1. Introduction
While preaching his way through the Book of Joshua, Origen found occasion to liken the New Testament authors to the trumpets that brought down the walls of Jericho. This analogy results in a full list of the New Testament books, with the possible exceptions of Revelation and 3 John.
sacerdotali tuba primus in Evangelio suo Matthaeus increpuit; Marcus quoque, Lucas et Iohannes suis singuli tubis sacerdotalibus cecinerunt; Petrus etiam duabus epistolarum suarum personat tubis; Iacobus quoque et Iudas. addit nihilominus adhuc et Iohannes tuba canere per epistolas suasFootnote 1 et Lucas Apostolorum gesta describens. novissimus autem ille veniens, qui dixit: ‘puto autem, nos Deus apostolos novissimos ostendit’ et in quatuordecim epistolarum suarum fulminans tubis muros Hiericho et omnes idolatriae machinas et philosophorum dogmata usque ad fundamenta deiecit.Footnote 2
Matthew first sounded the priestly trumpet in his Gospel; Mark also; Luke and John each played their own priestly trumpets. Even Peter cries out with trumpets in two of his epistles; also James and Jude. In addition, John also sounds the trumpet through his epistles, and Luke, as he describes the Acts of the Apostles. And now that last one comes, the one who said, ‘I think God displays us apostles last’ [1 Cor 4:9], and in fourteen of his epistles, thundering with trumpets, he casts down the walls of Jericho and all the devices of idolatry and dogmas of philosophers, all the way to the foundations.Footnote 3
Bruce Metzger assessed the importance of this list by noting that ‘it contains together, without mentioning any other books and without making any distinctions, the books that in ad 325 Eusebius would cite as “homologoumena” and “antilegomena” …, and Athanasius in 367 would enumerate as constituting the New Testament’.Footnote 4 However, scholars generally now seem to doubt the authenticity of this passage, both because it would be an unusually early catalogue of New Testament books and because Rufinus of Aquileia, whose Latin translation of Origen's Homilies on Joshua is alone extant, admittedly modified passages in Origen to make them more orthodox according to his later fourth-century standards.Footnote 5 According to Lee Martin McDonald, Origen's list of New Testament books ‘is very likely a creation of … Rufinus in the fourth century’.Footnote 6 The present paper takes up the issue of the authenticity of this Origenic passage by comparing Rufinus' translation of Eusebius' Ecclesiastical History at points where the topic focuses on the biblical canon.
Rufinus' methods of translation have received intense scrutiny; he is generally not trusted very far when it comes to the precise wording of his Greek Vorlagen.Footnote 7 Recently a more positive evaluation of Rufinus' methods has appeared, but this has little to do with trying to defend his accuracy in translation and more to do with attempting to understand sympathetically why Rufinus made certain changes to his base text.Footnote 8 When it comes to passages dealing with the books of Scripture, scholars have often found that Rufinus updates his Vorlage, whether Origen or Eusebius, so that the discussion reflects better the more rigid views at the end of the fourth century and the beginning of the fifth. As Francis Murphy wrote long ago, Rufinus ‘had some rather strong opinions of his own regarding the canonicity of the books of the New Testament, and, as in other matters, did not hesitate to correct Eusebius'.Footnote 9
2. The New Testament in Rufinus' Translation of Eusebius
One of the problems in dealing with this material is the difficulty in understanding Eusebius' own opinion on the New Testament canon, particularly in what he seems to have regarded as his clearest and most straightforward articulation of his views, HE 3.25. I find myself in agreement with Éric Junod's recent assessment of Eusebius' presentation: ‘A la première lecture, cette célèbre notice du livre iii frappe par son caractère systématique. Pourtant plus on l'examine, moins on est sûr de bien comprendre les classifications d'Eusèbe!’Footnote 10 The complexities of this discussion prevent us from entering fully into it here, but suffice it to say that Eusebius here divides Christian literature into three categories of books: the ὁμολογούμενα (universally acknowledged), the ἀντιλεγόμενα (disputed), and the heretical books, to which Eusebius does not here assign a title, though later he will call them παντελῶς νόθα, ‘completely spurious' (3.31.6).Footnote 11 Problematically he has already used the word νόθος in 3.25 to describe the ἀντιλεγόμενα in such a manner that in that passage νόθος constituted either an alternative name for the ἀντιλεγόμενα or the name of a sub-category of the ἀντιλεγόμενα.Footnote 12 At any rate they were not universally received, and so the distinctive adjective ἐνδιάθηκος or ‘encovenanted’ could not apply to them.
2.1 Canon Terminology
How does Rufinus handle these various labels and classifications? In English translations such as contained in the Loeb edition, the adjective ἐνδιάθηκος is rendered ‘canonical’.Footnote 13 Rufinus translates likewise infrequently. The term ἐνδιάθηκος is quite rare in patristic literature, appearing twice in Origen and eight times in Eusebius, all in the Ecclesiastical History, and then in no other author of the fourth and fifth centuries in discussions of the canon.Footnote 14 Actually, one of Eusebius' uses of the term overlaps with one of Origen's uses, since Eusebius is quoting Origen. This passage is at HE 6.25 (= Philocalia 3), where Origen, as quoted by Eusebius, introduces his canon list of the Old Testament with the observation that the Hebrews have twenty-two encovenanted books (ἐνδιάθηκοι βίβλοι). Rufinus' rendering of this as viginti et duo … libri in canone veteris testamenti is in agreement with the Loeb's translation. However, in only one other instance does Rufinus use the word canon to translate ἐνδιάθηκος. At HE 5.8.1, we read in Rufinus' Latin about the ordo canonis divinarum scripturarum, a translation of Eusebius' ἐνδιάθηκοι γραφαί. On other occasions Rufinus renders ἐνδιάθηκος as divina (6.14.1), or in auctoritate (3.3.3; 3.9.5), or certum (implied at 3.3.1), or he sometimes offers no precisely corresponding term (3.25.6; preface to book 6).
On the other hand, the word canon does appear in Rufinus' Ecclesiastical History in passages where Eusebius had not used ἐνδιάθηκος. For instance, at 3.25 where Eusebius introduces his own list of New Testament writings with the description that they are ‘the writings of the New Testament (καινὴ διαθήκη) which have been mentioned’, Rufinus describes the list as ‘the entire canon of the New Testament’. When Eusebius later summarises this discussion at 3.31.6 and speaks there of the ‘holy writings' (ἱερὰ γράμματα), Rufinus speaks of the scripturarum canon. Eusebius introduces the list of Old Testament books supplied by Melito of Sardis as the ὁμολογούμενα, which Rufinus renders in canone (4.26.12). At the introduction to Origen's Old Testament canon list, Eusebius' word κατάλογος comes across into Latin as canon. On one occasion, Rufinus translates Eusebius' term κανών with the Latin term canon, but whereas Eusebius used the word not of a catalogue of books but of a ‘rule’, namely, ‘the Church's tradition of accepting only four Gospels',Footnote 15 Rufinus makes clear that he is speaking about the canon Novi Testamenti (6.25.3; cf. 6.13.3). Rufinus also inserts the word where there is no corresponding term in Greek; for instance, at the end of Origen's list of Old Testament books, where Rufinus adds: ‘With these books concludes the canon of the divine volumes' (6.25.2; cf. 7.25.1). These appearances of the word canon generally give the impression of more established boundaries of scripture, which is appropriate for the era of Rufinus' revision.
2.2 Eusebius' Categories
This impression also comes across in the way Rufinus structures Eusebius' New Testament canon in 3.25 and the terminology he employs there. He introduces it as the ‘entire canon of the New Testament’, which should imply that some of the books on the fringe for Eusebius in the early fourth century would be accepted in Rufinus' early fifth-century context. Indeed, this is the case. What had been ὁμολογούμενα in Eusebius are listed similarly in Rufinus, with the concluding assertion, ‘concerning which no doubt ever existed at all’ (de quibus nulla umquam prorsus extitit dubitatio, 3.25.2). In Eusebius we then get the ἀντιλεγόμενα in 3.25.3 and the νόθα in 3.25.4–5. While the relationship between these two categories might be a little difficult to work out, Rufinus has worked it out. The ἀντιλεγόμενα of 3.25.3 now become those books ‘about which there used to be doubt by some’ (de quibus a nonnullis dubitatum est), and the books listed are basically the same as the ones Eusebius had listed, that is, five of the Catholic Epistles (James, Jude, 2 Peter, 2–3 John), books which were actually in the canon for Rufinus. The category called by Eusebius νόθος is now writings ‘about which there is the greatest doubt’ (de quibus quam maxime dubitatur), and here we have listed the Acts of Paul, the Shepherd of Hermas, the Apocalypse of Peter, the Epistle of Barnabas and the Didache (3.25.4). Some of these writings feature in what Rufinus calls in his Commentary on the Apostle's Creed the ‘ecclesiastical’ books – non-canonical writings useful to the church – namely, the Shepherd and the Didache, if that is what Rufinus means by Duae viae in his Commentary.Footnote 16 When Rufinus says in his translation of Eusebius that these books are doubted, he must refer to doubt about their place in the canon, not their orthodoxy, for ecclesiastical books are necessarily orthodox. But later he will class the Apocalypse of Peter among the apocryfa (6.14.1), a term Rufinus elsewhere uses for heretical works.Footnote 17 The books Rufinus lists at HE 3.25.4, then, contain, in Rufinus' mind, both ecclesiastical books and apocrypha. Possibly, Rufinus retained this grouping from Eusebius not because the books listed formed a homogeneous group but because he felt constrained by his Eusebian Vorlage.Footnote 18 These Greek and Latin categories can be presented in tabular form (see Table 1).
Rufinus thus divides what had been basically two categories for Eusebius (ὁμολογούμενα and ἀντιλεγόμενα) into three categories, acknowledging that there had been some doubt (note the perfect: dubitatum est) about certain of the Catholic Epistles, but nothing like the doubt that persists (dubitatur, present tense) in regard to the Acts of Paul and others. All of these had been ἀντιλεγόμενα for Eusebius, but Rufinus makes a firm distinction. In §6, Eusebius had said about the ἀντιλεγόμενα and νόθα together that ‘these are all ἀντιλεγόμενα’; Rufinus completely omits this summary phrase, signalling again a distinction. Rufinus' presentation is not dissimilar to a strong modern tradition, represented by Bruce Metzger, for instance, of interpreting Eusebius' New Testament canon as encompassing both the ὁμολογούμενα and the better-attested group of ἀντιλεγόμενα.Footnote 19
2.3 The Apocalypse of John
Earlier we noted that basically the same books appear in section three of both Eusebius and Rufinus, that is, Eusebius' first listing of the ἀντιλεγόμενα, before he introduces the term νόθος, and Rufinus' listing of the books about which there used to be some doubt. The lists are the same, except that Rufinus includes here the Apocalypse of John. Eusebius had treated the Apocalypse in a rather odd manner, including it among both the ὁμολογούμενα and the νόθα, each time with the note ‘if it should seem right’ (εἴ γε φανείη, §§2, 4). The second occurrence also includes the observation that some people athetise the book while others put it in the ὁμολογούμενα. For Eusebius, the Apocalypse is either universally accepted or spurious.Footnote 20 Rufinus, apparently, thought Eusebius' presentation made no sense, so he removed the Apocalypse from both of those categories and inserted it into the middle category (§3), the books about which there used to be some doubt. This repositioning of the Apocalypse prompted the early twentieth-century scholar J. E. L. Oulton to observe: ‘Rufinus, if he is unfaithful as a translator, is at any rate more intelligible.’Footnote 21
Eusebius promised a more lengthy discussion of the status and authorship of Revelation (3.24.18), which he delivers at 7.25 in the form of an extensive quotation of the opinion of Dionysius, the third-century bishop of Alexandria. Dionysius first mentions those ecclesiastical writers who have rejected the book, based on the ideas that it is not John's, that the meaning is obscured rather than unveiled (apocalypse), and that the book was actually written by the heretic Cerinthus. But, says Dionysius, these reasons do not persuade him to reject the book. He points out that while the writer calls himself John, he does not claim to be the apostle, and Dionysius asserts that the writer was certainly inspired. Up to this point, Rufinus translates the passage rather faithfully.
Then Dionysius argues in detail that the author of John's Apocalypse could not possibly be the author of the Fourth Gospel and the Catholic Epistle (1 John), due to several reasons: the Gospel and Epistle do not identify their author, but the John of the Apocalypse mentions his own name several times (HE 7.25.9–11); there were many people named John in the first century (7.25.12–16); and the literary style of the Gospel and Epistle clearly binds them together (7.25.17–21) and just as clearly separates the Apocalypse from them (7.25.22–7). All of this discussion against the common authorship of the Apocalypse and the Fourth Gospel occupies about three and a half pages in Schwartz’s edition of the Ecclesiastical History.Footnote 22 Rufinus severely abbreviates this discussion, reducing it to this bland comment:
Dionysius then makes an extended defence of the position that the Apocalypse was without doubt divinely inspired and written by a John, but that it does not seem clear to him that it is by the same John who wrote the Gospel, because the latter never makes mention of his own name nor references himself by name, but the former, who wrote the Apocalypse, makes mention of his own name about three times.Footnote 23 But also making a judgement from the very style of the writing he says that it could have been that in those times there was some other John, one of the saints, to whom God revealed these things.
What to make of this summary on the part of Rufinus? It is a clear case of Rufinus' freedom as a translator, and yet the translation does not distort the views expressed in the Greek, even though the views do not cohere with Rufinus' own or those of the fourth-century Latin church. While it is not a faithful translation, it is a faithful summary.
2.4 The antilegomena and nothaFootnote 24
Eusebius uses the word ἀντιλεγόμενα only seven times, and the verb ἀντιλέγω in reference to religious literature another three times.Footnote 25 A couple of times Rufinus offers no corresponding term (3.3.3; 3.25.6, first appearance), but mostly he alternates between three renderings: (1) books currently in doubt; (2) books received as canonical but about which some doubt had formerly been expressed; and (3) books that are not received as canonical but are used in the church. For the three appearances of the verb ἀντιλέγω Rufinus always opts for the first translation choice, retaining the sense of current doubt regarding the specified books (3.3.5, Hebrews; 3.3.6, Shepherd; 3.24.18, 2–3 John, Revelation).Footnote 26 Twice Rufinus renders ἀντιλεγόμενα according to the second option: at 3.25.3, as we have seen, and at 3.25.6, the summary statement to Eusebius' New Testament canon. In the latter passage, Rufinus describes the ἀντιλεγόμενα as those books ‘about which there had been some objection or hesitation, but since they were accepted by the vast majority of churches, they should be admitted’. The term ἀντιλεγόμενα at other times comes into Latin according to the third translation choice above, as something akin to non-canonical books useful to the church, thus corresponding to what Rufinus called in his Commentary on the Apostle's Creed the ‘ecclesiastical books' (§36). At HE 3.31.6, another summary statement where Eusebius had mentioned three groups – ‘holy writings', ἀντιλεγόμενα, and writings that are ‘completely spurious' – Rufinus instead speaks of the ‘books which are held in authority’ (i.e. canonical), the books ‘which are completely repudiated’ (i.e. apocryphal), and those which ‘held a middle position and were received by churches for instruction alone, not for an indisputable authority’. At 6.13.6, Rufinus says something similar (‘those books which seem not to be accepted by some’) and names Wisdom of Solomon and Sirach. Whereas Eusebius had classified these two books as ἀντιλεγόμενα here, along with Barnabas, Clement, Jude and Hebrews, Rufinus sets off Barnabas and Clement in a separate sentence, and then Jude in its own sentence, and he completely omits reference to Hebrews.Footnote 27
The last time Eusebius uses the term ἀντιλεγόμενα, Rufinus brings it into Latin as apocryfa and names the Apocalypse of Peter (6.14.1), which had been in Eusebius' νόθος category at HE 3.25.4.Footnote 28 In that passage Rufinus had brought νόθος into Latin as writings ‘about which there is the greatest doubt’ (de quibus quam maxime dubitatur), and his use of apocryfa at 6.14.1 to describe one of these same writings amounts to much the same thing.Footnote 29 Elsewhere Rufinus translates νόθος with words signalling doubt or rejection: 3.25.7, de quibus dubitari diximus, in reference to the earlier listing of dubious or spurious books; 3.31.6, penitus repudientur, where the subject is heretical writings; 6.31.1, ficta et aliena ab scriptura prophetica, as a description of Julius Africanus' accusations against the story of Susanna.Footnote 30 The verb νοθεύω appears once in the Ecclesiastical History. At 2.23.25, Eusebius says of James that ‘it is considered spurious' (νοθεύεται), whereas Rufinus' Latin asserts that ‘it is not received by some’ (a nonnullis non recipiatur). In the same passage, Eusebius says that James and Jude ‘have been read in public in most churches', but the Latin translation contends that ‘they are received in almost all churches' (ab omnibus paene ecclesiis recipi).Footnote 31 Here we can see clear adjustments by the Latin translator designed to update the thoughts presented in the older Greek work.
2.5 Summary
We have now gained some general impressions of Rufinus the translator when it comes to the issue of the biblical canon. Rufinus does sometimes adapt Eusebius' discussions of Scripture to his later early fifth-century context, after the boundaries of the canon had become more firm. These adaptations are subtle: the tweaking of a category name, the addition of a word, the slight rearrangement of material. Even while adapting Eusebius, Rufinus does not invent a biblical canon for the earlier historian, nor does he even add or subtract any books in the chief passage where Eusebius' views are found, HE 3.25. Rather, he works within the boundaries imposed on him by his Greek Vorlage, subtly adjusting some of the wording, especially with the result of hardening the categories.Footnote 32
3. The New Testament in Rufinus' Translations of Origen
We turn now to the specific criticisms voiced by Everett Kalin and others regarding Rufinus' translations of Origen on the canon of Scripture. Kalin seeks to establish the untrustworthiness of Rufinus' translations based on three alterations he makes to Origen's comments about certain New Testament books as preserved by Eusebius in HE 6.25.3–14: (1) Rufinus has ‘New Testament canon’ where Eusebius, in the introduction to his quotations from Origen, had ‘ecclesiastical canon’; (2) Rufinus gives the number of Pauline epistles as fourteen, whereas Origen specified no number; (3) Origen's reticence about 2 Peter and 2–3 John, his unwillingness to say more than that some Christians doubt their authenticity, becomes in the Latin translation a stronger approval of these writings with the admission that not everyone agrees.Footnote 33 Kalin does not mention that Origen omits reference to James and Jude in the passage from his Comm. Joh. 5 excerpted by Eusebius, even though these letters could have contributed to Origen's argument in that text that Scriptural writers typically produced few works, and despite his positive attitude towards James and Jude elsewhere. Rufinus also omits reference to them in his translation, resisting the urge to fill in the gaps in Origen's list.Footnote 34
Nevertheless, such evidence compels Kalin to suspect that the other passage in Origen's works containing a list of New Testament books, Homilies on Joshua 7.1, extant only in Rufinus' Latin translation, actually derives more from the translator than his Vorlage. It is not entirely clear what Kalin thinks Rufinus' Vorlage looked like. Kalin doubts the list is Origen's ‘in its present form’.Footnote 35 But which books in particular do not belong to such a list coming from the pen of Origen? What sorts of changes did Rufinus make? These questions Kalin leaves unaddressed. Certainly Origen accepted the fourfold Gospel, Acts, the Pauline letters, including Hebrews, and some of the Catholic Epistles, especially 1 Peter, 1 John, James and Jude.Footnote 36 Meinrad Stenzel wondered whether Rufinus has altered the sequence of the Gospels (Matthew, Mark, Luke, John), but he also realised that the same order appears in Origen's Comm. Joh. 1.4.6.Footnote 37 Origen also accepted Revelation, but Revelation probably did not form a part of this list in the Homilies on Joshua.Footnote 38 Second Peter appears in this list, and Origen's acceptance of it is certainly questionable, as is his acceptance of 2–3 John, at least one of which also appears on this list.Footnote 39 But this list says only that John authored letters, without supplying the number three, as we would expect if Rufinus were trying to update Origen for the fifth century. If we are looking for ways in which Rufinus may have distorted Origen's views on the New Testament canon, we can point only to his possible inclusion of the number two in relation to Peter's epistles.Footnote 40
There are signs that Rufinus exercised some restraint in this translation. First of all, such was his own claim about these homilies: in the epilogue to his translation of Origen's Comm. Rom., he reflects on his translation of the Joshua homilies, saying that he had accomplished these translations – unlike some other of his translations of Origen – ‘literally and without great effort’, that is, without the effort of filling in gaps he perceived in Origen's text or polishing the rough style.Footnote 41 Since Rufinus is elsewhere so candid about the alterations he introduces into his translations of Origen (as noted above), we should not lightly dismiss this claim to literalness, which Rufinus presents as a fault of his own translation work. Second, with regard to our passage specifically, the list of books probably omits the Book of Revelation, though its inclusion would have resulted in the full twenty-seven-book canon that Rufinus endorses. Third, Michael Kruger has pointed out in a recent article that the order of books here, with Acts and Paul coming at the end (mirrored at Hom. Gen. 13.2), diverges from the more traditional sequence that Rufinus presents in his own canon list (Comm. Symb. 35).Footnote 42 Admittedly, this example from the order of the books can be countered by the strange coincidence that the particular order for the Catholic Epistles (Peter, James, Jude, John) is known only in this translation by Rufinus and in Rufinus' own canon list.Footnote 43 Possibly Rufinus has made some adjustment here.
Kalin encourages us to consider whether Origen would have omitted the Shepherd of Hermas from a list of New Testament books; he suggests that its absence here may be due to the translator.Footnote 44 Certainly Origen did hold the Shepherd in high regard; in his Comm. Rom. 10.31, he even asserted its divine inspiration.Footnote 45 But this argument loses some force when one realises that the Commentary on Romans is preserved only in Rufinus' Latin. If Rufinus did not disguise Origen's high regard for the Shepherd there, it is not clear that he would have done so in the Homilies on Joshua. Rufinus himself included the Shepherd in his ‘ecclesiastical’ category of non-canonical books that are useful (Comm. Symb. 36).
Judging by the types of changes Kalin finds in Rufinus' translation of HE 6.25, we could guess that at Hom. Jos. 7.1 the Latin translator might have supplied the number ‘two’ for Peter's epistles and the number ‘fourteen’ for Paul's epistles, though why he did not also add the number ‘three’ to John's epistles remains peculiar. As Kruger says: ‘The vagueness of Origen's list on this point favors its authenticity.’Footnote 46 Such slight adjustments cohere with Rufinus' practice in his other translations. On balance it seems probable that Rufinus' translation of Origen's Homilies on Joshua contains a list of New Testament books because the translator found such a list – very similar to what he provided in Latin – in the Greek copy of the homilies lying before him.
Such a list is unusual in the third century, and unusual in the works of Origen. Origen's practice elsewhere is to admit the doubts about books not universally received, as for example in the comments extracted by Eusebius at HE 6.25. No such acknowledgements of doubt find expression in Hom. Jos. 7.1. Perhaps Rufinus has eliminated any expression of uncertainty about some of the books, but his habit in his translation of Eusebius is to soften the doubts expressed rather than omitting them completely. Probably a better explanation for the manner in which this list is presented, without any hesitation about any of the books, concerns the literary genre. A homily arguably does not provide the appropriate occasion for mentioning such doubts. Metzger reasonably suggests that
in the context of a sermon Origen enumerates writings which had not yet attained universal approval but which might be used perfectly well for the edification of the faithful, whereas in more detailed discussions he customarily differentiates between the two categories of books.Footnote 47
It may be that the feature of this list that we have just observed – the absence of any note regarding the reception of particular books – could explain why Eusebius did not include the passage in his compendium of Origen's thoughts on the canon (HE 6.25). Assuming the authenticity of the list, Eusebius may have omitted reference to it because he either overlooked it or considered it less suitable for his purposes. After all, Hom. Jos. 7.1 offers almost no comment on the status of the books listed, while the passages Eusebius preserves explicitly mention not only the four Gospels, but also that the church accepts these and no others, not only the letters of Peter, but also that one of them is received and the other is disputed. Eusebius found these latter passages more conducive to his purpose of chronicling the reception of writings in the church.
This list does not clearly represent Origen's attempt to produce an exclusive canon list, just as Hom. Num. 27.1.3 does not represent Origen's attempt to limit the Scriptural books without obscurities (i.e. those easy to understand) to Esther, Judith, Tobit and Wisdom of Solomon.Footnote 48 The passage from Hom. Jos. 7.1 might not carry the connotations of Athanasius' list from 367, in which the Alexandrian bishop very explicitly excluded all books not on his list. In other words, Origen did not necessarily know a definitive canon of Christian Scripture. After all, at least one book that Origen consistently regarded as an authentic composition of an apostle – the Apocalypse of John – apparently found no place in the list originally.
4. Conclusion
Our reflections on Rufinus' translational habits have encouraged us to question the grounds for suspecting his work on Origen's Hom. Jos. 7.1. This paper has suggested that Origen did provide a list of New Testament books very similar to the later list of Athanasius, with the possible absence of 2 Peter, 2–3 John and Revelation. If the argument favouring the basic authenticity of this list proves persuasive, then scholars will need to give consideration to this passage in their histories of the canon. The recent dominant view has maintained that lists of Christian Scripture began to appear only in the fourth century with the work of Eusebius and those who followed him, an idea that also plays a significant role in the fourth-century dating of the Muratorian Fragment.Footnote 49 The late dating of that text, though having gained popularity in the wake of the publication of Hahneman's book, has never won a consensus.Footnote 50 Perhaps the theory that the fourth century saw the beginning of the impulse to list the books of the New Testament has trumped some of the data. Origen's list is one piece of that data, and it now deserves renewed attention, for it does not seem that it can be ignored as merely the product of the translator Rufinus.
