtamad
A study of the Muʿtamad shows, first of all, that the Quranic verses are not spread evenly throughout the entire work. In fact, they are quite limited. Besides a few in the opening pages of the first part and at the very end of the single chapter from the fourth part (for the divisions of the Muʿtamad, see note 59), Quranic references in the main body of the book are limited to nine chapters, four chapters in the first part and five in the third part. If it is taken into account that the first part has twenty-two chapters and two independent sections, and that the third part has fourteen chapters, two of which have three subsections (sing. faṣl), to say nothing of the final fifty-page section on sects opposed to the Islamic doctrine of tawḥīd (that has no divisions into chapters and subsections); then it becomes clear that Ibn al-Malāḥimī uses the Quran as sparingly here as al-Zamakhsharī did in the Minhāj. As in the Minhāj, verses tend to be presented in clusters in chapters that range from one page to nearly one hundred.
In the main body of the Muʿtamad, Ibn al-Malāḥimī makes about 160 references to nearly 140 Quranic passages, either individual verses, several verses together, or simply parts of verses. Occasionally he refers to a verse more than once but rarely more than twice; the majority of the verses mentioned in the Muʿtamad are used only once. Ibn al-Malāḥimī indicates on several occasions that what he has given by way of Quranic references is not everything that could have been supplied, writing such things as wa-ghayr dhālika, ilā ghayr dhālika or ilā ghayr dhālika min al-āyāt after a number of Quranic citations.
Methodologically I will try to establish here whether any of the theological ideas Ibn al-Malāḥimī buttressed with Quranic passages in the Muʿtamad are repeated or reflected in al-Zamakhsharī's commentary on the same verses in the Kashshāf. Because of the large number of such references, it is not possible here to give an exhaustive comparison of their use in both works. As was the case with the more limited Minhāj fī uṣūl al-dīn, a few examples will be chosen to illustrate Ibn al-Malāḥimī's use of a verse or cluster of verses in the Muʿtamad; al-Zamakhsharī's comments on these same verses in the Kashshāf will then be presented. These verses will be drawn from an earlier study of approximately twenty verses (approximately 13% of the 140 or so) that were examined with respect to their use in the Kashshāf and the Muʿtamad.Footnote 61 A few remarks will then be offered as to the extent to which al-Zamakhsharī employed Ibn al-Malāḥimī's Muʿtamad when writing the Kashshāf.
2.2.2 Ibn al-Malāḥimī's Muʿtamad: a source for the Kashshāf?
Part 3, ch. 11 of the Muʿtamad deals with the meaning of describing God as being one and the meaning of divine unity. Ibn al-Malāḥimī here uses Q2:163/158: wa-ilāhukum ilāhun wāḥidun lā ilāha illā huwa l-raḥmānu l-raḥīmu (“Your God is One God; there is no god but He, the All-merciful, the All-compassionate”); it occurs alone in a discussion about what it means when the Muʿtazilite teachers (shuyūkhunā) describe God as being one (wāḥid). Ibn al-Malāḥimī says that one of three things is meant:Footnote 62
1. A single essence that cannot be divided or broken into parts (dhāt wāḥida lā yajūzu ʿalayhi l-tajazzu’ wa-l-tabʿīḍ). This is not applied to God in the way of praise (ʿalā jihat al-madḥ), since others too are described in this way.
2. No one shares with God the qualities that are attributed to him in the way that they are attributed to him; in this sense, it is said that he has no second.
3. God is the one god (ilāh wāḥid), who alone deserves worship (ʿibāda) and has no partners in it. Ibn al-Malāḥimī then quotes part of Q2:163/158: “Your God is One God; there is no god but He, the All-merciful, the All-compassionate”. Ibn al-Malāḥimī concludes the third point by referring to the last words of the Quranic passage he quotes: “By these two qualities He (May He be exalted) is praised”.
In this list we can see three aspects of divine unity underlined, or the three angles from which the idea is viewed. These can be described as: (i) internal unity, that is, indivisibility; (ii) uniqueness, that is, God is completely different from everything else (tanzīh); and (iii) external unity, that is, there is no god with God.
In the Kashshāf at Q2:163/158, al-Zamakhsharī mentions the third meaning, which Ibn al-Malāḥimī illustrated by a reference to the same verse, i.e. al-Zamakhsharī speaks of God's unity in terms of his being the only god, of being unique (sole, single or one of its kind, and not exceptional, inimitable or matchless, which is the second understanding given by Ibn al-Malāḥimī). He also refers to the two epithets (raḥmān and raḥīm), as did Ibn al-Malāḥimī; however, his comments are more in line with Ibn al-Malāḥimī's second point, that of God's incomparability with creatures in his attributes. He writes:
“One God”: unique with respect to divinity, with absolutely no one sharing it with him; it is not correct to call anyone other than him a god. “There is no god but He”: affirmation of unity, denying [other gods] and affirming him. “The All-merciful, the All-compassionate”: Lord of all benefits, both in their sources and in their developments; no one other than he is with this quality. Everything other than he is either a grace or object of a grace.Footnote 63
In al-Zamakhsharī's comments here, we see some similarity with the Muʿtamad. However, he gives a wider interpretation of the verse than Ibn al-Malāḥimī in that he incorporates ideas of tanzīh. Furthermore, while this would also have been an occasion for al-Zamakhsharī to give the summary of the shaykhs’ views that Ibn al-Malāḥimī did, he chooses not to do so.
Part 3, ch. 3 of the Muʿtamad, dealing with the impossibility of God's being a location for accidents, contains three references to the Quran, all of which are grouped together. Concerning the meaning of verse Q89:22/23: “and Thy Lord comes (wa-jā’a rabbuka)” (and that of parts of two other verses (Q2:210/206; 6:158/159) that speak of the Lord's coming (using the verbs jā’a and atā)), Ibn al-Malāḥimī says that, although this verse speaks of God's essence (dhāt), what is meant are his acts (afʿāl) and his punishments (ʿuqūbāt).Footnote 64 His answer involves listing other verses (Q59:2; 16:26/28) to explain how these three are to be understood, and to show that the true meaning of the verses is not the apparent one. Ibn al-Malāḥimī ends this section with a commentary on Q2:210/206: “What do they look for, but that God shall come to them in the cloud-shadows?” (one of the verses he has just referred to) that was transmitted on the authority of Ibn ʿAbbās, who is quoted as saying: “He will come to them with the promise and the threat (al-waʿd wa-l-waʿīd), and he will reveal to them what of the command (al-amr) was hidden from them”. It is interesting to note how this very Muʿtazilite tradition has made its way into Ibn al-Malāḥimī's treatise.Footnote 65
In the Kashshāf, al-Zamakhsharī makes some comments on Q89:22/23: “and Thy Lord comes (wa-jā’a rabbuka)”. However, while he raises essentially the same question as Ibn al-Malāḥimī concerning predicating motion of God, his answer is different. In the Muʿtamad, this verse was used in Ibn al-Malāḥimī's discussion of the essence of God, his movement, acts and punishments, and God's “coming” meant His acts and His punishments; in the Kashshāf al-Zamakhsharī sees it as an expression of incomparable divine majesty and glory. One might argue that al-Zamakhsharī's view is similar to Ibn al-Malāḥimī's here, in that he says that predicating the verb “to come” of God is “a comparison to the manifestation of the signs of [God's] might and the demonstration of the effects of his power and authority” and, therefore, in a sense he is saying that the verse speaks of the manifestion of something other than the essence (dhāt) of God. However, the rest of the passage indicates that al-Zamakhsharī is more concerned with the greatness of God's glory and majesty than with questions of the divine essence, for he writes: “In this, [God's] state is comparable to the state of a king who, if he is present himself, there appear with his presence effects of awe and command that are not evident with the presence of all of his armies and of all his ministers and leading personalities together without exception.’”Footnote 66
Reading Ibn al-Malāḥimī's Muʿtamad, it is striking how very little direct use he makes of the Quran in this six-hundred page book. When he does use the revealed text, the impression is the same as that left by al-Zamakhsharī's use of it in the Minhāj: the role is secondary. When the comparison is made with the Kashshāf at the twenty verses originally studied, practically nowhere in al-Zamakhsharī's commentary on these verses can specific ideas expressed in the Muʿtamad be found. It is only in his commentary on Q2:163 and Q5:116 (not mentioned in this article) that al-Zamakhsharī raises explanations that resemble material found in the Muʿtamad. Even in the former case, the similarities are counterbalanced by differences. Something similar is seen in the commentary on Q89:22 (above), where al-Zamakhsharī asks essentially the same question as Ibn al-Malāḥimī, only to give a different answer. For the remaining verses, there are no similarities between the Muʿtamad and the Kashshāf. Thus, if these twenty verses are indicative, the Muʿtamad played no role in the formulation of the ideas and explanations in the Kashshāf; where similarities are apparent they can probably be attributed to chance. Since Ibn al-Malāḥimī was al-Zamakhsharī's teacher, this probably means that al-Zamakhsharī had no access to the Muʿtamad.
The lack of common ground between the two works reflects the scarcity of shared ideas between al-Zamakhsharī's own work in Kalām, the Minhāj fī uṣūl al-dīn, and the Kashshāf, and recalls earlier remarks concerning al-Zamakhsharī's preparation as a theologian and his lack of interest in incorporating into his commentary even his own theological ideas. Furthermore, considering this comparison with the Minhāj, it would seem that even if al-Zamakhsharī had had access to the Muʿtamad before he began the Kashshāf, its ideas might not have been reproduced there. In any case, based on our survey of the use of the Quran in a number of chapters of the Muʿtamad, it is clear that this work cannot be considered a source for the Kashshāf.
3. The third approach: the matter of Khalaqa l-Qur’ān Footnote 67
3.1 BackgroundFootnote 68
The matter of the creation of the Quran is intimately linked to both Muʿtazilite theology and the Trial or Inquisition (miḥna) the ʿAbbāsid Caliph al-Ma'mūn set in motion four months before his death in 218/833, and which continued under his successors al-Muʿtaṣim (d. 227/842), al-Wāthiq (d. 232/847) and al-Mutawakkil (d. 247/861) until the latter finally put an end to it in 237/851–2.Footnote 69 Two phases in the development of the doctrine of the created Quran can be observed, the earlier of which did not directly involve the Muʿtazilites. Towards the middle of the eighth century, those who held that the Quran was created focused on the anthropomorphism (tashbīh) of the more traditionalist groups whose concept of God was based on such verses as Q4:164, which spoke of God's having spoken directly to Moses (kallama llāhu Mūsā taklīman) and Q4:125, which spoke of God's having taken Abraham for a friend (wa-ttakhadha llāhu Ibrāhīma khalīlan).Footnote 70 For those who held that the Quran was created, God could not be said to have spoken any more than he could be said to have taken a creature for a friend; God was above such human attributes as speech and friendship. God, therefore, created the sound of speech that was heard, but it was not God's speech in a literal sense. Those opposed to this position did not respond directly but stated that the Quran was God's speech; they sought to show the intimate connection between the Quran and God. This, in fact, was nothing new. What was new was for this group to add either a condemnation of those who said that the Quran was created, or a denial that the Quran was created, initially with the negative “It is not created (laysa bi-makhlūq)” rather than the more positive “It is uncreated (ghayr makhlūq)”. However, during this early period in the debate, the denial of the creation of the Quran was not meant to imply in any way that it was eternal (qadīm).
The traditionalists’ shift towards the doctrine of the eternal Quran was attributable to developments in the arguments of their opponents, the Muʿtazilites and other asserters of the creation of the Quran, in the decades leading up to the miḥna. The latter did not deny that God truly spoke, though they did argue that God produced speech without speech organs; furthermore, they held that the Quran was truly God's speech. What was new was the refusal to admit that anything could be co-eternal with God. To assert that the Quran was not created, i.e. that it was not temporal, was to say that it was eternal, which would destroy the unity of God by setting up something co-eternal with him. The traditionalist response rejected the Muʿtazilite view, noting, in particular, that it implied that God was originally ignorant (for the Quran contained the names and knowledge of God) and that there was a time when God did not speak.
The result of the miḥna, al-Ma'mūn's motives for which are still debated (cf. note 69), hinged on the acceptance by the ʿulamā’ (even under duress) of the statement that the Quran was created. The result of this Trial, however, was not the imposition of the doctrine of the created Quran, but its opposite: the acceptance of the doctrine of the uncreated, eternal Quran. As the doctrine of the created Quran was a Muʿtazilite tenet, they were discredited, and what was to become Sunnism developed unhindered. This did not mean, as Hinds says, that “the inspiration for the miḥna necessarily came from Muʿtazilīs or that its initial purpose was the imposition of Muʿtazilī doctrine”; rather, it was simply the case that, “in the context of the miḥna, Muʿtazilī interests overlapped considerably with those of al-Ma'mūn, for all that they were not identical”.Footnote 71 Neither did the result of the Trial mean that the Muʿtazilites and their doctrines disappeared: they continued to exist and even to thrive, influencing theological traditions both within and outside of Islam. Of Khwārazm, where al-Zamakhsharī was born more than two centuries after the end of the miḥna, Gimaret writes that Muʿtazilism continued to be the dominant ideology “to such an extent that outside this area khwārazmī was understood as a synonym of muʿtazilī”.Footnote 72 In such a milieu, then, beginning a Quran commentary with the words “Praise be to God who created the Quran (al-ḥamdu li-llāh alladhī khalaqa l-Qur’ān)” would not only immediately indicate the theological position of the author but would also call to mind the entire history of the miḥna and the suffering traditionalist scholars underwent at the hands of political authorities who had espoused this Muʿtazilite position.
The aforementioned anecdote, found in a number of biographical dictionaries, says that al-Zamakhsharī would have first written “Praise be to God who created the Quran (al-ḥamdu li-llāh alladhī khalaqa l-Qur’ān)” at the very beginning of the Kashshāf but then changed it to al-ḥamdu li-llāh alladhī jaʿala l-Qur’ān when told that his commentary would be shunned by the people if he did not do so.Footnote 73 Although none of the sources mentions copies of the Kashshāf that had either khalaqa or jaʿala, the latter of which, Ibn Khallikān says, was used by the Muʿtazilites with the meaning of the former anyway (wa-jaʿala ʿindahum bi-maʿnā khalaqa), those who give the report state that it was not al-Zamakhsharī who changed the expression to “Praise be to God who revealed the Quran (al-ḥamdu li-llāh alladhī anzala l-Qur’ān)” (which can be read today), even though they do not agree on who actually made the change.Footnote 74
Ibn Khallikān's remark that jaʿala and khalaqa have the same meaning merits comment. Wolfson has pointed out that the verb jaʿala differed from the verb khalaqa in that the former was used in the Quran in connection with itself while the latter was not. He illustrates this with a reference to Q43:2: “We have made it (jaʿalnāhu) an Arabic Quran”. Referring to Patton, Wolfson also says that, “in his defense of the uncreatedness of the Koran, [Ibn Ḥanbal] argued that the word jaʿal does not mean the same as ḫalaḳ”. In his first letter concerning the creation of the Quran and the Trial, al-Ma'mūn refers to Q43:2 as proof that God did create the Quran, for everything that God makes he also creates. In fact, during the Trial, an Ibn al-Bakkā l-Akbar from Baghdad admitted that the Quran was something “made” (majʿūl) and “newly produced” (muḥdath), but he would not say that it was “created” (makhlūq). When questioned by al-Muʿtaṣim's governor about the meaning of jaʿala in Q43:2, Ibn Ḥanbal replied by quoting Q105:5: “and He made them like green blades devoured (fa-jaʿalahum ka-ʿaṣfin ma'kūlin)”, which also has the verb jaʿala, and asking the governor if anything could be concluded from this text about their having been created.Footnote 75 Consequently, changing khalaqa to jaʿala (if any change ever took place) may not have been as unacceptable as Ibn Khallikān believed. In fact, in the second line of the muqaddima al-Zamakhsharī says that God made (jaʿala) the Quran begin with his high praise (wa-jaʿalahu bi-l-taḥmīd muftataḥan), and yet nowhere in these sources do we find any objections to the use of this word. So the question remains: Did al-Zamakhsharī begin the Kashshāf with this statement? In the next section, manuscript evidence will be examined to test the authenticity of this account.
3.2 Manuscripts of the Kashshāf: the main text and the marginal glossesFootnote 76
A study of 110 manuscripts of the Kashshāf containing the muqaddima revealed that only five had a comment on the khalaqa l-Qur’ān, and these were not all of equal weight. Even some of these glossators rejected the idea that al-Zamakhsharī had first written khalaqa l-Qur’ān and then changed it. In the margin of MS. Fatih 340, for example, there is a note next to the opening lines of the muqaddima that reads:
I have heard some people say that the author wrote khalaqa in the autograph copy (nuskhat al-aṣl), which is the umm al-Kashshāf, and then changed it to anzala out of fear (khawfan) but this is not so, for he was clear about his theological position (madhhab) throughout the entire introduction (khuṭba) and used not to conceal his position. I have studied the Kashshāf [written] in his handwriting (khaṭṭ) and I have seen neither erasure (kashṭ) nor change (taghyīr) in it.Footnote 77
A similar statement is found in the margin of MS. Fatih 344.Footnote 78
MSS. Beyezit 575 and Köprülü 129, however, support the claim that al-Zamakhsharī first wrote khalaqa l-Qur’ān and later changed it.Footnote 79 In the first of these manuscripts, the gloss refers not to a personal study of the Kashshāf but to someone else's observations:
Ibn al-ʿAmīd Amīr Kātib al-Itqānī (May God have mercy on him) said, “I saw in the author's (May God have mercy on him) copy the word anzala written in someone else's handwriting. It was as if in its place there had been khalaqa, in accordance with what was his (i.e. al-Zamakhsharī's) theological position (madhhabuhu)”.Footnote 80
The marginal note in the second manuscript, which has innumerable glosses throughout, reads: “Khalaqa was in the autograph (al-aṣl) but it was removed (kushiṭa) and so anzala was written”.
The fifth manuscript (MS. Shehid ʿAlī Pasha 153) had only the word jaʿala written above anzala.Footnote 81 An isolated word is hard to interpret but this one is not necessarily as neutral as it might appear if it is viewed in the light of the discussion of Ibn Khallikān's comment given above. It would have been far more significant if the main text had had jaʿala and the gloss had been anzala. As it is now, it appears to be either a protest that jaʿala had the same meaning as anzala, or else a half-hearted attempt to put forward what might be considered the Muʿtazilite position. In the latter case the glossator was careful not to write the more incendiary khalaqa.
An examination of the manuscript tradition shows that none of the manuscripts studied contained khalaqa l-Qur’ān or jaʿala l-Qur’ān, neither was there any indication that a later hand had changed a copyist's khalaqa or jaʿala to anzala. In 95.5% of the manuscripts the khalaqa l-Qur’ān received no attention at all. This question then was of interest to practically no one. In the 4.5% that did garner a few words from a glossator, only two said that al-Zamakhsharī had originally written khalaqa l-Qur’ān – less than 2% of the total: MS. Köprülü 129 which has a statement that cannot be substantiated; MS. Bayezit 575 which, on the other hand, is more interesting as the glossator cites from a known source, the aforementioned Ibn al-ʿAmīd Amīr Kātib al-Itqānī (d. 758/1357) who made his own copy of the Kashshāf in 718H.Footnote 82
The important question is: did al-Itqānī say what was attributed to him? He did see that the autograph of the Kashshāf at the Abū Ḥanīfa tomb-shrine for his own copy (MS. Lâleli 216, 718H) was made from it, and this was used by later copyists. However, only part of his copy was available (Q15–Q23) and so we do not know if he indicated that the introduction to the Kashshāf had been changed. While al-Itqānī's copy is of no interest for the matter at hand, a possible source for his views may be found in two complete copies of the Kashshāf made using al-Itqānī's copy. MSS. Ali Emiri Efendi 80–81 (840H) were copied from that belonging to al-Itqānī, and MS. Nurosmaniye 290/399 (1050H) was collated with the one belonging to him.Footnote 83 Each copy has the usual expression, anzala l-Qur’ān, in the introduction, and neither has marginalia to indicate that there had been anything different in either al-Itqānī's copy or al-Zamakhsharī's original. This indicates that al-Itqānī's copy had the standard text and makes one wonder why, if, as the glossator of MS. Bayezit 575 has claimed, al-Itqānī had said that anzala had been written by another hand, he did not write or otherwise indicate what had been in the original.
Al-Itqānī might have raised the issue of the opening words of the Kashshāf in his own muqaddima to his copy of the Kashshāf; this introduction appears to be given at the beginning of MS. Nurosmaniye 290/399.Footnote 84 In this short text, al-Itqānī says that al-Zamakhsharī received the title Jār Allāh because he had spent five years in Mecca; he supports this statement with lines from one of al-Zamakhsharī's qaṣīdas. He then says that al-Zamakhsharī wrote 20 works (mu'allafāt), and lists them.Footnote 85 However, he makes no reference to the opening words of the Kashshāf, either to say what al-Zamakhsharī had originally written or to indicate who had changed it. The conclusion, then, is not only that the remark given in MS. Bayezit 575 cannot be taken at face value, but that it is most likely simply false.
3.3 Sharḥs and ḥāshiyas on the Kashshāf
The margins of Kashshāf copies are not the only places where comments on al-Zamakhsharī's supposed declaration of his Muʿtazilism occur. The phrase al-ḥamdu li-llāh alladhī khalaqa l-Qur’ān received attention in at least three commentaries on the Kashshāf, where authors had more space to deal with the matter. In chronological order, those examined here are the sharḥs (or ḥāshiyas) by al-Ṭībī (d. 743/1342), al-Yamanī (d. 750/1348–9) and al-Jurjānī (d. 816/1413).Footnote 86 The first two are contemporaries and their comments on al-Zamakhsharī's khalaqa l-Qur’ān are similar, although not necessarily connected; they both reject the idea that al-Zamakhsharī ever wrote al-ḥamdu li-llāh alladhī khalaqa l-Qur’ān.
In the Futūḥ al-ghayb, al-Ṭībī's commentary, the author says that al-Zamakhsharī's opening words in his commentary are “the clarification of the order of the revelation (bayān tartīb al-nuzūl)”.Footnote 87 Al-Ṭībī's explanation of this order is traditional: first God “sent down” (anzala) the Quran from the Guarded Tablet (al-lawḥ al-maḥfūẓ) to the nearest heaven (al-samā’ al-dunyā); then he “sent it down” (nazzala) according to the requirements and events on earth, where it was established in a composed and organized way, as shown by the words mu'allaf and munaẓẓam in al-Zamakhsharī's commentary. As regards al-Zamakhsharī's having written and then changed khalaqa l-Qur’ān, al-Ṭībī says that anzala expressed what was originally intended and so no change had occurred; nor was there any need to use khalaqa l-Qur’ān, since later in the passage al-Zamakhsharī clearly indicates where he stood with respect to the nature of the Qur’ān. Al-Ṭībī writes:
This (i.e. the explanation he has just given) is what was intended and not what has been said, that he said first of all khalaqa l-Qur’ān and then changed it out of pious fear (taqiyyatan), because he was clear about that (i.e. his Muʿtazilite position) when he said, “They are only the attributes of a first creation, etc”.Footnote 88
A similar dismissal of the idea that anzala l-Qur’ān was a later emendation to al-Zamakhsharī's original khalaqa l-Qur’ān is given by al-Yamanī. He writes:
Know that God the Sublime “sent down” (anzala) the Quran all at once from the Preserved Tablet to the nearest heaven. Then, after that, he “sent it down” (nazzala) piecemeal, corresponding to requirements and matching events. That is why he (i.e. al-Zamakhsharī) said anzala and nazzala, taking into account the technique of tajnīs al-ishtiqāqī.Footnote 89 I have heard some people say that in the original copy (nuskhat al-umm) he wrote khalaqa and then changed it to anzala out of fear (khawfan) but this is not so, for he was clear about his theological position (madhhab) throughout the entire introduction (khutba). He used not to conceal his position but to make it known and to boast of it. I have studied (ṭālaʿa) the Kashshāf [written] in his handwriting (khaṭṭ) and I have seen neither erasure (kashṭ) nor change (taghyīr) in it.Footnote 90
The third witness to the debate on what al-Zamakhsharī originally wrote is al-Sayyid al-Sharīf al-Jurjānī (d. 816/1413).Footnote 91 At the beginning of his commentary he too specifically notes al-Zamakhsharī's theological position (madhhab) with respect to the Quran and adds that he went to trouble to show his Muʿtazilism and was proud of it. In his comments on anzala, however, al-Jurjānī does not explicitly deny that al-Zamakhsharī ever wrote al-ḥamdu li-llāh alladhī khalaqa l-Qur’ān; he does not address the question at all, but merely uses an anonymous khalaqa/anzala report as a way of introducing his own explanation of how brilliant al-Zamakhsharī's text is. He writes: “It is reported that in the original copy khalaqa occurred in the place of anzala, but that then the author changed it. If this is correct, then the change was beneficial”;Footnote 92 he then proceeds to give seven reasons why the change (if it did occur, as reported) had benefits (fawā’id). Al-Jurjānī, then, is not interested in confirming or denying the report; nor is he concerned with its authenticity. What is important is that the word anzala was much better suited to what al-Zamakhsharī was saying than khalaqa could ever have been, and this he amply demonstrates with the seven fawā’id that he supplies.
3.4 The evidence and the story
Manuscript evidence supports the view that al-Zamakhsharī never wrote khalaqa l-Quran, for in none of the 110 manuscripts containing the muqaddima was the reference to the creation of the Quran to be found; nor was there any indication that manuscripts had been tampered with, to erase or change what an earlier copyist had written.Footnote 93 This limited number of manuscripts may seem too small to be the basis for such a statement. However, further evidence is available: there are the biographical dictionaries (ṭabaqāt). While the story of the Kashshāf's opening words is to be found in a number of these works, none of the authors ever claimed to have seen a copy of the commentary containing the offending expression, or even one that had been changed. Moreover, there is no agreement in these ṭabaqāt works as to who would have changed khalaqa l-Qur’ān to anzala l-Qur’ān, if such a change did occur. Secondly, glossators corroborate the view that al-Zamakhsharī never wrote khalaqa l-Qur’ān by the general silence with which they pass over the account; furthermore, those who do say that khalaqa l-Qur’ān was written originally are very few and their evidence is extremely weak. Thirdly, the main commentators on the Kashshāf reject the idea that al-Zamakhsharī ever wrote khalaqa l-Qur’ān and take pains to point it out. Finally, there has been no debate on this matter in the sources. There was simply a story in circulation, which many biographers and most glossators ignored but which a few commentators took the time to refute. Nevertheless, the story seems to have taken on a life of its own, to the point that Brockelmann could remark parenthetically that “at the very beginning [of the Kashshāf al-Zamakhsharī] declares the Ḳur’ān created” and Jansen could state: “‘Glory to God who created the Koran’ are the opening words of Az-Zamakhsharī's Koran commentary”.Footnote 94 In the light of this study these statements are very hard to endorse. Not only have no manuscripts with khalaqa l-Qur’ān (or even jaʿala l-Qur’ān) been found; there is no reason to believe either that there ever were any, or that al-Zamakhsharī wrote khalaqa l-Qur’ān in the first place.
The question arises as to when and where the report about al-Zamakhsharī's having written khalaqa l-Qur’ān first appeared and who was behind it. Despite his risāla against al-Zamakhsharī's interpretation of the Quran, the aforementioned al-Itqānī cannot have been at the origin of this account. Not only does the evidence not support this position, but the story was in circulation before his time; Ibn Khallikān (d. 681/1282) had already reported the matter before al-Itqānī (b. 685/1286) was born. What is noteworthy about Ibn Khallikān's account is that it is one of two anecdotes he gives to illustrate how open al-Zamakhsharī was about his Muʿtazilism, a point already made by a number of other authors. Immediately before presenting the report about the opening words of the Kashshāf, Ibn Khallikān gives an anonymous report which says that, when visiting someone, al-Zamakhsharī would have himself announced with: “Abū l-Qāsim the Muʿtazilite is at the door (Abū l-Qāsim al-Muʿtazilī bi-l-bāb)”.Footnote 95 The Muʿtazilite Abū Yūsuf al-Qazwīnī (d. 488/1095) was attributed with a similar way of announcing himself, indicating that the story was either an invention applied to Muʿtazilites who were open about their theology, or perhaps a way of behaving that was not all that uncommon.Footnote 96 The report concerning the opening words of the Kashshāf, on the other hand, which Ibn Khallikān (d. 681/1282) places immediately after the report on al-Zamakhsharī's way of having himself announced and for which he himself says he found no evidence in the “many copies” (kathīr min al-nusakh) of the commentary he had seen (just as he found no evidence for saying that the emendation to the text was the work of the people and not that of the author), can only be a fabrication that appeared sometime in the late sixth/twelfth or early seventh/thirteenth century as an indirect means of saying not only that al-Zamakhsharī was a Muʿtazilite (which was well known) but also that his commentary was not free from his beliefs.
Such a “creation story” would have been in line with the thinking of Ibn Khallikān's (d. 681/1282) contemporary Ibn Abī Jamra (d. 699/1300?) who discussed who could study the Kashshāf and said that those who did so and who preferred it to the books of the “masters” were in error. This theme was continued by Ibn Ḥajar al-ʿAsqalānī (d. 852/1449) a century and a half later in perhaps stronger terms, for he forbad people to touch the Kashshāf and transmitted similar forebodings from al-Dhahabī (d. 748/1345). A less strident variation on the theme of the “Muʿtazilite Kashshāf” can, however, be seen in the biographies of such eighth/fourteenth century writers as Abū l-Fidā’ (d. 732/1331) and Ibn Kathīr (d. 774/1373). Most writers, though, had nothing but praise for the Kashshāf and its author, and passed over the “creation story” in silence. Nevertheless, this does not mean that they were unaware of it. Authors borrowed heavily from each other and no one had more borrowed from him than Ibn Khallikān (d. 681/1282). This story, then, was well known and continued to circulate (to the point of becoming an accepted fact in modern times), perhaps reflective of an effort either to fight a perceived resurgent Muʿtazilism or simply to destroy the popularity of the Kashshāf and to stop its spread. In the end, though, if the latter were the intention, the endeavour proved fruitless.Footnote 97
Conclusion
This article set out to examine in a systematic way al-Zamakhsharī's Kashshāf. From medieval times, this commentary was considered by some, at least, to be a Muʿtazilite commentary on the Quran and this perception has found its way into modern scholarship, even though, as the various references have shown, scholars do not define precisely what they might understand a Muʿtazilite commentary to be. Some modern-day scholars have shied away from considering the Kashshāf to be such a commentary, but even these do not seem to have studied the text itself to any great extent, and so what al-Zamakhsharī in fact said remains something of a mystery. While it is clearly not possible for every scholar to write a book on every medieval scholar or work before writing about them, it is necessary on occasion for individuals to examine a little more closely what many others may know from afar. This article comes out of such an endeavour with respect to al-Zamakhsharī and his Kashshāf.
The endeavour in question approached the Kashshāf from three angles: first, a close reading of a section of the text (Q44 and Q54), to see what it would yield by way of a Muʿtazilite interpretation. Only one passage clearly referred to a principle of this school; the commentary on Q54:17. What is interesting here, though, is that the Muʿtazilite interpretation was only one of three that al-Zamakhsharī offered; it was indeed the first but, it was only one of three. Just as infrequent was al-Zamakhsharī's slightly Muʿtazilite manipulation of a passage from an earlier source he was ostensibly quoting, al-Zajjāj's Maʿānī l-Qur’ān.Footnote 98 The relative scarcity of both Muʿtazilite interpretations of the Quranic text and Muʿtazilite alterations to the earlier source indicates that the Kashshāf was not primarily focused on the theology of its author. In fact, this detailed reading of these two sūras revealed the Kashshāf to be a traditional tafsīr musalsal, grounded in the earlier exegetical tradition and relying on the interpretations and tools of the trade that al-Zamakhsharī's predecessors had used.
The second approach involved studying two theological works whose ideas might be reflected in the Kashshāf, al-Zamakhsharī's own al-Minhāj fī uṣūl al-dīn and al-Muʿtamad fī uṣūl al-dīn by his teacher Ibn al-Malāḥimī (d. 536/1141). Unlike the first approach, which involved the continuous reading of the commentary on the two sūras in the hope that it might lead to Muʿtazilite ideas in the text, this was focused on verses whose commentary might reveal such ideas more readily. If a number of Quranic verses had been used by either al-Zamakhsharī or Ibn al-Malāḥimī to support, defend or even “prove” Muʿtazilite ideas in their theological treatises, it seemed reasonable to assume that some of these ideas might emerge in the commentary on the same verses in the Kashshāf. It was noted, however, that in these Kalām works, scripture was used relatively rarely to begin with, and that even then it appeared to have an ancillary role. With respect to the 28 Quranic passages quoted in the Minhāj, the commentary on approximately 40% of them had no Muʿtazilite content in the Kashshāf; the commentary on approximately 35% of them did reflect ideas presented in the theological treatise; and the commentary on the rest (25%) reflected Muʿtazilite positions that were not to be found in the Minhāj. These statistics may look good but do so only until it is realized that very little of the Quran made its way into the Minhāj. As for the Muʿtamad, this larger theological work used far more scripture than did the shorter Minhāj but, relatively speaking, it too did so exceedingly sparingly. The study showed that this book was simply not used by al-Zamakhsharī. In general then, even on the relatively rare occasions when scripture was used in these theological works, the connection between specific scriptural passages and the theological arguments they supported was weak, especially when it came to proofs, and this weakness was reflected in the lack of a connection between them and Muʿtazilite ideas in the Kashshāf. Understood in this light and on the basis of what has been studied here, it is not necessary to assume that al-Zamakhsharī would have had Muʿtazilite ideas to develop in his commentary on verses that had no connection to theology in the Kalām tradition since this would be pulling him out of the tradition, even if the connection between theology and scripture in the treatises often appeared weak.
The third approach was much more focused, looking at one line from the commentary, the first. For almost as long as the Kashshāf has been in existence, there has also been a story that it began with the blatantly Muʿtazilite words “Praise be to God who created the Quran (al-ḥamdu li-llāh alladhī khalaqa l-Qur’ān)”. By the beginning of the twentieth century, this story was considered to be fact. In conjunction with the few medieval sources that transmitted it, it may have been at the origins of the modern-day view that the Kashshāf was a “Muʿtazilite commentary” on the Quran. In any case, such a statement would have set the tone for the remainder of the commentary in the eyes of orthodox readers, and would probably have driven them away before they even turned the first page. This may, in fact, have been the intention, for the story is undoubtedly a fabrication set in place early by opponents of the Muʿtazila as a way of keeping people away from the Kashshāf which, they believed, had to be tainted by heresy since its author was a Muʿtazilite. The story did not, however, have the desired results, as the many manuscript copies and printed editions (20 as of today) of the Kashshāf show. In fact, a study of over 100 of these manuscripts gave no indication that this “creation story” was anything more than an invention, for there was no evidence to support it, and glossators and commentators openly rejected the story for the most part, if they spoke about it at all. Their silence reflects that of the medieval scholarly community in general, where the story was undoubtedly well known but probably not accepted, for it received limited exposure in the biographical dictionaries (ṭabaqāt) and does not seem to have had much effect on the reputation of either al-Zamakhsharī or his commentary. With the exception of al-Dhahabī (d. 748/1348) and Ibn Ḥajar (d. 852/1449) after him, none of the authors of the biographical works under study had anything critical to say about the Kashshāf.Footnote 99 On the contrary, the impression they give is that the work was held in high esteem, for it was invariably the first of al-Zamakhsharī's works to be listed. The so-called “Muʿtazilite” Kashshāf, then, received a much warmer reception from scholars than one might have expected if it had truly been steeped in heresy.
Al-Zamakhsharī's Kashshāf is a traditional Quran commentary written by a member of the Muʿtazilite school of theology, that sometimes incorporates Muʿtazilite comments into a traditional framework. Of the three approaches to the text followed here, the findings from the reading of Q44 and Q54 are probably more representative of the Kashshāf's Muʿtazilite contents than are those from the comparison of it to the Minhāj and the Muʿtamad where, despite the percentages with respect to the former, direct use of the Quran was sparse and the connection between Quranic passages and theology, on the level of “proof” at least, often appeared vague and unconvincing. It might be that their main role was simply to be a traditional confirmatur to the results of rational arguments (as Gardet said), no matter how they were presented in the Minhāj.Footnote 100 As for the “creation story”, it is a fabrication which attempted to draw a direct connection between the theology of al-Zamakhsharī and the contents of his commentary. To reject it for the forgery it is, is merely to concur with most medieval scholars, either directly or indirectly, even if the study of a limited number of these sources could and did lead a number of modern-day scholars astray. To say more about the Kashshāf, to describe it as a Muʿtazilite commentary on the Quran – and, as already noted, scholars today have not really defined what such a commentary might be, a difficulty that may have begun with al-Bayḍāwī – or to expect more from it than what this study has brought to light, would be to focus more on the author and the “creation story” that has dogged him throughout history than it would be to focus on the Kashshāf and what its author actually wrote in it.
Abbreviations
EI 1 = Encyclopaedia of Islam 1st ed., ed. M. Th. Houtsma et al. (Leiden: E. J. Brill, 1913–38).
EI 2 = Encyclopaedia of Islam 2nd ed., ed. H. A. R. Gibb et al. (Leiden: E. J. Brill, 1960–2005).
GAL 1 = C. Brockelmann, Geschichte der Arabischen Litteratur 1st ed. (Weimar: Verlag von Emil Felber, 1898–1902).
GAL = C. Brockelmann, Geschichte der Arabischen Litteratur 2nd ed. (Leiden: E. J. Brill, 1943–9).
GAL.Sp = C. Brockelmann, Geschichte der Arabischen Litteratur Supplementbände (Leiden: E. J. Brill, 1937–42).
