Is the Qur'an Written in Pure Arabic?
Foreign Words and Poor Grammar Disprove a Quranic Claim


Among Muslim claims for the Qur'an, the purity of the Arabic within the Quranic text is pointed to as a strong proof that the Qur'an was given as direct revelation from Allah. This is understood both in the sense of its being written solely in the Qurayshi-dialect of Arabic (the dialect of Mecca and surrouding areas) and thus free from foreign influences, and also in the sense of being perfect in its grammar and poeticity. The primary proof-text pointed to so as to prove this thesis is Surah 16:103,

"We know indeed that they say, "It is a man that teaches him." The tongue of him they wickedly point to is notably foreign, while this is Arabic, pure and clear." (16:103, Yusuf Ali translation)

Thus, this verse is addressing the challenge which early opponents of Mohammed had made against his preaching and claims of revelation, which was that he was receiving training and assistance in his work from a human guide and instructor. The verse pointedly addresses both of the issues mentioned above when it mentions that the unnamed human source claimed for Mohammed's revelations is "notably foreign", that is, widely known to be unskilled in the Arabic language (or at least the Qurayshi dialect). The proof of Mohammed's revelation is that the Arabic of the Qur'an is perfect and pure Arabic, free of foreign influences and lacking in errors which would suggest an imperfect human source. This claim is represented well by Ali below:

"The wicked attribute to Prophets of Allah just such motives and springs of action as they themselves would be guilty of in such circumstances. The Pagans and those who were hostile to the revelation of Allah in Islam could not and cannot understand how such wonderful words could flow from the tongue of the Holy Prophet. They must postulate some human teacher. Unfortunately for their postulate, any possible human teacher they could think of would be poor in Arabic speech if he had all the knowledge that the Qur'an reveals of previous revelations. Apart from that, even the most eloquent Arab could not, and cannot, produce anything of the eloquence, width, and depth of Quranic teaching, as if evident from every verse of the Book (Cf. 41:44).1

Indeed, the general understanding of commentators at this point, both Muslim and non-Muslim, has been just this: the purity and perfection of the Quranic Arabic proves it to be above human ability to produce or reproduce. The prevailing Muslim view is that the Arabic of the Qur'an is perfect because Arabic is the language of heaven, and thus revelation handed down from Allah to his final messenger Mohammed will necessarily be in flawless Arabic as well. So the question then becomes one of whether or not the Qur'an stands up to scrutiny directed toward this claim.

First, let us examine the purity of the Arabic from the perspective of vocabulary. In his groundbreaking work, Alphonse Mingana demonstrated that rather a large number of words used in the Qur'an, both religious and common, are derived from Syraic, a liturgical language used in Eastern Christian churches around the time of Mohammed (and even today), which originated from the pre-Hellenistic lingua franca, Aramaic. The Syraic influence in the time of Mohammed was strong throughout the Near East, especially in Syria and Palestine, which Mohammed had been to as a caravaner prior to his assuming the prophetship of Islam. Among the religious terms found in the Qur'an which are attributed to Syraic origin 2 can be found (Note - the Arabic terms are undeclined):

Mingana also furnishes a review of non-religious words, uncommon in Arabic but very common in Syraic, which also found their way into the Qur'an from Syraic influences. Included among these are:3

In addition to Syraic, Dr. Mingana also discerned foreign words introduced into the Quranic vocabulary from Ethiopic, Persian, and Greek (all of which would have had representative influences in the Hijaz during Mohammed's time). While Mingana's work is somewhat dated now, and not every one of his claimed Syraic influences on the Qur'an seem plausible today4, his work was an important step in delving the foreign influences upon the Quranic evolution, and in the main, remains foundational to this day.

Anis Shorrosh, a Palestinian Christian and native Arabic speaker, has also detailed a number of foreign words (excluding proper, personal names here) which exist in the Quranic text5:

As such, both Western scholars and native Arabs alike give evidence to a fairly great contingent of foreign words in this supposedly "pure" Arabic revelation. This is in spite of the fact, as Shorrosh points out, that there are perfectly fine native Arabic words which could have been used interchangeably. It is important to note that these words are not loan-words which entered the Arabic language at some distant point in the past, and gradually became genuinely Arabic. Rather, they are recent introductions whose origins are often quite easily traced to foreign influences being felt in Arabia at the time of Mohammed. The existence of these foreign words in the Quranic text would seem to point towards a fair amount of outside influence upon Mohammed during his composition of the Qur'an. Each of these languages were used by groups which had strong commercial, cultural, religious, and military influence over the Hijaz and nearby areas of the Arabian peninsula, so it is unsurprising (especially given the amount of Christian, Judaic, and Parsi influence demonstratably evident in the stories of the Qur'an) that these groups would have influenced Mohammed through their earlier, individual dealings with him.

Concerning the Arabic grammar of the Qur'an, examination has also shown it to be imperfect in places. Rather than transgress into the mechanics of a language in which I am as yet a mere dabbler, I will simply provide a list compiled by others more knowledgeable than myself, and direct the reader to their essay which covers the grammatical points in detail. Rafiqul-Haqq and Newton demonstrate in their foundational essay that the readings given below are errors in the Quranic Arabic grammar, and provide corrected readings based upon accepted rules of Classical Arabic grammar6:

The last of the above corrections is of special interest. Rafiqul-Haqq and Newton's argument on this point involves the use of the word inna in the manner stated by them in their article,

"The word haazaani was declined incorrectly because the word inna in the beginning of the nominal sentence causes a form of declension called "nasb" to the nominative and the "yeh" is the "sign of nasb". This is the third grammatical error.

Having previously considered this correction to be spurious because of the current reading of in that appears in the Qur'an instead of inna, I had abstained from including it on the list above. In response to a Muslim pushback, Rafiqul-Haqq and Newton have provided solid evidence from several classical Muslim sources for the appearance of inna in the older texts of the Qur'an. Their research demonstrates, then, TWO problems with the text of the Qur'an. First, the positive evidence that the present reading of in found in 20:63 is itself a corruption of the text. Second, their original claim of grammatical error in this ayah is vindicated because the older reading demonstrates the original problem which they had previously explained. As such, I have elected to include 20:63 into the list of errors, and invite the reader to review the evidence they provide which is linked above.

Muslim apologists have attempted rebuttal to the above information, usually by starting with the charactre assassination of Rafiqul-Haqq and Newton, and then proceeding on to defence of Quranic grammar through question-begging and the presentation of unsupported grammatical counterarguments. Lest one be tempted to suppose that greater credence be lent to Muslim counterarguers based upon the fact that they are Muslims, and thus supposedly more knowledgable of Arabic, a few things need to be kept in mind. Being a Muslim in no wise means that a person knows Arabic. The large majority of Muslims do not know this language. They may hear the Qur'an read in Arabic every Friday, but this does not necessarily mean they understand what they are hearing - many, perhaps most, do not.

Further, despite the sometime Muslim glorification of the Arabic language, it is a language no different or more difficult than any other - in fact, it is LESS difficult than languages in many non-Semitic linguistic groups (try comparing it to, say, Chinese or German). There is no reason at all why a non-Arab, non-Muslim scholar could not become extremely proficient in Arabic, and thus qualified to speak on topics relating to its grammar.

Also, it must be noted that Dr. Anis Shorrosh, a native Arabic speaker, confirms six of the above grammatical errors as being present in the Quranic text, though unfortunately he does not provide a great amount of in-depth explanation concerning the grammatical reasoning, due to the topical survey style of his book7.

From the above, we can see that the claims to a "perfect, pure Arabic Qur'an" are somewhat exaggerated, to say the least. This hyperbolic claim is typical of much of mainline Islamic treatment of the Qur'an, which tends to hold that the Qur'an was delivered directly to Mohammed from Allah, and that it has not changed an iota since it's revelation. Most traditional Muslim theologians are loathe even to consider that the Qur'an has a textual history behind it, or that it has been altered and edited to any degree at all. However, we can see that the presence of foreign loan-words (despite having perfectly acceptable Arabic counterparts) as well as certain grammatical errors embedded in the text, points to a more mundane point of origin for this book. This would be the logical conclusion both on the basis of the fact that Surah 16:103's claim is disproven, and also from the general supposition that a book handed down from God would presumably be free from grammatical errors.

End Notes

(1) - 'Abdullah Yusuf Ali, The Meaning of the Holy Qur'an, p. 665, note #2143 on Surah 16:103
(2) - A. Mingana, "Syraic Influence on the Style of the Koran", Bulletin of the John Rylands Library, 11 (1927), p. 84-5
(3) - A. Mingana, "Syraic Influence on the Style of the Koran", Bulletin of the John Rylands Library, 11 (1927), p. 87
(4) - e.g. his claim that the Syraic Allaaha influenced the Islamic name Allah has been supplanted by more more recent work demonstrating the native evolution of the name "Allah" in Arabia, as an elidation of the title "al ilah" meaning "the god". Likewise, Mingana's text can be construed as claiming that the early Arabic name for Jesus was "cIso", whereas more recent investigation has demonstrated that the more conventional "Yesu" was in use from the earliest Arabic records.
(5) - A. Shorrosh, Islam Revealed: A Christian Arab's View of Islam, p. 199
(6) - M. Rafiqul-Haqq and P. Newton,
The Qur'an: Grammatical Errors
(7) - A. Shorrosh, Islam Revealed: A Christian Arab's View of Islam, pp. 199-200