The Qurʾānic Subtext of Early Arabic Bible Translations

posted in: Blog | 7

by Miriam Lindgren Hjälm

In 2018, Gabriel Said Reynolds published a valuable reference work with the title The Qurʾān and the Bible: Text and Commentary.[1] In this book, Reynolds accumulates research on what is commonly known as the biblical subtext of the Qurʾān, i.e., Qurʾānic references to the Bible, or rather to the interpreted Bible, in use among Jews and Christians at the time.[2] In short, he argues that substantial portions of the Qurʾān should be studied, not only in light of what comes immediately after it (ḥadīth, Sīrat Muḥammad, tafsīr), but also in light of what comes before it, i.e. the reception of the Bible among Jews and Christians in Late Antiquity.[3]

The Qurʾān thus contains interpretations of an already interpreted Bible and as such it constitutes a now widely recognized treasure of biblical reception. The fact that religious interactions continued to be recorded in the production of sacred texts has received far less attention. As already noted by Sidney Griffith, the rise of the Bible in Arabic in its written form might be an attempt to “set the biblical record straight in Arabic.”[4] The emergence of Arabic translations then may respond not only to a general linguistic shift to this new lingua franca, but also to what can broadly be referred to as taḥrīf, i.e. the Muslim notion that Jews and Christians distorted the original version or meaning of the Bible. It is therefore of great interest to note that the earliest Syriac-based Arabic translations, i.e. those appearing during the long 9th century, often exhibit a notably Islamic/Qurʾānic-sounding language, as if their authors were in dialogue with an audience who knew and valued such references. It seems then that just as the Qurʾān reflects a biblical subtext, early Syriac-based Arabic translations in particular disclose what may be referred to as a Qurʾānic subtext.[5] Just as the audience of the Qurʾān was expected to catch and make sense of biblical references, so were allusions to a Qurʾānic subtext presumed to be meaningfully interpreted by the recipients of these translations.

The function of these references sometimes appear to be of literary character such as the sporadic insertion of the vocative allāhumma “O God!” in an early translation of Daniel.[6] Other references appear to be polemical: the same translation of Daniel uses the word ḥajj in the dedication of the statue that Nebuchadnezzar set up (Dan 3:1–3). Despite the proximity in meaning between ḥajj and its Semitic cognates ḥāg (Heb)/ḥaggā (Syr) “feast,” such a word choice seemingly linked biblical idolatry with the Muslim pilgrimage.[7] Yet other references are allusive, which is the case in an early Christian Arabic rendition of the Joseph narrative (Gen 37:9) where, by subtle additions, the translator makes the biblical rendition very similar to sūrat Yūsuf 4.[8]

Illustration 1: Sinai Ar. 1, fol. 124r. The oldest extant (Syriac-based) Arabic translations of Job, Daniel, Jeremiah including Lamentations, and Ezekiel. Ca. 9th c. © Saint Catherine’s Monastery, Sinai, Egypt. Courtesy of Father Justin.

In addition to using Islamic/Qurʾānic-sounding language and inserting subtle allusions to the Muslim scripture, similar methods of re-interpreting biblical material appear both in the Qurʾān/Muslim tradition and in early Syriac-based Arabic Bible translations. In both cases, they may go back to the Syriac heritage. Take for instance the well-known Muslim tradition that Ishmael and Abraham built the Kabʿa together, which (re)locates the biblical figures to the Ḥijāz. The idea that Abraham and his son were building a sanctuary together was not an originally Muslim idea but traceable to Jacob of Serugh (d. 521), who had Isaac helping his father to build a house.[9] The strategy of relocating events is apparently at play in a Christian Arabic translation of Job (42:17, i.e. the Syriac additions). Here the source text (prob. Syro-Hexapla) identifies the land of Uz as a place on the borders of Idumea and Arabia and Job’s wife as Arabian. Whereas the source text places Job in the southern parts of Transjordan and in Arabia, the Arabic translator moves Job to Damascus and Ḥawrān (modern Syria), and makes his wife a Ḥawrānite.[10] The “identification technique” we see in the Muslim tradition, which (re)locates a biblical narrative closer to the reader’s vicinity, appears to be similar to that detectable in this Arabic translation of Job. This technique probably goes hand in hand with the exegetical endeavor to identify geographical locations in the Bible and is ultimately part of the broader project of making biblical narratives relevant to new contexts and new generations of readers.

Another good example in this translation has been noted by Juan Pedro Monferrer-Sala. In Job 28:22, the Syriac word aḇdānā “destruction” translates the Greek ἡ πώλεια and Hebrew ăbaddōn with the same meaning. In Arabic, however, this word is rendered into al-malāk “the angel.” Monferrer-Sala argues that this non-literal rendition reflects the Jewish-Christian mythology where Abaddōn is the name of the angel of destruction (cf. Rev. 9:11).[11] The Arabic translator thus turned the original wording in Job 28:22 “Destruction and Death say,” into “Death and the Angel said” expecting the audience to understand that the “angel of destruction” was intended here and not an ordinary angel. It is worthwhile pointing out in this context that the Qurʾān sometimes refers to biblical figures by their epithets: Jonah is referred to as dhu-l-nūn “the one with the fish” (sūrat al-Qalam 48) and Saul as Ṭālūth which probably means “the tall one” and refers to the biblical account of Saul as tall and handsome (1 Sam. 9:2). The use of epithets in the Qurʾān may be motivated by style (to make verses rhyme).[12] In any event, in both the Christian Arabic case and in the Qurʾān, this is a homiletic technique that requires an audience submerged in biblical knowledge. Such renditions are often influenced by prevalent interpretative traditions and thus reflect an already interpreted Bible, which in one form or the other must be transmitted together with the holy scripture in order to provide “the full picture.”

In this connection it should be mentioned that early Arabic Bible translations use the Qurʾānic form of Saul, i.e. Ṭālūth and not, as expected, Šā[ʾ]ūl (the common form of the name in later traditions).[13] This word choice hence represents yet another of many examples where early Christian Arabic authors chose to dress their holy scriptures in a language that echoes the Qurʾān.[14]

Illustration 2: The Mingana Christian Arabic Additional 137, Cadbury
Research Library (10th c.) © and courtesy of the Cadbury Research Library.

In sum, especially early Syriac-based Arabic Bible translations occasionally reflect a pre-Islamic Syriac tradition wherein adaptions of biblical material took place. In other cases, the Qurʾānic subtext of these translations attests to vivid religious and literary interactions devised by creative translators who ultimately seemed motivated to appropriate the biblical legacy in an environment where the struggle over divine revelation was at its height.

Miriam L. Hjälm is Assistant Professor in Easter Christian Studies at the Stockholm School of Theology/Sankt Ignatios Academy (2017-). She primarily teaches courses on various aspects of how the Bible was used and transmitted in Eastern Christian traditions. She holds a Ph.D. in Semitic languages from Uppsala University (2015) where she focused on translation techniques in Christian Arabic versions of the book of Daniel. During her post doctoral position in the Biblia Arabica project at Ludwig Maximilian University of Munich (2015-2017), she worked primarily on translational and paleographical aspects of early Christian Arabic Bible translations. Currently, her research, funded by the Swedish Research Council, concerns the perception, use and interpretation of the Bible among early Arabic-speaking Christians, with a special focus on interreligious encounters.


[1]. Gabriel S. Reynolds, The Qurʾān and the Bible: Text and Commentary. Yale: Yale University Press, 2018.

[2]. Such interpretations appear especially in homiletic, liturgical and exegetical material. For more on the topic, see Sidney H. Griffith, The Bible in Arabic: The Scriptures of the “People of the Book” in the Language of Islam. Princeton: Princeton University Press: 2013, esp. 91–96.

[3]. For the relationship between tafsīr and the Qurʾān, see Gabriel S. Reynolds, The Qurʾān and Its Biblical Subtext. London: Routledge, 2010, esp. 3–36.

[4]. Griffith, The Bible in Arabic, 53.

[5]. I.e. as having the Qurʾān and early Islamic tradition in mind, or using the same interpretative strategies as those detectable in the Qurʾān. A larger study on the topic will be found in Miriam L. Hjälm, “The Bible in Muḥammad’s Ḥijāz and the Rise of Early Arabic Bible Translations,” submitted.

[6]. Miriam L. Hjälm, Christian Arabic Versions of Daniel: A Comparative Study of Early MSS and Translation Techniques in MSS Sinai Ar. 1 and 2. Leiden: Brill, 2016, 254. Allāhumma appears several times in the Qurʾān.

[7]. Hjälm, Christian Arabic Versions of Daniel,254.

[8]. Ronny Vollandt, Arabic Versions of the Pentateuch: A Comparative Study of Jewish, Christian, and Muslim Sources. Leiden: Brill, 2014, 189.

[9]. Joseph B. Witztum: “The Syriac Milieu of the Quran: The Recasting of Biblical Narratives.” Ph.D. dissertation submitted at Princeton University, 2011, 168; cf. Reynolds, The Qurʾān and the Bible, 69–70. The Syriac tradition developed several genres where the re-telling of biblical events took place. For instance, dialogue poems were composed wherein psychological dilemmas presumably experienced by biblical figures were put into play and by doing so the listeners were invited into the biblical world, see Sebastian Brock, The Bible in the Syriac Tradition. Piscataway: Gorgias Press, 2006, 81–88.

[10]. See fol. 11a–b in Sinai Ar. 1. Cf. the reproduction and discussion in Juan Pedro Monferrer-Sala: “Liber Iob detractus apud Sin. Ar. 1 Notas en torno a la Vorlage siriaca de un manuscrito árabe cristiano (s. IX),” Collectanea Christiana Orientalia, 2003, 1, 119–42, here 133. His transcription slightly differs from that of the present author.

[11]. Monferrer-Sala, “Liber Iob detractus apud Sin. Ar. 1,” 129–31.

[12]. Reynolds, The Qurʾān and Its Biblical Subtext, 198, 234, 241.

[13]. Cf. Miriam L. Hjälm: “Scriptures beyond Words: ‘Islamic’ Vocabulary in Early Christian Arabic Bible Translations,” Collectanea Christiana Orientalia, 2018, 15, 49–69; Adriana Drint: “Some Notes on the Arabic Versions of IV Ezra and Apocalypse of Baruch in Ms MT Sinai Arabic Codex 589,” Parole de l’Orient, 1999, 24, 165-77, here 171; Alphonse Mingana, Catalogue of the Mingana Collection of Manuscripts, Now in the Possession of the Trustees of the Woodbrooke Settlement, Selly Oak, Birmingham: Vol. 3, Additional Christian Arabic and Syriac Manuscripts. Cambridge: W. Heffer, 1939, 5.

[14]. For more examples, see Hjälm, “The Bible in Muḥammad’s Ḥijāz.”

7 Responses

  1. Blanca Villuendas

    I wonder why Monferrer did not comment over the possibility that the الملاك in BL Or. 1475 could be الهلاك (destruction). I haven’t accessed the manuscript to see whether this reading is supported from the paleography or it could only be considered as a possible corruption in the copying process. But in my humble opinion, this possibility makes much sense.

    • Miriam Hjälm

      I see your point in this case. In general, one need to have a look at the overall translation technique in order to decide whether an extant reading should be understood as a copying mistake or a deliberate interpretation.

  2. Steven Ring

    I am adept in Syriac but not (yet) in Arabic. I wonder to what extent the early Arabic translations of the gospel reflect the 2nd century AD Syriac Diatessaron gospel harmony (i.e. a Syriac gospel tradition older than the Peshitta)? I have gathered a great deal of textual and historical information about the Syriac Diatessaron, so if your research interests include the early Arabic gospel, it is possible that I may be able to help explain where some variant Arabic readings could have come from. From my Syriac based research, the Syriac Diatessaron remained an influential gospel text until the 13th century AD. Thus, it is possible that the Diatessaron may have influenced the early Arabic gospel translations and it is also possible that translations of the Diatessaron were made by Islamic scholars for polemical reasons, (e.g. by the religious commentator Al-Biruni in the 11th century). Best wishes, Steven.

    • Miriam Hjälm

      I definitely believe that there is a close correspondence between the Syriac/Aramaic and the Syriac-Arabic traditions, including the one you mention. Closer cooperation between these two fields is for sure needed. I have not worked much on Arabic gospel translations but if I recall correctly, there is influence from the Diatessaron on certain Arabic translations (cf. Hikmat Kashouh, Arabic Versions of the Gospels, and the comment below). Michael McCoy III and Robert Turnbull, who both posed on this site before me, probably know more than me in this matter. In general, early gospel translations seem to be more literal than early Pauline epistles and OT books, which often display a considerable amount of non-literal traits of various kinds. All my best, Miriam

  3. James David Audlin

    Steven Ring: I am a Bible scholar specializing in Galilean Aramaic, the extremely little-studied pre-Syriac dialect, and I work especially with the Johannine corpus, and I do read Arabic, though nowhere near as well. While I have made no formal study of the matter, I have had occasion to look at a number of these early gospel renderings in Arabic. Without exception my impression is that they are derived from either a gospel harmony like the Diatessaron, and indeed most likely the Diatessaron itself but we must remember that there were others; or else from an early Galilean “fourfold gospels” source, an evangeliarium. The wording is always extremely close to how gospel passages are phrased in, for example, Ephrem the Syrian or Ishodad of Merv, and Ephrem states that he uses those two sources, though he refers to his fourfold gospels source as “the Greek gospel” because of a colophon that these texts universally have that certify that they were back-translated into Aramaic from the original Greek manuscript prepared under the aegis of John the Presbyter, which was on public display in Ephesus for a few centuries. While I admire the efforts of Peshitta primalists, I do not agree; the Peshitta in large measure has been made to conform to the Greek Textus Receptus, and is in my view the latest of these sources. The two Old Syriac texts have in my view an origination independent from the Peshitta, and their Urtext was independently back-translated from the same Greek manuscript, but they too were brought to a degree in later generations of copies into conformation with the Greek and the much later dogmatic imperatives of the organized religion, but less so than the Peshitta. However in the Galilean text-type we find extremely little of such forced conformation; these mss. most notably include the several Palestinian Lectionaries and the Sinaiticus Rescriptus. This article may be of interest to you —

  4. James David Audlin

    Steven Ring, this little passage from the upcoming next (and I hope final) edition of my GOJ may also be of interest.

    A very early and reasonably faithful Arabic version of John 15:23-16:1 is found in Ibn Isḥāq’s Kitāb al’Maghāzī, compiled between 754 and 767. It is notable for predating all other extant translations of the gospels into Arabic, even those done by Arabic-speaking Christians, and as the only one not translated from either a Greek or a Syriac version. Notably, the Arabic transliterates ܡܢܚܡܢܐ rather than using a translation or transliteration of παρακλητος or ܦܪܩܠܝܛܐ. Sean W. Anthony in “Muhammad, Menahem, and the Paraclete” (Bulletin of SOAS, 79:3 [2016]) argues that the source was a Christian Palestinian Aramaic text, but as noted every extant version in that dialect or the Syriac dialect transliterates the Greek. Only the Palestinian Lectionaries have ܐܢܡܚܢܡ, which they must have taken from a Galilean Aramaic original, and so I think Ibn Isḥāq took his text from a Galilean copy of John.

    • Steven Ring

      Hello James, thank you for your interesting posts. There was a native Syriac word used instead of the Greek loan word Parakletos preserved by Isho`dad of Merv: ܡܒܝܐܢܐ which means ‘comforter’. This can be found in Margaret Gibson’s edition of the Syriac text of the commentary on John, page 188 line 5. Concerning your other points: In his Diatessaron commentary, Mar Ephrem refers 5 times to ‘the Greek’. These refer to a Greek translation of the Diatessaron done by Ammonius of Alexandria in c. 200 AD. Ammonius was probably responsible for naming the Greek version of this gospel harmony ‘Diatessaron’ since this name is Greek as you know. From my research it is clear to me that Tatian composed his gospel harmony in Syriac from Semitic language gospel sources. This is clear because of the many Hebrew language fragments preserved by Tatian in his gospel harmony (these could not have come from a Greek source) and even more fragments of Aramaic gospel text which have left no trace whatsoever in the Greek NT, demonstrating that Tatian is not likely to have used any Greek sources. Tatian’s source(s) were either written in the Hebraised Aramaic of Judea (the most likely option) or in Classical Hebrew, or some of his sources were in Hebrew and others were in Aramaic. I also do not accept the idea that the Peshitta gospels are a primary text. However, the Peshitta gospels do retain some traces of earlier (Hebraised) Judean Aramaic gospels. IMO, these traces were most likely inherited by the Peshitta gospels from Tatian’s Syriac gospel harmony. For example, the first word in John’s gospel in the Peshitta is in Hebrew, not in Syriac. If you can provide more information on Arabic Jn15:23 – 16:1 I should be able to comment on possible influence from the Diatessaron. Also, should you wish to continue our discussion, please either email me ( or contact me on Facebook as I do not get notified when posts appear on here. Best regards, Steven.

Leave a Reply