reader was evidently transmitting his own reading [rather than one transmitted to him from the Prophet].
Fourth, the fact that each of these [ten readers], as well as their followers, pro‌ duced arguments to prove the authenticity of their respective readings, and their rejec‌ tion of readings [other than] their own is an absolute proof that the readings were based on the personal judgment of the reader and the opinions of those who followed him. Otherwise, had the readings been received without interruption from the Prophet (peace be upon him and his progeny), there would have been no need for proof and arguments confirming their validity.
Fifth, the rejection of a number of readings by a number of prominent scholars is itself a clear proof of the lack of continuity in the transmission of the readings, for, otherwise, such a rejection would be incorrect. Ibn Jarir al-Tabar!,for example, re‌ jects the reading of Ibn Amir and contests, in numerous places, some other readings among the seven. Others have challenged the readings of f:lamza, Abu Amr, and Ibn Kathir. The scholars have generally agreed on denying the status of uninterrupted transmission to any reading which is not supported by one of the accepted levels of Arabic rhetoric, and have determined that some readers did indeed commit errors. ١
We have already mentioned in the biography off:lamza that his reading was rejected by the founder of the f:lanbalite school of jurisprudence, Al)mad b. f:lanbal, as well as by Yazid b. Harlin, Ibn Muhdi,٢ Abu Bakr b. Ayyash, and Ibn Durayd.
AI-Zarkashi, having adopted the view that the [accepted] readings are not open to choice, disagreed with a number of scholars, including al-Zamakhshari, who held that readings are [in fact] open to choice, and that this choice depends on the opinion of the masters of language and on the independent reasoning of the rhetoricians. AI‌ Zarkashi refutes the choice of f:lamza where he read the word ١١•a al-arbama ["the wombs," in verse ٤: I] with the declension i[i.e., al-arbami]. and then goes on to say:

Similarly, it has been reported that Abu Zayd, al-A maI,
l:lamza for reading ١m nul antum bi-musrikhiyya ["nor can you help me," in verse ١٤:٢٢] with the genitive declension I over the stressed ending [i.e., musrikhiyyl]. They also criticized Abu Amr for contracting the letter ra into the letter lam in yaghfir lakum ["will forgive you," a recurrent phrase in the Quran: i.e., he read it as yaghfir-rakum. instead of yaghflr lakum]. "This," al-Zajjaj said, "is an atrocious error."٣

Statements by the Schoiars Who Deny the Uninten-upted
T ransmiss١on of the Readings

To make this matter as clear as possible, it is appropriate to mention here [ten] ex‌ cerpts of what has been said by authorities in this field who deny that the readings were transmitted without interruption.

١. Ibn al-Jazari says:

Any reading that accords with one of the levels (wajh) of Arabic grammar, ٤ and com‌ plies with one of the Uthmanic codices, ٥ even if this were only a supposition,٦ and is supported by a sound chain of transmission, then it should be considered a sound read-