The claim: "The Qur'an presents itself as clear Arabic and says no one can produce its like. Yet the text contains iʿrāb (case-ending) errors: in 5:69 al-ṣābiʾūna is nominative, while the same word in the same construction is accusative in 2:62 and 22:17. In 2:177 a list running in the nominative suddenly turns accusative at al-ṣābirīna. In 20:63 we read in hādhāni, where after inna the dual should be hādhayni. In 4:162 everything around it is nominative, yet al-muqīmīna is accusative. These are grammatical mistakes, and they should not appear in a text that implies its own flawlessness."
What does the Qur'an say?
What is disputed is not meaning but the case ending at the end of a word. So the Arabic lines below are the evidence; the rendering is only there to show context. Most of these endings are invisible in the consonantal skeleton of Arabic script — they travel with recitation. That detail sits at the centre of the argument.
First site — al-ṣābiʾūna (5:69). The third item of the list ends in -ūna:
إِنَّ ٱلَّذِينَ ءَامَنُوا۟ وَٱلَّذِينَ هَادُوا۟ وَٱلصَّـٰبِـُٔونَ وَٱلنَّصَـٰرَىٰ
"Indeed, those who have believed [in Prophet Muḥammad (ﷺ)] and those [before him (ﷺ)] who were Jews or Sabeans or Christians - those [among them] who believed in Allāh and the Last Day and did righteousness - no fear will there be concerning them, nor will they grieve." (5:69)
The same grouping occurs twice more, and there the same word ends in -īna:
إِنَّ ٱلَّذِينَ ءَامَنُوا۟ وَٱلَّذِينَ هَادُوا۟ وَٱلنَّصَـٰرَىٰ وَٱلصَّـٰبِـِٔينَ
"Indeed, those who believed and those who were Jews or Christians or Sabeans [before Prophet Muḥammad (ﷺ)] - those [among them] who believed in Allāh and the Last Day and did righteousness - will have their reward with their Lord, and no fear will there be concerning them, nor will they grieve." (2:62)
وَٱلَّذِينَ هَادُوا۟ وَٱلصَّـٰبِـِٔينَ وَٱلنَّصَـٰرَىٰ وَٱلْمَجُوسَ
"Indeed, those who have believed and those who were Jews and the Sabeans and the Christians and the Magians and those who associated with Allāh - Allāh will judge between them on the Day of Resurrection. Indeed Allāh is, over all things, Witness." (22:17)
A second difference is easy to miss: in 5:69 the order of the list also changes — the Sabeans come before the Christians, whereas in 2:62 they come after.
Second site — al-ṣābirīna (2:177). At the close of a long "true righteousness" list, "those who fulfil their promise" ends in -ūna, and the phrase immediately after it ends in -īna:
وَٱلْمُوفُونَ بِعَهْدِهِمْ إِذَا عَـٰهَدُوا۟ ۖ وَٱلصَّـٰبِرِينَ فِى ٱلْبَأْسَآءِ وَٱلضَّرَّآءِ وَحِينَ ٱلْبَأْسِ
"[those who] fulfill their promise when they promise; and [those who] are patient in poverty and hardship and during battle." (2:177)
Third site — in hādhāni (20:63). The dual demonstrative appears as -āni after the particle:
قَالُوٓا۟ إِنْ هَـٰذَٰنِ لَسَـٰحِرَٰنِ يُرِيدَانِ
"Indeed, these are two magicians who want to drive you out of your land with their magic and do away with your most exemplary way" (20:63)
Fourth site — al-muqīmīna (4:162). "Those who establish prayer" ends in -īna, while "those who give zakāh" right beside it ends in -ūna:
وَٱلْمُقِيمِينَ ٱلصَّلَوٰةَ ۚ وَٱلْمُؤْتُونَ ٱلزَّكَوٰةَ
"But those firm in knowledge among them and the believers believe in what has been revealed to you, [O Muḥammad], and what was revealed before you. And the establishers of prayer [especially] and the givers of zakāh and the believers in Allāh and the Last Day - those We will give a great reward." (4:162)
Next to these four, the text's own statements about its language belong:
إِنَّآ أَنزَلْنَـٰهُ قُرْءَٰنًا عَرَبِيًّا لَّعَلَّكُمْ تَعْقِلُونَ
"Indeed, We have sent it down as an Arabic Qur’ān that you might understand." (12:2)
وَهَـٰذَا لِسَانٌ عَرَبِىٌّ مُّبِينٌ
"The tongue of the one they refer to is foreign, and this [recitation, i.e., Qur’ān] is [in] a clear Arabic language." (16:103)
بِلِسَانٍ عَرَبِىٍّ مُّبِينٍ
"In a clear Arabic language." (26:195)
وَلَوْ جَعَلْنَـٰهُ قُرْءَانًا أَعْجَمِيًّا لَّقَالُوا۟ لَوْلَا فُصِّلَتْ ءَايَـٰتُهُۥٓ
"And if We had made it a foreign [i.e., non-Arabic] Qur’ān, they would have said" (41:44)
وَمَآ أَرْسَلْنَا مِن رَّسُولٍ إِلَّا بِلِسَانِ قَوْمِهِۦ لِيُبَيِّنَ لَهُمْ
"And We did not send any messenger except [speaking] in the language of his people to state clearly for them" (14:4)
And the challenge the text issues:
فَأْتُوا۟ بِسُورَةٍ مِّن مِّثْلِهِۦ
"then produce a sūrah the like thereof and call upon your witnesses [i.e., supporters] other than Allāh, if you should be truthful." (2:23)
لَا يَأْتُونَ بِمِثْلِهِۦ
"If mankind and the jinn gathered in order to produce the like of this Qur’ān, they could not produce the like of it" (17:88)
What do we learn?
(interpretation) First observation: the four cases are not scattered at random. Three of them belong to a single structure — the long enumerative sentence (5:69, 2:177, 4:162); the fourth is a dual demonstrative (20:63). If one assumes an author who did not know Arabic, the fact that the lapses cluster in exactly two environments while thousands of other endings behave normally still needs explaining.
(interpretation) Second observation: Arabic script is consonantal. In 20:63 nothing in the skeleton shows whether the particle is inna or the light in; only vocalisation decides, and vocalisation arrives with recitation. So in that case what is under dispute is not strictly the written text but a reading applied to it. The same is not true of 5:69, where the difference is written into the consonants (wāw versus yāʾ).
(interpretation) Third observation: the rule books of Arabic grammar were compiled after the Qur'an, and the Qur'an was one of the main corpora from which the rules were drawn. Saying "the text breaks the rule" is therefore chronologically odd. But that does not settle the matter: the rules were also drawn from a wide pool of poetry and speech, and comparison with that pool is legitimate.
(interpretation) Fourth observation: the text's own linguistic claim is not "flawless iʿrāb" but intelligibility: "an Arabic Qur'an" (12:2), "clear Arabic" (16:103, 26:195), "in the language of his people" (14:4). 41:44 makes the frame explicit — the issue is whether the language can be understood. The objection may be imposing on the text a criterion the text never set for itself: exact conformity to one school's manual of case endings.
(interpretation) Fifth observation: the challenge (2:23, 17:88) was addressed to the first audience — native speakers, many of them hostile. The Qur'an contains no verse answering a charge of bad grammar. It does not follow that no such charge was ever made: that is an argument from silence, and it counts as evidence for neither side.
Different readings
(interpretation) The linguistic defence. The Arabic grammatical tradition supplies a named construction for each site. For 5:69 there is qaṭʿ — deliberately cutting an element out of the governing chain and making it the subject of an independent clause: "and the Sabeans likewise". The change in list order is read as a supporting signal of that break. A second account for the same site is coordination onto the maḥall, the position the noun would have held before the particle entered. For 2:177 and 4:162 the proposal is the accusative of praise or specification: shifting the ending to single an item out — "and especially those who are steadfast", "and above all those who uphold the prayer". For 20:63 there are two routes: read the particle as the light in, in which case the lām is distinguishing and the sense becomes "these are nothing but two magicians" — and then the nominative dual is required; or recall that some Arabian dialects keep the dual invariable in -āni in every case.
(interpretation) The critical reading. These are rescue devices produced after the fact. If every deviation is given a name, no deviation can ever count as an error and the rule loses its falsifiability. The dialect account also sits uneasily with the common claim that the text came down largely in one dialect: one cannot call it that dialect's highest specimen and also say this passage is another dialect.
(interpretation) The language-history reading. This one keeps its distance from both sides: how settled and uniform the case system actually was in the text's own period is itself an open question. It is argued that written standard and spoken Arabic diverged, that the invariable dual stabilised early in some regions, and that case drift inside enumerations also appears in poetry. On this reading both "error" and "rhetorical device" are too sharp; what we may be seeing is variation within a period.
(interpretation) The transmission reading. The text travelled as script and as recitation together. That 20:63 is transmitted with more than one reading is, on this view, a property of transmission rather than a corruption — and the question turns from "is it wrong?" into "which reading?".
An honest limit
Certain in the text: at these four sites the ending departs from the surrounding pattern. That is an observation, not a controversy. Also certain: the text's own linguistic claim is clarity and the language of the people (12:2, 16:103, 26:195, 14:4).
Left to interpretation: whether that departure is deliberate style or mistake. The Qur'an does not answer this itself. The answer depends on which grammatical framework is taken as the yardstick, which recitation is treated as primary, and what one assumes about the language of the period. These are linguistic judgements, not theological ones.
What this article is not: a verdict for one side. Whoever says "certainly an error" and whoever says "certainly a miracle" both claim more than the evidence carries.
Conclusion: The examples are real; at those four sites the text departs from the expected ending. But departure and error are not the same thing: each site can be matched to a named Arabic construction attested elsewhere, and the cases cluster in two narrow structures rather than scattering. On the other side, naming those constructions after the fact risks making the defence unfalsifiable. The honest residue is this: it is a linguistic dispute, not a verdict between faith and denial — and on the evidence available, neither side has grounds to declare a decisive win.
Related articles
- Are there contradictions in the Qur'an?
- Why was the Qur'an revealed in Arabic?
- The limits of reading a translation
- Was the Qur'an preserved?
Source: Qur'anic verses (M. Okuyan's Turkish rendering; Saheeh International in English). Presented with a text/interpretation distinction; not a legal ruling.