IN FACT THE ISSUE is not how the words were pronounced, but how they were written. The reason for this is due to the time when the Arabic language began to be written down, when there was no hamza (the letter alif ) when used at the beginning of a word – this was only introduced by grammarians in the eighth century AD. In the case of words which we pronounce with alif in the middle of the word, the copyists replaced the yet-to-appear alif with the letter yā’. So they wrote majrīhā instead of majrāhā. 

If we look at the Samarkand Qur’ān, the oldest copy of the Qur’ān known to researchers, we find that they wrote ‘and for the men’ as wa-lil-rajīl instead of wa-lil-rijāl. Similarly, they wrote ‘heavens’ as samawāt instead of samāwāt, and ya-ma‘shar instead of yā-ma‘shar, and dhalika instead of dhālika and so on. 

When grammarians began to work out grammar and spelling rules in the eighth century AD – because they were all Muslims close in time to the beginning of Islam and they believed that the Qur’ān was the word of God – they made the rules of spelling Arabic coincide with the Qur’ānic spelling. They thus argued that, 

The hamzat al-waṣl (the connecting hamza)^[2] is omitted from the second alif and replaced by the extended alif sign آ ‘ā‘ if there is an interrogative hamza such as:  أأنت رجل أم امرأة   a-anta rajul am imra’a (‘are you a man or a woman’)? And is written: آنت رجل أم امرأة   ānta rajul am imra’a. The same goes for أاية   or  ءاية    which is written as: آية , or ءادم  which is written as: آدم and so on. 

But the Muslim fuqahā’ who considered the Qur’ān to be God’s words brought down to earth, as if it were the direct dictation from God Himself, preserved the older spelling in the Qur’ān. And thus we still find  ءايات , ءالاء  and  ءادم .

The grammarians made the rules of spelling Arabic coincide with the Qur’ānic spelling

The hamzat al-waṣl is also omitted after the vocative yā’ and is written: يابن قومي تعال نتصالح  yābn qawmī ta‘āla nataṣālaḥ (‘O son of my people, come let us reconcile’) instead of  يا ابن قومي yā bn qawmī … and in the same way the hamza is omitted from the word اسم ism in the full basmala formula ‘In the name of Allah the gracious, the merciful’ which we write as: بسم الله الرحمن الرحيم bismi llāhi r-raḥmāni r-raḥīm. But it is not omitted when only part of the basmala formula is pronounced. Here we write it as: باسم الله مجراها ومرساها  bi-ismi. The hamza is also omitted when it follows the letter ل  lām governing the genitive case, such as: للقمر وجه ناصع  lil-qamar wajh naṣi‘ (the moon has a shining face’) instead of ل القمر وجه ناصع  li ‘l-qamar …

Mr. Hisham goes on to argue that:

when we say and yā-ma‘shar we find it in the Qur’ān as ya-ma‘shar with a small alif paced between the letters yā’ and mīm. We also find the Qur’ān writing ملائكة  malā‘ika (‘angels’) in a different spelling: ملئكة mala’ika, this time by placing the hamza below the ‘seat’ and not above it. Similarly we write مثواكم mathwākum (‘your resting place’); the original word is مثوى mathwā (without an alif) and the letter wāw here is part of the stem while the alif is a ‘soft’ letter which can be added to it. In the Qur’ān, however, it remains as it is and the ‘seat’ is written with a small alif written over it. But Arabic grammar does not generally do this, since the ‘seat’ is supplied for a dotted letter such as nūn, bā’, tā’ or yā’, and there is no ‘seat’ for madd (extending of the vowel sound).

The reason for the Qur’ān’s writing ya-ma‘shar instead of yā-ma‘shar  is that the scribes at the time did not know of the hamzated alif since this was only introduced into the orthography in the eighth century AD. At the time that the Qur’ān that we now have was written – about two to three centuries after the death of the Muḥammad – they placed a small alif above the yā’ so as not to effect a change in the Qur’ān’s spelling. The same applies to the word ‘angel’ in the Qur’ān, written as ملئكة mala’ika instead of ملائكة  malā‘ika, and to the word كتب kitab (‘book’) instead of  كتاب kitāb, and to سموات samawāt (‘heavens’) instead of سماوات samāwāt.

Some of the scribes may have had a knowledge of Syriac, or pronounced these words in Arabic the same way as Syriac speakers pronounced them

As for the reason for spelling  حيوة ḥyū(h) instead of حياة ḥayā(h) for the word ‘life’, or زكوة  zakū(h) instead of زكاة zakā(h) ‘zakat’, it is because some of the early scribes may have had a knowledge of Syriac, or pronounced these words in Arabic the same way as Syriac speakers pronounced them. The spelling of the words حيوة ḥyū(t) and زكوة  zakū(t) is identical to the way these words are spelled in Syriac: ܚܝܘܬ  (ḫyūṯ) ‘life’ and  ܙܟܘܬ‎ (zakūṯ); ‘zakat’.

Mr. Hisham states that:

There are many examples one can adduce where the Qur’ānic  spelling differs from the customary spelling.  So why are there such differences, and is there a sacred spelling that cannot be altered or substituted, of is this spelling intentional?  If the Qur’ānic spelling is sacred, why was the Qur’ān dotted according to the rules of punctuation set down in the eras when language and orthography was stabilised? And why were the orthographical rules not applied accordingly as well? ^[3]

And this is indeed something confusing. If they believed that the Qur’ān was the word of God and that God had preserved it from any alteration or change, why then did they insert dots around the letters while at the same time preserving the old spelling? I think the reason for this was their fear of committing errors when voicing the Qur’ān if it were not dotted.

For example, one might easily read the phrase lā rayba fīhi (‘there is no doubt in that’) as lā zayta fīhi (‘there is no oil in that’) if the letters were not dotted, as did the reciter Ḥamza, who was henceforth dubbed ‘Ḥamza the oilmaker’ after he made that very mistake. 

Mr. Hisham then goes on to ask: 

Are we able to suggest that a Qur’ānic word such as رحمة raḥma(h) (‘mercy’) is misspelled,  given that in some places it is spelled as رحمت raḥmat and in other places as رحمة raḥma(h)? Can we declare the spelling of the Qur’ānic term كلمة kalima (‘word’) to be erroneous, given that it is spelled  كلمت  in the phrase وَتَمَّتْ كَلِمَتُ رَبِّكَ صِدْقًا وَعَدْلًا And the word of your Lord has been accomplished truly and justly ?^[4] 

Moreover, to what extent have we made an orthographic error if we try to imitate the Qur’ānic spelling in our everyday writing, particularly given the fact that we write: بسم الله  bismi llāhi in the way the Qur’ān spells it as opposed to: باسم الله   bi-ismi llāhi which would be more correct according to the rules of orthography? We similarly write رحمن  raḥman in the way the Qur’ān spells it and not رحمان raḥmān which would be the more correct orthography.

If the rules on spelling dictate that one must place the letter alif after the letter wāw indicating the 3^rd person plural of the verb as when we write باءوا  bā’ū (‘they drew/incurred’), why do Qur’ān codices have باءو without the added alif?

As for the ‘open’ tā’ (ت) and the ‘bound’ tā’ (the tā’ marbūṭa – ة) those who copied the Qur’ān got themselves into a great muddle because they were not fluent in the Syriac language. Christophe Luxenberg argues that:

 ‘The Garden of Eden’ in Syriac is always singular and not plural, and is written with an ‘open’ tā’ – the letter ܬ (‘th’) – as  ܓܢܬ ܥܕܝܢ ganath ‘edein ^[5]

Now since the early scribes did not use the letter alif (the hamza) in the middle of words, they considered that the form جنت  j-n-t  was a sound feminine plural (i.e. janāt), and therefore its singular must be جنة . They kept to this form in many verses and wrote words sometimes with the ‘open’ tā’ (ت) and sometimes with the tā’ marbūṭa (ة). The non-use of the alif in the middle of the word led to the spelling رحمن  raḥman instead of رحمان raḥmān. Likewise, for the letter alif placed after the letter wāw for the 3^rd person plural of the verb, they omitted to maintain this because they did not have any rules of orthography at that time to guide them.

The copyists’ ignorance of Aramaic and Syriac made them miswrite, and misunderstand, many words

The copyists’ ignorance of Aramaic and Syriac made them miswrite, and misunderstand, many words. For example, in Aramaic the word מלאכה malãkah (‘angel’) with an h at the end is the source of the word ملاكه . However, the Arabs reproduced it as if it were a sound feminine plural and wrote it as ملائكة with a tā’ marbūṭa (ة) on the end. Thus, they took ملائكة to be female and the Qur’ān goes on to reproach the mushrikīn for saying that the angels were ‘daughters’. The Lisān al-‘Arab dictionary derives the word ملاك malāk from لاكَ lāk:

and its plural is ملائكة malā‘ika which they pluralised as malā‘ik, adding the letter hā’ as a feminine ending.

Suggested Reading

When did Islam emerge? – 1

In some cases we find that Muḥammad has preserved Syriac forms of numerals, forms which are inconsistent with the rules of Arabic grammar. For example, the verse that talks about the children of Israel: And We divided them into twelve tribes^[6]renders the number ‘twelve’ as  اثنتي عشرة   ithnatay ‘ashrata whereas the correct form in Arabic should be اثني عشر  ithnay ‘ashra. The former rendering would be correct in Syriac,^[7] which the Islamic commentators were of course unaware of.

Mr. Hisham then concludes that:

anyone who looks at the history of what is termed the ‘sciences of the Qur’ān’ finds that they came about to fill a need created by the conditions of expansion and the entry of non-Arabs into Islam. There was thus the difficulty of their receiving the Qur’ān and benefiting from it as easily as the Arabs could. Hence the first intimations of the so-called ‘linguistic miracles’ in the Qur’ān.  

Gone was the era when Muslims could take pride in what they thought was an eloquent Qur’ān, an eloquence which touched on the miraculous; now the era was one whereby they had to put aside the eloquence of the Qur’ānic text and focus on its interpretation and translation for non-Arabs who did not have the linguistic faculty to enable them to perceive these rhetorical and linguistic miracles in the Qur’ān.

More than that, the spread of Arabic grammatical errors in general, and in the Qur’ān in particular, was the primary motivation for developing the basic foundations of grammar, and it is for this reason that Abū Al-Aswad al-Du’alī and his student: Yaḥyā ibn Ya‘mar al-‘Adwānī proceeded to develop the basic rules of syntax and grammar that were familiar to the Arabs but not to the non-Arabs.  

I do not think that Mr. Hisham has done justice to the non-Arabs here. For it was the non-Arabs such as Abū Al-Aswad al-Du’alī, al-Farāhīdī, Sibawayh and others who developed the rules of the Arabic language. They were the ones who understood the Qur’ān for what it really was and began to criticize it on the basis of reason. Most of those tried for heresy were non-Arabs.

^[1] Qur’ān XI (Hūd), 41.

^[2] The connecting hamza is an ‘extra’ hamza at the beginning of the word, pronounced when starting a word, dropped when continuing from a previous word or a voweled letter. (Ed.)

[3] There are also interesting examples of the same word or expression being spelled in different ways within one and the same Qur’ānic text. See, for instance: قرءان [XXXIX,28] / قرءنا [XII,2]; بسم [I,1] / باسم [XCVI,1]; إبراهيم / إبرهم [II,124; I,125; II,260]; ابن أم VII,50] / يبنؤم [XX,94]; أريكم [XL,29] / أوريكم [VII,145; XXI,37]; عما [II,74; V,73] / عن ما [VII,166]; ميعاد [III,9; III,194; XIII,31; XXXIV,30; XXXIX,20] / ميعد [VIII,42]; الرسولَ [II,143] / الرسولا [XXXIII,66]; طغى [XX,24; XX,43; XX,45; LIII,17; LIII,52; LXXIX,17; LXXIX,37;XCVI,6] / طغا [LXIX,11]; ساحر LI,52] / سحر [XL,24]; سعوا [XXII,51] / سعو [XXXIV,5]; شجرة [XX,120; XXXVII,62; XXXVII,146] / شجرت [XLIV,43]; المهتدى [VII,178] / المهتدِ [XVII,97]; شركاء [LXVIII,41] / شركؤا [VI,94; XLII,21]; داخرين [XL,60] / دخرين [XXVII,87]; سراجا [LXXVIII,13] / سرجا [XXV,61]; الغمام [II,57; II,210] / الغمم [VII,160; XXV,25]; لدى [XII,25] / لدا [XL,18]; سيماهم [XLVII,29] / سيمهم [II,273 ; VII,46; Vii,48; XLVII,30; LV,41]; ربا [XXX,39] / ربوا [III,130]; شيء [XVI,14] / شاىء [XVIII,23]; وراء  [XXXIII,53] / ورائ [XLII,51]; أيكة [XV,78; L,14] / ئيكة [XXXVIII,13; XXVI,176]; أيها [LI,31] / أيه [XXIV,31]. (Ed.)

^[4] Qur’ān VI (al-An‘ām) 115.

^[5] C. Luxenberg, The Syro-Aramaic Reading of the Koran : A contribution to the decoding of the language of the Koran, p.47. For this work see the Almuslih library, listed under the rubric: The Qur’ānic Vorlage thesis. 

^[6] Qur’ān VII (al-A‘rāf), 160.

^[7] C. Luxenberg, The Syro-Aramaic Reading of the Koran, pp. 49-51. Luxenberg notes that the final letter hā’ in the Arabic ملئكه was reproducing the final Syriac letter ܐ in ܡܠܐܟܐ malāḵē ‘angels’, with the Syriac masculine plural ending. (Ed.)

Main image: إِنَّا نَحْنُ نَزَّلْنَا الذِّكْرَ وَإِنَّا لَهُ لَحَافِظُونَ Surely We have revealed the Reminder and We will most surely be its guardian [Qur’an XV (al-Hijr) 9], from an illuminated Qur’ān commissioned by the Mamluk sultan Rukn al-Dīn Baybars (1223-1277). British Library Add MS 22409 – folio f.9r.

See Part One of this essay here