Dialogue variants in the canonical Qirāʾāt readings of the Qurʼān
Previous article: A compilation of expert Twitter threads and tweets on the Qirāʾāt readings and textual transmission of the Qurʼān
The Qurʼān contains narratives in which various prophets, angels and other characters are quoted as saying certain things on specific historical occasions. As elsewhere in the Qurʼān, variant readings in these passages are common.
A unique feature of dialogue variants is that when they differ substantially in meaning, they cannot be reconciled by the usual (and often strained) reasoning that both readings are valid. While they should not be read as verbatim accounts, a historical story is being presented, and should internally be free from contradiction. Within a particular story dialogue passage, the character either said one thing or another at a certain point, not both. It seems likely that such variation would be unintentional in the most glaring cases, or even more likely, they were changes that arose during transmission.
In this article I present a dozen examples of conflicting dialogue variants found in the seven readings canonised by Ibn Mujāhid. In the final example, the variant angelic instructions moreover have a logical contradiction. In each case I provide a link to Corpus Coranicum (including a screenshot for the first example). In most cases they are also discussed in Tafsīr al-Jalālayn, to which links are also provided.
Stories of Moses (6 examples)
17:102 — In al-Kisāʾī’s reading, Moses said to Pharaoh, “I have known”, but in the other 6 readings he says “You have known”.
As is obvious in this and the other examples below, the fact that ancient language dialogue is presented in Arabic should not distract us. All the relevant ancient languages had pronouns.
7:113 Nāfiʿ, Ibn Kaṯīr and the transmission of Ḥafṣ from ʿĀṣim read that the magicians said to Pharaoh ʾinna lanā la-ʾaǧran (Indeed, for us is a reward), while the others read that they instead asked it as a question ʾa-ʾinna (or ʾā-ʾinna) lanā la-ʾaǧran (Is there indeed for us a reward…?). Pharaoh’s response in the next verse is the same in either reading (He said, “Yes, and, [moreover], you will be among those made near [to me]”).
Corpus Coranicum Tafsīr al-Jalālayn
A similar variant occurs in 10:81 where according to some readings Moses says to the magicians when they perfom their snake trick mā ǧiʾtum bihi s-siḥru (What you have brought is sorcery), whereas Abū ʿAmr read it as a question mā ǧiʾtum bihi ʾās-siḥru (What have you brought? Sorcery?). As well as the change at the end, mā here functions as an interrogative.
Corpus Coranicum Tafsīr al-Jalālayn
10:90 In the related Kufan readings of Ḥamza and al-Kisāʾī, at the point of drowning Pharaoh said in two sentences ʾāmantu ʾinnahū (I believe. Indeed…), whereas the others read that he said ʾāmantu ʾannahū (I believe that…). Confusion between ʾinna (indeed) and ʾanna (that) are quite common in the canonical readings.
Corpus Coranicum Tafsīr al-Jalālayn
20:96 Ḥamza and al-Kisāʾī read that Samiri said to Moses, lam tabṣurū (you did not perceive) whereas the others read that he said lam yabṣurū (they did not perceive). This is a consonantal dotting difference in the text.
Corpus Coranicum Tafsīr al-Jalālayn
20:97 Ibn Kaṯīr and Abū ʿAmr read that Moses replied to Samiri lan tuḫlifahū (never will you break it) whereas the others read him saying lan tuḫlafahū (never will you be broken it) i.e. active vs passive.
Corpus Coranicum Tafsīr al-Jalālayn
40.26 The canonical readers give a total of four permutations of what Pharaoh said here. They differ both on whether he said ʾau ʾan (or that) or wa-ʾan (and that) and on whether he said yuẓhira fi l-ʾarḍi l-fasāda (he will cause corruption in the land) or yaẓhara fi l-ʾarḍi l-fasādu (corruption may appear in the land).
Corpus Coranicum Tafsīr al-Jalālayn
Surah Yusuf (4 examples)
This surah contains a lot of dialogue in its story of Joseph.
12:12 Here we have two variants of what Joseph’s brothers said to their father. The three Kufans read yartaʿ wa-yalʿab (he way eat well and play), while the others read nartaʿ wa-nalʿab (we may eat well and play).
Corpus Coranicum Tafsīr al-Jalālayn
12:19 ʿĀṣim reads here that the man said yā-bušrā (Good news!), whereas the others read yā-bušrāya (My good news!) as a form of address.
Corpus Coranicum Tafsīr al-Jalālayn
12:49 Here Joseph tells the King his dream interpretation. Ḥamza and al-Kisāʾī read that he said taʿṣirūna (you will press), whereas the others read that he said yaʿṣirūna (they will press).
This kind of 2nd / 3rd person plural variation is common in the Qurʼān, but obviously does not work within the same dialogue account.
Corpus Coranicum (no comment on the variant in Tafsīr al-Jalālayn)
12:63 Ḥamza and al-Kisāʾī read that Joseph’s brothers said to their father yaktal (he will be given measure), whereas the others read naktal (we will be given measure).
Corpus Coranicum Tafsīr al-Jalālayn
Stories of Mary and Lot (2 examples)
19:19 This is a well known variant. The Angel Gabriel says to Mary li-ʾahaba laki ġulāman (that I may give you a boy) according to most readers, whereas Abū ʿAmr and the transmission of Warš (Warsh) from Nāfiʿ read that he said li-yahaba laki ġulāman (that he may give you a boy).
Interestingly, this variant is one of the few dozen that transgress the Uthmanic rasm standard (even its reported regional variants) and was discussed at the end of my previous article where we see the ya written in red ink on a manuscript. In the Ṣanʽā’ 1 palimpsest lower text we see a 3rd variant, li-nahaba (that we may give) on line 15 of Folio 22 B (see p.64 here).
Corpus Coranicum (no comment on the variant in Tafsīr al-Jalālayn)
11:81 Lot is told two conflicting instructions by the angels depending on the reading (see the Tafsīr al-Jalālayn link below). Ibn Kaṯīr and Abū ʿAmr read ʾilla mraʾatuka (except your wife) in the nominative case, meaning that Lot must set out with his whole family but not let anyone except his wife look back. The others read ʾilla mraʾataka in the accusative case, meaning that Lot must set out with his family except his wife, leaving her behind. These understandings of each variant are found in early scholars.
Later, when there was a shift towards treating every variant as divine, some held that even in the accusative reading the exception refers to none looking back (which would defy a grammatical rule followed in the Quran, whereby an exception must be nominative if the thing from which it is excepted is in the nominative position of a negative phrase, would make the variant pointless, and has various other problems). See this excellent thread on the variant by Dr Marijn van Putten.
Corpus Coranicum Tafsīr al-Jalālayn
What can we learn from this?
These variations in quoted dialogue betray their unintended origin. Each variant was intended by someone, but clearly they were not intended to vary in these ways from each other. While sometimes the Qurʼān retells a story afresh in another passage with different dialogue, it makes no sense for anyone to change an already existing passage in these ways. A particular telling of a story should be free of internal contradiction, yet in most cases one word or the pronoun for a verb in a quoted sentence is fundamentally changed, giving conflicting accounts of the dialogue meaning at that specific moment.
The next question that arises is whether these unintended variations were introduced unwittingly by the Prophet or in later transmission. An analogy from evolution in nature might hint at the most likely answer. Variations in species tend to thrive when they are separated into different habitats, such as islands, but more likely snuffed out when they arise within the original environment.
Next article: Superfluous variants in the readings of the Qurʼān
