Did Christians Write Old Testament Pseudepigraph That Appear to be Jewish?

James R. Davila
St. Mary’s College
University of St. Andrews

(c) 2004, presented in the Apocrypha and Pseudepigrapha Section

International Society of Biblical Literature meeting, Groningen

Did Christians write Old Testament apocrypha and pseudepigrapha in antiquity?  If so, did they always include references to explicitly Christian matters (hereafter, “Christian signature features”) or might they sometimes have written them strictly from an Old Testament, and therefore apparently Jewish, perspective?  On the face of it, there is no obvious reason why the answer to these questions must be no.  Nevertheless, scholars who study the Old Testament apocrypha and pseudepigrapha have often assumed not only that whatever is not obviously Christian is Jewish, but even that a work with only a few Christian signature features which can be excised without serious harm to the flow of thought was also originally Jewish.  This paper sets out, following a suggestion published by Robert Kraft some years ago, to test such assumptions empirically by looking at what verifiably Christian authors actually did when they wrote on Old Testament topics.  I should note that it is a painfully brief summary of much longer, exegesis rich, arguments presented in chapter four of a monograph I am currently writing on the problem of Christian transmission of Jewish apocrypha and pseudepigrapha.  I am grateful to the Arts and Humanities Research Board for a research leave fellowship that, along with a semester of leave granted by the University of St. Andrews, has made the research for this paper possible.  My primary aim is to demonstrate on empirical grounds that the danger of Christian works being mistaken for Jewish ones is real:  Christians could write works that had few Christian signature features that can be excised easily on redaction-critical grounds and they could write works that contained no Christian signature features whatever.

            We know that Christians did write Old Testament pseudepigrapha.  We have various Christian apocalypses attributed to the Old Testament figures Daniel and Ezra, and involving the prophet Elijah.  The Odes of Solomon is an early Christian liturgical work preserved in Syriac, which is pseudepigraphically attributed to Solomon.  We see that Christians wrote Sibylline Oracles from the interpolated material (if that is what it is) in Sib. Or. 1.324-400 and elsewhere, and in the short hymn to Christ cited by Lactantius and preserved as book six of the Sibylline Oracles .  It is now generally agreed that the Ascension of Isaiah , a work of rewritten scripture containing a visionary ascent, was composed by a Christian in Greek in the second century, although it may incorporate an earlier Jewish source.  The Testament of Solomon is a late-antique Christian work in Greek.

            The question then arises:  Did Christians write Old Testament apocrypha or pseudepigrapha with few enough and tangential enough Christian signature features that these features could be mistaken for secondary redactions? and Did Christians produce Old Testament apocrypha or pseudepigrapha without including any Christian signature features at all?   Again, there is no reason a priori why they should not have done either.  Depending on the agenda of an individual writer and the nature and length of an individual work, there may have been little or no reason to include explicitly Christian content.  There may also have been incentives not to include it.  An author writing in the name of an Old Testament character and wishing to convince contemporaries of its verisimilitude might well have avoided anachronistic references to Christian matters.  It is true that Christians sometimes put vaticinia ex eventu in the mouths of Old Testament figures in the name of prophecy (e.g., Vis. Ezra 38; Ascen. Isa . 3:13-31), but this is a risky strategy, and we need not assume it was always followed.

            The fundamental difficulty is, if we, as we must, exclude content as a criterion, how can we prove that an ancient work is a Christian composition?  Even if the work was transmitted with a title that attributes it to a known Christian author, even as part of a larger corpus by that author, one can always dispute the genuineness of the attribution, if for no other reason than the work shows no interest in Christian matters.  If the content is not explicitly Christian, we can only be certain that the work is by a Christian author if it is attributed to a named and well-known writer and the attribution is secure on other grounds, such as it being an integral part of a corpus (for example, one of a series of sermons on a biblical book), it conforming closely to the style of the other works of that writer, or it being mentioned or even quoted by contemporaries or near contemporaries as being by that author.  I have found no works whose authorship is this secure which lack Christian signature features entirely, although, as we shall see, one sermon by Augustine of Hippo comes extremely close.

            Another approach would be to find long works by Christian authors (whether anonymous or attributed to specific writers) which contain Christian signature features, but which also include episodic stretches of narrative which lack such features.  Evidence of this sort would establish that Christian authors sometimes told whole Old Testament stories without making reference to Christian matters, although it would not prove that they wrote complete works without doing so.  A study of these episodes would also shed light on how Christian authors interpreted and reformulated Old Testament stories and might give us a better understanding of the range of exegetical possibilities we can expect to find in Old Testament apocrypha and pseudepigrapha by Christians.  I analyse extended passages from three works of this kind below.

            A third approach would be to find anonymous (or pseudonymous) works on Old Testament subjects which lack Christian signature features but which can be argued to be of Christian origin on other grounds.  I shall consider one case of this type.

            First I look at two Christian sermons on Old Testament subjects whose Christian features can be redacted away without difficulty.  John Chrysostom, the fourth-century, Greek-speaking, priest of Antioch and Bishop of Constantinople, was one of the most prolific writers of the early church.  Among his many works are about one hundred fifty homilies or sermons on Old Testament topics, including two series of nine and sixty-seven, respectively, on Genesis.  The homily discussed here is from the second series, which may be the first set of sermons he delivered, while still a deacon.  Sermon number 64 in the series has a summary of Gen 41:46-49 as its rubric and it covers the story of Joseph from this point up to 45:24.  It consists of approximately 5500 Greek words.

            It is remarkable how much of this long homily consists of straightforward paraphrase and basic exegesis.  Chrysostom is also concerned to explain difficult details of the biblical text, and to speculate on the inner states of the characters, especially the motivations and strategies of Joseph.  He aims throughout to show how God guided events inexorably toward the fulfilment of Joseph’s dream.  There is some hermeneutical material, but it consistently derives its lessons narrowly from the Old Testament stories and does not attempt to Christianize the instruction until the very end, when Joseph’s attitude is held up as an example of the New Testament precept to love one’s enemies.  But this brief passage can be excised without interrupting the flow of thought and, under other circumstances, could easily be argued to be a secondary addition.  Likewise, the sermon concludes satisfactorily if the Christian doxology in the last sentence is deleted as a gloss.

            Augustine of Hippo preached an enormous number of sermons in his lifetime and more than five hundred of these survive in written form, most of them transcribed at the time of preaching by stenographers.  Sermon number 48 was preached at Carthage, apparently in May of a year late in Augustine’s life, perhaps 420.  It is brief for one of his sermons (about 1450 words, which the translator, Edmund Hill, suggests would have taken fifteen to twenty minutes to present).  It is a philosophical reflection on the question of why God allows the wicked to prosper, using Micah 6:6-8 and Psalm 72 (EV Psalm 73) as its texts. 

            This is a fascinating, and admittedly very rare, case of an ancient sermon whose attribution to a named and well-known Christian author is not in doubt, yet which does not refer once to a single explicitly Christian doctrine or quote from the New Testament or any other early Christian literature.  It is not a long sermon, but neither is it so very short.  In it Augustine speaks from within the world of the prophet and the psalmist and draws from their words only the lessons that he imagines they intended.

            The closest we come to a sure Christian signature feature is the line “But what is now in hiding, afterwards shall be in evidence.”  This alludes to the Synoptic saying of Jesus which appears in Luke 8:17 as “For there is nothing hidden that shall not be made evident.”  The adverbial phrase “in hiding” (in occulto ) is similar to the noun “hidden” (occultum ) in the Latin Vulgate and the adverbial phrase “in evidence” (in manifesto ) is similar to the verb “be made evident” (manifestetur ).  On the one hand, that Augustine is alluding to the Lucan passage can scarcely be doubted, given the overall context.  On the other hand, a scholar wanting to defend the Jewish origin of an Old Testament pseudepigraphon that contained this allusion to a New Testament passage could point out that the verbal similarity is not close Ð it involves only two word roots, neither in the same grammatical form as the New Testament verse Ð and the general sentiment is congenial to Jewish eschatological thought.  For example, the futility of hoping that one’s sins will be hidden at the final judgment is a theme of the Epistle of Enoch (1 Enoch 98:7-8; 104:7-8).  The apparent allusion could be a coincidence or it could be a scribal assimilation of the passage to the Lucan verse during copying.  Neither argument could be dismissed out of hand.  Early Christianity and ancient Judaism shared many themes and ideas and often these were expressed independently in similar language.  The parable of the sower in Mark 4:3-9 in the Latin Vulgate version and the comparison of human beings with the seeds sown by the farmer in 4 Ezra 8:41 make similar points and share some verbal similarities, yet they are completely independent.  Often we simply cannot be sure whether an apparent allusion to a New Testament passage is one in fact, and this particular one is sufficiently general in theme and dissimilar in wording for one Ð again, in another context Ð to regard it with scepticism.

            All this being the case, I was tempted to categorize this sermon as an example of a work on the Old Testament by a named Christian author which contained no Christian signature features, but I have kept to the route of prudence and accepted the phrase as an allusion rather than an echo, and therefore as a signature feature.  Nevertheless, it should be underlined that if this level of Christianization and no more were present in an unattributed narrative having to do with the Old Testament, it is virtually certain that some scholars would argue that it was a Jewish pseudepigraphon.

            Ephrem the Syrian was born in Nisibis in the first decade of the fourth century.  Late in life he moved to Edessa, and he is remembered for his time there as a zealous opponent of heresies.  He died in 373 while coordinating a relief effort for the poor of the city during a famine.  He appears to have written commentaries on most of the books of the Bible, although only those on Genesis and Exodus survive today and only the one on Genesis comments on the entire text.  Ephrem’s commentaries generally focus on straightforward exegesis of the text and references to explicitly Christian matters tend to be few and far between.  There are whole episodes that lack any Christian signature features.  I comment here on three such episodes:  Noah’s Flood, the story of Joseph, and the early life of Moses (the last covering Exodus 1-4).

            As with Chrysostom’s sermons, Ephrem’s commentaries show a strong interest in basic biblical exegesis and the explanation of difficult passages.  They being commentaries rather than sermons, they shows no particular hermeneutical interest.  The high density of parallels to Jewish, especially rabbinic, exegesis is striking.  It seems likely that Ephrem had access to Jewish targums or midrashic works and he drew on them frequently.  In addition, a number of times he alludes to Jewish customs or ideas of which he would have disapproved in the present, but which he presents in a positive light in their biblical context.  In the Joseph story Tamar longs to participate in the blessings of the circumcised Hebrews and is self-conscious of her own origin among the uncircumcised.  Likewise, Sephora is condemned for preventing the circumcision of Moses’ second son.  And in her prayer, MosesÕ mother refers to the line of the Hebrews as the seed that God blessed.  Christians had access to Jewish exegesis and were capable of drawing on it freely.  Allusions to Jewish exegetical traditions are by no means proof of Jewish authorship.  And Christians could make positive references to Jewish institutions and Jewish ethnicity if these were viewed in the context of the Old Testament period.

            A considerable corpus of epic poetry on biblical themes was produced in Latin in late antiquity.  Following Reinhart Herzog, we can divide it into three groups.  The first can be labelled “canonical” and consists of works by Juvencus, Sedulius, Arato, and Avitus.  The second consists of “noncanonical” works, those produced by named authors which survive only in a narrow manuscript base.  Third are the noncanonical and pseudepigraphic works on various New Testament and Old Testament subjects.  It is the third category that is of interest in this paper.  The pseudepigraphic Latin epics on Old Testament themes seem mostly to have been written in or around the fifth century.  Herzog believes that, whether or not their authors were known originally, once they had been rejected from the canon they circulated anonymously until they were collected together in the Carolingian era, when they were assigned to well-known pre-Ambrosian prose authors.

            The Heptateuchos is a five-thousand-line epic (about 32,500 Latin words) which paraphrases the Pentateuch, Joshua, and Judges.   Apparently the original work also included other Old Testament prose books.  Some manuscripts assign the work to a Cyprianus, but nothing more is known about the author and Herzog is probably correct in concluding that this is a pseudepigraphic attribution to Cyprian of Carthage.  It is generally agreed on various grounds that the Heptateuchos was composed during the fifth century, probably in Gaul.  It is an extended paraphrase of the Bible in ornate Latin poetry, and references in it to explicitly Christian matters are few and far between.  Space does not allow for a complete analysis, so I will limit my comments to the section on Leviticus , which consists of 309 lines (about 2000 Latin words).

            The internal evidence of the Leviticus portion of the Heptateuchos confirms the external evidence and leaves little doubt that it is a Christian work.  It is a careful summary of the contents of Leviticus, written with an astonishing sympathy for the material from a Christian author, but it ignores all references to sacrifice, sabbath, festivals, and the eternal nature of the Levitical laws.  A Jew might pass over some of these things, but it is hard to imagine one ignoring all of them.  Much of what is covered is of potential interest to Christians, a few passages may be doctored to make them more relevant to Christians, and the single use of the word “Jew,” in the retelling of story of the blaspheming half-breed Israelite in Leviticus 24, is quite uncomplimentary.

            All this granted, the Leviticus poem is an extraordinary and in some ways disquieting example of a Christian composition with an Old Testament theme.  It does not include a single Christian signature feature; it presents Jewish law and ritual in a very sympathetic light and offers criticism of them only obliquely, by omission or minor modification.  If it had been removed from its context and transmitted independently, one could envisage it being misread as a Jewish pseudepigraphon.  Indeed, one commentator, Jacques Fontaine, was tempted to read the Heptateuchos as a Jewish work.  Only a close and subtle reading of the Leviticus poem, combined with other information such as the Christian references elsewhere in the Heptateuchos and close readings of other parts of the whole poem, demonstrate its Christian origin and agenda.  It is thus a useful worst-case scenario:  an extended portion of a Christian work on the Old Testament which does not betray its Christian origin by any Christian signature features.  Any Old Testament apocryphon or pseudepigraphon that we suspect to be of Jewish origin should have its evidence measured against it.

            The last document to be considered is another Latin epic poem, De Martyrio Maccabaeorum , a work of 394 lines (about 2800 Latin words) which has never been translated into a modern language.  In the earliest surviving manuscript of the eighth centuries it is anonymous; the manuscripts of the tenth and eleventh century attribute it to “Hilariii,” although later tradition also ascribed it to “Marii Victorini Afri.”  It seems reasonable to assume that it circulated anonymously until it was attributed pseudepigraphically to a known author by Carolingian editors.  The date of composition is not certain but is probably roughly the same as the other Latin biblical epic pseudepigrapha, that is, sometime in the fourth to seventh centuries.

            This poem is based on the story of the seven martyrs and their mother found in 2 Maccabees 6-7 but reworks it very freely.  Although the surviving manuscripts associate the poem with the Christian feast day of the Maccabees (1 August), the work itself has no Christian signature features at all.  All other Latin biblical epics of this period contain such features, although sometimes very few of them, and can safely be accepted as Christian compositions.

            The closest one might come to finding a Christian signature feature is the repeated mention of “saints” (sancti ) throughout (lines 25, 62, 318, 358, 394), but this usage is found also at times in ancient Judaism.  On the other hand, there is a number of elements that might be taken at first glance to be Jewish signature features, most of them in the speeches of the mother.  She exhorts her sons to be mindful of their race, Israel (lines 92, 168, 286-88, 368, 375), and the holy seed of Abraham (lines 167-68); the people to which they were born (line 341); and their native land (line 375).  She also tells one son “to remember God and to hold onto the Law, not given by the king, but which God, the Judge of the globe and the Master of the world transmitted and accepted the prayers of Moses and gave it to the people” (lines 247b-50a).  In line 32 she also says “It is certain that by virtue with the Law it is given to me to conquer.”

            Nevertheless, a careful reading greatly weakens the case for taken these features as indications of a Jewish origin.  First, most are put in the mouth of the mother and are the sort of things an ancient Christian author would have expected her to say and such an author would regard them to have been valid in the mother’s time, before the coming of Christ, but not in the author’s time.  The narrator also refers to her people as “the holy people” (line 3) and “the people of God,” but according to Christian theology the Jewish people were both during the Old Testament period.  Second, the treatment of the Law is very generic and, indeed, is superficial.  A Christian author would agree that the Law was revealed to Moses and was valid at that time, although it was superseded by Christ.  Naturally, the last point would not be put in the mouth of the mother and the author evidently sees no need to make it either.  More tellingly, unlike either 2 Maccabees (6:18, 21, etc.) or 4 Maccabees (5:2 etc.), this version of the story never makes clear that the sin the martyrs resisted and died rather than commit was to eat pork that had been used in an idolatrous sacrifice.  Rather, we are told merely that the king required the mother to change her customs and adopt other ones (lines 11-12) and she urges her children to keep her customs (60) and precepts (85) and to hold onto the Law (247).  The characters and the narrator show a curious lack of interest in the specifics of the situation.  None of them ever make clear exactly what the king is demanding that the children do or the mother is urging them not to do.  This omission would be very strange in a Jewish work but it makes perfect sense in a Christian one.  The author wanted to encourage the audience to value martyrdom but did not wish to distract them with the fact that these martyrs died for a Jewish food law no longer required of Christians and looked down on by them.

            In addition, Dieter Kartschoke argues for direct dependence of De Martyrio Maccabaeorum not only on 4 Maccabees , which would not affect my case one way or another, but also on a sermon by Gregory of Nazianzus, whose presentation of the mother’s exhortations to the sons is strikingly similar to lines 52-56, 370 of the poem.  Kartschoke also argues that the “Freude in Leiden” theme of the poem is both decisively Christian and also characteristic of Gregory’s treatment of the story.

            The content of De Martyrio Maccabaeorum makes best sense if it is interpreted as a Christian composition.  In addition, it was transmitted as part of a corpus of epic texts by or attributed to Christian authors and was taken, at least by the Carolingian editors, to be itself a Christian composition.  If this is correct, the work is a remarkable example of a Christian writing that retells an Old Testament story without including any Christian signature features.

            In conclusion, this paper has given us a considerably more sophisticated and empirically based understanding of how ancient Christians used the Old Testament when they wrote about it.  It gives us good reason to believe that they may have written books about the Old Testament without alluding explicitly to Christianity and has also suggested some strategies for identifying such books as Christian nonetheless.  It has eliminated some of the traditional criteria for establishing Jewish authorship Ð such as extensive use of Jewish exegesis and rare, peripheral mention of Christian matters Ð by pointing to these features in indubitably Christian texts.  Much of the evidence has force not just in itself, but also a minori ad maius :  if we find Christian sermons, commentaries, and poems whose explicit purpose was to edify a Christian congregation or readership, which nearly entirely ignore explicitly Christian matters when dealing with the Old Testament, how much more might we expect anonymous authors of parabiblical Old Testament narratives and revelations to ignore such matters when they hoped to persuade their audience of the genuineness of the story or the revelation.


  1. The questions:  Did Christians write Old Testament apocrypha or pseudepigrapha with few enough and tangential enough Christian “signature features” that these features could be mistaken for secondary redactions? and Did Christians produce Old Testament apocrypha or pseudepigrapha without including anyChristian signature features at all?

Robert A. Kraft, “Setting the Stage and Framing Some Central Questions,” JSJ 32 (2001):  371-95, esp. p. 389

  1. Christians did write Old Testament pseudepigrapha:  apocalypses of Daniel, Ezra, Elijah; Odes of SolomonSibylline Oracles Ascension of Isaiah

  1. Methodological issues

  1. Christian works with few, easily redacted away, Christian signature features

            John Chrysostom, sermon 64 on Genesis (the Joseph story)

            Augustine of Hippo, sermon 48 on the Old Testament (on Micah 6:6-8 and
Psalm 72 [EV 73])


  1. Christian works with episodic units that lack Christian signature features entirely

            Ephrem Syrus, Commentaries on Genesis and Exodus

                        Noah’s Flood

                        The Joseph story

                        The early life of Moses (Exodus 1-4)

            The Heptateuchos , section on Leviticus

            Rudolph Peiper, Cypriani Galli Poetae Heptateuchos (CSEL 23; Prague,

                        1881), 1-211, esp. (Leviticus ) pp. 104-15

            Reinhart Herzog, Die Bibelepik der lateinischen Spätantike:  Formgeschichte einer

                        erbaulichen Gattung (Theorie und Geschichte der Literatur und der

                        schönen Künste, Texte und Abhandlungen 37; Munich:  Wilhelm Fink,


  1. A probably-Christian work that lacks any Christian signature features

            De Martyrio Maccabaeorum

                        Peiper, CSEL 23, 240-54 (cf. 255-69)

Contact details

St Mary’s College
The School of Divinity
University of St Andrews
South Street
St Andrews
Fife KY16 9JU
Scotland, United Kingdom

Tel: +44 (0)1334 462850 
Fax: +44 (0)1334 462852