Does ḥarrafa imply textual corruption?

When Muslims think of the previous scriptures, one particular word may come to mind: Taḥrīf (‘corruption’, ‘altering’). They will often think of textual corruption; i.e. that the Torah and Gospel we have in our hands today cannot be trusted, because it has been textually corrupted in the past.

The four ḥarrafa verses

The Qur’an does not use the word Taḥrīf, but it does use the verb from the same root (rf). The verb is used four times in the Qur’an:

So can you [believers] hope that such people will believe you, when some of them used to hear the words of God and then deliberately twist them [yuḥarrifūnahu], even when they understood them? (Q 2:75. Unless otherwise indicated, Qur’an citations are from the Abdel Haleem translation)

Some Jews distort the meaning of [revealed] words [yuḥarrifūna l‑kalima ʿan mawāḍiʿihi]: they say, ‘We hear and disobey,’ and ‘Listen,’ [adding the insult] ‘May you not hear,’ and ‘Ra’ina[ Look at us],’ twisting it abusively with their tongues so as to disparage religion. If they had said, ‘We hear and obey,’ ‘Listen,’ and ‘Unzurna [Look at us],’ that would have been better and more proper for them. But God has spurned them for their defiance; they believe very little. (Q 4:46)

But they broke their pledge, so We distanced them [from Us] and hardened their hearts. They distort the meaning of [revealed] words [yuḥarrifūna l‑kalima ʿan mawāḍiʿihi] and have forgotten some of what they were told to remember: you [Prophet] will always find treachery in all but a few of them. Overlook this and pardon them: God loves those who do good. (Q 5:13)

Messenger, do not be grieved by those who race to surpass one another in disbelief— those who say with their mouths, ‘We believe,’ but have no faith in their hearts, and the Jews who listen eagerly to lies and to those who have not even met you, who distort the meanings of [revealed] words [yuḥarrifūna l‑kalima min baʿdi mawāḍiʿihi] and say [to each other], ‘If you are given this ruling, accept it, but if you are not, then beware!’ — if God intends some people to be so misguided, you will be powerless against God on their behalf. These are the ones whose hearts God does not intend to cleanse— a disgrace for them in this world, and then a heavy punishment in the Hereafter— (Q 5:41)

The square brackets with English comments are original; I have added square brackets containing Arabic. Bold Emphasis added.

In all of my articles I use Abdel Haleem as my default translation; while I do sometimes disagree with his translation choices, his translation is well-respected. I stress that Abdel Haleem is the usual translation in my articles because in this case he adopts a translation that agrees with my own; his translation speaks of ‘twist[ing]’ words and ‘distort[ing] the meaning of words’ (emphasis added). He does not translate this phrase as indicating textual corruption.

This is not to say that he does not believe in textual corruption, or does not affirm it elsewhere. He may be suggesting such a belief in his footnote to Q 3:3-4, where he cross-references Q 5:48, for which see my previous article. But he does not seem to think that textual corruption is indicated by ḥarrafa.

Nor does Ali Quli Qarai’s Qur’an translation, used in Reynold’s (2018) The Qur’an and Bible: text and commentary. Nor even does Muhammad Asad in Q 4:46, 5:13, 5:41 (and perhaps even Q 2:75, for which issue cf. the footnote citing Jeremiah 23:36). He considers these verses to be discussing corruption of interpretation, even though he does state his belief in textual corruption elsewhere (Q 2:79, p. 45, footnote 64; Q 5:48, pp. 223-224, see his translation of muhaymin in Q 5:48 and see footnote 64).

Western scholarship and the ḥarrafa verses

According to Gabriel Reynolds (2010, 193), ‘[a]ccording to most Western scholarship, the Qurʾan is referring to textual alteration with the verb yuḥarrifūna.’ I have argued elsewhere (here and here) that it is not actually so clear what ‘most Western scholarship’ believes on the broader topic of whether the Qur’an alleges textual corruption of the previous scriptures. My own survey displayed a roughly even split. If this is the case for the broader issue, perhaps we can infer what ‘most scholarship’ does or does not hold on one of the constitutive arguments of that issue.

Reynolds then points us to Hava Lazarus-Yafeh’s definition of taḥrīf:

[C]hange, alteration, forgery; used with regard to words, and more specifically with regard to what Jews and Christians are supposed to have done to their respective Scriptures (yuḥarrifūna ʾl-kalima ʿan mawāḍiʿhi [sic], sūra IV, 46, V, 13; see also II, 75), in the sense of perverting the language through altering words from their proper meaning, changing words in form or substituting words or letters for others. [Hava Lazarus-Yafeh, “Taḥrīf,” in Encyclopaedia of Islam II, 10:111a]

After citing Lazarus-Yafeh in support of his statement, strangely then Reynolds makes precisely the opposite point by citing

Edwards W. Lane, who relies on medieval Islamic dictionaries…[and] defines the idiom Taḥrīf l-kalima ʿan mawāḍiʿihi as “the altering of words from their proper meanings.” He adds that in accordance “with this explanation, the verb [ḥarrafa] is used in the Ḳur[ān].’'[E. W. Lane, An Arabic-English Lexicon (London: The Islamic Texts Society, 19841), 1:549b].

This sounds like corruption of interpretation, not corruption of text. Reynolds’ (194) following linguistic discussion is worth quoting at length:

In order to understand what the Qurʾan means by yuḥarrifūna l-kalima ʿan mawāḍiʿihi, one might begin with the literal sense of mawāḍiʿ (sing. mawḍiʿ) ‘places’ (and not ‘meanings’, which might represent mawāḍīʿ [sing. mawḍūʿ]) and the primary meaning of the root ḥ-r-f ‘to move; to turn’. In classical Arabic the noun ḥarf means ‘letter’, a meaning that undoubtedly suggested to medieval Muslim scholars that with yuḥarrifūna the Qurʾān is concerned with an alteration of the very words of revelation. But ‘letter’ seems to be a secondary meaning of ḥarf, the primary meaning being extremity, verge, border, margin, brink, brow, side, or edge’. The only occurrence of ḥarf in the Qurʾan (22:11: man yaʿbudu llāha ʿala ḥarfin) evidently matches this primary meaning.

And so Reynolds concludes:

In other words, there is no compelling reason to associate Qurʾanic taḥrīf with an alteration of letters. Instead, the phrase yuḥarrifūna l-kalima ʿan mawāḍiʿihi seems to involve turning or shifting words out of their places or contexts. In other words, the Qurʾan intends scriptural falsification that involves reading or explaining scripture out of context, not erasing words and rewriting them.

Reynolds is not alone in this suggestion. He goes on to mention that Ignazio di Matteo made a similar point in response to Ignaz Goldziher. We have noted above that Lane’s lexicon, Abdel Haleem, Muhammad Asad (on most of the verses) and Qarai seem to have this understanding of the ḥarrafa verses. Additionally, my previous article noted a number of other scholars who seem to agree with Reynolds on the broader topic of whether or not the Qur’an teaches textual corruption of the former scriptures. But rather than simply picking scholarly sides, let us dive in to consider the verses in question.

Q 4:46

Some Jews distort the meaning of [revealed] words [yuḥarrifūna l‑kalima ʿan mawāḍiʿihi]: they say, ‘We hear and disobey,’ and ‘Listen,’ [adding the insult] ‘May you not hear,’ and ‘Ra’ina[ Look at us],’ twisting it abusively with their tongues so as to disparage religion. If they had said, ‘We hear and obey,’ ‘Listen,’ and ‘Unzurna[ Look at us],’ that would have been better and more proper for them. But God has spurned them for their defiance; they believe very little. (Q 4:46)

I begin with this verse as it has potentially more explanatory context than other verses. Most of the verse seems to be describing some form of oral corruption; they say certain words when they should say others, and they are ‘twisting it abusively with their tongues’.

The structure of the verse

How do we relate this oral corruption to the initial phrase that we are focusing on? It is possible that the verse is simply listing types of Jewish disobedience, first mentioning textual corruption and then moving on to discuss oral corruption. After all, one of the parallel verses (Q 5:13) may be combining two separate accusations of ḥarrafa and ‘forgetting’ the message they received.

It is probably more natural, however, to take Q 4:46 as providing an explanatory comment on the meaning of ḥarrafa, as is Droge’s (2013, 51, footnote 53) explicit understanding. Otherwise, it would be a very unbalanced combination of a brief accusation with no explanation, and a long explanation on a different (though admittedly related) accusation. Furthermore, it might seem strange to start with the stronger accusation of textual corruption, and then move on to discuss in detail the softer accusation of oral corruption, as if that were worth spending more time on.

The time of the action

Having argued that Q 4:46 should be interpreted as a continuous treatment of the same theme, let us now consider what we can learn about this action.

The action in question seems to be taking place in the present time. In v. 44 Muhammad is questioned: ‘[Prophet], do you not see how those who were given a share of the Scripture purchase misguidance and want you [believers], too, to lose the right path?’ These are people alive in Muhammad’s day. If textual corruption were in view, this would be wilful corruption during Muhammad’s day, those who ‘purchase misguidance and want you [believers], too, to lose the right path’. These are not sincere believers being misled by scriptures corrupted before their time. Coming to v. 46, the verb yuḥarrifūna is in the present tense, as is yaqūlūna (‘they say’), and in v. 47 goes on to address the present day ‘People of the Book’ in Muhammad’s audience.

It is true that ‘they said’ in ‘If they had said’ (wa‑law annahum qālū) is perfect tense; however, this may simply be due to the law conditional sentence. Though whether anna cancels out this expectation I do not know; I am happy to learn from one more advanced in Arabic grammar.

It is true that ‘Allah has cursed them’ (laʿanahumu allāhu, my translation) is in the past tense. But given the predominance of imperfect verbs and other time markers (see above), the weight of evidence is that the verse should be understood as describing malpractice by Jews in Muhammad’s own day. The perfect tense is not strictly incompatible with this; in response to their malpractice in the present, God, observing this, has pronounced his curse upon them. But it is also possible that this really is a curse that has occurred in the past; their present-day wickedness is a continuation of their past wickedness which God had cursed them for. Perhaps this curse resulted in a hard-heartedness that now explains their present-day rebelliousness.

‘We hear and disobey’

Even though Q 4:46 appears to be speaking in the present tense, Reynolds (2010, 162) notes the verbal parallel (‘We hear and disobey’) in Q 2:93, which occurs in the past. Again, we see the trans-historical connection between the Jews in Muhammad’s day and their forebears, as Muhammad addresses the Jews in front of him as if they had rebelled long ago:

91 When it is said to them, ‘Believe in God’s revelations,’ they reply, ‘We believe in what was revealed to us,’ but they do not believe in what came afterwards, though it is the truth confirming what they already have. Say [Muhammad], ‘Why did you kill God’s prophets in the past if you were true believers? 92 Moses brought you clear signs, but then, while he was away, you chose to worship the calf— you did wrong.’ 93 Remember when We took your pledge, making the mountain tower above you, and said, ‘Hold on firmly to what We have given you, and listen to [what We say].’ They said, ‘We hear and we disobey,’ and through their disbelief they were made to drink [the love of] the calf deep into their hearts. 94 Say, ‘How evil are the things your belief commands you to do, if you really are believers!’ Say, ‘If the last home with God is to be for you alone and no one else, then you should long for death, if your claim is true.’ (Emphasis added)

Given the connection between the Jews of the seventh century and their predecessors, we can respect the present tense indicators in Q 4:46 and conclude that ‘We hear and disobey’ in the verse is not a reference to a long past episode, but is the Jews of Muhammad’s own day walking in the crooked footsteps of their ancestors. The Qur’an is clearly still concerned about what people do or do not say in Muhammad’s own day, as Muhammad’s own community is instructed to say ‘We hear and obey’ (Q 24:51). Furthermore, the term ‘Rāʿinā’ seems to be an issue in the present day (see below).

But even if ‘We hear and disobey’ is not spoken by Muhammad’s Jewish contemporaries, Q 4:46 could be seen as a combined accusation of Jewish misdeeds across time. We shall have to consider the other accusations made in this verse.

‘Listen’, ‘May you not hear’

‘Listen,’ [adding the insult] ‘May you not hear,’ (wa‑ismaʿ ghayra musmaʿin) is translated differently by Qarai: ‘Hear without listening!’ Reynolds (2010, 162) suggests the Qur’an may have in mind Isaiah 6:9 (cited by Jesus in Matthew 13:14). If true, it could be alleged that the Old Testament is rebuking these scriptural passages, having been fabricated, that express such a sentiment. However, Q 4:46 seems to portray the entire set of words as an oral, not a written, corruption (‘twisting it with their tongues’). Furthermore, the term ‘Rāʿinā’ seems to be an issue in the present day (see below). Additionally, Abdel Haleem translates the words very differently, calling into doubt any parallel to Isaiah 6:9 or Matthew 13:14.

Rāʿinā

For ‘Ra’ina[ Look at us],’, Reynolds refers us to Q 2:104. In this verse the believers are instructed, similar to Q 4:46, not to say Rāʿinā but to say Unẓurnā (‘look at us’) instead. Reynolds (63) notes that a number of scholars highlight the closeness between ‘Ra’ina [Look at us]’ and the Hebrew word rāʿ, ‘evil’. As Reynolds notes:

Horovitz (“Jewish Proper Names,” 204, following Geiger, Was hat Mohammed, 17; Judaism and Islam, 12-13) argues that through the influence of Jews in Medina this word must have taken on a secondary, pejorative, sense because of its proximity to Hebrew rāʿ. Therefore Muḥammad commanded his followers to use unẓurnā instead.

He (63-64) notes that this approach ‘is close to the traditional Islamic explanation of this verse’, such as in al-Wāḥidī’s Asbāb al-Nuzūl:

This is because the Arabs used to employ this expression [raʿina], so when the Jews heard them using it with the Prophet, Allah bless him and give him peace, they liked it. This same expression in the parlance of the Jews had the connotation of vile abusive language. They said: “Before, we used to abuse Muḥammad secretly. Now, you can abuse him openly because this expression is used in their speech.

And according to The Study Qur’an (2015, 213), al-Zamakhsharī (d. 538/1144) suggested ‘that their deliberate mispronunciation of the word made it sound like a derogatory term in Hebrew’.

The nature of the malpractice

One might speculate that the Jews must have corrupted the text (e.g. at Exodus 24:7), which explains why the Jews in Muhammad’s own day are still pronouncing words incorrectly. But there is nothing in the Qur’an to suggest this; Q 4:46 itself explicitly says that it is a matter of ‘twisting it abusively with their tongues’, not inheriting a corrupted text. We have already noted how the correct use of these expressions is still an issue in the Qur’an’s own day (Q 2:104; 24:51). And interestingly, the next verse (Q 4:47) is one of the many Qur’an verses that says that the Qur’an has come ‘to confirm [muṣaddiqan, lit: ‘confirming’] what you already have’, which weighs against the view that the previous verse is alleging textual corruption.

Note also that Q 4:46, whatever action is being described, is clearly talking about the Jews, not the Christians (mina lladhīna hādū, ‘From those who are Jews’). The corruption of the Gospel, whether oral or written, is therefore not in view.

Boisliveau (2018) and Gobillot (2008) – corrupt translation?

Boisliveau (2018, paragraph 6) insists the aim of her article is not to try and demonstrate whether certain Qur’an verses do or do not teach textual corruption; nonetheless, she does decide to briefly deal with these verses as a reminder to the reader. But the reader should be alert to the fact that she argues that the whole of the Qur’an must be taken together. Any treatment by her of individual verses must be read in that light.

On Q 4:46 she (paragraph 12) introduces and summarises the work of Gobillot, Geneviève (“L’abrogation (Nâsihk Et Mansûhk) Dans Le Coran À La Lumière D’une Lecture Interculturelle Et Intertextuelle.” Al-Mawaqif, Avril (2008): 5-18). I will follow Boisliveau’s summary of Gobillot’s work as the latter is in French.

For the first set of words:

the Qurʾān says it has been badly transcribed or translated as samiʿnā wa-ʿaṣaynā, ‘we have listened and we have disobeyed’. What is probably implied by the text is that the reason for the mistake is the close connection between the Hebrew verb ʿasah, ‘to do, to accomplish’ and the Arabic verb ʿaṣā ‘to disobey’.’ (Boisliveau’s summary, paragraph 12)

I would argue that this translation could be oral, and not necessarily written. But if it is written, the Qur’an here would be challenging the accuracy of the Arabic text, not the Hebrew text.

Gobillot also suggests that the next word asmaʿ is a reference to the famous shema (Deuteronomy 6:4), which has been badly translated:

The qurʾānic text states that asmaʿ, ‘Listen’, has been badly translated as an asmaʿ‘which could not be heard’ (ġayra musmaʿ)—which, according to Gobillot’s hypothesis, leads to a mispronunciation of the šema following the form of the Arabic asmaʿ into something inaudible like šmaʿ.

In a footnote Boisliveau wonders if ‘it could be the opposite, something like an asmaʿ pronounced in the šema way, thus, sema; anyway the result is the same.’ She continues:

It seems to be an oral translation, but it could as well be a wrong written transcription that would render the word impossible to pronounce correctly—for Arabic speakers—such could be the meaning of ġayra musmaʿ.

Even if it is a ‘wrong written transcription’, note that this is a problem with the Arabic translation, not the Hebrew text. Finally, she deals with the words ‘ rāʿinā‘ being discouraged and ‘unẓurnā‘ preferred:

The verb rāʿinā means ‘favor us’, which, as Gobillot points out, refers to the idea of election of the Banū Isrā’īl; the text seems to disapprove of the presence this idea here. Now, it seems that an original Hebrew verb of the Semitic root r-’-y, rā’inā, meaning ‘Look at us’—probably with the meaning ‘Have mercy on us’, which appears in several biblical texts—has been mistranslated into rāʿinā, i.e. with the transformation of the hamza(’) into a ʿayn(ʿ). Thus what the Qurʾān says here is that it would be better to translate it by a synonym, for which no pronunciation mistake is possible: unẓurnā, ‘Look at us’, which, as well, is to be understood as ‘Have mercy on us’. Therefore, in this verse (Q IV, 46), the Jews here are accused of giving in the current language in use (Arabic) a meaning other than the one originally present in the written Torah in its original language. On the one hand, the expressions ‘do not say’/‘say’ seem to refer to an oral translation. Most plausibly, it indicates that when reading this written translation, one may pronounce it wrongly because the text has been wrongly transcribed (written down) by the Jewish translators from the Hebrew cognate to the Arabic corresponding letters. Gobillot’s hypothesis makes clear that here the qurʾānic text prompts one not to ‘transcribe’ the Torah directly by using the Arabic words having the same root as the Hebrew, but to ‘translate’ them, because a direct transcription from Hebrew script to Arabic script would falsify the meaning of the initial text or may lead to its unintentional falsification.

One might see here a written corruption in Arabic, which might mislead the reader. Alternatively, it could be postulated that there is an oral corruption (as Boisliveau concedes ‘do not say’/’say’ seems prima facie to suggest), that the one speaking is aware of the exclusivist connotations of ‘favor us’, and that this is being consciously affirmed (cf. Q 2:111). The latter would fit better with Q 4:46, which stresses their conscious deviance in their error (‘twisting it abusively with their tongues so as to disparage religion’).

How can we choose between Reynold’s approach and Boisliveau’s (following Gilliot’s) approach on Q 4:46, whether the issue is the Hebrew word rāʿ (evil) that concerns the Qur’an, or whether the Qur’an is concerned with the exclusivist (rāʿinā which could mean in Arabic ‘favour us’) claims of the Jews? For the former option, presumably, someone would have to tell Muhammad the way in which the Jews were playing a pun in Hebrew, which is not impossible given Muhammad’s frequent interactions with the Jews. This approach seems to be hinted at in al-Wāḥidī as mentioned above. But the latter is also plausible, being a theme attested in the Qur’an (e.g. Q 2:111), and not requiring any postulation of Qur’anic awareness of the Hebrew word rāʿ (though the closeness to rāʿ then effectively becomes coincidental).

I cannot choose between these two options, but neither necessitates textual corruption, and especially not in the Hebrew text.

Tafsīr

I have already mentioned al-Wāḥidī’s Asbāb al-Nuzūl (as pointed out by Reynolds, 2010, 63-64), which sees rāʿinā as being exploited by Muhammad’s Jewish contemporaries as a means of mocking him.

It is also always of interest to see what Muqātil ibn Suleymān (d. 150/767), who wrote our earliest complete extant Qur’an commentary, has to say:

They alter words from their places [yuḥarrifūna l‑kalima ʿan mawāḍiʿihi] – it means with the alteration [bi-l-taḥrīf] of the description of Muḥammad [SAW], of its places, of its declaration in the Torah, twisting with the tongues.

This would seem to be an action by Jews at the time of Muhammad. The accusation of ‘alteration’ (taḥrīf) could refer to either verbal or written alteration; the former might be in view given that he goes on to mention ‘twisting with the tongues’ (as does Q 4:46), though the latter is alleged in Q 2:79 (Nickel, 2011, 101) and Q 3:78 (ibid, 98). Given that Q 3:78 seems on its surface to speak of verbal corruption, yet is understood by Muqātil to describe textual corruption, one might think this is also his view in commenting on Q 4:46. But Muqātil is more explicit in those verses in speaking of how they ‘erased’ (maḥā) Muhammad’s description and of writing something else (ibid, 98, 101). And there does seem to be a distinction in Muqātil’s commentary in Q 3:78 between the ‘twisting of the taḥrīf of the tongues’, commenting on ‘they twist their tongues in the book’, and ‘they write the meaning of what is in the Torah other than the description of Muhammad (SAW)’, which is prompted by the phrase ‘And it is not of the Book’. Therefore the taḥrīf of Q 4:46 may well be oral taḥrīf.

As for Al-Ṭabarī (d. 310/923) Nickel (124) helpfully summarises and translates:

On the phrase, “they tamper with the words,” Ṭabarī offers a definition similar to the one he offered at Q 2:75. This phrase means “they change (baddala) their meaning (maʿnan) and alter (ghayyara) them from their interpretation (taʾwīl).” The exegete also gives an explanation of “out of their places,” glossing it as “out of their places (amākin) and their meanings (wujūh)” He notes Mujāhid’s view that “words” (kalim) means the Torah, but says no more about this line of interpretation.

He too, like al-Wāḥidī, envisages Jews in Muhammad’s own day being verbally rebellious.

Tafsīral-Rāzī

An article by Bassam Zawadi at Call to Monotheism highlights the interesting words of Fakhr al-Dīn al-Rāzī (d. 606/1210) in his commentary known both as al-Tafsīr al-kabīr and Mafātīḥ al-ghayb. He helpfully provides a translation:

In regards to Allah’s statement “They pervert the words from their proper places” (Surah 4:46), it means that they mention the corrupted interpretations for those verses, and there is no proof that they take the actual statements out of the book.

And as for the verse in Surah 5:41, this is evidence that they have combined between the two (textual corruption and misinterpreting the text), they used to mention their corrupted interpretations, and they also used to remove the statements from the book. Allah’s statement “They pervert words” indicates misinterpreting the text and Allah’s statement “after their being put in their right places” indicates that the statements were removed from the book. (Fakhar ad-Din ar-Razi, Tafsir Al Kabir, Commentary on Surah 4:46, Source)

We shall discuss this more in our consideration of Q 5:41 below. For now, we simply note that al-Rāzī thinks corruption of interpretation is in view in Q 4:46, as Zawadi accepts.

Q 2:75

So can you [believers] hope that such people will believe you, when some of them used to hear the words of God and then deliberately twist them [yuḥarrifūnahu], even when they understood them? (Q 2:75. Unless otherwise indicated, Qur’an citations are from the Abdel Haleem translation)

When and what is this referring to? Could this be referring to an episode in the distant past, like much of Q 2, such as Q 2:58 when the Israelites were commanded to say one thing when entering a town, but said another? Or perhaps in Q 2:93 when they said ‘We hear and disobey’ when given the covenant, when they should have said ‘We hear and obey’ (cf. Q 4:43)? Or perhaps the ‘words of God’ hear are their hearing of the Qur’an, which they then distort (and perhaps even mock)?

In favour of an episode in the past, up to v. 74 events in the past have been described. It seems like vv. 74-75 may be acting as a bridge, connecting past wrongdoing to present disbelief:

74 Even after that, your hearts became as hard as rocks, or even harder, for there are rocks from which streams spring out, and some from which water comes when they split open, and others which fall down in awe of God: He is not unaware of what you do. 75 So can you [believers] hope that such people will believe you, when some of them used to hear the words of God and then deliberately twist them, even when they understood them?

While v. 75 does address the present situation of ‘such people’ not believing in the Prophet, it roots their present disobedience in their previous misdeeds (v. 74, 75b). The reason they will not believe in the present is because even in the past ‘some of them used to hear the words of God and then deliberately twist them’. This may well be, therefore, an allusion to some such episode as Q 2:58 or Q 2:93. Those were originally acts of oral corruption of God’s word. While one might speculate that such oral corruption would have then been written down, and turned into written corruption, this is a deduction not made in the Qur’an, and it is not a necessary deduction. We have already seen that Q 4:46 envisages a willful act of intentional verbal corruption, not an unintentional mispronunciation due to corrupted texts.

It is also possible that Q 2:75 refers to a past action of verbal corruption, but an action in the recent past. Q 2:75 would then effectively be asking: ‘How can you hope that they will believe you, given that some of them would even go so far as to hear the words of God [in the revelation of the Qur’an] and then twist them?’ An action in the recent past might fit better with Q 4:46, though it is not implausible that the Qur’an might accuse the same group of the same type of corruption at different times in history.

Boisliveau (2018, paragraph 10) comments thus on this and the surrounding verses:

Q II, 75‒79 deals with the writing down of the Word of God that was orally transmitted to some bad people: this writing down was perverted (taḥrīf al-naṣṣ). Verse 78 also mentions people ‘from them’ (from the Jews?) accused of being pagans and of mulling over it: it seems that these people are accused of taḥrīf maʿnawī or, at least, of ignorance of the true meaning of the scripture.

I agree that some form of textual corruption is occurring in Q 2:79, viz. making false copies, which can be done for any text, even the Qur’an. I would also stress that this is only a subgroup (vv. 78-79) of the subgroup of vv. 75-77. This does not, therefore, impugn the entire manuscript tradition before the time of Muhammad.

Putting that discussion aside, it seems like Boisliveau may see in Q 2:75-78 taḥrīf maʿnawī, which regrettably was then written down (taḥrīf al-naṣṣ) (v. 79). This is the immediate focus of my article, whether the ḥarrafa verb itself implies oral or textual corruption, and it seems that Boisliveau may be affirming the former. It is possible, however, that she finds corruption of interpretation only in Q 2:78, and that ḥarrafa in v. 75 pre-empts the conclusion of taḥrīf al-naṣṣ in v. 79.

Tafsīr

Muqātil understood Q 2:75 to be referring to seventy leaders who request to hear God’s words through the mediation of Moses (Nickel, 2011, 76). Some of these leaders then faithfully pass on these instructions to the rest of the people, while others add to it the saying: ‘If you are not able to give up what he has forbidden you, then just do what you are able.’ (Nickel’s translation). Nickel notes that a similar story is given in Ibn Isḥāq’s Sīrat al-Nabī (pp. 251-252, Guillaume’s translation).

It could be speculated that this act of verbal corruption was then written down, and thus corruption of text also occured. But the accounts do not tell us this happened; it is equally possible that this ruse was discovered and corrected by God and his messenger Moses. The Qur’an does seem to think the Torah in Muhammad’s own day was still reliable enough to judge by (Q 5:43).

Al-Ṭabarī is more diverse in the exegetical options that he presents (Nickel, 2011, 120-122). One is that the Jewish leaders had another (more lenient) book alongside the Torah, from which they would judge if a bribe was offered by a guilty party. Another opinion is that it is not the reading of the Torah that is in view, but those who asked Moses to see God, and were struck with lightning. According to Nickel, al-Ṭabarī prefers the second option. Nickel notes that al-Ṭabarī explains ‘then they tampered with it’ as ‘then they changed (baddala) its meaning (maʿnan)’.

I should highlight to the reader the words of al-Rabīʿ: ‘They heard this like the prophets heard; then they altered it after they had understood it, wittingly’ (J. Cooper’s translation, 1987, 402, summarised but not fully translated by Nickel, 2011, 121). There is no more commentary on this statement, but it may be alluding to some episode such as that of the Jewish leaders in the time of Moses, which immediately follows it. One could interpret this, though I assume my Muslim friends would not wish to interpret it, to say that individuals who themselves received revelation as prophets altered that revelation; this would conflict strongly with Muslim views of prophethood.

Q 5:13

But they broke their pledge, so We distanced them [from Us] and hardened their hearts. They distort the meaning of [revealed] words [yuḥarrifūna l‑kalima ʿan mawāḍiʿihi] and have forgotten some of what they were told to remember: you [Prophet] will always find treachery in all but a few of them. Overlook this and pardon them: God loves those who do good. (Q 5:13)

As with Q 2:75, it is not entirely clear when this action of ḥarrafa is supposed to have taken place. The previous verse is clearly describing the past:

12 God took a pledge from the Children of Israel. We made twelve leaders arise among them, and God said, ‘I am with you: if you keep up the prayer, pay the prescribed alms, believe in My messengers and support them, and lend God a good loan, I will wipe out your sins and admit you into Gardens graced with flowing streams. Any of you who now ignore this [pledge] will be far from the right path.

But v. 13 seems to move, at some point, from past to present:

But they broke their pledge [lit: ‘their breaking of their pledge’], so We distanced them [from Us] [lit: ‘We cursed them’, perfect tense] and hardened [perfect tense] their hearts. They distort the meaning of [revealed] words [imperfect tense] and have forgotten [perfect tense] some of what they were told to remember: you [Prophet] will always find treachery in all but a few of them…

‘They distort the meaning…’ might seem like the shift to the present because of the imperfect verb, but it is immediately followed by a perfect tense verb, ‘have forgotten…’ As discussed before, this could be the recent past rather than the distant past, however. We clearly do, however, have a shift in v. 12 from the past, to the present at the end of v. 13 where Muhammad will still find the Jews to be largely treacherous. Where on the timescale we place the intermediate verbs will be a matter of debate. But we have already seen that, whether in Muhammad’s day or before, there is no need to see ḥarrafa as implying textual corruption.

One could argue that the combination of ḥarrafa and ‘have forgotten some of what they were told to remember’, suggests that this is a textual corruption, resulting in certain teachings or portions of scripture being forgotten because they have been lost.

In response, however, we note first that ‘forgotten’ is used and not ‘lost’ or ‘erased’. Typically we use ‘forgot’ to describe something that one can, and perhaps should, remember but have not. And, secondly, it is in fact this wilful form of forgetting that is hinted at in this verse: ‘forgotten’ is immediately followed by ‘you will always find treachery in all but a few of them…’ Indeed, most of the actions in the verse are those of wilful disobedience. Thirdly, we can see this wilful use of ‘forget’ elsewhere in the Qur’an. In the immediate context (Q 5:13-14), Christians are accused of both forgetting and concealing, the latter suggesting that the truth still exists in scripture to be concealed. Elsewhere in the Qur’an, in Q 2:44, the Qur’an is dismayed that people can ‘forget’ piety themselves, even as they encourage it for others and recite the Book. ‘Forgetting’ here is moral and wilful, not solely epistemological. Similarly, the ‘forgetting’ of the day of judgement in Q 7:51 goes hand in hand with the rejecting of God’s signs.

Tafsīr

Nickel (2011, 81) helpfully summarises Muqātil’s understanding of this verse:

On “tampering with words out of their places” – the phrase that first appeared at Q 4:46 – Muqātil offers no new information about the verb ḥarrafa. But he writes, as he did in his exegesis of Q 4:46, that “the words (kalim) are the description (ṣifa) of Muḥammad.”

We discussed at Q 4:46 whether this might be textual or non-textual corruption. As for the verb ‘forgotten’ (nasiya), Muqātil writes:

This is about how God, powerful and exalted, made a covenant with Banū Isrāʾīl in the Torah that they would believe (āmana) in Muḥammad (PBUH), and give credence (ṣaddaqa) to him. He is written [in what is] with them in the Torah. [Footnote – Nickel notes the exact parallel to Q 7:157, except for the reference to Gospel] Then when God, powerful and exalted, sent him, they disbelieved (kafara) in him and envied (ḥasada) him, and said, “This one is not from the descendant of Isḥāq, but rather he is from the descendents of Ismāʿīl. (Nickel’s translation)

Nickel (81-82) notes that though most of the Jews act in ‘treachery’ (khāʾina) and ‘faithlessness’ (al-ghishsh), a few do believe in Muhammad.

It would seem that for Muqātil, this is not an innocent forgetting of past truths because of the previous corruption of scripture. This active disbelief in Muhammad stems from their having ‘envied’ (ḥasada) him, and having disliked his ethnicity. But the truth of Muhammad’s prophethood is still there ‘with them in the Torah’ for those sincere few (e.g. ʿAbd Allāh ibn Salām) to recognise him.

Significantly, Nickel (126) notes that Q 5:13 is one of the few places where al-Ṭabarī alleges some kind of textual corruption. Commenting on our key phrase yuḥarrifūna l‑kalima ʿan mawāḍiʿihi al-Ṭabarī says:

They tampered with (ḥarrafa) the word (kalām) of their Lord which he sent down upon their prophet Moses (PBUH), and it is the Torah. So they changed (baddala) it and wrote with their hands other than what God, exalted and powerful, sent down upon their prophet. They said to the ignorant of the people, “This is that word (kalām) of God which he sent down upon his prophet Moses” – (PBUH) – “and the Torah which he revealed to him.” And this characterized the Jews in the centuries after Moses, some of whom reached the era of our prophet Muḥammad (PBUH). (Nickel’s translation)

There is here an accusation of some kind of textual corruption, with language (‘wrote with their hands other than what God…sent down’) reminiscent of Q 2:79 and perhaps Q 3:78 (as noted by Nickel, 2011, 141):

So woe to those who write something down with their own hands and then claim, ‘This is from God,’ in order to make some small gain. Woe to them for what their hands have written! Woe to them for all that they have earned! (Q 2:79, emphasis added)

There are some of them who twist the Scripture with their tongues to make you [people] think that what they say is part of the Scripture when it is not; they say it is from God when it is not; they attribute lies to God and they know it. (Q 3:78, emphasis added)

This textual corruption began in the past, and has continued over the centuries, ‘some of whom reached the era of our prophet Muḥammad’. Is this latter phrase suggesting that only some (though perhaps most) of the Jews corrupt scripture, whereas others (a minority) do not? Perhaps, perhaps not.

Al-Ṭabarī on Q 5:13 – oral tradition written down?

Nickel (141) notes Abdullah Saeed’s (2002, 425) statement that al-Ṭabarī seems in Q 5:13 to be referring to written down false interpretations being attributed to God. Though Saeed does not explicitly state so, he may be envisaging the Mishnah or Talmud. This is possible, though hardly evident, at least here.

But we do have evidence elsewhere that al-Ṭabarī knew of such a position, as in Q 2:79 we find (his own?) opinion that ‘those who tampered with (ḥarrafa) the kitāb of God from the Jews Banū Isrāʾīl, and wrote a book according to how they interpret it from their interpretations…‘ (Nickel’s translation, 154, emphasis added). He does also transmit traditions of the Jews ‘adding’ and ‘removing’ as they wished, and ‘erasing’ the description of Muhammad; so he is also aware of the perspective that the Torah itself is textually corrupted. But as for the concept of writing a separate book of interpretations, Nickel is not the only one to notice this perspective in al-Ṭabarī, and he (158) cites numerous other scholars who have made similar observations (Adang, Rippin, Goldziher, Hirschfeld, Watt and Lazarus-Yafeh). Nickel (158-159) provides other citations from al-Ṭabarī to make this point (Q 2:42, 75, 102; 3:71), which are beyond the scope of this blog post.

Despite this, Nickel (161) does still think al-Ṭabarī is making an accusation of textual corruption of the Torah itself in Q 5:13. But, as Nickel (141) reflects, it is not clear why al-Ṭabarī makes this accusation in Q 5:13 and not the other ḥarrafa verses. Nickel (162) does later wonder whether the scriptural context of ḥarrafa in Q 5:13 might be the reason; the word ‘treachery’ is given particular emphasis in his commentary, and Nickel wonders whether ‘in the context of a portrayal of the ultimate treachery of the Jews towards Muḥammad, Ṭabarī chose to offer an interpretation of ḥarrafa which emphasized the extremes to which faithless people might go?’

At the end of the day, we are not bound to follow al-Ṭabarī in seeing written corruption in Q 5:13, whether of Torah manuscripts themselves (as Nickel prefers) or the writing of books of interpretations (as seemingly Saeed). This is especially so as al-Ṭabarī does not elsewhere consider ḥarrafa to necessitate textual corruption, and it is not clear why he alleges such in Q 5:13 (assuming that he indeed does).

Q 5:41

Finally, we arrive at our last verse:

Messenger, do not be grieved by those who race to surpass one another in disbelief— those who say with their mouths, ‘We believe,’ but have no faith in their hearts, and the Jews who listen eagerly to lies and to those who have not even met you, who distort the meanings of [revealed] words [yuḥarrifūna l‑kalima min baʿdi mawāḍiʿihi] and say [to each other], ‘If you are given this ruling, accept it, but if you are not, then beware!’ — if God intends some people to be so misguided, you will be powerless against God on their behalf. These are the ones whose hearts God does not intend to cleanse— a disgrace for them in this world, and then a heavy punishment in the Hereafter— (Q 5:41)

Most naturally this would seem to be taking place during Muhammad’s own day; the accusation of taḥrīf is preceded and succeeded by present actions of Jewish perfidy and sayings. The verses before and after also deal with Muhammad’s contemporaries.

As mentioned, the taḥrīf accusation is sandwiched between two seperate actions of verbal deception, suggesting that yuḥarrifūna l‑kalima min baʿdi mawāḍiʿihi is also some non-textual misdeed.

Tafsīr

Nickel (2011, 82) notes that Muqātil devotes a comparatively large amount of space to Q 5:41. Much of it is the retelling of a lengthy narrative, similar to that found in Ibn Isḥāq’s Sīrat al-Nabī. For the phrase we are focusing on (yuḥarrifūna l‑kalima min baʿdi mawāḍiʿihi), Muqātil considers these ‘words’ to be referring to the command of stoning in the Torah (84). He explains ‘out of their places’ as ‘out of its declaration (bayān) in the Torah.’ (Nickel’s translation). Muqātil doesn’t provide a gloss for yuḥarrifūna, but dives into a story about an adulterous Jewish couple and the legal punishment (flogging or stoning) that might be applied to them. Ibn Ṣūriyā feels compelled by Muhammad to admit that stoning is indeed the penalty to be found in the Torah. Additionally, he goes on to acknowledge that the Jews recognise Muhammad as a true prophet, but that they reject him out of envy (85-86). There is nothing in the story about textual corruption, and, indeed, at least on the matters of stoning for adulterers and the prophethood of Muhammad the scripture seems to be textually preserved.

al-Ṭabarī also has a lengthy discussion on Q 5:41 (Nickel, 2011, 129). This includes a similar narrative to that found in Muqātil and in the Sīrah (129-130). Once again, Ibn Ṣūriyā admitS that stoning is the penalty in the Torah, and acknowledges that the Jews recognise the prophethood of Muhammad (130). al-Ṭabarī also narrates a different narrative, where a scholar at first denied, but then admits, that stoning is the penalty in the Torah (130-131). Although the scholar was well aware of this, he explained that due to the frequency of adultery amongst their elites, stoning was replaced with a lesser penalty. A third story is told to similar effect. al-Ṭabarī (132) prefers the story involving Ibn Ṣūriyā. Interestingly in commenting on the first part of Q 5:41, ‘do not be grieved by those who race to surpass one another in disbelief’, we are told that they recognise the prophethood of Muhammad, indeed ‘we know that for certain, through our discovery (wujūd) of your description (ṣifa) in our book.’

Commenting on our phrase in question, al-Ṭabarī explains:

These listeners to the lie tamper with (ḥarrafa) – Jewish listeners to other folk who do not come to you – al-kalim. Their taḥrīf was this: their changing (taghyīr) the judgement (ḥukm) of God, almighty his mention, which he sent down in the Torah concerning married women and married men (muḥṣina) of adultery by stoning, to flogging and blackening. So he said,… “they tamper with the words,” meaning: these Jews; and the meaning: the judgement (ḥukm) of the words (kalim). (Nickel’s translation, 134)

As was noted above, when al-Ṭabarī recounts stories explaining how this ‘judgement of God’ was changed, it is a matter of oral misrepresentation and concealing rather than of textual corruption.

Nickel (135) notes that al-Ṭabarī does give another explanation of this verse concerning blood money and retaliation rights amongst Jewish tribes. He also includes a saying attributed to Ibn Zayd that ‘they tamper with the words out of their places’ means ‘they do not impose (waḍaʿa) what God sent down.’ (135).

Tafsīral-Rāzī

As mentioned above, Bassam Zawadi draws out attention to the words of Fakhr al-Dīn al-Rāzī (d. 606/1210) in his commentary, and helpfully provides a translation:

In regards to Allah’s statement “They pervert the words from their proper places” (Surah 4:46), it means that they mention the corrupted interpretations for those verses, and there is no proof that they take the actual statements out of the book.

And as for the verse in Surah 5:41, this is evidence that they have combined between the two (textual corruption and misinterpreting the text), they used to mention their corrupted interpretations, and they also used to remove the statements from the book. Allah’s statement “They pervert words” indicates misinterpreting the text and Allah’s statement “after their being put in their right places” indicates that the statements were removed from the book. (Fakhar ad-Din ar-Razi, Tafsir Al Kabir, Commentary on Surah 4:46, Source)

Now I concede that al-Rāzī does seem here to be alleging textual corruption in Q 5:41, and we shall discuss why shortly. But the reader should be aware that al-Rāzī is somewhat complex on this issue. Elsewhere he discusses texts that are so geographically dispersed that they could not realistically be textually corrupted, and Nickel (2015, 52-53) points out that such is even attributed to the Torah:

[Commenting on whether Q 5:13 speaks of corruption of interpretation or of text] We have shown earlier that the first sense is preferable, because alteration of the word does not occur in a scripture transmitted without interruption (biʾl-tawātur). (Nickel’s translation, 52)

And more strikingly still, his comments on Q 6:91:

The Torah is a book which has reached the people of the West and the East, and the majority of the learned men (ahl al-ʿilm) know and have memorized it (ḥafiẓūhu). In such a book it is impossible to insert (idkhāl) additions (ziyāda) or omissions (nuqṣān). And the proof is that if anyone now wants to insert additions or omissions into the Qurʾān, he could not do it: the same goes for the Torah. (Nickel’s translation, 53)

How do we reconcile these statements with al-Rāzī’s comments about Q 5:41, made in his commentary on Q 4:46? (It is strange that these comments are not made in his commentary on Q 5:41 itself). Perhaps he has contradicted himself, or changed his mind, or perhaps concerning Q 5:41 he speaks only of limited corruption of the manuscripts in Medina at the time of Muhammad. Or perhaps some other suggestion.

Why is textual corruption (of some kind) proposed for Q 5:41, but not for the near identical Arabic phrase in Q 4:46 and Q 5:13? It would seem to be due to a finely grained difference in the Arabic:

Some Jews distort the meaning of [revealed] words [yuḥarrifūna l‑kalima ʿan mawāḍiʿihi]: … (Q 4:46, emphasis added)

…who distort the meanings of [revealed] words [yuḥarrifūna l‑kalima min baʿdi mawāḍiʿihi]… (Q 5:41, emphasis added)

I have placed in bold the difference between these otherwise identical phrases; one sentence uses the preposition ʿan and the other the prepositional phrase min baʿdi. Abdel Haleem clearly sees no great difference affecting the sense of the sentence, nor seemingly does al-Ṭabarī, according to Nickel’s (2011, 134) summary:

The exegete [al-Ṭabarī] also provides an explanation for the phrase, “from its places.” He writes that this means, “after God had put them into context.”[Footnote – ‘Burton’s translation in “The Corruption of the Scriptures,” 102. Literally “after God placed that its places.”] As for the words min baʿd in the phrase “from (min baʿd) its places,” this can be taken to mean “out of (ʿan) its places” – which is the wording in Q 4:46 and 5:13. Ṭabarī offers an example to support this reading. He writes that when anyone says, “I came to you out of (ʿan) my leisure from the activity,” they really intend, “after (baʿd) my leisure from the activity.”

Conclusion

In conclusion, we have argued that Q 4:46 is something of an exegetical key, unpacking in more detail than the other verses that some kind of oral or non-textual corruption is envisaged by our key phrase yuḥarrifūna l‑kalima min baʿdi mawāḍiʿihi. We noted that even certain Muslim translations (e.g. Abdel Haleem, Qarai and mostly, if not entirely, Muhammad Asad) tend to have this understanding, seeing the phrase as describing a corruption of the meaning of words.

When we turned to the tafsīr we saw that the word ḥarrafa mostly suggested to al-Ṭabarī non-textual forms of corruption, though for some reason Q 5:13 may be an exception (though Saeed, but not Nickel, thinks that al-Ṭabarī may refer to the writing down of oral traditions). Muqātil’s commentary is much briefer, and it is not always clear what he is alleging (Q 4:46, 5:13), though elsewhere the word seems not to evoke a charge of textual corruption (Q 2:75, 5:41). While one might have wondered whether the entire phrase yuḥarrifūna l‑kalima ʿan/min baʿdi mawāḍiʿihi (as in Q 4:46, 5:13) might be more suggestive of textual corruption than ḥarrafa/yuḥarrifūna without this phrase (Q 2:75), Q 5:41 also has the phrase yet evokes from Muqātil no such accusation.

Even if we did see textual corruption in these verses, they seem to be speaking about the Jews and not the Christians (and hence not the Gospel). It then becomes important to determine whether this textual corruption happened in the past or the present, which is not always easy to determine; Q 5:41 and Q 4:46 seem to refer to the present day, and Q 2:75 and 5:13 are ambiguous. If the corruption occurred before Muhammad’s time, then our entire manuscript tradition may be accused of corruption; if during Muhammad’s time, then no such accusation is made.

Bibliography

Asad, Muhammad. “The Message of the Quran.” http://www.muhammad-asad.com/Message-of-Quran.pdf

Boisliveau, Anne-Sylvie. “Qurʾānic Discourse on the Bible.” MIDÉO [Online] 33 (2018). https://journals.openedition.org/mideo/1819

Cooper, J. The Commentary on the Qurʾān by Abū Jaʿfar Muḥammad B. Jarīr Al-Ṭabarī. Oxford: Oxford University Press, 1987.

Guillaume, Alfred. The Life of Muhammad: A Translation of Isḥāq’s Sīrat Rasūl Allāh. Karachi: Oxford University Press, 1955.

Nasr, Seyyed Hossein, Caner K. Dagli, Maria Massi Dakake, Joseph Lumbard, and Mohammed Rustom. The Study Quran: A New Translation and Commentary. New York: HarperOne, 2015.

Nickel, Gordon. The Gentle Answer: To the Muslim Accusation of Falsification. Calgary: Bruton Gate, 2015.

Nickel, Gordon. Narratives of Tampering in the Earliest Commentaries on the Qurʾān. Leiden/Boston: Brill, 2011.

Reynolds, Gabriel Said. “On the Qurʾanic Accusation of Scriptural Falsification (Taḥrīf) and Christian Anti-Jewish Polemic.” Journal of the American Oriental Society 130, no. 2 (2010): 189-202.

Saeed, Abdullah. “The Charge of Distortion of Jewish and Christian Scriptures.” Muslim World 92, no. 3-4 (2002): 419-36.

Al-Tafsir.com – https://www.altafsir.com/

Zawadi, Bassam. “Evidence That Islam Teaches That There Was Textual Corruption of the Christian and Jewish Scriptures.” https://www.call-to-monotheism.com/evidence_that_islam_teaches_that_there_was_textual_corruption_of_the_christian_and_jewish_scriptures

Leave a Reply

Your email address will not be published. Required fields are marked *