The Significance of the Non-Muslim Evidence for Qur’anic Studies, Part 2

Ibn Warraq

With this article, Ibn Warraq continues the examination of significant figures in the medieval West’s appraisal of Islam that he began in the December 2007/Jan uary 2008 issue. In this installment, he focuses on Robert of Ketton (who probably died in the second half of the twelfth century) and Mark of Toledo (fl.1193–1216).—Eds.


The Abbot of Cluny, Peter the Venerable (1092/94–1156) “was a learned man who recognized frankly that little trustworthy information about Islam existed in Latin and blamed Christian ignorance on the general loss of zeal for the study of languages.” Stimulated by his visits to Cluniac abbeys and priories in Spain in 1141 and 1142, Peter planned an ambitious project to translate Arabic texts into Latin. He picked two translators, Robert of Ketton and Herman of Dalmatia, the latter having translated, for example, the Kitab Nasab al-Rusul by Said ibn Umar and the former the Qur’an and a collection of Judeo-Islamic legends. One of the first works translated was Al-Kindi’s Risala. Peter himself produced a refutation of Islamic doctrine, out of love, not hatred, and using reason instead of heavy-handed rhetoric.

Robert had already been translating scientific works from the Arabic in Spain and is now known for his Latin translation, Liber algebrae et Almucabola, of al-Khwarazmi’s manual of algebra, al-Kitab al-mukhtasar fi hisab al-jabr wa-al-muqabalah. Translating the Qur’an was another matter entirely, but Robert set about it in a dedicated manner. He completed his translation of the Qur’an in 1143 and was well paid for his pains. Then he went back to translating scientific works.

His was the first complete translation of the Qur’an into a Western language, and it became a medieval best-seller. Its accuracy has been challenged ever since. Juan de Segovia [c.1393–1458] objected to the cavalier manner in which Robert had translated and to his redivision of the Qur’an into more than the standard 114 suras. Furthermore, Robert “had moved what was at the beginning of many Quranic passages to the end, and vice versa; he had altered the meaning of Quranic terms as he translated them; he had often left out what was explicitly in the text, but incorporated into his Latin version what was only implicit in the Arabic original.”

Ludovico Marracci likewise found Robert’s effort more of a paraphrase than a faithful translation. In the eighteenth century, George Sale (1697–1736), in the preface to his own translation, wrote that “the book deserves not the name of a translation; the unaccountable liberties therein taken and the numberless faults, both of omission and commission, leaving scarce any resemblance of the original.”

However, Thomas Burman, in a series of lucid and convincing articles, conference papers, and books, has argued that Robert’s rendering is worthy of respect and fares well in comparison with, for example, the more literal translation of Mark of Toledo [fl. 1193–1216]. Burman contends that

there is no denying that Robert was an exuberant paraphraser who simply could not leave well-alone, at least when it came to the Qur’an, and his paraphrasing certainly did do violence to the Arabic text at points. Rather; what I intend to quarrel with is the assertion that simply because his Lex Mahumet is a paraphrase it is therefore a poor and misleading translation. There are several grounds for disputing that assertion, not least because scholars of translation theory and practice—both ancient and modern—have long argued that in many cases literal translations are much less faithful to the original texts than well-constructed paraphrases. But the specific argument that I intend to make here is that Robert compensated for his elaborately paraphrasing approach in a very surprising way: at the same time as he was recasting nearly every sentence that he translated, Robert was also going to remarkable lengths to insure that his paraphrase nevertheless reflected what Muslims themselves thought to be the meaning of the Qur’an. The most vivid signs of this are the numerous passages in all parts of his Latin Qur’an where Robert has incorporated into his paraphrase glosses, explanations, and other exegetical material drawn from one or several Arabic Quranic tafsirs or commentaries.

In other words, Robert’s version often reflects accurately Muslims’ understanding of their own holy book, more so than Mark of Toledo’s literal effort. Burman further shows that Mark of Toledo sometimes turned to Arabic exegetical literature to make sense of that opaque text that is the Qur’an.

It is also interesting to note that Robert rearranged the order of the passages. This is after all the same basic principle employed by Richard Bell nearly eight hundred years later in his famous translation of the Qur’an that came out between 1937 and 1939 and which I shall discuss in Part 3.

It is undeniable that the Qur’an is a difficult text, and all translators have had recourse to tafsirs or commentaries, not to mention lexicons and manuals of rare and difficult words. Even Sale, who showed nothing but contempt for Robert’s rendering, was obliged to smuggle in extraneous exegetical matter to complete his own translation.

It is altogether another matter, however, to decide that by using these commentaries one is any closer to what the Qur’an really means. Robert’s reading may indeed reflect Muslims’ own understanding of their scared scripture, but is this understanding accurate? If Luxenberg’s thesis is anywhere near correct, then the answer to my rhetorical question is “No.”

Furthermore, if, as Gerd Puin once said, one-fifth of the Qur’an makes no sense, and, if the Qur’an is indeed an abstruse allusive scripture that no one has understood, then, surely, a literal translation such as the one by Mark of Toledo, which reflects the obscurities of the original, is also valuable. Mark’s version does not pretend to smooth over the difficulties with arbitrary and sometimes far-fetched glosses or commentaries of the Muslims and thus can teach us perhaps more about the language and syntax of the original. Here is what Mark himself wrote about the style of the Qur’an: “. . . sometimes he [Muhammad] speaks like a crazy man, sometimes however like one who is lifeless, now inveighing against the idolators, now menacing them with death, occasionally indeed promising eternal life to converts, but in a confused and unconnected style. . . .” (Emphasis added.)

To be continued.

Ibn Warraq

Ibn Warraq, Islamic scholar and a leading figure in Qur’anic criticism, was a senior research fellow at the Center for Inquiry. He is the author of many books, including What the Qur’an Really Says (Prometheus Books, 2002) and Which Koran? Variants, Manuscripts, Linguistics.


With this article, Ibn Warraq continues the examination of significant figures in the medieval West’s appraisal of Islam that he began in the December 2007/Jan uary 2008 issue. In this installment, he focuses on Robert of Ketton (who probably died in the second half of the twelfth century) and Mark of Toledo (fl.1193–1216).—Eds. The Abbot …

This article is available to subscribers only.
Subscribe now or log in to read this article.