Old Hijazi

It’s been taken for granted by both academics and Muslim scholars that Classical Arabic was the language of the Quran and Arabs before Islam and during the first three centuries of Islam. This is reflected in the way Muslims recite the Quran and Hadith today. It’s even reflected in all Arabic historical movies and TV works depicting the early centuries of Islam such as the movie “Al-Risālah” about the life of Muhammad where all the actors are shown speaking in Classical Arabic with its two main hallmarks: full case inflection (final short vowels and nunation or Tanwīn) and full use of the Hamzah (glottal stop). This belief is reinforced by the popular claim by Muslim scholars that the readings tradition of the Quran, which the Quran is recited based on, were transmitted to us verbatim from the mouth of Muhammad. And all of these readings employ a full case system and heavily use the Hamzah. But recent academic research pioneered by Ahmad Al-Jallad of Ohio University and Marijn Van Putten of the University of Leiden, shows that the Quran was actually composed in a different language, which they call Old Hijazi, the ancient vernacular dialect of the Hijaz region which includes Makkah and Medina.

Old Hijazi differs markedly in pronunciation and grammar from the later classical Arabic that is imposed upon the Quran. This imposition led to the mismatch between the pronunciation and the text, which means the Quran was originally written phonetically in Old Hijazi. Old Hijazi sounds more like most modern Arabic vernacular as most modern Arabic dialects are completely devoid of the case system and devoid of the Hamzah to varying degrees.

The characteristics of Old Hijazi have been revealed by (1) early Arabic texts written in Greek and Hebrew letters which showed what the early defective Arabic script couldn’t, (2) the investigation into the Quranic Consonantal Text (QCT) which is the underlying consonantal skeleton (in Arabic, rasm رسم) of the Qur'an that originally lacked dots and other signs.

Main characteristics of Old Hijazi

The first five characteristics are proven by (a) the early Arabic scripts written in Greek and Hebrew letters, (b) the linguistic analysis of the Quranic consonantal text (QCT) which is the original text of the Quran that’s devoid of dots and signs that were added later.

1- Lack of nunation and final short vowels except in construct[1].


هذا كتاب جديد (This is a new book)

Classical Arabic pronunciation: hādhā kitābun jadīd

The word kitāb (book) is the subject of the sentence so it took a final ‘u’ short vowel: kitābu. The word is also indefinite so it also took nunation: kitābun.

The word jadīd (new) should also take a final “un” but since that the word is in a pausal position (last word in the sentence) it remains in its original form without any suffix.

In Old Hijazi, the word kitāb remains in its original form: hādhā kitāb jadīd.

The only case where final short vowels are retained in Old Hijazi is in construct, for example:

هذا كتاب محمد (This is Muhammad’s book)

Old Hijazi and Classical Arabic pronunciation: hādhā kitābu Muḥammad.

since that the word kitāb (book) is the subject of the sentence, it takes the ‘u’ final short vowel. The two words “kitābu Muḥammad” are in construct (book of Muhammad). The possessed noun retains the final short vowel in Old Hijazi.

Note: Old Hijazi retains case inflection in the following situations where case is expressed with long vowels: The five nouns, the Dual and Sound masculine plural[2].

2- The feminine ending is always “ah” and it only turns to “at” in construct or when a pronoun gets attached to it[3]. E.g. :

المدرسة جديدة (the school is new)

Classical Arabic: al-madrasatu jadīdah.

Old Hijazi:           al-madrasah jadīdah

Example for construct:

هذه مدرسة الحي (This is the neighborhood’s school)

Classical Arabic: hādhihī madrasatu l-ḥay

Old Hijazi: hādhih madrasatu l-ḥay

3- The indefinite accusative marker is always an ‘ā’ long vowel[4]. E.g. :

اشتريت كتابا جديدا (I bought a new a book).

Classical Arabic: ishtaraytu kitāban jadīdā.

Old Hijazi: ishtarayt kitābā jadīdā.

4- The third person masculine singular pronoun is always a mere ‘-h’ with no vowel attached to it[5]. E.g.

كتابه جديد (His book is new).

Classical Arabic: Kitābuhū jadīd.

Old Hijazi: Kitābuh jadīd.

In plural, the ‘-h’ pronoun only takes the “-hum” form as opposed to classical Arabic which also allows another form: “-him”. The same goes for the dual: Old Hijazi only has “-humā” while Classical Arabic has “-humā” and “-himā”.



Classical Arabic: ʕalayhim

Old Hijazi: ʕalayhum

5- The Alef Maqsūrah ى is pronounced as ē[6]. E.g. :


Classical Arabic: hudā

Old Hijazi: hudē

6- Lack of Hamzah (glottal stop)[7].

This characteristic is proven by (a) and (b) mentioned earlier. It’s also proven by early Arab grammarians’ description of the dialect of Quraysh which Muhammad belonged to.

A glottal stop naturally occurs in every language when the first word to be uttered begins with a vowel. The glottal stop in Arabic is called “Hamzah” and it has the symbol: ء . This symbol wasn’t invented yet at the time of Muhammad. The Hamzah can occur at the beginning, middle or end of a word.

Classical Arabic: رأس raʾs

Old Hijazi: راس rās

Classical Arabic: ذئب dhiʾb

Old Hijazi: ذيب dhīb

Note: QCT evidence based on rhyme shows that the Hamzah is retained when it's the last letter of a word while preceeded by the long vowel ā. E.g. : samāʾ سماء

7- The ض  letter (ḍ) sounds very similar to the sound of ظ  (ḍh) as apposed to the modern pronunciation fo ض as ḍ (emphatic d)[8].

This characteristic is proven by QCT analysis (specifically, evidence based on rhyme) and early Arab grammarians’ description of the sounds of the Arabic letters.

Quranic Comparison between Classical Arabic and Old Hijazi

Verses of Surah 104 Classical Arabic Pronunciation Old Hijazi Pronunciation
وَيْلٌ لِكُلِّ هُمَزَةٍ لُمَزَةٍ waylun likulli humazatin lumazah wayl likulli humazah lumazah
الَّذِي جَمَعَ مَالًا وَعَدَّدَهُ alladhī jamaʕa mālan wa ʕaddadah alladhī jamaʕ mālā wa ʕaddadah
يَحْسَبُ أَنَّ مَالَهُ أَخْلَدَهُ yaḥsabu ʾanna mālahū ʾakhladah yaḥsab an mālah akhladah
كَلَّا لَيُنْبَذَنَّ فِي الْحُطَمَةِ kallā layunbadhanna fil-ḥuṭamah kallā layunbadhan fil-ḥuṭamah
وَمَا أَدْرَاكَ مَا الْحُطَمَةُ wa mā ʾadrāka mal-ḥuṭamah wa mā adrēk mal-ḥuṭamah
نَارُ اللَّهِ الْمُوقَدَةُ nāru l-lāhi l-mūqadah nāru l-lāh ǝl-mūqadah
الَّتِي تَطَّلِعُ عَلَى الْأَفْئِدَةِ allatī taṭṭaliʕu ʕala l-ʾafʾidah allatī taṭṭaliʕ ʕala l-afidah
إِنَّهَا عَلَيْهِمْ مُؤْصَدَةٌ innahā ʕalayhim muʾṣadah innahā ʕalayhum mūṣadah
فِي عَمَدٍ مُمَدَّدَةٍ fī ʕamadin mumaddadah fī ʕamad mumaddadah

Introduction to the Iʕrāb

In order to understand how the language of the QCT differs from the later classical Arabic it is now read in, it's important to understand the Iʕrāb.

The Iʕrāb system in Arabic is a set of grammatical endings attached to words to convey aspects such as case, mood, and voice in a sentence. This system has its origin in classical Arabic as formulated by the classical Arabic grammarians after the 8th century, and it continues to be used, with very little change, in Modern Standard Arabic (MSA), known in Arabic as Fuṣḥā, the lingua franca of the Arab world and the language of books and official media such as government proclamations and news media.

In Classical Arabic and MSA, nouns and adjectives can be marked for three grammatical cases: nominative (مرفوع), accusative (منصوب), and genitive (مجرور). The markings consist of either a short vowel, a short vowel and an “n” sound, or (occasionally) a long vowel. Nouns can also be in the state of definiteness or indefiniteness, which each take different endings. The system of Iʕrāb specifies the vowel endings for nouns in these different states (Fischer, 86).

Here is a basic outline of i'arab for singular nouns:

1. Nominative case: the noun takes a short -u for definite and -un for indefinite nouns. This case is typically used for the subject of a sentence. E.g. :

al-baytu jadīd البيتُ جديد (The house is new) : The word “al-bayt” (the house) is the subject so it takes the ‘u’ final short vowel.

2. Accusative case: the noun takes a short -a for definite and -un for indefinite nouns. This case is often used for the direct object of a sentence. E.g. :

raʾaytu baytan jadīdā رأيتُ بيتاً جديدا (I saw a new house) : The word “bayt” (house) is the object so it takes the ‘a’ final short vowel. The word is also indefinite so it also take a final ‘n’.

3. Genitive case: the noun takes a short -i for definite and -in for indefinite nouns. This case is commonly used for objects of prepositions and to express ownership or relation of one noun to another (Karing Ryding, 183-184). E.g. :

Marartu ʕala l-bayti l-jadīd مررتُ على البيتِ الجديد (I passed by the new house) : The word “l-bayt” (the house) is in the genitive so it takes the ‘i’ final short vowel.

These are the basic form for regular, singular nouns, there are other variations on these three cases for other types and classes of noun.

As for verbs, they can take different endings based on tense, mood, and voice.

Present tense (المضارع) verbs can have different endings based on the mood:

1. Indicative mood (مرفوع): The verb ends in -u. This is used when the verb is in an independent clause.

2. Subjunctive mood (منصوب): The verb ends in -a. This is used after certain particles or in other cases.

3. Jussive mood (مجزوم): The verb ends in a sukūn, indicating a full stop/lack of vowel. This is used in commands or after certain particles (Karin Ryding, 445).

The endings for nouns are thus u, a, and i, and for verbs they are u, a, and sukūn (silent, no vowel).

A lot of words and particles take a fixed final short vowel that doesn’t change with case, the most commonly used one is ‘a’, which is attached in many instances such as:

- Singular masculine verbs in the past tense: dhahaba ذهبَ “(he) went”.

- The final “ūn/īn” of the sound masculine plural nouns and second/third person masculine verbs, such as: muʾminūna مؤمنونَ (believers). Yaktubūna يكتبونَ “(they) write”.

-The singular masculine possessive pronoun -k: kitābuka كتابكَ  (your book).

- Many particles such as: kayfa كيفَ (how) , ʕinda عندَ (at), ayna أينَ(where), thumma ثمَّ  (then), hunāka هناكَ (there),

Almost all Arabic words take a final short vowel. The exceptions are:

- Words that end with a long vowel (though many of them can take nunation).

- The jussive case which doesn’t require the addition of a final vowel.

- imperatives, such as: uktub اكتب  (write)

- A few particles such as kam كم (how much), min من  (from).

Modern Arabic dialects have completely lost these endings, and MSA may be spoken with or without them and be understood. All of the classical Islamic reading traditions feature full use of the Iʕrāb system. Despite the presence of the diacritical markings on every word indicating the presence of these short vowels, these endings are not pronounced at the end of a line of Quranic recitation. If the Iʕrāb were to be pronounced at the end of all lines, the Qur’an would cease to rhyme; meanwhile, if the Qur’an is read without the Iʕrāb, hundreds of internal rhymes emerge.

The Quranic Consonantal Text

Comparison of modern Arabic Print to the QCT

The Qur’anic consonantal text (QCT) is the original consonantal skeleton of the text of the Qur’an. It is derived from the vast Uthmanic corpus of copies of the Qur’an created after the third caliph Uthman standardized the Quran and ordered the destruction of all other different versions. The QCT was written without most (but not without all) of the diacritical marks and dots which now typify Arabic texts, including hamzahs, consonant dots and short vowel marks (rarely used in modern Arabic writing but used in the Quran and Hadith to ensure proper pronunciation). Other differences between QCT and modern Arabic writing include the mid-word long ā vowel which is usually unwritten in QCT.

The QCT shows a number of differences from both the later interpretation of it in the Islamic tradition and later medieval norms around writing Arabic.

Final Yā’

Some Arabic words such as raʾā رأى and fatā فتى are spelled with a final “y” (ى) but pronounced as a long vowel ā. The ‘y’ that is pronounced as ā is called Alif Maqṣūrah and in modern spelling it doesn’t take the bottom two dots which are reserved for a genuine ‘y’ or ‘ī’, i.e: in modern prints of the Quran, a word-final ى is pronounced as ā, while a word-final ي  (with the two dots) is pronounced as ‘y’ or ‘ī’. The QCT analysis, particularly evidence based on rhyme, shows that the Alif Maqṣūrah in the Quran wasn’t pronounced as ā. Early Arabic texts written in Greek show that this letter was pronounced as ē. It’s also pronounced as ē in some canonical readings. When Arabic was Classicized, the ‘ē’ sound merged into ‘ā’.

The development of the triphthongs in Quranic and Classical Arabic, 2017

The Hamzah

In most cases where later forms of Arabic and interpretations of the QCT would have a hamzah (the letter ء, in later Arabic used to represent the glottal stop) the QCT does not spell word with a hamzah in any position. The orthography of the QCT seems to indicate a total lack of the glottal stop in all cases save one : When a final ‘ā’ is followed by a hamzah, such as سماء samāʾ. Unlike with the other hamzahs, where rhyme seems to indicate the glottal stop was not pronounced, the rhymes involving this position seem to indicate the hamzah may have been pronounced, though it was never written.

Nunation Lost

The QCT never writes out the tanwiin, the addition of a nūn to the Iʕrāb ending of a noun, with one exception, where the expression kaʾayyin min “oh how many of” is written in all its 7 attestations in the Quran as كأين من , which is an older form where the nunation got fossilized into the expression. Otherwise the 3rd person masculine accusative ‘an’ is written just as a long ‘ā’ ا or else the tanwīn is not written at all.

Enforcement of Classical Arabic on Early Arabic Texts

Modern scholars have generally taken for granted the antiquity and universality of the Arabic of the grammarians (Classical Arabic). Earlier written texts, such as the papyri from the seventh and early eighth centuries ce and the Quran, the earliest manuscripts of which precede the grammatical tradition by more than a century, are conventionally interpreted according to much later norms, without the need for justification. Any reader of these texts will notice that the oral component differs from the written in significant ways. To illustrate, consider the word ملىكه in Q66:6. All reading traditions instruct that this word should be pronounced as [malāʔikatun]; these traditions go back to the middle of the eighth century at the earliest, while the true seventh-century form is the written artifact, mlykh, lacking the final syllable tun. Despite the fact that the written in these cases is demonstrably older than the reading traditions, the oral is given default preference, and the differences are a reduced to orthographic convention. Indeed, most scholars have assumed that the language behind the most ancient component of the Quran, its Consonantal Text (QCT), is more or less identical to the language recited in the halls of Al-Azhar today.

The careful and dispassionate study of Arabia’s ancient epigraphy reveals a picture quite dissimilar from that presented in Muslim historical sources. The Arabic of the grammarians is not met with; instead, the peninsula displays a dazzling degree of linguistic diversity. The Old Arabic dialects differ in ways not recorded by the grammarians, while features that figure prominently in the grammatical manuals are nowhere to be found. Consider nunation (tanwīn)—this is a standard feature of Classical Arabic, but in the consonantal South Semitic writing systems, Greek transcriptions, and the Graeco-Arabic inscription A1, the feature is completely absent. While the absence of nunation in Arabic orthography is usually written off as a convention, there is no reason to assume such conventions when Arabic is written in other scripts, much less before the development of the Arabic script itself. These attestations can mean only one thing: nunation had disappeared in most forms of Old Arabic.

Early transcriptions of Arabic in Greek and Hebrew scripts

Since that short vowels aren’t represented in Arabic writing, early Arabic texts written in non-Arabic scripts provide important pronunciation details as these scripts are free from Arabic spelling rules and do show short vowels.

One of the great challenges of understanding the linguistic history of Arabic in the early Islamic period is the highly defective spelling of early Arabic. It is ambiguous in terms of phonetic features such as the short vowels, the hamzah, and a general disagreement whether a written text is supposed to represent the vernacular or rather a form approximating Classical Arabic, or something in between, make it difficult to establish much of a baseline of expectations of the Arabic of this period. Historically, scholars interested in the history of Arabic have relied on the descriptions of the language by the Arab Grammarians who started their effort to standardize a high Arabic language around the end of the 8th century. The form of Arabic they describe, however, is highly idealized, and certainly rather artificial. Any data there is about the spoken vernacular in such works is, as Rabin (1951, p. 4) put it, seen “only through the veil of the literary Arabic used by their speakers”. Recent advances in the field of Arabic historical linguistics, spearheaded by Ahmad Al-Jallad, have made it clear that in the Pre-Islamic period, Arabic was much more diverse than was previously thought.
The lack of explicit prescriptivism in the early grammatical tradition concerning a large amount of phonological, morphological and syntactic variation should not be understood as evidence that the data presented by the grammarians is an uncurated representation of the dialects of Arabic. In fact, if we compare what the grammarians describe to contemporary Arabic texts written in scripts other than Arabic, we find one very striking difference: The Arabic of this period, not filtered through the grammarian lens, lacks the full ʔiʕrāb and tanwīn system which so quintessentially marks Classical Arabic.

First Islamic century Greek transcriptions

These texts are mainly official documents belonging to the Umayyad caliphate which was founded by Muʕāwiyah, a companion of Muhammad. Although the Greek texts in these documents contain short Arabic phrases (mainly names and titles), they reveal that the documented dialect has the following features:

1-The loss of final short vowels and nunation.[9]

E.g.:The name banī saʕd بني سعد is written without the final short vowel ‘i’ and without nunation (tanwīn):

Β(ανι) Σααδ β(εν) Μαλεχ / B(ani) saad b(en) malek / بني سعد بن مالك Classical Arabic pronunciation: Banī saʕdin ibni mālik

You can view the papyri here.

2- Final short vowels are retained In construct.


Ομμου Ιωσεw / ommu yūsef/ أم يوسف [10]

But if the possessive noun begins with the definite article, the final short vowel of the possessed noun is replaced with the vowel of the ‘al’ article. E.g.: αβδαλλα/abdalla / عبد الله (Classical Arabic: Abdullah).

3- The feminine ending “ah” changes to “at” only in construct[11]. Which proves the lack of final short vowels in non-construct.

Example for feminine “at” in construct:

The (female) servant of God أمة الله

αμαθαλλα : amatalla[12]

4- Case inflection with long vowels is retained.

The word “father” in the nominative appears as abū while in the genitive appears as abī[13].


Αβου Σαειδ /abū saʕīd/ أبو سعيد

Οβαιδαλλα β(ιν) Αβιλαας / ʕobaydallāh b(in) ʾabī l-ʕās ̣/ عبيد الله بن أبي العاص

5-  The alef maqsūrah ى is pronounced as ē instead of the Classical Arabic pronunciation ā.

E.g. [15]

Μαυλε /mawlē/مولى

ιαειε /yaḥyē/ يحيى

ιαλε /yaʕlē/يعلى

An example of these Umayyad Greek-Arabic texts:

A Greek Inscription from Jordan Dated 42 AH / 662-63 CE

A Greek inscription containing an Arabic line written in Greek letters which reveals that final short vowels weren’t used: ABDALLA MAAUIA AMIRAALMUMENEN (The servant of Allah Muʕāwiyah the commander of the faithful)

This inscription includes the Arabic pronunciation of the title and name of Muʕāwiyah, the first Umayyad caliph.

“In the days of the servant of God Muʕāwiya, the commander of the faithful, the hot baths of the people there were saved and rebuilt…”

In this inscription, not a single Arabic word recieved a final short vowel:

“The servant of God Muʕāwiya the commander of the faithful”

The Greek transcription:                ABDALLA MAAUIA AMIRAALMUMENEN

Classical Arabic pronunciation: ʕabdullāhi muʕāwiyatu amīru l-muʾminin

عبدُ اللهِ معاويةُ أميرُ المؤمنين

In classical Arabic, the first 4 words receive the following final short vowels:

“ʕabd” receives ‘u’. “Allah” receives ‘i’. Muʕāwiyah receives ‘u’ which turns the ‘ah’ to ‘at’: muʕāwiyatu. “amīr” receives ‘u’.

You can view the inscription here.

Another Umayyad Greek-Arabic text:

A Bilingual Umayyad Document From The Year 54 AH / 674 CE

A Bilingual Umayyad Document where a number of names were written in both Arabic and Greek. The names in Greek lack nunation and final short vowels.

This is an Umayyad Note to the people of the city of Neṣṣana demanding that they pay their due of the Jizyah (Tax on non-Muslims). It’s written both in Arabic and Greek. The Greek portion includes the following Arabic names that lacked any case inflection:

Alaaret b(en) Abd الحارث بن عبد

Classical Arabic: Al-ḥārithi bni ʕabd (The name in the document is in the genitive case, hence taking the ‘i’ final short vowel)

Adie B(en) Kaled عدي بن خالد

Classical Arabic: Adeyyi bni khālid (The name in the document is in the genitive case)

بني سعد بن مالك

B(ani) saad b(en) malek

Classical Arabic: Banī saʕdin ibni mālik

You can view the document here.

The Damascus Psalm Fragment

This document, dated to the third Islamic century, was discovered in the Umayyad Mosque in Damascus in 1900. It includes a translation of a portion of “The Book of Psalms” of the bible (Psalm 77). This Arabic translation is written with Greek letters. The translation is literal with strict adherence to the syntax and wording of the original language*, which caused parts of it to sound awkward and hard to understand.

The phonology and morphology of the Psalm Fragment reflect the contemporary vernacular, while its syntax follows the Greek.

The dialect of the Psalm Fragment has the following features:

1- The loss of final short vowels and nunation[16].


yuheyyī māy(i)deh li-šiʕb-hu(hi)[17]

ὑεϳει μάϳδεὑ λιχχειγβὑϳ

يهيِّي (يهيء) مايدة (مائدة) لشعبه

Classical Arabic:

yuhayyiʾu māʾidatan li-šaʕbih

2- In construct, final short vowels are retained in some cases and lost in others[18].

Example for the loss[19]:

حول خيامْهُم

ḥawl ḫiyēm-hum

χαυλ χηέμὑμ

Example for retention[20]:



βη αυθάνϳὑμ

3- The Alef maqṣūrah is pronounced as [ā] in backed and labial environments, but as [ē] otherwise[21].


أعطى  aʕṭā


أتى  atē


4- The “L” of the definite article doesn’t assimilate to the following coronal consonant[22].





Note: In the Greek transcriptions from the first Islamic century, the L is assimilated:



5- The pronominal suffix of the 3rd person masculine plural takes only the “hum” form. While classical Arabic has both “hum” and “him”.

In the following example[25], the final pronoun should take the “him” form in accordance with classical Arabic rules. The psalm fragment instead uses the “hum” form.



βη αυθάνϳὑμ

6- The indefinite accusative is marked with ā instead of classical Arabic “an”.

This is attested twice in the word γεδδα [ǧeddā][26] which means “very”.

7- The Feminine Ending is “eh” instead of “ah” which matches modern Levantine Arabic[27]

οελευδιεὑ [wel-ʾʔewdiyeh] والأودية[28]

χαϳμετ σεϳλουμ [ḫaymet seylūm] خيمة سيلوم [29]

8- ā is realized as [ē] unless there is an inhibiting factor, that is, an emphatic or a labial[30].





The text of the Damascus Psalm Fragment

fa-sēlet mayyah wel-ʔewdiyeh fāḍat leʕal wa-ḫubz yeqdir yuʕtī

ʔeu yuheyyī māy(i)deh li-šiʕb-hu(hi) [sic] [*li-siʕbi-h(?)]

فسىلت (فسالت) مَيَّه والأودية فاضت لعل وخبز يقدر يعطي أو يهيِّي (يهيء) مايدة (مائدة) لشعبه.

li-dhālik semiʕ el-rab fa-ʔamtenaʕ wel-nār ʔeshteʕalet fī yaʕqūb wa ruǧz ṣaʕ(ad)

ʕalā ʔisrāel

لذلكْ سمع الرب فأَمتَنَع والنار اشتعلت في يعقوب ورُجُز صعد على إسراييل

li-ʔen(nahum) (la)m yūmi(nū) billāh wa-lā (tawa)kkelū ʕalā khalāṣ-h

لأنهم لم يومنوا بالله ولا توكلوا على خلاصه

wa ʔamar el-siḥēb min fawq wa ʔabwāb el-se…samā fateḥ

وأَمَر السحىب (السحاب) من فوق وأبواب السما فتح

wa ʔamṭar lehum m(ann)a liyā(kul)ū (wa) (ḫub)z min el-(semā) ʔaʕṭā-hum

وأمطر لهم منَّا لياكلوا و خبز من السما أعطاهم

(ḫub)z el-melēyke (ʔak)el ʔinsēn (ša)ba(ʕ) baʕaṯ la-hum ley(i)temellew

خبز الملايكة أكل إنسان شبع بعث لهم ليتمَلَّوْا.

ʔahāǧ el-teym(an) min el-semā wa ʔatē bi-quwwet-uh el-ʕāṣif

أهاج التيْمَن* من السما وأتى بقوته العاصف

* Al-Jallad Notes: The name of the south wind in Classical Arabic is al-ǧanūb. The use of Teym[an] here might be an Aramaicism, tayman “south.” An identical term is used in the Hebrew Bible, têmān. (p.83)

wa ʔamṭar ʕaley-hum mithl el-turāb luḥūm wa mithl raml el-buḥūr ṭiyūr


وأمطر عليهم مثل التراب لحوم ومثل رمل البحور طيور مِجْنِحَة

fa-waqaʕat fī wasaṭ ʕasker-hum ḥawl ḫiyēm-hum

فوقعت في وسط عسكرهم حول خيامهم

fa-ʔakelūwa šebiʕū ǧeddā wa šehwet-hum ǧēb la-hum

فأكلوا وشبعوا جدا وشهوتهم جاب لهم

(la)m yuʕdemū (š)ehwet-hum wa ʕindmā kēn el-ṭaʕām fī fāh-hum

لم يُعدموا شهوتهم وعندما كان الطعام في فاهم

wa ʔabtelew wa marmarū el-ʔilēh el-ʕālī wa šehād(ā)t-uh lam yeḥfaḏ̣ū

وابتلوا ومرموا الإله العالي وشهاداتُه لم يحفظوا

fa ʔanqalebū wa ġadarū miṯl ābāy(i)-hum ʔanqalebū miṯl el-qaws el-ʕawǧē

فأَنقلبوا وغدروا مثل آبايهُم أَنقلبوا مثل القوس العوجى

wa (ʔa)sḫaṭū-h bi-ʔawθāni-hum wa bi-menḥūtēti-hum ʔaġārū-h

وأسخطوه بأوثانهم وبمنحوتاتهم أغاروه.

semiʕ allāh wa teġāfel (wa) ʔafsel ǧed(dā)—li-isra(il)

سمع الله وتغافل وأفسل جدا لإسراييل

wa ʔaqṣā ḫaymet seylūm el-mesken elleðī ʔesken fil-bašer

وأقصا خيمة سيلوم المسكن الذي أسكن في البشر

wa ʔas(l)e(m) lilseb(ī) (q)oe(t-hum)

وأسلم للسبي قوتهم.

Judaeo-Arabic Texts

A collection of papyri from Egypt includes Arabic texts written with Hebrew characters. These papyri predate 900 AD[32].

“The Arabic of these papyri has lost case and mood categories in the noun and verb. The breakdown of the case system is indicated by several features. Had tanwīn existed in the type of Arabic studied here, there can be no real doubt that it would have been marked by final nun; the fact that these texts use an acoustically based orthography, quite free from the influence of literary Arabic spelling, makes this virtually certain. The adverbs terminating in aleph must therefore be regarded as reflecting the ending ā, not the literary tnwīn.”

Although the Hebrew script is defective and doesn’t write short vowels and many long ‘a’ vowels, these Judaeo-Arabic texts are still valuable as they don’t abide by Arabic orthography rules. E.g. A word pronounced as “kalbun”(with final short vowel and nunation) is written in Arabic as “klb” without the suffix in accordance with Arabic orthography rules. But when this word is written in Judaeo-Arabic as “klb”( instead of “klbn”) then this means it’s pronounced without the suffix because these texts are phonetic and don’t abide by Arabic orthography rules[33].

In the following example[34] from the Judaeo-Arabic Papyri, the word عافية ʕāfiyah is written as ʕfyh which means the word lacked nunation otherwise it would have been written as ʕfytn (ʕāfiyatin).

The same example also shows the word محمود  maḥmūd is written as mḥmwd which means the word lacked nunation otherwise it would have been written as mḥmwdn (maḥmūdun).

ונחן פי עפיה ואלה מחמוד עלא דלך

ونحن في عافية والله محمود على ذلك

(We are in good health thanks to Allah)

Litteral transcription: wnḥn fy ʕfyh walh mḥmwd ʕlā dlk

Pronunciation: wanaḥn fī ʕāfiyah wallāh maḥmūd ʕalā dhālik

Classical Arabic: wanaḥnu fī ʕāfiyatin wallāhu maḥmūdun ʕalā dhālik

Besides the loss of final short vowels and nunations, other features of these texts include:

1- Frequent loss of Hamzah[35]:

אלרדיה (The cloaks الأردية )

Litteral transcription: alrdyh

Pronunciation: alardiyah

Classical Arabic: alʾardiyah

Note: The Hebrew letter א can be used to express the long vowel ā or a glottal stop. If the word for “the cloaks” was pronounced with a glottal stop, it would have been written with two א :  אלארדיה

תכוד (تأخذ  you take)

Litteral transcription: tkwd

Pronunciation: tākhudh

Classical Arabic: taʾkhudh

Note: If the word was pronounced with a glottal stop, it would have been written as: תאכוד

2- The indefinite accusative is marked with ‘ā’ instead of classical Arabic “an”[36].


(and also) (وأيضا)

Litteral transcription: wydā

Pronunciation: wēḍā

Classical Arabic: waʾayḍan


(and tomorrow) (وغدا)

Litteral transcription: wgdā

Pronunciation: waghadā

Classical Arabic: waghadan

3- The pronominal suffix of the 3rd person masculine is ‘h’ with no vowel after it as opposed to classical Arabic forms: hū/hī, hu/hi

After consonants the pronoun is spelled as wh, to be pronounced uh or oh. And after vowels the pronoun is spelled h.


His son ولده

Litteral transcription: wldwh

Pronunciation: waladuh

Classical Arabic: waladuhū

מין אכיה

From his brother من أخيه

Litteral transcription: myn akyh

Pronunciation: min ʾakhīh

Classical Arabic: min ʾakhīhi

4- The pronominal suffix of the 3rd person masculine plural takes only the “hum” form[38]. As opposed to classical Arabic which has both “hum” and “him”.

In the following example, the pronominal suffix should take the “him” form in accordance with classical Arabic rules. But it’s written as “hum”.

עלא חאלתהום

In their condition على حالتهم

Litteral transcription: ʕlā ḥālthwm

Pronunciation: ʕalā ḥālatihum

Classical Arabic: ʕalā ḥālatihim

5- Loss of verbal moods


They are going to flee يهربوا

Litteral transcription: yhrwbw

Pronunciation: yahrubū

Classical Arabic: yahrubūn

Final short vowels and Nunation in the Quranic Readings Tradition

The third caliph’s standardization of the Quran unified the Quranic consonantal text. But as this standardized Quran lacked diacritics, this allowed for the emergence of readings that differ in short vowels, dotting and linguistic characteristics.

In their recitation of the Quran, Muslims follow early Quran scholars whose readings became named after them, hence called eponymous readings. In the third Islamic century, the Quranic readings were canonized into 7 readings. And in the ninth Islamic century another 3 were added making the total number of canonical readings 10.  Today the reading of Ḥafṣ is the most popular in the world.

Final short vowels and nunation (Iʕrāb) are fully employed in all of the canonical readings. This doesn’t necessarily mean Iʕrāb was part of the original language of the Quran since that the linguistic and historical analysis reveals that these readings were linguistically reworked[40], none of them represents natural language[41], they were full of innovation and didn’t strictly adhere to a supposed oral tradition that goes back to Muhammad. An example of this is the following report from Al-Kisāʾī, the founder of a one of the canonical readings. In his reading, he treats the word thamūd in the accusative as triptotic (thamūdan: i.e. it can take nunation and one of three possible final short vowels). He also treats the word in the genitive and nominative as diptotic (thamūda/thamūdu: i.e. can’t take nunation, and can take one of two possible final short vowels instead of three). Al- Kisāʾī breaks his own rule in verse Q11:68 where the word thamūd is mentioned twice: first in the accusative and second in the genitive. According to his rule, the first word in the verse should be treated as triptotic (thamūdan), and the second word as diptotic (thamūda), but instead he treats both of them as triptotic (thamūdan, thamūdin, respectively):

ʔa-lā ʔinna thamūdan kafarū rabba-hum ʔa-lā buʕdan li-thamūdin (Q11:68)”.

﴿أَلَا ‌إِنَّ ‌ثَمُوداً كَفَرُوا رَبَّهُمْ أَلَا بُعْدًا لِثَمُودٍ (٦٨)﴾  هود

Al-Kisāʾī was asked about this, and his response showed that his reasoning had no regard to oral transmission, he said: “It is ugly to have a word occur twice in two places (within the same verse) and then have them disagree [on triptosy/dipotsy], so I treated it [ṯamūdin] as a triptote because of it being close to it [ṯamūdan]." [42]

Marijn Van Putten comments on this report saying:

While this account of course does not prove that the Quran was once composed without ʔiʕrāb, what it does show is how readers themselves thought about their role in applying ʔiʕrāb in recitation. Their role was not to faithfully verbatim the ʔiʕrāb as had been taught to them, but rather to argue and rationalize why a word should have the ʔiʕrāb that they would give it. In such cases even purely aesthetic arguments such as the one cited, was apparently enough to deviate from the way their teacher taught it (Ḥamzah, al-Kisāʔī’s direct teacher reads ṯamūda and li-ṯamūda in the relevant verse). As such the application of ʔiʕrāb by these readers can tell us nothing at all about the use of ʔiʕrāb of the original language of the qct. However, given that the choice of ʔiʕrāb was a rational endeavour explicitly based on both the rasm (QCT) and aesthetic preference rather than prophetic example, it becomes quite easy to envision that the presence of this very system was not original to the text, but was rather imposed on it sometime after the standardization of the QCT by ʕuṯmān.

Another sign that the application of Iʕrāb wasn’t based on oral tradition, is the pseudo correct use of Iʕrāb, as in the following example[43]:

The question word أيَّانَ ʾayyāna (when?) which is used in the Quran several times (such as Q7:187), is a merge between two words: أي  ʾayy (which) and  آنʾān (time). The original shape with Iʕrāb is: ʾayya ʾānin. The Hamzah is lost which allows the merge: ʾayyānin. Yet all of the readings have the word as “ʾayyāna” instead of “ʾayyānin” although ʾān in this structure should take the final short vowel ‘i’ and nunation according to classical Arabic rules: The word ʾayy is the possessed مضاف, and the word ʾān is the possessor مضاف إليه which means it should take the “in” suffix. What explains the readers giving “ayyān” a final -a is that the two words originally had no final short vowels: ayy ān, or ayya ān. And when the two words merged as “ayyān” and got treated as a single word, the readers gave it a final -a because they thought the word is similar to other words such as:

- Question words such as أينَ  ʾayna which only takes a fixed final -a and can’t take nunation nor any other final short vowel.

- Denotations of time which grammatically usually take a final -a, such as يومَ  yawma ‘on the day’ and حينَ ḥīna ‘at the time’.

The mysterious letters

Unlike what the readers did with the entirety of the Quran, the mysterious letters are the only verses they didn’t force case inflection on. These letters appear at the beginning of 29 Surahs, such as Surah no.2 which begins with the three letters ألم (ALM) recited as “Alef lām mīm”. No one knows what these letters mean or why they were employed. Muslim scholars have many different explanations that are based on pure speculation.

In Hadith, which classical Arabic has been imposed on, the mysterious letters are inflected for case:

مَنْ قَرَأَ حَرْفًا مِنْ كِتَابِ اللَّهِ فَلَهُ بِهِ حَسَنَةٌ وَالْحَسَنَةُ بِعَشْرِ أَمْثَالِهَا لَا أَقُولُ: آلم حَرْفٌ. أَلْفٌ حَرْفٌ وَلَامٌ حَرْفٌ وَمِيمٌ حَرْفٌ

“If anyone recites a letter of God’s Book he will be credited with a good deed, and a good deed gets a tenfold reward (Al-Qur’ān, 6:160). I do not say that A.L.M are one letter, but alif is a letter, lām is a letter and mīm is a letter.”[44]

In the Arabic text of the Hadith, the three letters are case inflected in the last part of the Hadith:

“alifun is a letter, lāmun is a letter and mīmun is a letter.”

Also when modern Muslim scholars cite this Hadith they pronounce the letters at the last part of the Hadith with inflection as can be heard in this video:

Since that these letters should be inflected in Classical Arabic, why were they left without any inflection in the readings tradition? Van Putten answers:

The form of the mysterious letters is fairly easy to understand from a situation that started out as lacking inflectional endings, which were classicized. As these mysterious letters have no obvious syntactical function, it is difficult to classicize these into an inflectional paradigm. The inverse, however, is more difficult to understand. There is no reason why the mysterious letters would be uninflected, if the base language of the Quran was inflected.

Evidence based on the Consonantal Text of the Quran

most scholars have assumed that the language behind the most ancient component of the Quran, its Consonantal Text (QCT), is more or less identical to the language recited in the halls of Al-Azhar today. It is only in very recent years that the QCT has enjoyed study as an epigraphic document, interrogating the text for linguistic facts rather than assuming them. These efforts have led to a radically different view of the language of its composition.
When looking to answer the question what the language of the Quran is, the reading traditions fail to give a consistent answer. They are linguistically diverse, none of them look like natural language, and they must be considered to be a concerted effort to beautify the recitation of the Quran through the use of exotic linguistic features from a variety of different dialects, augmented with completely innovative forms that do not seem to have been part of anyone’s natural speech. However, there is a source of the Quran that carries linguistic information that does go back to the very first decades of Islam: the written text itself.

The mismatch between Classical Arabic pronunciation and Spelling

(for a detailed explanation see this article)

There exists a mismatch between Arabic spelling and classical Arabic pronunciation. For example:

هناك كتاب جديد hunāka kitābun jadīd (that is a new book)

The word Kitābun is spelled as if it were pronounced as kitāb. The final short ‘u’ vowel cannot be spelled as Arabic doesn’t write short vowels. But the final ‘n’ can be spelled, yet nunation is never written in Arabic. If it was, then the sentence would look like this:

هناك كتابن جديد

Arab grammarians tried to explain the mismatch between spelling and classical Arabic pronunciation by saying that every Arabic word is spelled in the pausal form even if the word wasn’t in a pausal position.

Pausal forms

A pausal position means the word at the end of an utterance such as the last word of a sentence. When a word is in a pausal position, it receives special treatment in Classical Arabic:

- Neither a final short vowel nor nunation can be attached to the word.

-Due to the lack of a final short vowel, the feminine ending “ah” doesn’t change to “at”: al-madrasatu => al-madrasah.

- The “an” marker for the indefinite accusative becomes a long ‘a’ vowel: Kitāban => Kitābā.

- The dropping of the vowel of the third person masculine singular pronoun: Kitābuhū => Kitābuh.

Evidence Against the Pausal Spelling Rule

Historical linguists Van Putten and Phillip Stokes note that such a spelling convention is unique among the languages of the world[45] . They also challenge the pausal spelling convention by the following arguments based on the linguistic analysis of QCT:

1- Internal Rhymes.

(for a detailed explanation see this article)

There are Internal rhymes in the Quran that only show up if every word is pronounced in the pausal form. This means that what was thought to be a special treatment for the pronunciation of pausal words was actually the norm for almost all words in the original language of the Quran. The mismatch between the Quranic spelling and the Classical Arabic pronunciation is the result of imposing classical Arabic on a text that wasn’t written in Classical Arabic. Which means the QCT was written phonetically in Old Hijazi, a language that lacked nunation and final short vowels.

An example of these internal rhymes is the following attribute of Allah used as a verse ending in 15 verses such as 4:26, 8:71:

عليم حكيم ʕalīmun ḥakīm (Knowing, Wise)

The first word of the pair, ʕalīm, takes the final short vowel ‘u’ plus nunation. While the second word ḥakīm is in a pausal position (the end of a verse) and hence according to Classical Arabic rules the word doesn’t take the expected ‘un’ suffix. But suffix of the first word is dropped, the two words rhymes with each other:  

ʕalīm ḥakīm

Which means that in the original language of the Quran, final short vowels and nunation weren’t only lost in pausal positions, but they were lost in all words (except in construct).

There are hundreds of similar examples of Old Hijazi internal rhymes. You can see all these examples here.  

2- The treatment of the final -ī

Word-final -ī in QCT is written in some cases and omitted in others.

A thorough analysis of all words with final -ī revealed that -ī is omitted mostly in pausal positions, while kept mostly in context. This important find reveals that the spelling of the Quran isn’t based on the pausal rule. Otherwise, the treatment of the final -ī wouldn’t have changed between pausal and non-pausal positions.

An example of the treatment of final -ī:

The possessive -ī is omitted in 143 pausal positions and never omitted in context. While it’s written in 531 context positions and 21 pausal positions.[46]

3- The feminine ending ‘-ah’ is only spelled as ‘-at’ in construct.

Most of Arabic singular feminine nouns and adjectives end with “-ah”. As in: madrasah مدرسة (school). But when these words are in a construct position, or when anything is attached to the end of these words such as a final short vowel, the feminine ‘-ah’ turns into ‘at’ in pronunciation but it remains spelled as ‘ah’.


al-madrasatu hunāk. المدرسة هناك (the school is there).

The word madrasah is the subject of this sentence so it received a final ‘u’ vowel. And since that something was attached to the end of the word, the feminine ‘ah’ of “al-madrasah” turns into ‘at’: al-madrasatu. Yet in spelling, the word is still spelled with a final ‘h’: المدرسة . That’s because in Classical Arabic spelling, the feminine ending is always spelled as ‘h’ even in construct which is a position where the feminine ending is always pronounced as ‘t’ and never as ‘h’. The QCT adheres to this spelling rule but shows some instability in many instances where the feminine ending is spelled as ‘-t’ ـت .

Van Putten and Stokes found that all instances where the feminine ending is spelled as ‘-t’ occurred in construct, a position where the feminine ending is always pronounced as ‘-t’ in all forms of Arabic including Classical Arabic and modern Arabic dialects. The incidence of the ‘-t’ spelling is 47 (22%) out of 218 total of feminine constructs present in the QCT. The rest of the feminine words in construct were spelled as ‘-h’.

“The best way to understand these spellings then, is as inconsistencies of orthography by the scribe, who would occasionally write the construct feminine the way he pronounced it, rather than the non-phonetic orthographic practice to write it with ‘-h’ ـه.”

The fact that this hāʔ only shows up in pausal position, is yet another piece of evidence that ‘pausal spelling’ is not a governing principle in Quranic orthography. Had that been the case, all apocopates and imperatives should have received a final h, not just the one that stand in a pausal position.

As for the other thousands of instances where feminine words aren’t in construct, QCT spells them all with ‘-h’ although most of them are in positions where they should receive final short vowels which turns the feminine ending to ‘-t’. The 22% incidence of the ‘-t’ spelling in construct and the total lack of the ‘-t’ spelling in non-construct proves that in the language of QCT, the construct is the only position that allows turning the feminine ending into ‘-t’. Which means the QCT language lacked final short vowels which explains why there wasn’t any incidence of the ‘-t’ spelling in non-construct.

4- The pausal hāʾ هاء السكت

This hāʾ is used in the Quran several times for rhyming purposes, except in two instances:

- The word ‘yatasannah’ (yatasanna+h) in 2:259

فَٱنظُرۡ إِلَىٰ طَعَامِكَ وَشَرَابِكَ لَمۡ ‌يَتَسَنَّهۡۖ

The original form of the word is ‘yatasannā’ يتسنى (yatasannē in Old Hijazi), but the final long vowel is shortened because the word is preceded by the jussive negating particle ‘lam’, yielding the jussive form: yatasanna. Which makes the word an apocopate (a word whose final sound is omitted).    

- The imperative word ‘iqtadih’ (iqtadi+h) in 6:90:

أُوْلَٰٓئِكَ ٱلَّذِينَ هَدَى ٱللَّهُۖ فَبِهُدَىٰهُمُ ‌ٱقۡتَدِهۡۗ  

The previous two words, yatasannah and iqtadih, are the only two times an apocopate and an imperative occur in in a pausal position. And both of them received a final ‘h’. Which means in the language of QCT, an apocopate or an imperative receives a final ‘h’ in pause.

The fact that this hāʔ only shows up in pausal position, is yet another piece of evidence that ‘pausal spelling’ is not a governing principle in Quranic orthography. Had that been the case, all apocopates and imperatives should have received a final h, not just the one that stand in a pausal position.

The Hamzah in the Quranic Reading Traditions

The use of Hamzah is among the differences in linguistic characteristics between the 10 canonical readings. A lot of the readings, including the reading of Ḥafṣ, heavily use the Hamzah while other readings use it moderately. The Hamzah isn’t an original part of the Quran (except in word-final Hamzah that’s preceded by the long vowel ā). There are reports clearly showing that the Hamzah was a later addition. One of these reports says that Nāfiʕ, the founder of one of the ten canonical readings, was asked if it’s possible to introduce the Hamzah to (the two words in the Quran) “al-dhīb” and “al-bīr”. He replied: “If there are Arabs who use the Hamzah with these words then you can use the Hamzah”[47].

Khalaf, another founder of one of the ten canonical readings, says: “Quraish (Muhammad’s tribe) doesn't use the Hamzah. It's not in their dialect. The eponymous readers took the Hamzah from non-Quraishi dialects.”[48]

Khalaf’s statement agrees with what early grammarians report. Mukhtār Al-Ghawth says in his book "The dialect of Quraish": “Since that the Hamzah is hard to pronounce, some early Arabic dialects leaned towards dropping the hamzah. This was most notable in the dialect of Quraish as all early sources agree that this dialect lacked the hamzah.” [49]

The enforcement of Hamzah on the Quran is apparent in the instances of pseudo-correct application of the Hamzah. Van Putten lists 12 examples[50] of such cases, such as the word kās كاس (cup) which is read as kaʾs in all canonical readings.

Comparative Semitic evidence shows that kaʔs ‘cup’ must have a pseudocorrect hamzah in Arabic. The reflexes in Hebrew כוס kos (spelled without aleph) and Aramaic kās as well as Ugaritic ⟨ks⟩ leave no doubt that the reconstruction of this noun in Proto-West Semitic is *kās and the hamzah in the Quranic reading traditions must be pseudocorrect.

The Hamzah and the Consonantal Text of the Quran

A table showing how the Hamzah is written in standard and pre-standard Arabic

The consonantal text of the Quran provides evidence that the original language of the Quran lacked the Hamzah.

A- The introduction of the Hamzah breaks the rhyme in the following verses[51]:

1- In Surah Ar-raḥmān (No.55), all verse-final words end with “ān”:

ar-raḥmān, al-qurʾān, al-ʾinsān…etc. الرحمن، القرآن، الإنسان

Verse no.29 ends with the word shaʾn شأن  (with the Hamzah). By dropping the Hamzah from the word it becomes shān شان and thus the word fits the Surah’s rhyme.

Al-Farrāʾ, an early Arab grammarian, noted this when he said[52]: You can use the Hamzah with the word “shaʾn” in the entirety of the Quran except for the one in Surah ar- raḥmān (No.55) because the word comes in the middle of verses that lack the Hamzah.

2- In Surah no.19, the rhyme is a short vowel + yyā: zakariyyaā, khafiyyā, shaqiyyā...etc.

زكريَّا، خفيَّا، شقيَّا

Verses 9, 42, 60 and 67 end with the word: shayʾā شيئا.

By dropping the hamzah, the word becomes: shayyā شيَّا .

The reading tradition of ḥamzah reads the word shayʾā without the hamzah.

3- Verse 47 of the same Surah ends with the word riʾyā رئيا . Without the Hamzah the word becomes: riyyā. It’s read as such in the reading of ḥamzah.

4- In Surah no.96, verses from 15 to 18 end with the following words:

nāṣiyah, khāṭiʾah, nādiyah, zabāniyah.

كَلَّا لَئِنْ لَمْ يَنْتَهِ لَنَسْفَعًا بِالنَّاصِيَةِ (15) نَاصِيَةٍ كَاذِبَةٍ خَاطِئَةٍ (16) فَلْيَدْعُ نَادِيَهُ (17) سَنَدْعُ الزَّبَانِيَةَ (18)

By dropping the Hamzah from the word khāṭiʾah, the word perfectly rhymes with the final words of the surrounding verses:

nāṣiyah, khāṭiyah, nādiyah, zabāniyah.

The reading of Abū jaʕfar reads it as khāṭiyah.

5- In surah no.69, verses 8-10 end with following words:

Bāqiyah, khāṭiʾah, rābiyah.

فَهَلْ تَرَى لَهُمْ مِنْ بَاقِيَةٍ (8) وَجَاءَ فِرْعَوْنُ وَمَنْ قَبْلَهُ وَالْمُؤْتَفِكَاتُ بِالْخَاطِئَةِ (9) فَعَصَوْا رَسُولَ رَبِّهِمْ فَأَخَذَهُمْ أَخْذَةً رَابِيَةً (10)

By dropping the Hamzah from khāṭiʾah, it becomes: khāṭiyah خاطية. It’s read as such in the reading of Abū jaʕfar.

The second evidence the QCT provides for the lack of Hamzah:

B- The shapes the Hamzah takes in the QCT.

The sign for the Hamzah in Arabic is ء. But the Hamzah in Arabic orthography is rarely written with the Hamzah sign ء alone. In most cases, the Hamzah is written by adding the Hamzah symbol to one of the three vowel letters like this:

  أ ؤ ئ

The letter for a long 'a' vowel is Alef ا . As in: kitāb (book). كتاب

The letter for a long 'u' vowel or the 'w' sound is: و . As in: rʕḥ (soul) روح

The letter for a long 'i' vowel or the 'y' sound is: ي , يـ . As in: fī (in) في

All these different forms: ء , ئ , ؤ, أ are pronounced the same: a glottal stop.

The Hamzah takes all these different shapes because Arabic orthography was standardized based on the QCT which represents a dialect that lacks the hamzah.[53]

In many cases, the dropping of Hamzah leads to the creation of a long vowel, the w sound or the y sound. Prior to the standardization of Arabic orthography, the Alef used to be the symbol of the Hamzah or the long vowel ā. For example, the word for "believer" is مؤمن muʾmin. Before the first half of the first century of Islam, Arabs who had the Hamzah in their dialects wrote this word as مامن . The Alef is the sign for the Hamzah in this word. But for Arabs who didn't have the Hamzah like Quraish, they wrote the word like this: mūmin مومن . It’s written as such because the dropping of the Hamzah in this word creates the long vowel ū. That's why in the QCT the word is written as مومن  instead of مامن.

Nearly 25 years after the death of the prophet, the third caliph Uthman decided to create a standard copy of the Quran to put an end to disputes over the different readings of the Quran. He assigned the job to a team of scribes most of whom belonged to the Quraishi tribe. This standard copy that Uthman assembled is called the Uthmanic Quran. All Qurans in the world today follow the exact script of the Uthmanic Quran (QCT) except for signs that were later invented like the dots, the Hamzah and diacritics short vowels.

In the original Uthmanic text that lacks diacritics (QCT), the Hamzah is only written when it's located at the beginning of a word. That's because in the dialect of Quraish, the Hamzah is only pronounced when it's at the beginning of speech.

Example:اغفر "ighfir" . Note that the Hamzah is written as an Alef because the Alef was the symbol for Hamzah or the long vowel ā.

After Uthman created the standard copy of the Quran, he sent copies to all different regions of the caliphate. The inhabitants of these regions started basing their Arabic orthography on these Qurans. So an Arab who had the Hamzah in his dialect started writing the word muʾmin as مومن  instead of مامن although he kept pronouncing the Hamzah in it. One century later when the sign of Hamzah ء  was created, those Arabs who had the Hamzah in their dialects added the Hamzah sign over the long vowels, turning words like

Mūmin muʾmin

Yastahzī yastahziʾ

Rās raʾs

مومن to مؤمن.

يستهزي to يستهزئ

راس to رأس

That's why Arabic today writes the Hamzah in four different shapes: ء, أ, ؤ, ئ

The Iraqi Quran scholar and linguist Ghānim Qaddūrī says:

After Uthman sent copies of the standardized Quran to the different regions of the caliphate, these copies became the reference not only in recitation but also in orthography. (Note: The "newly formed societies" in the next line means the cities in Iraq that were formed after the Muslim conquest of Iraq, especially the cities of Kūfah and Baṣrah which were the capitols of Arab grammarians).

The Arabic language in the newly formed societies went through a phase of linguistic mixing between the dialects of the people of the Arabian peninsula (who migrated to these newly formed cities). The Arabic language there started adopting the hamzah. This was boosted by:

1- The adoption of Hamzah by the scholarly movement of Iraq because the scholars tended to study the Arabic of the tribes of central and eastern Arabia (whose dialects use the hamzah).

2- Iraq is open and connected to central Arabia.

3- Many central Arabian tribes migrated to Iraq.

People started writing the Hamzah as one of the three letters ا ي و  following the steps of the Uthmanic script. It became forgotten that the Alef is the supposed shape of hamzah. And it became forgotten that the Uthmanic Qurans were written in the dialect of the people of Hijaz who drop the hamzah. People were careful to follow the Uthmanic Quranic text which was agreed upon by the prophet's companions. This made people stick to the shapes of words as written in Uthmanic text. The Hamzah in the Uthmanic text was written as the vowels  و ي ا . So when people copied this orthography, they added dots over these letters to indicate the hamzahs. Then the dot changed to the ء symbol after Al-Khalīl invented it.
Ghānim Qaddūrī, Rasm Al-Miṣḥaf, 1982, p.575-577


  1. Marijn Van Putten, Quranic Arabic, p.282
  2. Marijn Van Putten, Quranic Arabic, p.282
  3. Marijn Van Putten & Phillip Stokes, Case in the Quranic Consonantal Text, p165
  4. Marijn Van Putten & Phillip Stokes, Case in the Quranic Consonantal Text, p158
  5. Marijn Van Putten, Quranic Arabic, p.275
  6. Marijn Van Putten, Quranic Arabic, p.24
  7. Van Putten, Hamzah in the Quranic Consonantal Text, p.97
  8. Van Putten, Inferring the Phonetics of Quranic Arabic from the Quranic Consonantal Text, p.4-5
  9. Ahmad Al-Jallad, The Arabic of the Islamic conquests, 2017, p11]
  10. Ibid, p12
  11. Ibid, p14
  12. Kaplony, Andreas, The orthography and pronunciation of Arabic names and terms in the Greek , p.16
  13. Ahmad Al-Jallad, The Arabic of the Islamic conquests, 2017, p11
  14. Ibid, p12
  15. Ibid, p13
  16. Ahmad Al-Jallad, The Damascus Psalm Fragment, 2020, p.21
  17. Ibid, p.79
  18. Ibid, p.22
  19. Ibid, p.84
  20. Ibid, p.90
  21. Ibid, p.48
  22. Ibid, p.49
  23. Ibid, p.80
  24. Ahmad Al-Jallad, The Arabic of the Islamic conquests, p.428
  25. Ibid, p.90
  26. Ibid, p.22
  27. Ibid, p.51
  28. Ibid, p.79
  29. Ibid, p.91
  30. Ibid, p.51
  31. Ibid, p.79
  32. Blau and Hopkins, Judaeo-Arabic papyri, 1987,  p. 90
  33. Joshua Blau, A Handbook of Early Middle Arabic, 2002, p.137
  34. Joshua Blau, A handbook of early Middle Arabic, 2002, p.140
  35. Blau and Hopkins, Judaeo-Arabic Papyri, 1987, p.126
  36. Blau and Hopkins, Judaeo-Arabic Papyri, 1987, p. 149
  37. Ibid, p.151
  38. Ibid, p.152
  39. Ibid, p.155
  40. Marijn Van Putten, Quranic Arabic, p.214
  41. Ibid, p.99
  42. Ibid, p.189
  43. Ibid, p191-192
  44. Mishkat al-Masabih 2137, Book 8, Hadith 28
  45. Marijn Van Putten & Phillip Stokes, Case in the Quranic Consonantal Text, p149
  46. Ibid, p16
  47. السبعة في القراءات لابن مجاهد، ص346، تحقيق شوقي ضيف Al-Sabʕah Fil-Qirāʾāt, Tahqiq by Shawqi Ḍayf, p.346,
  48. رسم  المصحف لغانم قدوري ص357 Ghanim Qadduuri, Rasm Al-Mishaf, p357
  49. لغة قريش، مختار الغوث، ص39
  50. Marijn Van Putten, Quranic Arabic, p.158-167
  51. Marijn Van Putten, Hamzah in the Quranic Consonantal Text, 2018, p.101
  52. معاني القرآن للفراء، ج3 ص116، دار المصرية Al- Farrāʾ, maʕāni l-qurʾān, vol.3 p.116
  53. Marijn Van Putten, Hamzah in the Quranic Consonantal Text, 2018, p.94