Explaining Contradictions in Exegetical Hadith

In a recent thread on the website formerly known as Twitter, a user named D1mashqi criticised one of the points I outlined in my podcast discussion with Javad Hashmi regarding some of the problems with the authenticity of Hadith. The point in question concerns the unreliability of exegetical Hadith in particular, which I summarised on the podcast (here) as follows:

Problem # 19, which you have already touched upon, is that, usually, exegetical reports about the context of the Quran and even sometimes the meanings of words are actually exegesis in disguise. These claim to be historical reports about what the Quran was saying and why it was revealed, but actually, this material was inferred from the Quran itself. The way to establish this is pretty straightforward. If you [look at the exegesis of] any given verse or surah of the Quran, you find all this variation. It is really, really common. It is typical that you find all these contradictions. But each contradictory report explains the verse by itself, vaguely plausibly. The way to explain that is that the material is being inferred from the Quran, but different people are inferring, elaborating, and extrapolating in different ways. So, you end up with all this variation—but each one by itself would kind of explain the verse.

A simple example is Surah 74, Verses 49-51, which talk about how people are turning “away from the reminder like frightened asses fleeing from a qaswarah.” So, what does that mean? What is a qaswarah? According to some reports, it is “a lion”, or it is specifically an Abyssinian loanword meaning “lion”; or, alternatively, it means “a party of men” and not “a lion”; or, alternatively, it means “hunters”, or it means “a party of hunters with bow and arrow”, or it means “archers”, or it means “arrows”, or it means “human voices”. And, by the way, most of these are attributed to Ibn ʿAbbās. Ibn ʿAbbās is made to say a bunch of different conflicting opinions here, and others as well.

So, what is going on? Why is there all this variation? Each one by itself makes sense in the context of the verse, but they are all conflicting. The simple explanation is [the following]: people looked at the verse and then they came up with an explanation that would seem to match the verse, [an explanation] that explains it based on the context of what is being described; but different people come up with different explanations—thus, the contradictions.

So, these reports about the Quran are actually just exegesis in disguise, being turned into reports. This is particularly clear when all of these conflicting explanations are put in narrative form: [at that point,] people are just creating narratives, but most of them just have to be wrong anyway, because they are all conflicting. [Each] will give you a different context [for] the revelation of the [same] verse and a different meaning for the [same] verse. So, these people are not only creating—coming up with—the meanings just based on the text itself and creating scenarios; they are [also] putting [them] into narrative form, which again brings us back to the storytellers.

This view of exegetical Hadith is widely held in secular, critical scholarship: the notion that meanings and contexts conveyed in exegetical hadiths were inferred or extrapolated from the Quran itself has been affirmed—to varying degrees—by scholars such as Michael Cook, Patricia Crone, Andrew Rippin, François de Blois, Lawrence Conrad, Gabriel Reynolds, Harald Motzki, and Andreas Görke,[1] whilst the notion that oral storytellers played a key role in the creation of historical and exegetical material has likewise been affirmed—again, to varying degrees—by scholars such as Michael Cook, Gautier Juynboll, Patricia Crone, Andrew Rippin, Robert Hoyland, Andrew Bannister, Neal Robinson, Gregor Schoeler, Harald Motzki, and Andreas Görke.[2] (Henceforth, for the sake of convenience, I will describe this view as the “critical” explanation of the data.)

However, D1mashqi is having none of it: “This is a very disappointing argument and to consider it a major point is atrocious.”[3] D1mashqi’s core objection is simple: traditional Islamic scholarship already has at its disposal equal or even superior explanations for the occurrence of conflicting exegetical reports regarding both (1) the meaning of obscure words in the Quran and (2) the historical context of the revelation of particular surahs or verses.

Are the traditional explanations indeed equal or superior to those articulated by critical scholars? Do the former even explain the evidence at all? In what follows, we will explore some of the relevant evidence, summarise the competing explanations therefor, and systematically compare their relative merits.

Explaining the Meanings of Obscure Words

According to D1mashqi,[4] there are far better traditional explanations for the coexistence of multiple reports asserting conflicting meanings for rare or unique words in a Quranic verse—often attributed to the same figures—than the explanation that has been proposed in secular, critical scholarship:

Dr. Musā‘id al-Tayyār lists the many reasons why there is Ikhtilāf in Tafsīr:

1. Ishtirāk: When a word refers to multiple meanings in the Arabic language (like “Qaswara” above).

2. A difference over the object referred to by a pronoun (he/it/they), what is it referring to?

3. That there is a word dropped out from a verse (this is called مجاز حذف in Balāgha, where there is a word that is not used in a sentence, but it is part of it implicitly). So the Ikhtilāf would stem from a difference on what the word that was dropped is.

4. That the differed-on word has multiple possible different roots in language.

5. That the word is used in different ways by the Arabs.

6. That this verse revolves around abrogation. And thus one Mufassir sees it as being abrogated, and another sees it as the one abrogating.

7. That there is a general verse which is specified elsewhere.

8. That a description in a verse applies to multiple things/people.

9. That a word in a verse has multiple narrations from multiple Qirā’āt, thus leading to a possible range of meanings.

Fusūl fī Usūl al-Tafsīr, Pg. 63-69.

The great Mufassir Ibn Juzayy al-Kalbī writes in his introduction to his Tafsīr:

“The reasons for the Ikhtilāf between the Mufassirīn are 12:

1. Difference in the Qur’ān.

2. Difference in the I‘rāb of a word.

3. The Ikhtilāf of linguists on the meaning of a word.

4. Ishtirāk of a word where it has multiple meanings.

5. ‘Umūm & Khusūs: General phrase that is specified elsewhere.

6. Itlāq & Taqyīd: Unrestricted phrase that is restricted elsewhere.

7. Haqīqa & Majāz: is a word literal or metaphorical?

8. Idmār & Istiqlāl.

9. The possibility of an added word (مجاز حذف).

10. Taqdīm & Ta’khīr: which verse came first?

11. Abrogation.

12. A difference in Tafsīr from the Prophet ﷺ, his companions, or their Tābi‘īn.

al-Tashīl, Pg. 12.

As we can see, this is not some hidden key to uncover that the Hadīth corpus is all fake.

“Qaswara” being transmitted to have different meanings is because it *does* have multiple possible meanings in the Arabic language.

And a companion mentioning different meanings is to outline this difference, and offer possible meanings (all of which could be true, and this goes back to the Mujtahid)

To evaluate whether any of this is actually a better explanation for the evidence in question, or whether indeed any of this explains the evidence at all, it is helpful to turn to a concrete example. In Sūrat al-Muddaṯṯir (Surah 74), we find the following verses (49-51):

[49] What is [the matter] with them, turning away from the warning

[50] as if they are frightened asses

[51] fleeing from a qaswarah?

The issue here is the word qaswarah in verse 51: the context of the verse makes it clear that it must be something that would frighten wild Arabian asses (i.e., onagers), but what exactly does it mean? In the extant corpus of exegetical Hadith,[5] we find the following interpretations of the word qaswarah attributed to the following figures:

[1] “arrows” (al-nabl)

• Qatādah

[2] “archers” (al-rumāh)

• Muqātil
• ʾAbū Mālik
• Yazīd b. ʾabī Ḥabīb
• ʿAṭāʾ
• Ibn Kaysān
• ʿIkrimah
• Mujāhid
• al-Ḍaḥḥāk
• Ibn Jubayr
• Ibn ʿAbbās
• ʾAbū Mūsá

[3] “archers-cum-hunters” (al-rumāh al-qunnāṣ or al-qunnāṣ al-rumāh)

• Qatādah
• Mujāhid
• Ibn ʿAbbās

[4] “hunters” (al-qunnāṣ or rijāl al-qanṣ)

• Ibn Jubayr
• Ibn ʿAbbās

[5] “the snares of hunters” (ḥibāl al-ṣayyādīn)

• Ibn ʿAbbās

[6] “[a] group[s] of men” (jamʿ al-rijāl or ʿuṣab/ʿuṣbat al-rijāl)

• Ibn ʿAbbās

[7] “strong men” (rijāl ʾaqwiyāʾ)

• Zayd b. ʾAslam

[8] “the sound/shouting of people” (laḡṭ al-qawm or rikz al-nās)

• ʾAbū al-Mutawakkil
• Ibn ʿAbbās

[9] “predators” (al-sibāʿ)

• ʿAbd al-Raḥmān b. Sābiṭ

[10] “lion” (al-ʾasad)

• ʿAbd al-Raḥmān b. Zayd
• Zayd b. ʾAslam
• ʾAbū Hurayrah
• Ibn ʿAbbās

[11] “the darkness of the night” (ẓulmat al-layl)

• ʿIkrimah

In sum, we have ten or eleven different definitions given for the word qaswarah in Q. 74:51: each plausibly fits the context of the verse; each is attributed to an early Muslim authority; and several are attributed simultaneously to the same specific authorities (i.e., Zayd b. ʾAslam, Qatādah, Mujāhid, Ibn Jubayr, ʿIkrimah, and Ibn ʿAbbās).

With our example thus outlined, we can now turn to the question of the best explanation therefor. To begin with, let us examine in more detail how critical scholars explain such data:

• The key point is this: these exegetical reports are quite contradictory overall, yet each by itself perfectly matches the verse. The simple explanation for this peculiar combination of facts is that early Muslim exegetes did not know the meaning of the word and were thus forced to infer or extrapolate its meaning from—or in relation to—the verse itself. However, given the ambiguities of the verse and the inevitable idiosyncrasies of individuals, each exegete—quite predictably—arrived at different conclusions through this process. Thus, the appearance of contradictory yet equally plausible interpretations.[6]
• Additionally, if such interpretations only exist in exegetical contexts,[7] that is highly suspect: this is exactly what it would look like if such meanings were the product of exegetical speculation rather than actual usage.[8]
• Alongside this, many of these conflicting Quran-derived interpretations are attributed to the same figures, which is most easily explained by widespread false ascription. This is especially likely in the case of ascriptions to Companions, since the general loss of meaning entailed by all of this fits more comfortably with later generations who were not part of the original Quranic milieu. Finally, this is most likely in the case of Ibn ʿAbbās, who attained a reputation for expertise in tafsīr and thus became a magnet for exegetical retrojections, resulting in a vast quantity of highly contradictory exegetical hadiths being attributed to him.[9]
• In short, the most straightforward explanation for contradictory yet equally plausible exegetical reports that are often ascribed to the very same figures is that later exegetes variously inferred or extrapolated the meanings of obscure words in the Quran from the Quran itself and then falsely ascribed these inferences or extrapolations back to earlier figures.

With both sets of explanations—traditionalist and critical—thusly defined, we can now test each of those cited by D1mashqi and see whether they really make good sense—or sense at all—of the data.

The first set of traditional explanations cited by D1mashqi are the nine outlined by Musāʿid b. Sulaymān al-Ṭayyār in his Fuṣūl fī ʾUṣūl al-Tafsīr,[10] as follows:

“Amongst the causes of the disagreement between the exegetes [amongst] the Salaf [are the following]:

1 – Homonymity (al-istišrāk), when a word denotes more than [one] meaning in the Arabic language.”

Some words do indeed have multiple meanings: kitāb means writing, book, document, letter, and contract; waliyy means friend, ally, patron, and legal guardian; kāhin means both pagan soothsayer and Christian priest; etc. However, as these examples immediately illustrate, Arabic nouns usually have multiple closely related meanings: the alleged semantic range present in the exegesis on qaswarah—whilst certainly possible—is far beyond the norm. It other words, it is implausible that qaswarah meant “arrows” and “archers” and “hunting archers” in particular and “hunters” more generally and “the snares of hunters” and also “a group of men” and “strong men” in particular and also “the voices of men” and also “predators” and “lion” in particular and also “darkness”, etc. Already, al-Ṭayyār’s first explanation begins to founder, in light of our general background knowledge on the usual semantic ranges of specific nouns.[11]

Additionally, this hypothesis simply does not explain the amazing coincidence that all of these definitions equally match the verse context. To appreciate just how unlikely that is, consider the other examples just cited: when a text speaks of an Arabian kāhin who cavorts with jinn and engages in divination, it clearly means “pagan soothsayer”, not “Christian priest”; when a text speaks of a girl’s requiring the consent of her waliyy to get married, it clearly means a legal guardian, not a friend, ally, etc.; when a text describes how someone wrote a kitāb about ancient Arabian customs, it clearly means “book”, not “letter”, “contract”, etc. In other words, it is very rare that all or even most of the possible meanings of a word—especially a word with as many alleged meanings as qaswarah—would equally match a given context of usage. By contrast, the critical hypothesis perfectly explains this remarkable fact: most of these different meanings were inferred from the verse itself or in relation to the verse, so they all equally match the verse. As such, the critical hypothesis has greater explanatory scope than al-Ṭayyār’s first explanation.

Furthermore, the word qaswarah must mean one specific thing (i.e., one definition or the other) in the verse under consideration, so how are we to make sense of, for example, Ibn ʿAbbās’ having given multiple definitions for the same word? In each report, it is specified that the definition was given in relation to this verse in particular, so it cannot be the case that Ibn ʿAbbās was simply listing off, in the abstract, different possible meanings of the word at different times: in each case, he is presented as clarifying what the word means here in this instance, yet in each case, he is made to say something different. Al-Ṭayyār’s first explanation simply does not explain this fact, unless he posits an additional hypothesis: Ibn ʿAbbās simply did not know the specific meaning of this verse (i.e., which sense of qaswarah was meant here) and thus changed his mind repeatedly, asserting something different in each instance. However, this would also mean that, in the space of a few decades, the Companions had forgotten or lost the specific meaning of this word and many others like it—a bullet that a traditionalist would be unlikely to bite, I would guess.

It could be hypothetically argued that the definitional differences under consideration reflect differences in Arabic tribal dialects. In other words, we might imagine that different Arabic tribal dialects had different definitions for the same word—thus, different Arab tribesmen mistakenly assumed their own definitions when they heard the verse. Even in this instance, however, we would face the problem of numerous definitions that equally fit the verse: how likely is such an occurrence (i.e., that each different tribe’s definition would happen to equally match the context), especially given that there are numerous cases like this (i.e., similar to the qaswarah situation)? Moreover, if indeed such disagreement was simply an issue of tribal dialect, we would expect the early Arabic grammarians to record these dialectal differences, as they did so often for other differences. What then are the different tribal-dialectal definitions of qaswarah recorded in the early grammatical literature? There appear to be none,[12] which is straightforwardly inconsistent with such an explanation.

It could also be argued—as indeed does D1mashqi, here—that differences over the definitions of obscure words could be the result of such words’ having multiple possible roots in Arabic. However, this would not actually explain the disagreement: what difference would it make if there are multiple possible roots for the same word? In context, it should be clear in most cases which one is meant. For example, the word حبّ (written without any vowels) could yield either “love” (from the ḥubb root) or “grain” (from the ḥabb root): context alone would tell us which word (and root) was meant in almost any instance. In other words, this ‘explanation’ runs into the same problem as the rest: it would not explain the unexpected way in which, time and again, all the proposed words and meanings equally match the relevant verses. Additionally, as above, it would require the supposition that the earliest Muslims had already forgotten the specific meaning of the word, such that they had to try and infer the possible meanings from different roots. Of course, this explanation is usually not applicable in the first place: in most cases like this, there simply are not multiple viable Arabic roots for the given word. For example, take qaswarah: what are the multiple different Arabic roots that yield each of the definitions that were adduced therefor? There is only one conceivable Arabic root for this word (qsr),[13] rendering this variant of al-Ṭayyār’s first explanation irrelevant.

In sum, al-Ṭayyār’s first explanation (as applied to the issue of qaswarah): (1) posits an implausibly wide semantic range for the word qaswarah; (2) does not explain the remarkable coincidence that all of the alleged meanings of qaswarah equally match the verse in question; and (3) is inconsistent with the fact that Ibn ʿAbbās is talking about the meaning of the word specifically in this verse (i.e., he repeatedly contradicted himself), not merely listing the different definitions of the word in general. By contrast, the critical hypothesis explains all of this data effortlessly. So much for al-Ṭayyār’s first explanation.

“2 – Disagreement over the [intended] referent of a pronoun [i.e., to what it is referring].”

This is completely irrelevant here, and to most of the examples like qaswarah that critical scholars have cited.

“3 – That there is an omitted [word] in the sentence.”

Again, irrelevant.

“4 – That the word has more than [one possible] morphology in the [Arabic] language.”

This appears to be about whether a verb is active or passive, etc., which is irrelevant to most cases (and certainly to the issue of qaswarah).

“5 – The Arabic usage of the word varies in [what is] intended: close [i.e., more literal] meanings and distant [i.e., more metaphorical] meanings.”

This would still leave unexplained the remarkable fact that all of the numerous possible meanings of words like qaswarah equally match the context of the relevant verses. In most cases, however, this metaphorical vs. literal distinction does not even apply to begin with: in the case of qaswarah, for example, the issue is not whether it means (for example) “lion” in a literal sense or “lion” in a metaphorical sense, but rather, whether it means “lion” (whether literal or metaphorical) or “archers” (whether literal or metaphorical) or “darkness” (whether literal or metaphorical) or “snares” (whether literal or metaphorical) or “shouts” (whether literal or metaphorical), etc. In most cases like this, the issue is not one of metaphorical versus literal usage, but rather, the basic meanings of words.

“6 – [It could be] that [the debate over] the judgement of this verse revolves around enforceability or abrogation [i.e., whether the verse retains legal applicability].”

This is again irrelevant in most cases.

“7 – [It could be] that [the debate over] the judgement of this verse revolves around [whether the meaning is] general or specific.”

Once again, irrelevant in most cases.

“8 – [It could be] that the description [in the verse] plausibly refers [simultaneously] to more than [one] referent.”

What exactly would this mean in this case? Is qaswarah supposed to mean all of the definitions simultaneously? Did the author want us to imagine an explosion of lions, hunters, archers, strong men, arrows, snares, shouts, and darkness all bearing down on some poor wild asses at the same time, like the crazed climax of an action film? Are we instead to imagine some kind of Lovecraftian chimaera of all of these things combined together? Are we supposed to keep multiple separate images in our minds simultaneously? All of this seems highly implausible: based on overwhelming standard usage, it is reasonable to expect that the author intended a single, concrete thing when he said qaswarah. We can thus safely set aside al-Ṭayyār’s eighth explanation (at least as applied to something like qaswarah).

“9 – [It could be] that there is an aspect (ḥarf) in the verse that has two [possible] readings (qirāʾatān).”

This is irrelevant in most cases, and certainly in the present case.

Summary of al-Ṭayyār

To summarise al-Ṭayyār’s nine points, as applied to the issue of qaswarah: (1) al-Ṭayyār’s first explanation is implausible, possesses less explanatory scope than the critical explanation, and is either contradicted by the evidence or requires the assumption that the very earliest Muslims had forgotten or lost the specific meaning of qaswarah; (2) al-Ṭayyār’s second explanation is irrelevant; (3) al-Ṭayyār’s third explanation is irrelevant; (4) al-Ṭayyār’s fourth explanation is irrelevant; (5) al-Ṭayyār’s fifth explanation is irrelevant; (6) al-Ṭayyār’s sixth explanation is irrelevant; (7) al-Ṭayyār’s seventh explanation is irrelevant; (8) al-Ṭayyār’s eighth explanation is wildly improbable; and (9) al-Ṭayyār’s ninth explanation is irrelevant. In short, al-Ṭayyār’s alternative explanations—at least as applied to examples like qaswarah—variously: go against our established background knowledge in general; contradict the very evidence that they are supposed to explain; contradict other specific evidence; explain the evidence poorly; and/or don’t explain the evidence at all.

In addition to al-Ṭayyār’s nine explanations, D1mashqi cites an additional twelve from Muḥammad b. ʾAḥmad b. Juzayy al-Kalbī’s al-Tashīl li-ʿUlūm al-Tanzīl,[14] to which we will now turn:

[1] “Disagreement over [whether one part of] the Quran [explains another].”

This is irrelevant in practically all cases (and certainly in this case).

[2] “Disagreement over the ways to decline [the word] [i.e., its grammar], even if the readings [thereof] are all in agreement.”

Again, irrelevant in most cases.

[3] “Disagreement between the language experts over the meaning of a word.”

This is either the very phenomenon that we are trying to explain in the first place, and thus not an explanation at all; or it is simply a variant of al-Ṭayyār’s first explanation and thus subject to the same problems.

[4] “A word’s having two or more meanings [i.e., homonymity].”

This is the same as al-Ṭayyār’s first explanation; see above.

[5] “The possibility of general [versus] specific [meanings].”

This is the same as al-Ṭayyār’s seventh explanation; see above.

[6] “The possibility of a lack of restriction [in meaning] or a restriction [in meaning].”

Again, irrelevant.

[7] “The possibility of literal [meaning] or metaphorical [meaning].”

This is the same as al-Ṭayyār’s fifth explanation; see above.

[8] “The possibility of ellipsis [i.e., something is missing from the text] or independence [i.e., the text stands alone].”

Again, irrelevant.

[9] “The possibility of an added word.”

This is the same as al-Ṭayyār’s third explanation; see above.

[10] “The possibility of a text’s being [correctly] ordered, or of [its being] fronted or post-positioned [instead].

Again, irrelevant.

[11] “The possibility that the [legal] ruling [in the verse] is abrogated or [still] prescribed.”

This is the same as al-Ṭayyār’s sixth explanation; see above.

[12] “Disagreement in the transmission of exegesis from the Prophet and from the Salaf.”

What exactly would this mean in practice (i.e., applied to a case like qaswarah)? Did the leading Followers forget what the leading Companions told them about the meaning of the verse? Did the leading Companions forget what the Prophet had told them? Automatically, this kind of scenario requires an assumption—an ultra-rapid loss of knowledge—that a traditionalist is unlikely to accept. Moreover, even if the Followers and the Companions had lost knowledge of the specific meaning of the word, how exactly did they arrive at their respective, conflicting opinions on its meaning? If the answer is supposed to be that the word had different meanings, we run into the problems facing al-Ṭayyār’s first explanation. If the answer is supposed to be tribal-dialectal differences, we run into the problems facing a possible variant of al-Ṭayyār’s first explanation. If the answer is supposed to be the existence of several possible roots for the word, we run into the problems facing another possible variant of al-Ṭayyār’s first explanation. In other words, there is no avoiding the critical explanation: even if the reports all go back to the Followers and the Companions, it still looks like they variously inferred or extrapolated the meaning of qaswarah from or in relation to the verse. Of course, the notion that such inferences or extrapolations genuinely go back to the Followers and the Companions runs into the problem of conflicting ascriptions: did the same figures repeatedly contradict themselves? It seems far simpler—and it certainly conforms to our established background knowledge on the ubiquity of false ascription in general—to suppose not merely that the meanings of qaswarah were largely extrapolated from the Quran, but also, that these extrapolations were then retrojected back to the Followers and the Companions.

Moreover, if it is assumed that the Companions would have known what the word meant (i.e., that the Prophet would have explained to his immediate disciples what his pronouncements meant and/or that his immediate disciples had access to the same linguistic knowledge as he), then it follows that the attribution to them of guesswork- or inference-based interpretations are very likely false (i.e., are far more likely to be the product of later generations—farther removed from the original Quranic milieu—who had actually lost knowledge of the meanings of such words).

In short, Ibn Juzayy’s twelfth explanation (1) requires the assumption of an ultra-rapid loss of knowledge and (2) only works—as an alternative to the critical explanation—in combination with one of the previous, already-criticised explanations. Additionally, (3) this explanation fits poorly with the numerous ascriptional contradictions, which are far more easily explained by false ascription.

Summary of Ibn Juzayy

To summarise Ibn Juzayy’s twelve points, as applied to the issue of qaswarah: (1) the first is irrelevant; (2) the second is irrelevant; (3) the third is either not an explanation at all or simply a variant of al-Ṭayyār’s first explanation; (4) the fourth is the same as al-Ṭayyār’s first explanation; (5) the fifth is the same as al-Ṭayyār’s seventh explanation; (6) the sixth is irrelevant; (7) the seventh is the same as al-Ṭayyār’s fifth explanation; (8) the eighth is irrelevant; (9) the ninth is the same as al-Ṭayyār’s third explanation; (10) the tenth is irrelevant; (11) the eleventh is the same as al-Ṭayyār’s sixth explanation; and (12) the twelfth only works in tandem with some of the preceding (already-criticised) points and in any case poorly explains recurring contradictions in these kinds of ascriptions.

Explaining the Contexts of Revelation

Having thus repeatedly demonstrated the superiority of the critical explanation over those adduced by D1mashqi regarding obscure words in the Quran, we can now turn to D1mashqi’s final point:

There is another issue I wanted to touch on (not mentioned by Little) related to the above that orientalists also love: Why are there different reasons of revelation for the same verse? The late polymath Shaykh Dr. Nūr al-Dīn ‘Itr answers: “The difference in Riwāyāt in Asbāb al-Nuzūl is for the following reasons…”

1. A weakness in the transmitters. Thus the weak transmitter might narrate something that is incorrect and contradicts what is authentic.

2. There being a variety of reasons for the revelation, with only 1 text revealed for all of them together.

3. That a text was revealed multiple times (meaning: the Prophet ﷺ was commanded to recite a verse in multiple different occasions for different reasons).

Badr al-Dīn al-Zarkashī said: “Something might occur, or a question might be asked, about an issue that has already been addressed in a past verse, so the same verse would be given to the Prophet ﷺ once again, to remind them of it, and that it applies to this.”

‘Ulūm al-Qur’ān, Pg. 49-52.

(Pace D1mashqi, this second point was in fact mentioned by Little in the podcast discussion; see the transcription at the outset.)

Once again, to evaluate whether any of this is actually a better explanation for the evidence in question, or whether indeed any of this explains the evidence at all, it is helpful to turn to a concrete example. What follows is an excerpt from an unpublished (but hopefully forthcoming) study of mine concerning the context of Q. 5:3, summarising the various “contexts of revelation” (ʾasbāb al-nuzūl) reported for this verse (or for Sūrat al-Māʾidah as a whole, including that verse).

• It is reported via a mixed Baghdado-Kufan SS that Ibn ʿAbbās asserted that no one knows the day referred to in Q. 5:3.
• The Egyptian CL Ibn Lahīʿah reported, from an Egyptian SS, that Ibn ʿAbbās asserted that Q. 5:3 and Sūrat al-Māʾidah in general were revealed on a Monday, just as the Prophet’s birth, the commencement of his prophethood, his departure from Makkah, his entrance into Madinah, the Battle of Badr, the lifting of the Black Stone, and the Prophet’s death all occurred on Mondays.
• It is reported via a mixed Iraqo-Hijazo-Egyptian SS that Yazīd b. ʾabī Ḥabīb asserted that Q. 5:3 was revealed on a Monday, just as the Prophet’s birth, the commencement of his prophethood, his death, and the Day of al-Furqān all occurred on Mondays.
• According to the Eastern CL ʾIbrāhīm b. Ṭahmān, from a Basran source, from ʾUmm ʿAmr bt. ʿAbs, her uncle asserted that Sūrat al-Māʾidah was revealed when he and the Prophet were on a journey (fī masīr); this uncle and others (an unspecified “we”) discerned the revelation by the way that the foot (kaff) or shoulder (katif) of al-ʿAḍbāʾ, Prophet’s female riding camel (rāḥilatu-hu), was crushed under the weight (min ṯiqal) of the descending sūrah.
• It is reported via a mixed Baghdado-Egyptian SS that ʿAbd Allāh b. ʿAmr asserted that Sūrat al-Māʾidah was revealed when the Prophet was riding his female riding camel (rākib ʿalā rāḥilati-hi); she was unable to bear him (i.e., due to the weight), forcing him to dismount (nazala).
• It is reported via an Eastern SS that al-Rabīʿ b. ʾAnas asserted that Sūrat al-Māʾidah was revealed when the Prophet was riding his female riding camel (rākib rāḥilata-hi) on the way (fī al-masīr) to the Farewell Pilgrimage (fī ḥijjat al-wadāʿ), causing the camel to buckle under the weight (min ṯiqali-hā).
• al-Quraẓī, Sūrat al-Māʾidah was revealed when the Prophet was riding his she-camel (ʿalá nāqati-hi), somewhere between Makkah and Madinah, during the Farewell Pilgrimage (fī ḥijjat al-wadāʿ), cracking the she-camel’s shoulder (katif) and forcing the Prophet to dismount (nazala).

• According to the Kufan CL Layṯ b. ʾabī Sulaym, Šahr b. Ḥawšab reported that his patron ʾAsmāʾ bt. Yazīd asserted that Sūrat al-Māʾidah was revealed all at once when the Prophet was riding al-ʿAḍbāʾ, his she-camel (nāqah), whilst ʾAsmāʾ was leading it by the reins; the she-camel’s upper leg (ʿaḍud) was almost crushed under its weight (min ṯiqali-hā).
  • According to another version reported from Layṯ, Šahr himself asserted that Sūrat al-Māʾidah was revealed when the Prophet had stopped (wāqif) at ʿArafah whilst mounted on his female riding camel (ʿalá rāḥilati-hi), who was forced to kneel to prevent her foreleg (ḏirāʿ) from being crushed (i.e., due to the weight).
• According to the Kufan CL ʾAsbāṭ b. Naṣr, al-Suddī asserted that Q. 5:3 was revealed on the Day of ʿArafah, after which no more verses regarding ḥalāl and ḥarām were revealed, and after which he soon died.
  • According to one version from ʾAsbāṭ, al-Suddī reported that ʾAsmāʾ bt. ʿUmays asserted that this occurred when she and the Prophet were travelling (nasīru) during the Ḥajj. Gabriel appeared to the Prophet and the latter slumped upon his female riding camel (ʿalá rāḥilati-hi), whilst the camel buckled under the weight (min ṯiqal). Consequently, ʾAsmāʾ covered the Prophet in a cloak and led his camel.
• According to the Basran CL Dāwūd b. ʾabī Hind, al-Šaʿbī asserted that Q. 5:3 was revealed when the Prophet was standing (wāqif) on ʿArafah, when pre-Islamic polytheism and its rituals—such as the naked circumambulation of the Kaʿbah—had been abolished.
• The Basran CL Qatādah asserted that Q. 5:3 was revealed when God had dedicated to the Muslims their dīn/ḥajj and forbade the polytheists from visiting the Kaʿbah. Additionally, Qatādah claimed to have heard that Q. 5:3 was revealed on the Day of ʿArafah, on a Friday.
  • According to the PCL ʿAbd al-Razzāq, Qatādah specifically stated that the Day of ʿArafah coincided with (wa-wāfaqa) a Friday.
• It is reported via a mixed Mecco-Harranian SS that Mūsá b. ʿUbaydah al-Rabaḏī reported, from an unnamed šayḵ of the Banū Ḥāriṯah, that an unnamed man of that tribe was praying with the Prophet (i.e., in Madinah) when the qiblah shifted from Jerusalem to the Kaʿbah of Makkah; the Prophet then informed him that the Ḥajj was the last of the religious obligations to be revealed, whereupon Q. 5:3 was revealed.
• It is reported via a mixed Eastern-Kufan SS that the Alid scion Muḥammad al-Bāqir asserted that Q. 5:3 refers to the Day of the Greater Pilgrimage (yawm al-ḥajj al-ʾakbar), which is the Day of Immolation (yawm al-naḥr).
• According to the Kufan CL Muḥammad b. Fuḍayl, from Hārūn, the latter’s father ʿAntarah asserted that Q. 5:3 was revealed on the Day of the Greater Pilgrimage (yawm al-ḥajj al-ʾakbar). Thereafter, ʿAntarah reported that ʿUmar wept on the day that Q. 5:3 was revealed and explained that this was because Islam, having been perfected, could only diminish henceforth—a prediction that the Prophet confirmed.
• It is reported via a mixed Eastern-Iraqian SS that ʾAbū al-ʿĀliyah reported that some Followers (an unspecified “they”) reported that they were with ʿUmar when he mentioned Q. 5:3, whereupon a kitābiyy declared that its revelation to his own community would have been commemorated by them. This prompted ʿUmar to declare that the day in question was the Day of ʿArafah, and also the Day of Immolation (al-naḥr). God had thereby perfected Islam, and it could only diminish henceforth.
• According to the Yemenite CL ʿAbd al-Razzāq, from a Meccan SS, ʿIkrimah reported that ʿUmar asserted that Sūrat al-Māʾidah was revealed on the Day of ʿArafah, which coincided with (wa-wāfaqa) a Friday.
• According to the Kufan CL Yaḥyá b. al-Ḥimmānī, from a Kufan SS, from Ibn al-Ḥanafiyyah, his father ʿAlī asserted that Q. 5:3 was revealed when the Prophet was standing (qāʾim), on the Eve of ʿArafah.
• According to the Madino-Baghdadian CL Yaʿqūb b. ʾIbrāhīm, from a mixed Iraqian SS, Samurah asserted that Q. 5:3 was revealed when the Prophet was standing (wāqif) on ʿArafah.
• According to the Levantine CL ʾIsmāʿīl b. ʿAyyāš, ʿAmr b. Qays reported that Muʿāwiyah gave a sermon on the minbar and asserted that Q. 5:3 was revealed on a Friday, on the Day of ʿArafah.
• According to the Basran PCL ʿAbd al-Wahhāb, the above-mentioned Basran CL Dāwūd informed al-Šaʿbī that Jews (according to one version) or an unspecified people (according to another version) were chiding the Arabs for not having memorised the day referred to in Q. 5:3; al-Šaʿbī responded by asking Dāwūd why he had not done so already, and when Dāwūd asked about the day, al-Šaʿbī clarified that Q. 5:3 was revealed on the Day of ʿArafah.
• According to the Basran CL Ḥammād b. Salamah, ʿAmmār reported that Ibn ʿAbbās recited Q. 5:3, whereupon a Jew who was with him informed him that, had Q. 5:3 been revealed to his own community, they would have made that day a holiday. In response to this, Ibn ʿAbbās declared that Q. 5:3 had in fact been revealed on two holidays: it was revealed on both the Day of ʿArafah and a Friday.
• It is reported via an Eastern SS that ʿĪsá b. Jāriyah reported that he had been sitting with others in the Dīwān when a Christian informed them that, had Q. 5:3 been revealed to his own community, they would have commemorated that day and even the hour, even if only two of them remained. No one answered the Christian (i.e., because none of them knew the day in question), but when ʿĪsá later met al-Quraẓī, the latter informed the former that ʿUmar had said Q. 5:3 was revealed when the Prophet was standing on the mountain (wāqif ʿalā al-jabal) on the Day of ʿArafah, a day that would remain a holiday for the Muslims even if only a single Muslim remained.
• al-Quraẓī reported that a Jew asked ʿUmar if he knew the day that Q. 5:3 had been revealed, chastising ʿUmar and the Muslims for disregarding it and asserting that, had it been revealed to his own community in the Torah or the Gospel (!), they would have commemorated that day. When ʿUmar asked the Jew to clarify the verse and learned that it was Q. 5:3, he chastised the Jew for his ignorance and informed him that it had been revealed when the Prophet was standing (wāqif) on ʿArafah, on the Day of the Farewell Pilgrimage (yawm ḥijjat al-wadāʿ), which will remain a holiday for Muslims as long as Islam endures.

• According to the Basro-Levantine CL Rajāʾ b. ʾabī Salamah, from a Levantine source, ʾIsḥāq b. Qabīṣah reported that Kaʿb al-ʾAḥbār informed ʿUmar that he knew of another religious community who, had a certain verse been revealed to them, would have taken note of the day and made it a holiday. This prompted ʿUmar to inquire about the verse in question, and when Kaʿb informed him that it was Q. 5:3, ʿUmar declared that he knew the day on which it had been revealed: it was both a Friday and the Day of ʿArafah, both of which happen to be holidays.
• According to the Kufan CL Qays b. Muslim, Ṭāriq b. Šihāb reported that a Jew informed ʿUmar that, had Q. 5:3 been revealed to his own community, they would have made that day a holiday. In response to this, ʿUmar claimed to know the day and place of its revelation: ʿArafāt (in some versions) or the Day of ʿArafah (in other versions), on a Friday.

(There are also Šīʿī reports that link Q. 5:3 to Ḡadīr Ḵumm, which I have not yet incorporated into my research on this topic.)

With our example thus outlined, we can now turn to the question of the best explanation therefor. To begin with, let us examine in more detail how critical scholars explain such data:

• Once again, we have reports that contradict each other, but that equally explain the context of the verse or surah in question. To this extent, the reasoning outlined previously applies here as well.
• However, in the case of ʾasbāb al-nuzūl, the reports are usually narratives, which means that exegetes were not merely guessing at the context of a given verse or surah—they were fleshing out their guesswork in narrative form, which automatically exposes such narratives as false creations.[15]
• However, in addition to contradicting each other, such narratives often overlap in content, sharing various elements in common. Given that it is extremely unlikely that the same specific scenarios, statements, sequences, etc., occurred over and over in historical reality (of the sort outlined above, like the ever-repeating scenario of a smug kitābiyy asking about Q. 5:3),[16] these recurring elements and commonalities must be tropes, motifs, archetypes, schemata, and other such artificial narrative structures, which are ubiquitous in ancient literature. This in turn means that narratives were being created from other narratives and/or that narratives were being created from a common stock of narrative material.[17]
• The fact that such exegetical guesswork was being expressed in the form of stories automatically points to the agency of oral storytellers[18]; and this is corroborated by the fact that such stories exhibit recurring narrative commonalities or appear to be variations of common themes, etc., which is typical of storyteller compositions (i.e., strongly conforms to our established background knowledge of storytelling in oral cultures and contexts more broadly).[19]
• In short, the most straightforward and broadly corroborated explanation for equally viable yet contradictory yet overlapping ʾasbāb al-nuzūl narratives is that such narratives are the product of the exegetical guesswork of oral storytellers in particular.[20]

With both sets of explanations now defined, we can once again test each of the three cited by D1mashqi from Nūr al-Dīn Muḥammad al-Ḥasanī ʿItr’s ʿUlūm al-Qurʾân al-Karīm[21] and see whether they really make good sense—or sense at all—of the data.

“1 – The weakness of the transmitters.”

Putting aside the various problems with proto-Sunni Hadith criticism that have been outlined repeatedly in previous secular scholarship, does this explanation really account for the evidence?

Let us suppose, for the sake of argument, that there is only a single strong chain in the vast spectrum of material outlined above, which would thus represent the single true account of the verse’s context: this would be the report of Qays b. Muslim, which was judged to be “sound” (ṣaḥīḥ) and incorporated into the Sunnī Hadith canon. Even if we accept this, an obvious question remains: how did all the other accounts come into being? Clearly, the overwhelming majority were falsely created in some fashion: each story contradicts others, yet each story overlaps with another, or the same elements keep appearing across different accounts. As noted already, the best explanation for this is popular, oral storytelling: these parallel, alternative contexts of revelation must derive from a vast process of the combination and recombination of a common stock of narrative elements, and from the continuous reworking and remixing of material through successive retellings. Even if there is a single true version, this must have happened in general: storytellers freely created and recreated contexts for the Quran.

Now comes the problem: the so-called one true version fits seamlessly into this vast spectrum of storyteller material; it looks equally like a product of exactly the same processes that produced the rest, sharing the same recurring elements, etc. This already seems extremely odd: to get around this, we would have to suppose that this one true version somehow spawned all the rest. If so, however, we face another puzzle: if the one true version was so widespread and influential that it spawned all the rest, why did every single other version deviate from it? Alternatively, if it was a common historical event that inspired both the one true version and all the false stories alike, how is it that every single other story deviated from such a famous and widely known event, leaving only the one true version intact? Are we really supposed to imagine that a single succession of reliable tradents managed to accurately preserve what really happened amidst a colossal ocean of storyteller distortions swirling around them? It is not impossible, but it does seem extraordinary—surely it is easier to suppose that all of this material, which all looks similar in character, equally reflects a similar storyteller origin?

There is however a more serious problem in this case: by taking into account the standard insights of textual criticism and form criticism, it is absolutely clear that the so-called one true version is a secondary or even tertiary development within the narrative material under consideration, an outgrowth from more basic themes and narratives. The point is simple: in the course of the creation, redaction, and transmission of narrative material, elements and details tend to accrue.[22] Thus, over time, some iterations of a basic theme will be combined with other themes; and then some versions of this combination will acquire further details; and so on, so forth. The more primitive forms often survive alongside these more developed versions, however, such that this development—and the original discreteness or individuality of the forms in question—can still be detected in hindsight.[23] This is exactly the situation we have with the material on Q. 5:3: originally, there was a basic theme about ʿArafah, amongst others; then, at some point, some versions of this theme acquired the Friday element, even as other versions survived without it; then, at some point, some versions acquired the motif of the smug kitābiyy, even as other versions survived without it; and here, at this tertiary stage of development, we find the version of Qays, amongst others. The conclusion here is unavoidable: Qays’ version is not only a coequal product of the same exegetical and storytelling process that produced the rest, but also a version that is more altered and more artificial than many others.

All of this is completely inconsistent with the notion that “weak” tradents are to blame for the contradictions: the transmissions of so-called “reliable” tradents are indistinguishable from those of so-called “weak” tradents. To put it another way, there are no reliable tradents per se (at least for the early period), and all of the material is the same: the distinction presupposed by this first explanation from ʿItr simply does not apply to such material.

“2 – There are a number of causes [for] a single revelation.”

This makes no sense as an explanation for the data under consideration: each of the reports is claiming to convey the specific context (the time, the place, the people present, the activity of the Prophet) at the moment that a given verse or surah of the Quran was revealed. It makes no difference if the verse or surah was revealed, in some broader sense, for a variety of reasons: there would still only be one specific moment of revelation. This ‘explanation’ is thus completely irrelevant to the data under consideration, which concerns conflicting reports about the specific moment of revelation.

“3 – [It could be] that the revelation of the [same] text occurred a number of times at a number of different occasions.”

Here, in the final explanation from ʿItr, we encounter what is perhaps the most implausible proposition thus far, at least as applied to the kinds of examples being adduced in critical scholarship: that the wild array of different scenarios depicted in reports concerning, for example, “the revelation of Q. 5:3” or “the revelation of Surah 5” more broadly actually refer to different yet simultaneously true events. This is the Muslim-apologist equivalent of the Christian-apologist claim that Judas died both by falling over and splitting his belly open and by simultaneously hanging himself, or the Christian-apologist claim that Jairus’ daughter was raised from the dead twice, on two different occasions. All of this is possible; and all of it is absurd. The overwhelmingly more probable meaning of a given claim that X verse was revealed in Y circumstance, or that Y was the circumstance of X verse’s revelation, is that X verse was first revealed in Y circumstance. In short, ʿItr’s third explanation requires a tortured, implausible interpretation of the intended meaning of the reports under consideration.

Such an explanation becomes even more tortuous when the specific scenarios in these reports are considered: are we really to believe that the same verse—or the entire surah including the verse—was revealed on a Monday and again on a Friday; that it was revealed on the Day of Immolation/the Greater Pilgrimage and again on Day of ʿArafah; that it was revealed in Madinah and again somewhere between Madinah and Makkah and again at Mount ʿArafah; that it was revealed whilst the Prophet was praying towards Jerusalem, and again whilst he was riding a camel during a journey, and again when his camel was being led by ʾAsmāʾ bt. Yazīd, and again when his camel was being led by ʾAsmāʾ bt. ʿUmays, and again whilst he was mounted on a camel at Mount ʿArafah, and again whilst he was standing on Mount ʿArafah; and so on, so forth? Surely it is easier to suppose that these represent not just conflicting attempts to identify the same specific sabab al-nuzūl, but also different iterations of a common set of narrative material. Certainly, such an explanation conforms to our established background knowledge on oral storytelling practices.

In fact, this established background knowledge generates a further problem with this already wildly implausible explanation from ʿItr: in general, patterns of evidence like this are explained by artificial narrative creation and storytelling. In other words, it cannot be denied that the vastly more likely explanation for the recurrence of similar elements in the different accounts is not that the Prophet repeatedly revealed the same verse in similar situations, but rather, that these similarities are stock tropes, motifs, etc., that were used and reused in the course of the composition of these stories.

In short, ʿItr’s third explanation—at least as applied to examples like Q. 5:3—strains credulity and contradicts our established background knowledge on the intended meanings of words (i.e., the most obvious intended meaning of “X was revealed in Y context”); it requires the strange assumption that the same verse or surah was repeatedly re-revealed in numerous extremely similar yet different situations; and it is vastly less likely, as an explanation for such patterns of evidence in general, than artificial narrative construction and oral storytelling.

Summary of ʿItr

To summarise ʿItr’s three explanations, as applied to the kinds of reports cited by critical scholars (such as those concerning the sabab al-nuzūl of Q. 5:3): (1) the first appeals to a distinction—between weak and reliable tradents—that is completely incompatible with the internal indications that all of the material equally reflects a common process of exegetical speculation, artificial narrative construction, and oral storytelling; (2) the second is irrelevant; and (3) the third is wildly implausible and contradicts our established background knowledge on multiple points.

Broader Implications

After summarising the preceding traditionalist explanations, D1mashqi concluded: “With that, we cross off this criticism of the meticulous system of Hadīth that our scholars have put together.”[24] In this, we see what is really at stake here: the reliability of the Hadith corpus more broadly, and of traditional Islamic Hadith scholarship.

Of course, it could be argued that conclusions about exegetical Hadith cannot be generalised to the rest of the corpus—after all, even within traditional scholarship, exegetical Hadith (along with several other genres) were regarded as generally inferior to legal and doctrinal Hadith. For example, the famous Baghdadian Hadith critic ʾAḥmad b. Muḥammad b. Ḥanbal (d. 241/855) reportedly declared:

When we transmit from the Prophet [reports] about what is permissible (al-ḥalāl), what is forbidden (al-ḥarām), precedents (al-sunan), and legal rulings (al-ʾaḥkām), we are stringent (tašaddadnā) regarding isnads; but when we transmit from the Prophet [reports] about the merits of good deeds (faḍāʾil al-ʾaʿmāl), or that which neither establishes nor abolishes a legal ruling (ḥukm), we are lax (tasāhalnā) towards isnads.[25]

Similarly, the infamous Syrian Ḥanbalī jurist and Hadith scholar ʾAḥmad b. Taymiyyah (d. 728/1328) cited the following from Ibn Ḥanbal concerning exegetical Hadith and other such non-doctrinal material:

As for hadiths [regarding] the cause of revelation (ʾaḥādīṯ sabab al-nuzūl), most of them are mursal [i.e., discontinuous in their isnads], not musnad [i.e., continuous or unbroken in their isnads]. In this regard, Imam ʾAḥmad b. Ḥanbal said: “Three [bodies] of knowledge have no isnad (ṯalāṯ ʿulūm lā ʾisnāda la-hā)…” And in [another] wording: “[Three bodies of knowledge] have no [proper] source (laysa la-hā ʾaṣl): Quranic exegesis (al-tafsīr); Prophetical biography (al-maḡāzī); and [End Times] tribulations (al-malāḥim).” Meaning: that the hadiths thereon are mursal.[26]

In a sense, therefore, even traditional Muslim Hadith scholars recognised that exegetical Hadith are problematic. However, despite this recognition, the problems with exegetical Hadith do have broader implications for the reliability of Hadith and traditional Islamic Hadith scholarship, in several ways: (1) many specific exegetical hadiths—despite the general reservations just cited—were explicitly judged to be “sound” (ṣaḥīḥ), yet are manifestly false; (2) the widespread reliance upon guesswork, inference, and extrapolation in the creation of exegetical Hadith indicates that early Muslims in general were fundamentally disconnected in some way from the original Quranic milieu (i.e., had lost most of the context and even some of the meaning of the Quran)[27]; and (3) the distinctive signs of storyteller construction manifest in exegetical Hadith can be found all across the Hadith corpus, even within doctrinal material.[28]

Conclusion

The results of this exercise are unambiguous: time and again, the critical explanations for contradictions in exegetical Hadith have (1) actually explained the evidence, (2) done so in a simple fashion, and/or (3) conformed to our established background knowledge; and time and again, the traditional explanations have failed on one or more of these counts. In other words, none of the alterative traditional explanations have survived contact with the evidence.

All of this is a perfect illustration of a common problem in religio-apologetical interactions with secular scholarship that I identified in the podcast mentioned at the outset (here), as follows:

At the end of the day, you have to engage with the evidence and the arguments. And this is the problem: I think most of the time people are not engaging with the actual arguments and all of the actual evidence, and I have a problem with that. I don’t mind people disagreeing. I’ve got no problem with that. But I want people to actually understand what they are disagreeing with.

It is not sufficient, as a response to critical scholarship on a given issue, merely to point to a traditionalist attempt to rationalise the issue in some vague or abstract way: the actual argumentation of critical scholars’ needs to be addressed, along with the actual evidence being cited. More fundamentally, it is not sufficient merely to come up with an alternative explanation; the alternative has to actually explain the evidence, and to do so better than other explanations.[29]

As things stand, the best explanation overall for the occurrence of wildly conflicting exegetical reports regarding the meaning of obscure words in the Quran is that the meanings thereof were largely forgotten, leading early Muslim exegetes to variously infer or extrapolate the meaning from or in relation to the Quran itself; and the best explanation for conflicting exegetical reports regarding the historical context of the revelation of particular surahs or verses is that such reports were created through the imaginings and reimaginings of later exegetes and especially storytellers.

Of course, all of this still leaves a final question unanswered: what does qaswarah actually mean? Were any of the eleven interpretations expressed in the relevant exegetical Hadith correct? If so, how exactly did the other erroneous interpretations arise? Were they the product of guesswork based purely upon the context of Q. 74:51, or did some of them arise through a misunderstanding of the word and/or erroneous extrapolations from related words? All of this will be discussed in a second, forthcoming article, in which I subject all of the relevant exegetical Hadith to a combined isnad-cum-matn, geographical, linguistic, and historical-critical analysis.

* * *

I owe thanks to J.B., abcshake, Marijn van Putten, Mehrab, CJ Canton, and especially Yet Another Student, for their generous support over on my Patreon. I also owe special thanks to Marijn van Putten and Lameen Souag, for their critical insights and helpful feedback on a draft version of this article.

---

[1] Michael A. Cook, Muhammad (Oxford, UK: Oxford University Press, 1983), 70-73; Patricia Crone, Meccan Trade and the Rise of Islam (Princeton, USA: Princeton University Press, 1987), ch. 9; Andrew L. Rippin, “The Function of Asbāb al-Nuzūl in Qurʾānic Exegesis”, Bulletin of the School of Oriental and African Studies, Volume 51, Issue 1 (1988), 1-20; Patricia Crone, “Two legal problems bearing on the early history of the Qurʾān”, Jerusalem Studies in Arabic and Islam, Volume 18, Number 1 (1994), 1-37; Michael A. Cook, The Koran: A Very Short Introduction (Oxford, UK: Oxford University Press, 2000), 136-137; François de Blois, “Naṣrānī (Ναζωραίος) and ḥanīf (έθνικός): Studies on the religious vocabulary of Christianity and of Islam”, Bulletin of the School of Oriental and African Studies, Volume 65, Issue 1 (2002), 17-18; Gabriel S. Reynolds, The Qurʾān and Its Biblical Subtext (Abingdon, UK: Routledge, 2010), e.g., 21-22, 99, 231; Harald Motzki, “The Origins of Muslim Exegesis. A Debate”, in Harald Motzki, Analysing Muslim Traditions: Studies in Legal, Exegetical and Maghāzī Ḥadīth (Leiden, the Netherlands: Koninklijke Brill NV, 2010), 270-271; Lawrence I. Conrad, “Qurʾānic Studies: A Historian’s Perspective”, in Manfred S. Kropp (ed.), Results of Contemporary Research on the Qurʾān: The Question of a Historico-Critical Text of the Qurʾān (Beirut, Lebanon / Würzburg, Germany: Orient-Institut Beirut / Ergon Verlag, 2007), 13; Harald Motzki, Reconstruction of a Source of Ibn Isḥāq’s Life of the Prophet and Early Qurʾān Exegesis: A Study of Early Ibn ʿAbbās Traditions (Piscataway, USA: Gorgias Press, 2017), 124; Andreas Görke, “Between History and Exegesis: The Origins and Transformation of the Story of Muḥammad and Zaynab bt Ǧaḥš”, Arabica, Volume 65, Issue 1-2 (2018), 34-35, 48-49, 62; Harald Motzki, “Historical-Critical Research of the Sīra of the Prophet Muhammad: What Do We Stand to Gain?”, in Josephine van den Bent, Floris van den Eijnde, & Johan Weststeijn (ed.), Late Antique Responses to the Arab Conquests (Leiden, the Netherlands: Koninklijke Brill NV, 2022), 85-86.

[2] Cook, Muhammad, 66-67; Gautier H. A. Juynboll, Muslim tradition: Studies in chronology, provenance and authorship of early ḥadīth (Cambridge, UK: Cambridge University Press, 1983), 11-14, 17, 23, 74; Crone, Meccan Trade, esp. ch. 9; Rippin, “The Function of Asbāb al-Nuzūl”, 19; Crone, “Two legal problems”, Gregor Schoeler (ed. James E. Montgomery & trans. Uwe Vagelpohl), The Biography of Muḥammad: Nature and Authenticity (New York, USA: Routledge, 2011), 12 and ch. 2, esp. 74, 79; Motzki, “The Origins of Muslim Exegesis. A Debate”, in Motzki, Analysing Muslim Traditions, 265-266; Robert G. Hoyland, “History, fiction and authorship in the first centuries of Islam”, in Julia Bray (ed.), Writing and Representation in Medieval Islam: Muslim horizons (Abingdon, UK: Routledge, 2006), 31-32; Andreas Görke, Harald Motzki, & Gregor Schoeler, “First Century Sources for the Life of Muḥammad? A Debate”, Der Islam, Volume 89, Number 2 (2012), 28-29; Andrew G. Bannister, An Oral-Formulaic Study of the Qur’an (Lanham, USA: Lexington Books, 2014), 45-46, 60 (n. 19); Andreas Görke, “Authorship in the Sīra literature”, in Lale Behzadi & Jaakko Hämeen-Anttila (eds.), Concepts of Authorship in Pre-Modern Arabic Texts (Bamberg, Germany: University of Bamberg Press, 2015), 84; id., “Between History and Exegesis”, 36, 62; Neal Robinson, “Muhammad, ‘The Prophet like Moses’, and the editing of the Qur’an”, research presented at the Oriental Institute, University of Oxford (23^rd/October/2018); Motzki, “Historical-Critical Research of the Sīra”, in van den Bent et al. (ed.), Late Antique Responses to the Arab Conquests, 83.

[3] https://twitter.com/D1mashqi/status/1687064342045323264?s=20

[4] https://twitter.com/D1mashqi/status/1687064342045323264?s=20

[5] A comprehensive citation and analysis of all of this material will be given in a follow-up article, but for now, see Muḥammad b. Jarīr al-Ṭabarī (ed. ʿAbd Allāh b. ʿAbd al-Ḥasan al-Turkī), Tafsīr al-Ṭabarī: Jāmiʿ al-Bayān ʿan Taʾwīl ʾÂy al-Qurʾān, vol. 23 (Cairo, Egypt: Dār Hajar, 2001), pp. 455 ff.; ʾAḥmad b. Muḥammad al-Ṯaʿlabī (ed. Ṣalāḥ b. Sālim Bā-ʿUṯmān), al-Kašf wa-al-Bayān ʿan Tafsīr al-Qurʾān, vol. 28 (Jeddah, KSA: Dār al-Tafsīr, 2015), pp. 88 ff.; ʿAbd al-Raḥmān b. ʾabī Bakr al-Suyūṭī, Tafsīr al-Durr al-Manṯūr fī al-Tafsīr al-Maʾṯūr, vol. 8 (Beirut, Lebanon: Dār al-Fikr, 2011), p. 339.

[6] Previous scholarship has mostly appealed to mere contradictions to this end, e.g., Crone, Meccan Trade, 204 ff.; ead., “Two legal problems”, 1-2. In other words, to my own surprise, the further lines of argumentation that I have adduced here appear to be my own, or else I acquired them from a source now forgotten. With that said, there is a hint of the sort of thing that I am arguing in ead., Meccan Trade, 212: “Taken in isolation, each suggestion sounds convincing.”

[7] I.e., including: if linguistic sources are only ever able to cite exegetical reports as evidence for the interpretations, as opposed to pre-Islamic poems, etc. The degree to which such a scenario applies to qaswarah will be elaborated in a forthcoming article.

[8] For a similar point, see Görke, “Between History and Exegesis”, 48.

[9] On this last point, see Herbert Berg, “Ibn ʿAbbās in ʿAbbāsid-Era Tafsīr”, in James E. Montgomery (ed.), ʿAbbasid Studies: Occasional Papers of the School of ʿAbbasid Studies (Leuven, Belgium: Peeters Publishers & Department of Oriental Studies, 2004), 129-146.

[10] Musāʿid b. Sulaymān al-Ṭayyār, Fuṣūl fī ʾUṣūl al-Tafsīr (Riyadh, KSA: Dār al-Našr al-Dawliyy, 1993), pp. 65 ff.

[11] I expect this point to be serially misunderstood by some, so allow me to reiterate here what I am actually saying in excruciating detail: I am not saying that an Arabic root, nor even a specific Arabic word, cannot encompass a wide range of meanings. Take for example the word nabl: it means both “arrows” and “noble, exalted, superior, etc.”, i.e., two completely different meanings. The problem with qaswarah is not merely that it has completely different meanings, but rather, that it has numerous completely different meanings: this is clearly extremely rare for a noun and thus unexpected, i.e., suspicious in this instance. The key word here, incidentally, is noun: obviously, a verb with a basic meaning and its corresponding first-form participles (e.g., faʿala and fāʿil; daḵala and dāḵil; ḵaraja and ḵārij) can have a vast range of meaning. However, qaswarah is not from a root with such a broadly applicable basic meaning, nor is it a verb with such a basic meaning, nor is it a participle derived from such a verb (a verbish kind of noun, so to speak): it is a non-participle type of noun (a nounish kind of noun, so to speak), and one in a very rare form (faʿwalah) at that. It is extremely rare, and thus unexpected, that such a noun would have such a broad range of disparate meanings. This is all the more unexpected given that qaswarah is a rare word, i.e., one that almost never appears outside of discussions of and references to Q. 74:51. The idea that such a rarely used word would have such a huge range of meaning is totally unexpected, i.e., the opposite of what we would expect: it is frequent usage that allows radically diverse meanings to arise. There is no denying the obvious here, absent special pleading: the alleged semantic range of qaswarah presented in the relevant exegetical Hadith is implausible. (I owe special thanks to Marijn van Putten for correcting my formulations on this issue.)

[12] Thus, the most important early sources for pre-Islamic Arabic dialects, such as Sībawayh and al-Farrāʾ, make no connection between these different definitions and different dialects.

Two much later sources do relate qaswarah to dialect, but only for one or two definitions, and in a contradictory way at that: compare ʿAbd Allāh b. al-Ḥusayn b. Ḥasnūn al-Muqriʾ (ed. Ṣāliḥ al-Dīn al-Munajjid), Kitāb al-Luḡāt fī al-Qurʾân, 2nd ed. (Beirut, Lebanon: Dār al-Kutub al-Jadīd, 1972), p. 12, with Salamah b. Muslim al-Ṣuḥārī (ed. ʿAbd al-Karīm Ḵalīfah, Nuṣrat ʿAbd al-Raḥmān, Ṣalāḥ Jarrār, Muḥammad Ḥasan ʿAwwād, & Jāsir ʾAbū Ṣafiyyah), Kitāb al-ʾIbānah fī al-Luḡah al-ʿArabiyyah, vol. 1 (Muscat, Oman: Wizārat al-Turāṯ al-Qawmiyy wa-al-Ṯiqāfah, 1999), p. 102. According to Ibn Ḥasnūn, qaswarah “means “lion” (al-ʾasad) in the dialect of Qurayš and the dialect of ʾAzd Šanūʾah.” However, according to al-Ṣuḥārī, qaswarah is the Ethiopian word for “lion”, and “in the dialect of ʾAzd Šanūʾah, [it means] “archers” (al-rumāh).” Even if it were accepted that qaswarah was ʾAzdī and/or Qurašī dialect for “lion”, or else that it was ʾAzdī dialect for “archers”, that would still leave all of the conflicting definitions unaccounted for in terms of tribal dialect. Thus, already at this point, an appeal to dialectical differences to explain away the numerous contradictory definitions for qaswarah would fail.

Of course, the notion that qaswarah was Qurašī and ʾAzdī dialect for “lion” (only attested by Ibn Ḥasnūn, quite late) is directly contradicted by a report that can be traced back at least as early as Šuʿbah b. al-Ḥajjāj (d. 160/777), reportedly from ʾAbū Ḥamzah, from Ibn ʿAbbās: “I am not aware of its [meaning] “lion” (al-ʾasad) in the language of any of the Arabs. It [means] “a band/bands of men” (ʿuṣbah/ʿuṣab al-rijāl).” See Saʿīd b. Manṣūr (ed. Saʿīd b. ʿAbd Allāh al-Ḥumayd & Ḵālid b. ʿAbd al-Raḥmān al-Juraysī), Sunan Saʿīd bn Manṣūr, vol. 8 (Riyadh, KSA: Dār al-ʾAlūkah, 2012), pp. 213-214; Ṭabarī (ed. Turkī), Tafsīr, XXIII, pp. 458.

In short, the evidence that the disagreements over the meaning of qaswarah reflect tribal-dialectal differences is extremely weak at best (covering only one or two definitions) and non-existent at worst (i.e., unattested or contradicted by much earlier sources).

[13] This is the putative root for qaswarah that is cited over and over in the tafsīr works, e.g., Maḥmūd b. ʿUmar al-Zamaḵšarī (ed. ʿĀdil ʾAḥmad ʿAbd al-Mawjūd & ʿAlī Muḥammad Muʿawwiḍ), al-Kaššāf, vol. 6 (Riyadh, KSA: Maktabat al-ʿUbaykān, 1998), p. 263, ad Q. 74:51: “[The word] al-qaswarah [means] “a group of archers who are hunting them [i.e., the wild asses]” (jamāʿat al-rumāh allaḏīna yataṣayyadūna-hā). However, it is also said [that it means] “lion” (al-ʾasad); lions (luyūṯ) are called qasāwir. It [i.e., the word qaswarah] is the faʿwalah [form] of [the root] al-qasr, which [means] “subjugation” (al-qahr) and “defeating” (al-ḡalabah). Amongst the [other] names for “lion”, [the word] al-ḥaydarah is [likewise modelled] on the same form [i.e., faʿwalah]. And [it is also transmitted] from Ibn ʿAbbās [that qaswarah means] “the sound of men and their voices” (rikz al-nās wa-ʾaṣwātu-hum), and from ʿIkrimah [that it means] “the darkness of the night” (ẓulmat al-layl).”

[14] Muḥammad b. ʾAḥmad b. Juzayy al-Kalbī (ed. Muḥammad Sālim Hāšim), al-Tashīl li-ʿUlūm al-Tanzīl, vol. 1 (Beirut, Lebanon: Dār al-Kutub al-ʿIlmiyyah, 1995), p. 12.

[15] In general, see Crone, Meccan Trade, ch. 9.

[16] Likewise, ibid., 219; Stijn Aerts, “‘Pray with Your Leader’: A Proto-Sunni Quietist Tradition”, Journal of the American Oriental Society, Volume 136, Issue 1 (2016),41.

[17] See esp. Cook, Muhammad, 66; Crone, Meccan Trade, ch. 9.

[18] Ibid., 216.

[19] Ibid., 215; Hoyland, “History, fiction and authorship”, in Bray (ed.), Writing and Representation in Medieval Islam, 31-32; Bannister, An Oral-Formulaic Study of the Qur’an, 45-46, 60 (n. 19). More generally, see Bart D. Ehrman, Jesus Before the Gospels: How the Earliest Christians Remembered, Changed, and Invented Their Stories of the Savior (San Francisco, USA: HarperOne, 2016).

[20] For a whole separate line of argumentation (i.e., appealing the inherent implausibility of the historical transmission required for some ʾasbāb al-nuzūl stories, such that they must be exegetical guesswork in narrative form), see Motzki, Reconstruction, 124 ff. However, since this applies to only a subset of exegetical narratives, I have opted not to address it here.

[21] Nūr al-Dīn Muḥammad al-Ḥasanī ʿItr, ʿUlūm al-Qurʾân al-Karīm (Damascus, Syria: Maṭbaʿat al-Ṣabbāḥ, 1993), pp. 49 ff.

[22] E.g., Cook, Muhammad, 63-64; Crone, Meccan Trade, 223-224; Pavel Pavlovitch, The Formation of the Islamic Understanding of Kalāla in the Second Century AH (718–816 CE): Between Scripture and Canon (Leiden, the Netherlands: Koninklijke Brill NV, 2015), 37-39. More generally, see John P. Postgate, “Textual Criticism”, in The Encyclopædia Britannica: A Dictionary of Arts, Sciences, Literature and General Information: Eleventh Edition, vol. XXVI (Cambridge, UK: Cambridge University Press, 1911), 713; Bruce M. Metzger & Bart D. Ehrman, The Text of the New Testament: Its Transmission, Corruption, and Restoration, 4th ed. (Oxford, UK: Oxford University Press, 2005), 166, citing Johann Jakob Griesbach.

[23] E.g., Ehsan Roohi, “A Form-Critical Analysis of the al-Rajīʿ and Biʾr Maʿūna Stories: Tribal, Ideological, and Legal Incentives behind the Transmission of the Prophet’s Biography”, Al-ʿUṣūr al-Wusṭā, Volume 30 (2022), 270-271.

[24] https://twitter.com/D1mashqi/status/1687064515655991296?s=20

[25] ʾAḥmad b. ʿAlī al-Ḵaṭīb al-Baḡdādī (ed. ʾIbrāhīm b. Muṣṭafá al-Damyāṭī), al-Kifāyah fī Maʿrifat ʾUṣūl ʿIlm al-Riwāyah, vol. 1 (Cairo, Egypt: Dār al-Hudá, 2003), p. 399.

[26] ʾAḥmad b. Taymiyyah (ed. Muḥammad Rašād Sālim), Minhāj al-Sunnah al-Nabawiyyah fī Nqḍ Kalām al-Šīʿah wa-al-Qadariyyah, vol. 7 (Riyadh, KSA: Jāmiʿat al-ʾImām Muḥammad bn Saʿūd al-ʾIslāmiyyah, 1986), p. 435.

[27] For more on this, see Arthur Jeffery, The Foreign Vocabulary of the Qurʾān (Leiden, the Netherlands: Koninklijke Brill NV, 2007); 3-4; Patricia Crone, Slaves on Horses: The Evolution of the Islamic Polity (Cambridge, UK: Cambridge University Press, 1980), ch. 1; Cook, Muhammad, 70-73; Crone, Meccan Trade, esp. 210, 225-230; ead., “Two legal problems”, passim; Gerald R. Hawting, The Idea of Idolatry and the Emergence of Islam: From Polemic to History (Cambridge, UK: Cambridge University Press, 1999), passim; Cook, The Koran, 138-141; Chase F. Robinson, “Reconstructing Early Islam: Truth and Consequences”, in Herbert Berg (ed.), Method and Theory in the Study of Islamic Origins (Leiden, the Netherlands: Koninklijke Brill NV, 2003), 117; Conrad, “Qurʾānic Studies”, in Kropp (ed.), Results of Contemporary Research on the Qurʾān: The Question of a Historico-Critical Text of the Qurʾān, 12-13; Claude Gilliot, “Reconsidering the authorship of the Qurʾān: Is the Qurʾān partly the fruit of a progressive and collective work?”, in Gabriel S. Reynolds (ed.), The Qurʾān in Its Historical Context (Abingdon, UK: Routledge, 2008), 98; Reynolds, The Qurʾān and Its Biblical Subtext, passim; Fred M. Donner, Muhammad and the Believers: At the Origins of Islam (Cambridge, UK: Harvard University Press, 2010), esp. 56; Gabriel S. Reynolds, “Variant readings”, The Times Literary Supplement (5^th/August/2015): https://www.the-tls.co.uk/articles/variant-readings/; Nicolai Sinai, “The Unknown Known: Some Groundwork for Interpreting the Medinan Qur’an”, Mélanges de l’Université Saint-Joseph, Volume 66 (2015-2016), 47-48 (incl. n. 4), 80; Pavlovitch, Formation, esp. ix-xi, ch. 2, 512 ff.; Patricia Crone (ed. Hanna Siurua), The Qurʾānic Pagans and Related Matters: Collected Studies in Three Volumes, Volume 1 (Leiden, the Netherlands: Koninklijke Brill NV, 2016), passim; Jack Tannous, The Making of the Medieval Middle East: Religion, Society, and Simple Believers (Princeton, USA: Princeton University Press, 2018), passim, esp. 295 ff.; Juan Cole, Muhammad: Prophet of Peace amid the Clash of Empires (New York, USA: Bold Type Books, 2018), e.g., 97, 100, 142, 145, 199, 281 (n. 26), 282 (n. 28), 285-286 (n. 47).

However, the amnesia was certainly not complete, and archaic (i.e., original, accurate) understandings of the Quran can still be found, even amidst later widespread confusion and misconception. For a notable example thereof, see Saqib Hussain, “The Bitter Lot of the Rebellious Wife: Hierarchy, Obedience, and Punishment in Q. 4:34”, Journal of Qur’anic Studies, Volume 23, Issue 2 (2021), 66-111.

[28] See esp. Crone, Slaves on Horses, 7; ead., Roman, provincial and Islamic law: The origins of the Islamic patronate (Cambridge, UK: Cambridge University Press, 1987), 32-33; ead., Meccan Trade, 225, n. 96; ead., “Two legal problems”; ead., “Muhammad and the origins of Islam. By F. E. Peters”, Journal of the Royal Asiatic Society of Great Britain and Ireland, Volume 5, Issue 2 (1995), 270-271.

[29] Of course, this kind of problem is not limited to religio-apologetical interactions with secular scholarship; but it is a recurring feature of such interactions.