Hostname: page-component-74656f9884-7drmj Total loading time: 0 Render date: 2026-11-04T05:31:28.671Z Has data issue: false hasContentIssue false
  • English
  • Français

Clement of Rome in Syriac Biblical Manuscripts

Published online by Cambridge University Press:  06 October 2026

Dan Batovici
Dan Batovici*
Affiliation:
University of Vienna; danb76@univie.ac.at
Rights & Permissions [Opens in a new window]

Abstract

This article revisits the evidence for the inclusion of Clementine writings in Syriac Biblical manuscripts—1 and 2 Clement, the two epistles De virginitate, and the Eight Books of Clement—introduces a series of new witnesses, identifies a likely cause for this inclusion (medieval rather than early Christian), and contends that, despite current assumptions in the field to the contrary, this is a case where the later, pseudepigraphic works influence and determine the reception of the older, “genuine” writing, rather than the other way around.

Keywords

Clement of Romereceptionparatextthe Eight Books of ClementDe virginitateJacob of Edessa

Information

Type
Article
Information
Harvard Theological Review , Volume 119 , Issue 3 , July 2026 , pp. 405 - 427
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2026. Published by Cambridge University Press on behalf of the President and Fellows of Harvard College

Introduction

Addressing the fact that the two letters De virginitate attributed to Clement of Rome are found in the fifteenth-century Syriac New Testament manuscript Amsterdam, Allard Pierson Reading Room ms. III H 25—the only such witness known at the timeFootnote 1 —J. B. Lightfoot suggested in 1869 that “[t]his is possibly due to a mistake. A Syrian transcriber, finding the ‘Two Epistles of Clement’ mentioned at the end of some list of canonical books, might suppose that the two letters with which alone he was acquainted were meant, and thus assign to them this quasi-canonical position in his ms.”Footnote 2 Studying afresh the broader reception of Clementine literature in medieval Syriac biblical manuscripts both confirms and disproves Lightfoot’s assessment. The canonical list could perhaps be identified in the form in which the 85th Apostolic Canon seems to have circulated in the West Syrian context, yet the emerging picture is more complex than the mere blunder of an isolated scribe. If this was a mistake it was one which was repeated from the thirteenth century on in at least eight other witnesses that have surfaced after Lightfoot’s passing. This article revisits the evidence for Clementine writings in Syriac Biblical manuscripts—1 and 2 Clement, the two letters De virginitate, and the Clementine Octateuch—and their context, introduces a series of new witnesses, and contends that, despite current assumptions in the field to the contrary, this is a case where the later, pseudepigraphic works influence and determine the reception of the older, “genuine” writing, rather than the other way around. As such, instead of the persistent reception of one important early Christian text and its satellites, these data point to the generative reception of an early Christian character—in this case ܩܠܝܡܝܣ ܬܠܡܝܕܐ ܕܦܐܛܪܘܣ ܫܠܝܚܐ, “Clement, the disciple of Peter the apostle.”Footnote 3

To that end, in the following a series of witnesses will be discussed, aiming to understand better the presence of Clementine writings in these biblical manuscripts. If, in Greek or Coptic, connecting Clementine books with biblical texts are isolated events,Footnote 4 in the West Syrian milieu this is a more developed phenomenon with a significant number of late medieval witnesses: the epistles De virginitate are copied in seven New Testament manuscripts starting in the thirteenth century, 1 and 2 Clement are copied in one twelfth-century New Testament manuscript, and the composite work known in Syriac as the Eight Books of Clement is also copied in several biblical pandects (manuscripts collecting the Old and the New Testaments, in varying configurations) beginning in the twelfth century.Footnote 5

Catholic and Liturgical: 1 and 2 Clement in Cambridge University Library Add. 1700

Cambridge University Library Add. 1700 is a Syriac New Testament manuscript copied in Edessa in 1169/70Footnote 6 in which, after the six catholic epistles, 1 and 2 Clement are included (the former with the superscript title “the catholic epistle of Clement, the disciple of Peter the apostle, to the church of the Corinthians”), followed by the fourteen Pauline epistles. This collection does not include Revelation.

The colophon describes the whole collection as “the book that is full of life of the gospel, and of the acts of the holy apostles and the two epistles of Clement, along with the teaching of Paul the apostle according to the recension (ܬܘܪܨܐ) of Thomas of Ḥarqel,” i.e. the Syriac version completed in 615/616 and characterised by a marked formal equivalence.Footnote 7 This description is certainly accurate for the New Testament texts in the manuscript. However, given that “the text of the Clementine letters shows a far greater degree of wordiness by comparison,”Footnote 8 the translation of 1 and 2 Clement is normally assigned to a different Syriac translation context: the circle of Jacob of Edessa in the second half of the same seventh century.Footnote 9

The collection is also liturgical (meaning it has designated liturgical readings embedded in the continuous text), and the index of lections preserved at the beginning, while incomplete, divides it in three corpora, each with its own continuous numbering of lections: 1) the gospels, 2) the “acts of the holy apostles”, and 3) the Pauline section. To the gospels section is appended a Passion Harmony,Footnote 10 and the middle corpus—which contains Acts, the seven catholic epistles, and 1 and 2 Clement—bears at the top of the reading table on fol. 7r the following general title: “table of the lections which are in the acts of the holy apostles.” A similar title is then appended to the end of the table of lections from this section (that includes readings not only from Acts, but also the seven catholic epistles, and 1 and 2 Clement), on fol. 8v: “finished to write the table of these lections which are in this book of acts, that came before.”

The liturgical apparatus of Cambridge Add. 1700 thus includes 1 and 2 Clement. Even though Wright notes that “only the first leaf remains” (fol. 7r-v) of the table of lections from the “book of the acts of the apostles,”Footnote 11 upon examination in situ it was possible to verify that the last part of this section is in fact still extant on fol. 8r and 8v, ending at the bottom of the first column. Fortunately, this includes the list of readings from 1 and 2 Clement: the last three entries on the second column of f. 8r and the first twelve entries on the first column of f. 8v display the titles of the fifteen Clementine lections that are dependent on the date of Easter, in chronological order. These entries show that, starting with the twenty-fifth Sunday after the Resurrection through the thirty-ninth Sunday after the Resurrection excerpts from the two letters would have been read out on fifteen consecutive Sundays, broadly organised around the themes of obedience, love, humility and concord.Footnote 12

The presence of 1 and 2 Clement in this Syriac New Testament collection of the twelfth century is usually taken as a “remarkable example of a late appreciation of the letters of Clement in terms of canon history.”Footnote 13 Lightfoot, who had described the presence of 1 and 2 Clement in the manuscript at length, concludes: “As the result of these facts, it will be evident that, so far as regards the scribe himself, the Clementine Epistles are put on an absolute equality with the Canonical writings. Here for the first time they appear, not at the close of the volume, as in A [i.e. Codex Alexandrinus], but with the Catholic Epistles—the position which is required on the supposition of perfect canonicity.”Footnote 14 Nevertheless, he proceeds then to qualify this statement by pointing out that this particular manuscript is an isolated witness in this regard among contemporary Syriac New Testament manuscripts of that century, and even among its contemporaries copied in Edessa,Footnote 15 speaking against a broad acceptance of the Clementine letters as canonical. I would note, however, that the standing of 1 and 2 Clement here—however conceivedFootnote 16 —should probably be extended, beyond the “scribe himself,” to the individual who commissioned the production of the manuscript in this form, the liturgical compiler that selected the readings from the Clementine letters, and perhaps the immediate context in which the liturgical cycle of Cambridge University Library Add. 1700 would have been performed. Whether the immediate context is a community or a more private setting is unclear, but the sum of these actors can already be conceived of as a community in a broader sense.

There may be another degree of separation in the manuscript itself: if the general title of the second liturgical section of the manuscript, comprising Acts, the six catholic epistles and 1 and 2 Clement, is “the acts of the holy apostles” (both in the superscript and subscript titles, on fol. 7r and 8v), the description of the contents of this collection in the colophon on fol. 216v at the same time sets the Clementine letters apart “this book is full of life of the gospel and of the acts of the holy apostles and the two epistles of Clement, along with the teaching of Paul the apostle.” The formulation is conceivably still tripartite: the book 1) of (-ܕ) the gospels, 2) of (-ܕ) the acts of the apostles and (-ܘ) the Clementine letters, 3) along with (ܥܡ) the teaching of Paul. However, the letters of Clement are at the same time markedly mentioned in the colophon separately from, or as a separate part within, “the acts of the holy apostles.”

To briefly conclude, 1 and 2 Clement are part of the New Testament collection in Cambridge University Library Add. 1700 in all meaningful ways: they are included in the collection as well as in its liturgical cycle, receiving the same scribal treatment as the other texts, being grouped with the catholic epistles before the Pauline corpus and, in the case of 1 Clement, being explicitly identified as a catholic epistle. There is little to suggest some separation from the other books in it, except the different translation source for the Clementine letters, and perhaps their separate mention in the colophon. However, should one be inclined to see this as indication of canonical status for 1 and 2 Clement in this manuscript, that status is not accurately explained by a strict, clear-cut, post-Reformation understanding of the “canon” as a closed list with broad theological, exegetical, and practical implications. Thus, the inclusion of 1 and 2 Clement in such a manuscript should be further assessed in its adequate context—medieval West Syrian manuscript culture—questioning in particular whether this is due to a medieval dynamic (the inclusion of works attributed to Clement of Rome in West Syrian biblical manuscripts), rather than one of early Christianity (i.e. something akin to Codex Sinaiticus).

3 Clement, 4 Clement: The Epistles De virginitate in Syriac New Testament Manuscripts

The two epistles De virginitate attributed to Clement of Rome (henceforth 3 and 4 Clement for brevity) seem to have been composed in Greek but only a number of excerpts survived in this language, last edited in 1913,Footnote 17 retrieved from the unmarked quotations identified in Pandecta scripturae sacrae of Antiochus of Mar Saba.Footnote 18 Another series of fragments from 3 Clement are extant in Coptic,Footnote 19 but the complete text has only survived in Syriac.

Notably, Lightfoot’s suggestion, quoted at the beginning of this essay, that the inclusion of 3 and 4 Clement in a fifteenth century New Testament manuscript might be due to a mistakeFootnote 20 was based on the only Syriac witness of these letters known then to western scholars: olim Amsterdam Remonstrant Seminary ms. 184—now Amsterdam, Allard Pierson Reading Room ms. III H 25. Since then a good number of witnesses have emerged,Footnote 21 but they are still to be included in future scholarly treatment of these letters in Syriac.Footnote 22 The following discusses the manuscript known to Lightfoot, introduces briefly the other witness known to date, and brings to the fore the types of collections (mostly New Testament) in which 3 and 4 Clement are included.

A. The First Witness: Amsterdam, Allard Pierson Reading Room ms. III H 25 (15th cent.)

Brought to Amsterdam seemingly from Aleppo in 1752 by one James Porter, ms. III H 25 was copied around Mardin in 1470Footnote 23 and is the second volume of a two-part codex (or the second half of one mutilated codex) of the New Testament. The Clementine letters it contains were published in the same year by Wettstein (who argued for their authenticity) in the second volume of his edition of the Greek New Testament, at the end of the book and from right to left, in the Syriac order of pages.Footnote 24 This manuscript is also the sole witness used for the latest edition of 3 and 4 Clement, published in 1856 by Beelen (who also argued for their authenticity).Footnote 25

The first part of the codex would have contained the four gospels, and the second contains the Peshitta textFootnote 26 of Acts, of the so-called “major” catholic epistles (James, 1 Peter, 1 John), and of the 14 Pauline letters, then the Philoxenian text (the version of 507/508)Footnote 27 of the so-called “minor” catholic epistles (2 Peter, 2 and 3 John, Jude), followed by 3 and 4 Clement.

The “minor” catholic epistles were not part of the Peshitta translation (nor was Revelation),Footnote 28 but for the purposes of this collection they were added from the Philoxenian recension after Paul for comprehensiveness, as the superscript title of 2 Peter written in red ink at the top of fol. 145v indicates: “We add to these epistles of Paul those epistles of the apostles that are not found in all manuscripts; first, the second letter of Peter.” This is then followed by 2 John on fol. 148r, 3 John on 148v, James on 149r and, in the same outlook (i.e. one line left empty, then red ink superscript title, and red subscript title at the end), 3 Clement on 150r and 4 Clement on 157r, and—rarely mentioned—a series of exegetical proemia from fol. 163r through 166v (on Acts, including a note on the Gospel of Luke, mentioned as the first book by this author, on the catholic epistles, on the letters of Paul).Footnote 29

The question would then be whether the superscript to 2 Peter on fol. 145v, where the scribe states that “we add to these letters of Paul those letters of the apostles that are not found in all manuscripts,” is meant to imply, for the accumulative aim of this collection, the “minor” catholic epistles only (2 Peter, 2–3 John, Jude), or 3 and 4 Clement as well.Footnote 30 One might feel inclined to lean towards the latter, given that no explicit difference is marked in the manuscript, and that they are of the epistolary genre, unlike the exegetical proemia that follow the letters.

Interestingly, the exegetical proemion devoted to the catholic epistles (fol. 164r)—non-Euthalian—mentions neither the Clementine letters nor their author. It states that only three catholic epistles were translated in Syriac from the beginning (James, 1 Peter, 1 John), and later on the other four (2 Peter, 2–3 John, Jude), and concludes by noting that Jacob, the brother of the Lord and bishop of Jerusalem, “is of the seventy evangelists, and not of the twelve apostles.”

Perhaps one could imagine this being the common ground for the inclusion of Clement of Rome in Amsterdam ms. III H 25, in a collection project that is already accumulative in intentFootnote 31 (in as much as it explicitly states the aim to add in, from a different version, the “minor” catholic epistles which are missing in other codices), namely as the letters of another early Christian figure from the broader circle of evangelists: one other bishop appointed by one of the twelve apostles. This is consonant with, and perhaps explains why, the connection to Peter is stressed in the superscript titles, similarly to 1 and 2 Clement in Cambridge Add. 1700 (and, as will be shown presently, in all other witnesses): 3 Clement is introduced in Amsterdam ms. III H 25 as ܬܘܒ ܐܓܪܬܐ ܩܕܡܝܬܐ ܕܛܘܒܢܐ ܩܠܡܝܣ ܬܠܡܝܕܗ ܕܦܛܪܘܣ ܫܠܝܚܐ, “also, the first epistle of the blessed Clement, the disciple of Peter the apostle” (on fol. 150r).

B. More Data: 3 and 4 Clement in New Testament Collections (13th–20th cent.)

The most important “new” witness to surface in western scholarship for these letters is significantly older: Damascus, Syriac Orthodox Patriarchate ms. 39 (olim 2/11) [=SOP 39] predates Amsterdam ms. III H 25 by two centuries. Completed in 1249/50 (fol. 128r), this is a New Testament collection with the Harklean text of the four gospels, Acts, the seven catholic epistles (James, 1–2 Peter, 1–2–3 John, Jude), the fourteen Pauline letters, followed by 3 and 4 Clement (fol. 311r-312v).Footnote 32 Unfortunately, due to the loss of several folios, the beginning of 3 Clement is missing, and so is most of 4 Clement: after the subscript title of the former and the superscript title of the latter, only thirteen lines of 4 Clement remain accessible, on the second column of the page which is now the last surviving one of the manuscript (fol. 312v). Even incomplete, SOP 39 will be central for any future critical edition.

This is, however, only one of a series of new Syriac New Testament manuscripts in which 3 and 4 Clement are included, and in all of them Clement is introduced as the follower of Peter. The list below presents chronologically the updated list of witnesses, marking those that are New Testament manuscripts. The relevant details about each manuscript are set in footnotes.

NT 1. Damascus, Syriac Orthodox Patriarchate ms. 39 (olim 2/11) [=SOP 39] 1249/50
NT 2. Damascus, Syriac Orthodox Patriarchate ms. 2 (olim 1/2)Footnote 33 [=SOP 2] 1418/19
NT 3. Qaraqosh, Syriac-Orthodox Church of St George ms. 1Footnote 34 15th cent.
NT 4. Amsterdam, Allard Pierson Reading Room ms. III H 25 (olim 184) 1470/71
5. Istanbul, Syriac Orthodox church of Yoldat Aloho, Maryemana 4Footnote 35 15/17th cent.
4 Clement only
NT 6. Birmingham, Cadbury Research Library, Mingana syr. 480Footnote 36 1713
NT 7. Pampakuda, India, Konat Collection, (APSTCH) ms. 1.237Footnote 37 1898
NT 8. Manchester, John Rylands Library, Syr. 11Footnote 38 1898
9. Mardin, Church of the Forty Martyrs (CFMM) ms. 321Footnote 39 20th cent.
4 Clement only

This list reflects the manuscript dissemination of 3 and 4 Clement from the thirteenth century through the twentieth. It is notable that two of these witnesses show that 4 Clement has made its way separately from 3 Clement in other types of collections (hagiographic and legal in this case, as shown above in the two corresponding footnotes), a development that is not paralleled, for instance, in the transmission of 2 Clement, which remains entirely dependent on 1 Clement.Footnote 40 But the most interesting feature remains the fact that 3 and 4 Clement are repeatedly included in New Testament collections as the first and the second letter of Clement: instead of an isolated mistake, this is a recurrent phenomenon from the thirteenth century on (without later notes to indicate that this is a mistake in any way, so far as I was able to verify), repeated as we will see presently in different collections, with more witnesses than the Greek, Coptic, and Latin manuscript reception of 1 Clement taken together.

Seven of the nine witnesses are indeed New Testament manuscripts, however with varying configurations with regard to 1) the type of biblical text (Harklean, Peshitta, Philoxenian), 2) the order in which the catholic epistles are copied, and 3) the position of 3 and 4 Clement in the manuscript. The following table therefore presents the succession of epistles in these seven witnesses, with dates and places of copying to the extent that they are known. It also details the order of the catholic epistles (Cath.), and notes the type of text—whether Harklean (Hrk), Peshitta (Pe), or Philoxenian (Phl) when this is the case for the “minor” catholic epistles—whenever this information was available or was possible to verify or conjecture. Finally, Cambridge, University Library, Add. 1700, the sole witness for 1 and 2 Clement in Syriac, is added on the first column for comparison:

Add. 1700
1169/70
Edessa		 SOP 39
1249/50
—		 SOP 2
1418/19
Tur Abdin		 Qaraqosh 1
15th cent.
(Mosul?)		 III H 25
1470/71
Mardin		 Ming. 480
1713
Mardin		 Konat 1.237
1898
(Mosul?)		 JRL 11
1898
Mardin
7 Cath. Hrk:
James
1–2 Peter
1–3 John
Jude
1–2 Clem
14 Paul Hrk		 7 Cath. Hrk:
James
1–2 Peter
1–3 John
Jude
14 Paul Hrk
3–4 Clem		 3 Cath. Pe
4 Cath. Phl:
James
1–2 Peter
1–3 John
Jude
3–4 Clem
14 Paul Pe		 [3 Cath. Pe]
[4 Cath. Phl]
Jude
3–4 Clem
[14 Paul Pe]		 3 Cath. Pe
Jas 1Pt 1Jn
14 Paul Pe
4 Cath. Phl:
2 Pt 2 Jn
3 Jn Jude
3–4 Clem		 3 Cath. Pe
4 Cath. Phl:
James
1–2 Peter
1–3 John
Jude
3–4 Clem
14 Paul Pe		 3 Cath. Pe
4 Cath. Phl:
James
1–2 Peter
1–3 John
Jude
3–4 Clem
14 Paul Pe		 3 Cath. Pe
4 Cath. Phl:
James
1–2 Peter
1–3 John
Jude
3–4 Clem
14 Paul Pe
[ii] [i] [iv] [iv] [iii] [iv] [iv] [iv]

The table shows the Clementine letters copied in New Testament manuscripts over nine centuries in a space that encompasses Edessa, Tur-Abdin, Mardin, and Mosul—all being important ecclesiastic, monastic, and manuscript centres of the Syrian Orthodox church in northern Mesopotamia throughout the medieval period,Footnote 41 for which these artefacts are witnesses of biblical, liturgical, and exegetical contexts. Three of them (Cambridge UL Add. 1700, SOP 39, and Amsterdam III H 25) are primarily New Testament manuscripts, with various apparatuses added in: tables of liturgical readings, introductions to biblical books or group of books, lists of chapters (including Euthalian apparatus for SOP 39 and Add. 1700), Ammonian sections and Eusebian canons (SOP 39), occasional exegetical passages in the margins, and occasionally other texts (e.g. the Harmony of the Passion in Add 1700, Eusebius’ letter to Carpianus in SOP 39,Footnote 42 or the exegetical fragments at the end of Amsterdam III H 25).

The other five witnesses (SOP 2, Qaraqosh ms. 1, Mingana syr. 480, Konat [APSTCH] ms. 1.237, and JRL Syr. 11) are significantly more complex biblical-exegetical collections, with extensive introductory material before the gospels (about 30 folios in Mingana syr. 480 and JRL Syr. 11)Footnote 43 as well as throughout. The gospels are copied in one column in a larger script, the second column being reserved for an extensive commentary in a smaller script excerpted from various authors. The text of Acts and the epistles (including 3 and 4 Clement) is written on both columns, with extensive exegetical notes copied in the margins (excluding 3 and 4 Clement). Manchester, John Rylands Library Syr. 11 seems to be a slightly simplified version of the collection.Footnote 44 Nevertheless, the others are likely directly connected, with suggestions being made that SOP 2 is the antigraph of Qaraqosh ms. 1 and Mingana syr. 480,Footnote 45 and Konat [APSTCH] ms. 1.237 the apograph of the latter.Footnote 46

While the exact relationship between them and the history of this biblical-exegetical collection is still to be accounted for, it is clear for our purposes that all five manuscripts reflect the same arrangement of the catholic, Pauline, and Clementine epistles. Moreover, the colophon of SOP 2—preserved in Yale Syriac ms. 8 and found then copied at the end of Mingana syr. 480 (fol. 428v) and Konat [APSTCH] ms. 1.237 (fol. 230v; presumably it would have also been present in Qaraqosh ms. 1)—describes the contents of the collection that are relevant here in the following manner:Footnote 47

And I completed the Gospel of life of Matthew and Mark, along with Luke and also John, and also the act of the apostles, that is praxis, and also the catholic epistles in order of the evangelists and also of divine apostles: one of James, and two of Peter, three of John, and of Jude one, and two of Clement the disciple of Peter, and also the fourteen epistles of Paul the apostle, with requisite and necessary commentaries in between.

This is interesting in two ways. First, similar to Cambridge ms. Add. 1700, the colophon seems to list the Clementines as catholic epistles; second, similar to the exegetical fragment on the catholic epistles in Amsterdam ms. III H 25, this colophon also distinguishes between the main apostles (presumably John, or Peter) and the broader circle of evangelists (presumably James, or Jude) and says that the book includes catholic epistles from both. Clement, specifically described as “the disciple of Peter,” would then be set in the latter group.

Grouping these five manuscripts in one category leaves us with four overall epistolary configurations that involve Clementine letters (which are also marked [i] through [iv] at the bottom of the previous table), reflecting four separate occasions in which a Syriac compiler would have included them in New Testament manuscripts:

[i] Harklean text of the catholic and Pauline epistles, followed by 3–4 Clement (13th cent.)

[ii] Harklean text of the catholic and Pauline epistles, with 1–2 Clement before Paul (12th cent.)

[iii] Peshitta “major” catholic and Pauline epistles, followed by Philoxenian “minor” epistles and 3–4 Clement (15th cent.)

[iv] Peshitta catholic and Pauline epistles (Philoxenian “minor” epistles) with 3–4 Clement before Paul (15th–20th cent.)

In 1987, Jeffrey Siker published a discussion of the peculiar dynamic surrounding the translation and use of the catholic epistles in Syriac and the implication for the shape of the New Testament canon in Syriac.Footnote 48 He addresses the fact that the Peshitta never incorporated the “minor” catholic epistles, not even after subsequent Syriac translations, including these letters, would have been in circulation (Philoxenian, Harklean), asking: “Is it because the Peshitta version was more authoritative than the Minor Catholic Epistles themselves? Did the version exercise authority over the books themselves and thus determine their canonicity?”Footnote 49 Based on a survey of manuscripts and quotations from the catholic epistles in Syriac literature, Siker responds in the affirmative, concluding that the knowledge and use of the “minor” catholic epistles was limited to West Syrians, where they “probably accorded. . . some kind of canonical status inasmuch as they formed a part of the Philoxenian and Harclean versions”; but to the extent that the Peshitta prevailed as the “authoritative version of the books of the New Testament” the result was that “the authority of the Minor Catholic Epistles met with a stalemate.”Footnote 50

However, rather than an indication that the authority of the Peshitta (however conceived or reified) ruled out or blocked the canonical status of the “minor” catholic epistles, what the data presented by Siker seem to indicate is simply the fact that the Peshitta was used and copied more than the other Syriac versions in circulation. The strength of that contribution lies in problematising the status of the catholic epistles in general and of the “minor” epistles in particular as New Testament books which, like Revelation, were translated and integrated in Syriac New Testament collections only later.Footnote 51

More interesting for the present enquiry, however, is the fact that Siker’s example for the intermediary stage of adding the “minor” catholic epistles to the Peshitta after Paul, where the scribe’s superscription to 2 Peter expresses the plan of adjoining to Paul the letters that are not found in all manuscripts,Footnote 52 is also the first witness that came to be known to western scholars where two Clementine letters were included in the Syriac New Testament, Amsterdam ms. III H 25. There are in fact several other Peshitta manuscripts where the “minor” catholic epistles are added to the “major” ones from the Philoxenian text,Footnote 53 and occasionally also from the Harklean text.Footnote 54

As such, I would provisionally conclude that the data presented here so far indicate that to this space of variation around the catholic epistles, characterised by accumulation and restoration, belongs too the inclusion of two pairs of Clementine letters found in the four configurations known to date, listed above from [i] through [iv]—again as additional letters attributed to a prominent follower of a main apostle. Before drawing more general conclusions, the next section explores a possible reason for why Clement of Rome would have been selected for inclusion with the New Testament epistles, and adds to the conversation a new set of witnesses for Clementine writings in Syriac biblical manuscripts, normally considered separately.

Lists, Biblical Pandects, and the Syriac Octateuch of Clement

Lightfoot’s suggestion of 1869 was that the presence of 3 and 4 Clement in Amsterdam ms. III H 25 “is possibly due to a mistake” of a “Syrian transcriber [who] finding the ‘Two Epistles of Clement’ mentioned at the end of some list of canonical books,” added the two letters attributed to Clement to the New Testament epistles.Footnote 55 However, rather than an isolated occurrence, this inclusion was performed in at least four different New Testament configurations, one of which—[iv]—was then copied again at least four more times (from the fifteenth through the twentieth century) in the “handsome and sumptuous”Footnote 56 codices with the biblical-exegetical collections briefly presented above. As shown in the following, this can perhaps be linked to the peculiar form in which the canonical list in the 85th Apostolic Canon has circulated in Syriac.

A. The 85th Apostolic Canon in Syriac Context

The list of canonical books could perhaps be identified in the most obvious candidate: the 85th Apostolic Canon—the Apostolic Canons themselves being the 47th chapter of the 8th book of the so-called Apostolic Constitutions, composed in GreekFootnote 57 —which presents the prescribed list of βιβλία σεβάσμια καὶ ἅγια, “venerable and holy books,” of the old and new covenants.Footnote 58 The list, formulated in the first person as coming from Clement, includes the two epistles of Clement with the letters of the new covenant, as well as the Eight Books of Clement, in the following manner:

Παύλου ἐπιστολαὶ δεκατέσσαρες, Ἰακώβου μία, Ἰωάννου τρεῖς, ᾽Ιούδα μία, Πέτρου δύο, Κλήμεντος δύο, καὶ αἱ Διαταγαὶ ὑμῖν τοῖς ἐπισκόποις δι᾽ ἐμοῦ Κλήμεντος ἐν ὀκτὼ βιβλίοις προσπεφωνημέναι, ἃς οὐ χρὴ δημοσιεύειν ἐπὶ πάντων διὰ τὰ ἐν αὐταῖς μυστικά. καὶ αἱ Πράξεις ἡμῶν τῶν ἀποστόλων.

Fourteen epistles of Paul, one of James, three of John, one of Jude, two of Peter, two of Clement, and the constitutions addressed to you, bishops, from me, Clement, in eight books, which are not to be published for everyone, given the mysteries in them. And the Acts of us, the apostles.

It is surprising that Lightfoot does not explore this possibility for the Syriac side, considering that elsewhere he makes the connection between the order of the New Testament books in Codex Alexandrinus and the shape in which the canonical list in the 85th canon seems to have circulated in Coptic: “It is interesting to observe. . . that the treatment in the Alexandrian ms. exactly accords with the language of the 85th Apostolical Canon as read in the Coptic Churches.”Footnote 59 Indeed, the list fluctuates slightly from one language to another, and even within one language. The table below presents the list in Coptic, Greek and in Syriac, the latter with the two lists that seem to be in circulation:

Coptic
Tattam 1848Footnote 60 		Greek
Metzger 1987Footnote 61 		Syriac
Diyarbakır 1/1 (1498)Footnote 62 Deir al-Surian 31 (7/8 cent.)Footnote 63
4 gospels of M M L J
Acts of the Apostles
2 letters of Peter
3 letters of John
1 letter of James
1 letter of Jude
14 letters of Paul
Apocalypse of John
2 letters of Clement		 4 gospels of M M L J
14 letters of Paul
1 letter of James
3 letters of John
1 letter of Jude
2 letters of Peter
2 letters of Clement
8 books of Clement
Acts of the Apostles		 4 gospels of M M L J
14 letters of Paul
1 letter of James
2 letters of Peter
3 letters of John
1 letter of Jude
2 letters of Clement
8 books of Clement
Acts of the Apostles		 4 gospels of M M L J
14 letters of Paul
2 letters of Peter
3 letters of John
1 letter of James
1 letter of Jude
2 letters of Clement
8 books of Clement
Acts of the Apostles

Compared to the Coptic where the Clementine epistles come after Revelation, and to the Greek where they come after the epistles of Peter (with no variance in the critical apparatus in this regard),Footnote 64 the peculiarity of the Syriac version is that in both lists in circulation in this space, the two letters of Clement are linked to the Epistle of Jude. The same can be found in three of the four separate configurations identified above, where the Clementine letters are indeed linked to Jude and follow its position in the New Testament: either before the Pauline corpus as in [ii] for 1 and 3 Clement and [iv] for 3 and 4 Clement, or after the Pauline corpus with the other “minor” catholic epistles, as is the case for 3 and 4 Clement in [iii].

If the connection suggested by Ligthfoot between the Coptic form of the 85th Apostolic Canon and the order of the New Testament books and the position of 1 and 2 Clement in Codex AlexandrinusFootnote 65 holds any water, a similar connection is even more likely here, where the Syriac form of the same canon list seems to have provided the link to Jude as a place of insertion for adding in both sets of letters, 1 and 2 Clement on the one hand, and 3 and 4 Clement on the other, in several pre-existent Syriac New Testament collections. The exception—[i]—where the Clementine letters follow the Pauline corpus, is the oldest witness of 3 and 4 Clement (SOP 39) and could perhaps be seen as the intermediary step in which these were added first at the end of a New Testament,Footnote 66 before they were eventually linked to Jude, as we can find them in all other witnesses.

Until further evidence surfaces, the fact that the Clementine Octateuch eventually became an important juridical collection in West Syrian context (which was “often copied, excerpted, and translated” and “widely influenced canon law in Eastern Christian traditions”Footnote 67 ) seems the most probable basis for the repeated inclusion of Clementine letters in Syriac New Testament manuscripts, especially in compilation projects where other letters (the “minor” catholic epistles) were being reinstated and added to the then available New Testament collections.

B. The Clementine Octateuch in Biblical Pandects (12th–17th cent.)

It remains to verify whether and to what extent the title Eight Books of Clement found in the list of the 85th Apostolic Canon in Syriac context might have generated a similar effect. In Greek this title likely designates the Apostolic Constitutions, composed in eight books, which contains the Apostolic Canons and the canonical list of books in the 85th canon. In Syriac, however, a slightly larger collection (which includes the Testament of our Lord Jesus Christ, the Apostolic Church Order, as well as the Apostolic Constitutions, inclusive of the Apostolic Canons in book 8)Footnote 68 came to be in circulation as the Eight Books of Clement or the Octateuch of Clement.

This composite collection seems to have been translated from Greek within the same translation movement as 1 and 2 Clement towards the end of the seventh century; for instance, the translation of the Testament of our Lord Jesus Christ, which forms books 1 and 2 of the Clementine Octateuch, is one of the few Syriac translations from Greek for which we know both the translator and the date of translation: Jacob of Edessa in 686/7,Footnote 69 for whom a case can be made that he translated the whole collection,Footnote 70 and furthermore “who might be responsible for compiling the Clementine Octateuch in its present form.”Footnote 71

Moreover, the transmission of the Syriac Octateuch is formed by mainly two types of collections. On the one hand, extended reworked sections from the Books of Clement were included in ecclesial canonical collections of conciliar acta and documenta (usually grouped with other legal works like the Didascalia Apostolorum at the beginning, before the canons of the councils). It is possible that Jacob of Edessa is responsible for compiling this “larger Syrian Orthodox canonical collection containing many texts translated from Greek that is preserved in BnF syr. 62 and other Syriac manuscripts”Footnote 72 as well.

On the other hand—forming the third example of Clementine inclusion in Syriac biblical manuscripts—the Octateuch is found incorporated as a whole in Syriac biblical pandects starting with the twelfth century:Footnote 73

The Clementine Octateuch was included (although this is usually now forgotten), at some point prior to the twelfth century, in the canon of the Syrian Orthodox New Testament—as recorded by Bar Hebraeus in his thirteenth- century Nomocanon—and so was included in Syriac biblical pandect manuscripts, such as the twelfth-century “Buchannan Bible” in Cambridge, Diyarbakır Meryem Ana 1/1 of AD 1496, and the Mosul manuscript of AD 1652 used by Raḥmani for his edition.Footnote 74

Bar Hebraeus’ canonical list in Nomocanon 7.9 in fact preserves the text of the 85th Apostolic Canon in a somewhat abbreviated form, listing after the one letter of Jude and the two of Clement, “the eight books of mysteries of the same Clement, and our acts of the apostles,”Footnote 75 hence leaving aside the information that these eight books are meant for the bishops.Footnote 76

The oldest and most interesting example for this is Cambridge, University Library ms. Oo. 1.1,2, stemming from Tur Abdin, or Antioch, or—perhaps more likely—from around Edessa.Footnote 77 Dated towards the end of the twelfth century and donated in 1809 to the University of Cambridge,Footnote 78 it contains broadly speaking the Old Testament, the New Testament, and “the Six Books of Clement, documents coinciding in the main with those which are known as the ‘Syriac Octateuch.’ ”Footnote 79

As in all other witnesses discussed so far, the presumed author of the Clementine corpus is introduced as the disciple of Peter on fol. 310v, both in the superscript title and in the inscription to the left and to the right of the iconographic depiction of Clement. Incidentally, a particularity of this manuscript is that it features illuminations of several biblical characters and authors (e.g. Ruth on fol. 188r, Susanna on 188v, Esther on 189v, Clement on 310v), as well as representations of a certain episode as, for example, the election of Matthias portrayed on fol. 295v at the beginning of Acts.Footnote 80

The New Testament in Cambridge ms. Oo. 1.1,2 is the Peshitta text with the Philoxenian for the “minor” catholic epistles,Footnote 81 and the configuration is the following: the gospels, the Pauline corpus (14 epistles), Acts and the catholic epistles in the following order: James, 1 Peter, 1 John, then 2 Peter, 2–3 John, and Jude, with a subscript title that speaks about the “seven catholic epistles” as a whole.Footnote 82 This is then followed by the “Six Books of Clement.”

The ending of Cambridge ms. Oo. 1.1,2 is unfortunately lost, but the canonical list from the 85th Apostolic Canon is preserved in the other Syriac biblical pandect that includes the Octateuch of Clement: Diyarbakır, Meryem Ana Kilisesi ms. 1/1 [=HMML DIYR 2] of the 15th century (copied in 1498), in which the list as it is given in the previous table can be found on the bottom of the fourth column of fol. 258v and the top of the first on fol. 259r.Footnote 83 The same list appears in the third Syriac biblical pandect that includes the Octateuch of Clement mentioned by Taylor: Mosul, Mar Benham ms. 387 of the seventeenth century, on fol. 351v at the bottom of the middle column.Footnote 84 Both include the admonition to the bishops, for whom the Eight Books of Clement are explicitly meant.

To conclude, the Clementine books follow the Old and New Testament in these manuscripts as the legislative corpus of ecclesiastic law meant for bishops described in the 85th Apostolic Canon. This is also reflected in the iconographic cycle of Cambridge ms. Oo. 1.1,2: even though Clement is portrayed on fol. 310v in a very similar manner to all the other authors who are portrayed in the New Testament part of the manuscript (i.e. seated and copying a book from a book stand of different shapes; see Mark on 247v, Luke on 255r, John on 265v, or James on 306r), at the same time his garments are markedly different: only Clement is wearing an episcopal omophorion, i.e. a blue band with yellow crosses on his shoulders:Footnote 85 an ecclesiastical garment for the claimed author of the ecclesiastical corpus.

Conclusion

At some point after the translation, at the end of the seventh century, of a series of legislative collections (and the formation of the Octateuch of Clement and of the ecclesial canonical collections, both preserving the 85th Apostolic Canon), and before the oldest witnesses of the twelfth and thirteenth century, a number of works attributed to Clement of Rome (1 and 2 Clement, 3 and 4 Clement, and the Eight Books of Clement) were incorporated in West Syrian biblical manuscripts.

In accordance with the outlook of the 85th canon that is specific to Syriac, they are mostly linked to the catholic epistles, as 1 and 2 Clement in Cambridge Add. 1700 (copied in 1169/70) and the Clementine Octateuch in Cambridge Oo. 1.1,2 (dated to the twelfth century). The exception, SOP 39 (dating from 1249/50, where 3 and 4 Clement are copied after the Pauline corpus) only serves to underline the same tendency: in all subsequent witnesses (six manuscripts from the fifteenth through the twentieth century) 3 and 4 Clement are moved up and linked to the epistle of Jude, which they now follow: either at the end of the New Testament collection (in Amsterdam III H 25, completed in 1470, where the “minor” catholic epistles were copied after Paul), or before the Pauline corpus (in SOP 2, copied in 1418/19, and all other witnesses).

Furthermore, the letters are occasionally labelled “catholic” in paratext: 1 Clement in the superscript title in Cambridge Add. 1700 (where both letters are also fully embedded in the liturgical cycle). The same is implied for 3 and 4 Clement in the colophon of SOP 2 and the other witnesses in group [iv].

At the same time, these witnesses represent relatively limited occurrences among Syriac biblical manuscripts and, perhaps more importantly, there are also paratextual clues in the manuscripts which set the Clementine books apart to various degrees. For instance, the colophon of Cambridge Add. 1700 and the episcopal garment of Clement in Cambridge Oo. 1.1,2. Moreover, in group [iv] of biblical-exegetical collections, 3 and 4 Clement do not receive any exegetical comments, which the New Testament books receive. In this context, it seems very likely that the always-stressed connection of Clement to Peter cuts both ways, with the moniker “Clement, the disciple of Peter” working as a legitimising reason for adding Clementine books to biblical collections, while at the same time functioning as a qualifier that sets the Clementine books apart from the books with which they are copied.

As such, a rigid definition of what “canonical” is (or is not) does not fully explain the data. Instead, what we see in the available evidence is a medieval dynamic involving accumulative collections wherein the Philoxenian or Harklean “minor” catholic epistles are reinstated in various configurations. In turn, such accumulative collections cohabit with the more wide-spread and in a sense more conservative Peshitta collections. To this context of restoration and accumulation also belongs the inclusion of Clementine writings in Syrian Orthodox biblical manuscripts (at some point after the end of the seventh century), in close resemblance to the canonical list preserved in Syriac of the 85th Apostolic Canon, and in the wake the mounting prominence of the two main legislative bodies of literature in circulation—the Octateuch itself and ecclesial canonical collections for which the earliest witnesses date from the eighth and ninth centuries.Footnote 86

In turn, the evidence discussed here serves as a two-fold example of how the reception of a later, pseudepigraphical work can influence and even determine the reception of older, perhaps more “genuine” books. First, the obvious result of this study is that the reception of the Octateuch of Clement influences the reception of the Clementine letters, including the more “genuine” books like 1 and 2 Clement: since for the most part the Clementine letters seem to only have survived because of the inclusion in New Testament collections,Footnote 87 this direction of influence is significant precisely because it may feel counter-intuitive in a field where we tend to center on the older books as the most important and influential.Footnote 88

Second, with regard to the interplay of 1 and 2 Clement and 3 and 4 Clement, the two sets of letters have previously crossed paths in Syriac context, as evidenced in one of the earliest anthologies of patristic quotations.Footnote 89 In British Library Add. 12156, dated to the sixth century, there are three short excerpts from 3 and 2 Clement one after another on fol. 69v: 3 Clement 5–6, introduced in red ink as ܕܩܠܝܡܝܣ ܐܦܣܩܦܐ ܕܪܗܘܡܐ ܡܢ ܐܓܪܬܐ ܩܕܡܝܬܐ ܕܥܠ ܒܬܘܠܘܬܐ, “of Clement the bishop of Rome, from the first letter about virginity”; 2 Clement 1 with the red ink superscript ܕܝܠܗ ܟܕ ܕܝܠܗ ܡܢ ܫܘܪܝܐ ܕܐܓܪܬܐ ܕܬܠܬ, “of the same from the beginning of the third letter”; and 2 Clement 9.5 with the red ink title ܕܝܠܗ ܟܕ ܕܝܠܗ, “of the same (author and) the same (text).”

The way the Clementine letters are counted in the only witness they intersect is therefore peculiar. If 3 Clement is “the first letter de virginitate,” and 2 Clement the “third letter of Clement,” what other unquoted Clementine epistles are implied in BL Add. 12156? 4 Clement as the second letter de virginitate, but not 1 Clement? Or perhaps 1 Clement as a fourth letter of Clement? A possibility would be that the compiler knew 3 and 4 Clement as one letter (which is sometimes argued to have been the case initiallyFootnote 90 ), counted in BL Add. 12156 as the first (in which case “about chastity” would simply describe the topic of that first letter, rather than indicating that there are two letters on that topic), after which 2 Clement would be then the third, with 1 Clement implied and counted as a second letter of Clement.

In any case, in all subsequent Syriac florilegia the same excerpt from 2 Clement 1 is designated as coming from “the second letter of Clement,”Footnote 91 and by the time of the manuscripts studied in this article any knowledge of such intersection seems lost. From this perspective, the comparatively more developed reception in New Testament collections of 3 and 4 Clement from the thirteenth century on, over against the isolated witness of 1 and 2 Clement in the twelfth century, could perhaps be then conceived as a case in which the former two epistles replace the latter two in the West Syrian church.

Footnotes

*

Funded in part by the Austrian Science Fund (FWF) [grant DOI 10. 55776/ STA214], this research was made possible through a Fritz Thyssen Project Grant (Ref. 30.19.0.039TR) in Leuven and a British Academy Visiting Fellowship at Cambridge (2023–2024). I am extremely grateful to Madalina Toca, Andy Hilkens, and the two reviewers of the Journal for corrections and suggestions.

References

1 This is olim Amsterdam Remonstrant Seminary ms. 184, which is how it is mostly known in scholarly literature to date.

2 J. B. Lightfoot, S. Clement of Rome: The Two Epistles to the Corinthians; A Revised Text with Introductions and Notes (London: Macmillan, 1869) 16; repeated identically in J. B. Lightfoot, S. Clement of Rome: A Revised Text with Introductions, Notes, Dissertations, and Translations (1.1 of The Apostolic Fathers; London: Macmillan, 1890) 408.

3 Unless otherwise specified, all translations are mine.

4 1 and 2 Clement are copied in British Library ms. Reg. 1 D VIII, known in biblical studies as Codex Alexandrinus, and the Akhmimic Coptic version of 1 Clement is copied in Strasbourg, Bibliothèque nationale et universitaire, copte 362–385, seemingly together with the Epistle of James and the Gospel of John, both dated palaeographically to the 5th century. For the former see Patrick Andrist, “Au croisement des contenus et de la matière: l’architecture des sept pandectes bibliques grecques du premier millénaire,” Scrineum Rivista 17 (2020) 5–106, esp. 78–79. On the latter see more recently Dan Batovici, “The ‘Apostolic Fathers’ in Coptic: Problems and Overview,” in Parabiblica Coptica (ed. Ivan Miroshnikov; Parabiblica 3; Tübingen: Mohr Siebeck, 2024) 103–26, esp. 107–8.

5 Apart from the works discussed in this article, Clement was present in the Syriac context through the broad circulation of the Pseudo-Clementines—present in the oldest dated Syriac manuscript, BL Add. 12150, copied in 411 CE—and later reworkings such as the Life of Clement, most recently published in Alphonse Mingana, “Some Early Judaeo-Christian Documents in the John Rylands Library,” BJRL 4.1 (1917) 90–108; these writings too emphasise Clement’s Petrine connection. A useful overview of witnesses is available in F. Stanley Jones and Patricia Duncan, “Pseudo-Clementines,” e-Clavis: Christian Apocrypha (nasscal.com/e-clavis-christian-apocrypha/pseudo-clementines/).

6 The colophon giving the date (AG 1481) and the name of the scribe—“Sahda, the meanest of the monks in Edessa”—is found on fol. 216v, at the top of the second column. The manuscript is described in William Wright, A Catalogue of the Syriac Manuscripts Preserved in the Library of the University of Cambridge (vol. 1; Cambridge: The University Press, 1901) 6–16. See also The Ḥarḳlean Version of the Epistle to the Hebrews Chap. XI. 28–XIII. 25 (ed. R. L. Bensly; Cambridge: The University Press, 1889) 7–9; Lightfoot, The Apostolic Fathers, 1.1:130–31; and Barbara Aland and Andreas Juckel, Das Neue Testament I. Die grossen katholischen Briefe (ANTF 7; Berlin: De Gruyter, 1986) 37–38, where it is listed as H3.

7 A recent presentation is available in Andreas Juckel, “Introduction to the Ḥarklean text,” in Matthew (ed. George Anton Kiraz; vol. 1 of Comparative Edition of the Syriac Gospels; Piscataway, NJ: Gorgias, 2004) xxxi–lxxxii, where it is described as “the extremely literal translation of the Greek text (‘mirror translation,’ S. P. Brock),” at xxxiii.

8 Christopher Tuckett, 2 Clement: Introduction, Text, and Commentary (Oxford Apostolic Fathers; Oxford: Oxford University Press, 2012) 5. A discussion of the peculiarities of the translation is available in R. L. Bensly, The Epistles of S. Clement to the Corinthians in Syriac (Cambridge: The University Press, 1899) vii–xviii; see also Lightfoot, The Apostolic Fathers, 1.1:136–42, likely based on Bensly’s notes made for Lightfoot in 1876/77, for which see H. R. Kennett, “Preface,” in Bensly, The Epistles of S. Clement in Syriac, v.

9 E.g. Bensly, The Ḥarḳlean Hebrews, 8; Lightfoot, The Apostolic Fathers, 1.1:135, mentioning Bensly; Wright, Cambridge Catalogue I, 10; Anton Baumstark, Geschichte der syrischen Literatur mit Ausschluß der christlich-palästinensischen Texte (Bonn: Marcus&Weber, 1922) 261; Aland and Juckel, Das Neue Testament I, 26 n. 26. Sebastian Brock, “The Chronological Development of Adjectives in -aya up to 725,” Hugoye 24.1 (2021) 53–125, simply assigns it to the 7th cent., at 113 and 117.

10 The list of the lections in this Passion Harmony is given in Wright, Cambridge Catalogue I, 8–9.

11 Wright, Cambridge Catalogue I, 12.

12 For a detailed discussion see Dan Batovici, “Out of Order: Reading the Argument of 1 Clement in the Sequence of the Liturgical Cycle of Cambridge, University Library, MS Add. 1700,” in 1 Clement as an Argumentative Text (ed. David du Toit, Jacob N. Cerone, and Kathrin Hager; NovTSupp 196; Leiden: Brill, 2025) 296–316.

13 Aland and Juckel, Das Neue Testament I, 26 n. 26 (translation mine).

14 Lightfoot, The Apostolic Fathers, 1.1:134 (emphasis in original).

15 Ibid., 135–36, listing as examples at 135 n. 1 Oxford, New College ms. 333 (12th/13th cent.), Bodleian Syr. 24 (Or. 130, 12th cent.), BL Add. 14474 (9th cent.), and “an Amsterdam ms.,” probably referring to Allard Pierson ms. III H 25 (15th cent.).

16 Lightfoot’s understanding of “canon” here can be circumscribed and contextualised. For an example of how the modern understanding of literary and canonical categories is itself historicised, see for instance Annette Yoshiko Reed, “The Afterlives of New Testament Apocrypha,” JBL 133.2 (2015) 401–25.

17 Franz Diekamp, Patres Apostolici (vol. 2; Tübingen: Laupp, 1913) x–xi and 1–49. For a more recent assessment, see Sara Giorgetti, “Le epistole pseudo-clementine Ad Virgines nella tradizione manoscritta copta e nel Pandectes di Antioco di San Saba,” SZRKG 117 (2023) 207–26.

18 J.-P. Migne, Patrologiae cursus completus, Series Graeca (vol. 89; Paris: Migne, 1865) col. 1428–1849.

19 L.-Th. Lefort, Les Pères Apostoliques en copte (CSCO 135; Leuven: Durbecq, 1955) xv–xix and 35–43, and Enzo Lucchesi, “Compléments aux Pères apostoliques en copte,” AnBoll 99 (1981) 395–408, at 405–408. See also Batovici, “The ‘Apostolic Fathers’ in Coptic,” 108–10, and Giorgetti, “Le epistole pseudo-clementine Ad Virgines,” 207–26.

20 Lightfoot, The Apostolic Fathers, 1.1:408.

21 See S. P. Brock, “Notes on Some Texts in the Mingana Collection,” JSS 14.2 (1969) 205–26, at 209 for Mingana syr. 480; J. F. Coakley, “A Catalogue of the Syriac Manuscripts in the John Rylands Library,” BJRL 75 (1993) 105–207, at 130–31, for John Rylands Library, Manchester, Syr. 11.; Dan Batovici, “Apostolic by Proxy: Corpora, Editiones Minores, and Networks of Texts,” ETL (2022) 503–24, at 520–22, for Damascus, Syriac Orthodox Patriarchate ms. 39 and ms. 2, and Qaraqosh, Syriac-Orthodox Church of St George ms. 1.

22 This stands also for text-critical studies, e.g. most recently Marco Pavan, “La traduzione degli scritti attribuiti a Clemente di Roma in ambito siriaco—Il caso delle epistole Ad Virgines,” SZRKG 117 (2023) 227–44, based on the Amsterdam and the Mingana manuscripts.

23 Brock, “Notes on Some Texts in the Mingana Collection,” 209.

24 Johann Jakob Wettstein/Johannes Jacobus Wetstenius, Ἡ Καινὴ Διαθήκη. Novum Testamentum Graecum (vol. 2; Amsterdam: Dommeriana, 1752).

25 Jan Theodor Beelen/Johannes Theodorus, Sancti Patris nostri Clementis Romani epistolae binae de virginitate (Leuven: C. J. Fonteyn, 1856). For an overview of authors arguing for the genuineness of 3 and 4 Clement see Lightfoot, The Apostolic Fathers, 1.1:407–14.

26 Gwynn, Remnants, xlix; David Taylor, “Répertoire des manuscrits syriaques du Nouveau Testament,” in Le Nouveau Testament en syriaque (ed. J.-C. Haelewyck; Études syriaques 14; Paris: Geuthner, 2017) 291–313, at 305: “[Pesh.: Actes; Jc, 1 P, 1 Jn; 14 épîtres].”

27 So Lightfoot, The Apostolic Fathers, 1.1:408, quoting the written memorandum he received from John Gwynn. Taylor, “Répertoire,” 305 too lists this manuscript as “Leiden 2791 (olim Cod. Soc. Remonstrant., Amsterdam 184), ‘ms. Wetstein’” under the group of witnesses for the “Petites Épîtres catholiques (Philoxéniennes).” For a brief presentation of the Philoxenian and Harklean versions see Peter J. Williams, “The Syriac Versions of the New Testament,” in The Text of the New Testament in Contemporary Research (ed. Bart D. Ehrman and Michael W. Holmes; NTTSD 42; Leiden: Brill, 2013) 143–66, at 154–55.

28 Jeffrey S. Siker, “The Canonical Status of the Catholic Epistles in the Syriac New Testament,” JTS 38.2 (1987) 311–40, at 312. On the canonical status of the catholic epistles in Syriac see also Gerard Rouwhorst, “Les lectionnaires syriaques,” in La lecture liturgique des Épitres catholiques dans l’Église ancienne (ed. Christian-Bernard Amphouz and Jean-Paul Bouhot; Lausanne: Zèbre, 1996) 105–40, esp. 109–11. A survey of the translations into Syriac can be found in Williams, “The Syriac Versions,” 143–66.

29 Edited and translated in Beelen, Clementis epistolae binae, 295–313. Further research will be needed, but this material seems not to be Euthalian.

30 Siker, “The Canonical Status,” 317, notes that “this heading seems to accept as the status quo that these epistles are not regularly found in all copies of the Syriac New Testament”—which is indeed confirmed by the comparatively high number of Peshitta New Testament manuscripts with only three catholic epistles—and asks: “Does this indicate that the canonical status of these epistles [the ‘minor’ catholic epistles] was likewise occasional?” Unfortunately, he does not discuss the significance, in this regard, of the presence of the Clementine letters in the cluster of added letters.

31 This reflects a broader phenomenon where the Philoxenian “minor” catholic epistles are added in the collections with the Peshitta New Testament. Cf. the list of witnesses between the 9th and the 18th cent. in Taylor, “Répertoire,” 304–5.

32 Yuhanna Dolabani et al., “Catalogue des manuscrits de la bibliothèque du patriarcat Syrien Orthodoxe à Homs (auj. à Damas),” Parole de l’Orient 19 (1994) 555–661, at 562 has the following description: “2/11 [Ar.: 2/10]: [Commentaire du] Nouveau Testament jusqu’à l’épître aux Hébreux, selon la version héracléenne.. . . À la fin, les deux lettres de Clément, la 2e incomplète. Copié peut-être au XIIIe siècle.” This likely implies 1 and 2 Clement. The most recent SOP catalogue, however, identifies the letters correctly; see the forthcoming Roger Akhrass, Ephrem Aboud Ishac, and Erich Renhart, Catalogue of the Syriac Orthodox Patriarchate Manuscript Library, Harrasowitz, 2026. I thank Roger Akhrass for kindly providing me with the digital images of the letters and with the pre-publication description of the manuscript, on which I am drawing here.

33 This is a complex exegetical collection that includes the New Testament with extensive commentary on parallel columns. A brief description is available in Dolabani et al., “Catalogue,” 559. Ignatius Aphram I Barsoum, The Scattered Pearls: A History of Syriac Literature and Sciences (Piscataway, NJ: Gorgias, 2003) 499, indicates this to be the antigraph for Mingana syr. 480, which is the sixth in our list. If this is correct, then the Pauline letters and “major” catholic epistles are in the Peshita text, and the “minor” ones are in Philoxenian text; cf. Taylor, “Répertoire,” 305. From the digital images of 3 and 4 Clement it was possible to verify that the introduction to Paul’s letters, following immediately 4 Clement, is indeed the same as in Mingana syr. 480. Andy Hilkens recently identified the Yale Syriac ms. 8 as in fact the lost final part of SOP 2. 3 and 4 Clement are copied on fol. 394v-405r.

34 What seems to survive of this manuscript are the few photographs taken by Vööbus, now digitised at HMMLVoobus.org. Arthur Vööbus, Handschriftliche Überlieferung der Mēmrē-Dichtung des Jaʿqōb von Serūg III (CSCO 421/Subsidia 60; Louvain, CorpusSCO: 1980) 131–33, briefly describes the manuscript, mentioning “den ersten und zweiten Klemensbrief” at 132 n. 9 (likely referring to 1 and 2 Clement), dating the script to the 13th or 14th cent., at 133. However, Barsoum, The Scattered Pearls, 499, indicates that it was copied at end of the 15th century, and as a copy of SOP 2 (the second in our list), as is Mingana syr. 480 (the sixth). This suggests that Qaraqosh ms. 1 would have had the same text and configuration for the Pauline and catholic epistles, confirmed by the presence of the same introduction to Paul after 4 Clement in the extant photos. 3 and 4 Clement are copied on fol. 362r–370r.

35 This is a collection of mostly hagiographies, letters (including 4 Clement), and homilies, the photos of which, taken by Vööbus, are also available at HMMLVoobus.org. Description in Arthur Vööbus, “Important Discoveries for the History of the Monastery of Qartamin: New Light on the Literary Traditions Regarding its History,” OLP 8 (1977) 223–27, at 225–26. 4 Clement is copied on fol. 184v–190r.

36 Like SOP 2, this is also a large-format exegetical collection including the New Testament with commentary. Copied in 1713, it is described in Alphonse Mingana, Catalogue of the Mingana Collection of Manuscripts I. Syriac and Garshūni Manuscripts (Woodbrooke Catalogues 1; Cambridge: Heffer and Sons, 1933), col. 863–82, which mentions on col. 873 the “first” and “second epistle of Clement of Rome,” then identified as the two letters De virginitate in Brock, “Notes on Some Texts in the Mingana Collection,” 209. It also contains the Peshitta text of the Pauline letters and the “major” catholic epistles, and the Philoxenian text for the “minor” ones; cf. Taylor, “Répertoire,” 305. 3 and 4 Clement are on fol. 344r–351r.

37 This manuscript is available online at vhmml.org/readingRoom/view/539758, where it is described as a likely copy of Mingana syr. 480, completed in 1898 in Mosul, Iraq. It has the same configuration as the previous manuscripts. 3 and 4 Clement are copied on 259v–264v.

38 The same type of large-format exegetical collection as the previous witnesses and dating from 1905 with the same configuration, this manuscript is described in Coakley, “Catalogue,” 130–31. 3 and 4 Clement are on fol. 259v–264v.

39 The manuscript is digitised at vhmml.org/readingRoom/view/503668, and described in Arthur Vööbus, The Didascalia Apostolorum in Syriac I (CSCO 401/Scriptores Syri 175; Louvain: CorpusSCO, 1979) 30*–32*. It broadly contains books attributed to early Christian figures with claimed apostolic connection, some of which are legal in nature. Apart from Didascalia Apostolorum and the Epistola ad Timotheum de morte apostolorum attributed to Dionysius the Areopagite, it includes three Clementine writings: the Octateuch of Clement and two letters, one of which is 1 Clement, p. 243–94. This is thus one of the four modern witnesses of 1 Clement in Syriac; the others are Harvard, Houghton Library ms. Syriac 99 (1889), Mingana syr. 4 (1895), and Charfet, Rahmani ms. 79 (1901). The last text in the manuscript, after the pseudo-Dionysian letter to Timothy and listed in the vhmml.org description with the title found in the manuscript, “Letter of Clement, disciple of Peter,” is in fact 4 Clement, copied on pages 365–80.

40 Tuckett, 2 Clement, 6, correctly notes: “The manuscript evidence is thus unanimous (as far as it goes) in linking 2 Clement closely with 1 Clement.”

41 Now found in the south of Turkey (Edessa, Tur-Abdin, Mardin) and the north of Iraq (Mosul).

42 Akhrass et al., Catalogue, forthcoming.

43 See Mingana, Catalogue I, cols. 863–68, and Coakley, “Catalogue,” 129: the introductory material is found in Manchester, John Rylands Library, Syr. 10, which is the first volume to JRL Syr. 11, containing the gospels.

44 Coakley, “Catalogue,” 129 n. 45: “There is some relationship between this [Syr. 10] (and the next [Syr. 11]) manuscript and ms. Mingana Syr. 480,” indicating then the exegetical and introductory sections that they have in common.

45 Barsoum, The Scattered Pearls, 499. The same information is given in Gabriel Rabo, Dionysius Jakob Bar Salibi: Syrischer Kommentar zum Römerbrief (Göttinger Orientforschungen Syriaca 56; Wiesbaden: Harrassowitz, 2019) 31 n. 28. See also Takahashi, “Also via Istanbul to New Haven,” 168 n. 38. In a forthcoming book, Andy Hilkens describes the compositional dynamic of the exegetical material in these collections, and the relationship between them.

46 See the description at vhmml.org/readingRoom/view/539758. Out of this group, the most detailed description available is that of the Mingana manuscript; see Mingana, Catalogue I, cols. 863–82.

47 The colophon (found on p. 201) and the rest of the Yale pages of the manuscript are digitised and available online at https://collections.library.yale.edu/catalog/2054609; the transcribed text starts on line 6 and continues over nine lines.

48 Siker, “The Canonical Status,” 311–40.

49 Ibid., 312.

50 Ibid., 327–29.

51 On the reception of Revelation, see more recently T. C. Schmidt, The Book of Revelation and its Eastern Commentators: Making the New Testament in the Early Christian World (Cambridge: Cambridge University Press, 2021); Adrian Pirtea, “Manuscripts, Paratexts, and the NT Canon: Arethas of Caesarea and the Reception of St John’s Apocalypse in Byzantium,” in Religious Rhetoric of Power in Byzantium and South-Eastern Europe (ed. Ivan Biliarsky et al.; Brăila: Editura Istros, 2021) 37–61.

52 Siker, “The Canonical Status,” 317.

53 Taylor, “Répertoire,” 304–5, lists nineteen examples between the 9th and the 18th cent., including Amsterdam ms. III H 25 and Mingana syr. 480. We can add also Konat (APSTCH) ms. 1.237, Manchester JRL Syr. 11, and very likely SOP 2, as well as Qaraqosh ms. 1.

54 Taylor, “Répertoire,” 298, lists BL Add. 14474 of the 9th century.

55 Lightfoot, The Apostolic Fathers, 1.1:408.

56 Mingana, Catalogue I, col. 863.

57 For a brief overview on the Apostolic Canons and this list in particular see Edmond L. Gallagher and John D. Meade, The Biblical Canon Lists from Early Christianity (Oxford: Oxford University Press, 2017) 134–41.

58 Marcel Metzger, Les constitutions apostoliques. Tome III. Livres VI et VIII (Sources chrétiennes 336; Paris: Cerf, 1987) 306–8.

59 Lightfoot, The Apostolic Fathers, 1.1:372. See now Mina Monier, “The History of Codex Alexandrinus: New Evidence from Arabic Paratexts,” NovT 67.4 (2025) 501–26.

60 Henry Tattam, The Apostolical Constitutions, or Canons of the Apostles in Coptic (London: Oriental Translation Fund, 1848) 209–11.

61 Metzger, Les constitutions apostoliques III, 306–8.

62 The manuscript is digitised, with a brief description, at www.vhmml.org/readingRoom/view/122717. See also A. Vööbus, “Nouvelles Sources de l’Octateuque Clémentin Syriaque,” Le Muséon 86 (1973) 105–9, at 107. This is also the order of books printed in François Nau, La version syriaque de l’Octateuque de Clément (Paris: P. Lethielleux, 1913; repr. Piscataway, NJ: Gorgias, 2012) 128.

63 Sebastian P. Brock and Lucas Van Rompay, Catalogue of the Syriac Manuscripts and Fragments in the Library of Deir al-Surian, Wadi al-Natrun (Egypt) (OLA 227; Leuven: Peeters, 2014) 237. The same order seems to be in Damascus, Syriac Orthodox Patriarchate 8/11; copied in 1204, now apparently lost (see the list of manuscripts missing from the Patriarchal collection at https://dss-syriacpatriarchate.org/patriarchal-library/manuscripts/?lang=en), the manuscript was published in Arthur Vööbus, The Synodicon in the West Syriac Tradition I (CSCO 367/Scriptores syri 161; Louvain, CorpusSCO: 1975) 71. See also the description in Dolabani et al., “Catalogue,” 594.

64 Metzger, Les constitutions apostoliques III, 308.

65 Lightfoot, The Apostolic Fathers, 1.1:372.

66 The digital images of the pages that preserve the incomplete remains of 3 and 4 Clement in Damascus, SOP 39 (olim 2/11) do not preserve any ancient numbering (apart from the modern one in pencil) and therefore it was not possible to check whether or not these folios were displaced at the end of the manuscript. The digital images seem to suggest that the two extant folios are pasted in or at the very least reinforced in the binding during a previous conservation effort. In any case, 3 and 4 Clement were at the end of the manuscript in 1994 as well; see Dolabani et al., “Catalogue,” 562.

67 Jeff W. Childers, “Clement of Rome and Pseudo-Clementine literature,” in Gorgias Encyclopedic Dictionary of the Syriac Heritage (ed. Sebastian P. Brock et al.; Piscataway, NJ: Gorgias, 2011) 100–101, at 101.

68 Brief overview of the parts of the Octateuch of Clement in Syriac and their legislative standing can be found in Hubert Kaufhold, “Sources of Canon Law in the Eastern Churches,” in The History of Byzantine and Eastern Canon Law to 1500 (ed. Wilfried Hartmann and Kenneth Pennington; Washington, DC: The Catholic University of America Press, 2012) 215–342, at 241–43; see also Gallagher and Meade, Biblical Canon Lists, 252 n. 75.

69 Brock, “The Chronological Development of Adjectives in -aya up to 725,” 63; the list of datable translations is given on 58–63. An overview of ancient versions and modern editions of this book can be found in Simon Corcoran and Benet Salway, “A Newly Identified Greek Fragment of the Testamentum Domini,” JTS 61.2 (2011) 118–35, at 124–27.

70 Herman G. B. Teule, “Jacob of Edessa and Canon Law,” in Jacob of Edessa and the Syriac Culture of His Day (ed. Bas ter Haar Romeny; Monographs of the Peshitta Institute Leiden 18; Leiden: Brill, 2008) 83–100, at 84–86, gives an overview of the debate, concluding that “there is enough circumstantial evidence to credit Jacob of Edessa with the Syriac translation of the Octateuch,” at 85.

71 David Taylor, “The Authorship of the Apocalypse of the Testament of Our Lord Reconsidered,” in Le calame et le ciseau: colophons syriaques offerts à Françoise Briquel Chatonnet (ed. Simon Brelaud et al.; Cahiers d’études syriaques 8; Paris: Geuthner, 2021) 421–40, at 426.

72 Taylor, “Authorship,” 426. On the activity of Jacob of Edessa as translator and canonist see Teule, “Jacob of Edessa and Canon Law,” 84–100. The contents of Paris, Bibliothèque nationale de France syr. 62 of the 9th cent. are described in detail by Flavia Ruani at https://archivesetmanuscrits.bnf.fr/ark:/12148/cc1018002, and the manuscript is digitised at https://gallica.bnf.fr/ark:/12148/btv1b531152203.

73 Overviews of composition, editions and the manuscript transmission of the Octateuch are available in Nau, La version syriaque de l’Octateuque de Clément, 5–7; ArthurVööbus, “Nouvelles sources de l’Octateuque clémentin syriaque,” Le Muséon 86 (1973) 105–9. An updated list of manuscripts is available in MichaelKohlbacher, “Wessen Kirche ordnete das Testamentum Domini nostri Jesu Christi?,” in Zu Geschichte, Theologie, Liturgie und Gegenwartslage der syrischen Kirchen (ed. Martin Tamcke and Andreas Heinz; Studien zur orientalischen Kirchengeschichte 9; Münster: LIT, 2000) 55–137. A critical edition of the Syriac Octateuch is currently in preparation, as mentioned in Kaufhold, “Sources,” 243.

74 Taylor, “Authorship,” 425.

75 Paul Bedjan, Nomocanon Gregorii Barhebraei (Paris, Leipzig: Otto Harrassowitz, 1898) 103–4; Julius Y. Çiçek, Nomocanon of Bar-Hebraeus (Glane-Losser: Bar Hebraeus Verlag, 1986) 62.

76 The Nomocanon 7.9 too includes the two letters of Clement in the New Testament list, as reflected in the witnesses presented in the previous sections.

77 As argued for in Lucy-Anne Hunt, “The Syriac Buchanan Bible in Cambridge: Book Illumination in Syria, Cilicia and Jerusalem of the later Twelfth Century,” OCP 57 (1991) 331–69.

78 The manuscript is fully digitised at https://cudl.lib.cam.ac.uk/view/MS-OO-00001-00001/1. See the description in Wright, Cambridge Catalogue II, 1037–44; Gwynn, Remnants, xliv–xlv; Gallagher and Meade, Biblical Canon Lists, 252.

79 Gwynn, Remnants, xliv.

80 A detailed study of the illuminations in this manuscript, on the basis of which the proposal is put forward that it was copied in Edessa, is available in Hunt, “The Syriac Buchanan Bible in Cambridge.”

81 Taylor, “Répertoire,” 300 and 305.

82 Gwynn, Remnants, xliv, adds: “This tacit recognition of their canonicity is hardly weakened by the fact that next after them it places ‘the Six Books of Clement.’”

83 The folio numbering is that of the digital information at https://www.vhmml.org/readingRoom/view/122717.

84 The manuscript is also digitised at https://www.vhmml.org/readingRoom/view/131685. The folio numbering is that of the digital project.

85 Rima Smine, “L’art au service de l’hagiographie: la representation des saints dans la tradition syro-occidentale,” in L’hagiographie syriaque (Études syriaques 9; Paris: Geuthner, 2012) 247–70, at 255: “Saint Clement, ‘disciple de Pierre,’ est représenté dans son habit d’évêque, avec l’omophorion.” The image can be seen at https://cudl.lib.cam.ac.uk/view/MS-OO-00001-00001/1.

86 Mardin, Church of the Forty Martyrs manuscripts 309 and 310, and Paris, BnF syr. 62. Looking forward, a promising direction of inquiry will be the reception of the figure of Clement in Syro-Arabic, Copto-Arabic, and Ethiopic circles, where various configurations of the Clementine Octateuch are present and popular.

87 2 Clement and 3 Clement are not preserved in any other type of Syriac manuscripts than biblical. For 1 Clement Cambridge ms. Add. 1700 is the only ancient Syriac witness we have. Only 4 Clement seems to have survived independently in pre-modern times (in Istanbul, Syriac Orthodox church of Yoldat Aloho, Maryemana ms. 4 of the 15th/17th cent.).

88 Another example is the sole Latin witness of 1 Clement. A very interesting recent contribution, Donatella Tronca, “The Latin Translation of the Letter of Clement to the Corinthians in Manuscript Namur, Grand Séminaire 37: A Miscellany on Papal Primacy,” ETL 98.3 (2022) 379–89, shows how this manuscript (containing Rufinus’ prologue to the Clementine Recognition, the Recognition in Rufinus’ translation, Clement’s letter to James in the translation of the same and, at the end, 1 Clement) is in fact a construct meant to consolidate ecclesiastical construction of papal primacy in late 11th and early-12th cent. Latin ecclesial context. Unless one prefers to see the whole collection as an introduction to 1 Clement, it seems that we owe the Latin 1 Clement to the cumulative nature of the Namur manuscript and its papal-primacy project, as an added satellite to the Latin Recognition. For the pivotal role of the prefatory Epistula Clementis in the reception of the Recognitions, see also Warren Campbell, “The Death of ‘Jewish Christianity’ in the Afterlife of the Clementine Recognitions,” HTR 117.3 (2024) 456–74.

89 Albeit in different translations than those found in the manuscripts discussed so far; see Lightfoot, The Apostolic Fathers, 1.1:182.

90 The latest contribution to do so is Pavan, “La traduzione degli scritti attribuiti a Clemente di Roma in ambito siriaco.”

91 BL Add. 17214, f. 76v: 2Clem 1 and 2Clem 9 (7th cent.); BL Add. 12157, f. 200v: 2Clem 1 (7th/8th cent.); BL Add. 14532, f. 214v: 2Clem 1 and 2Clem 9 (8th cent.); BL Add. 17191, f. 58v: 2Clem 1 and 2Clem 9 (9th/10th cent.); and BL Add. 14538, f. 20r: 2Clem 1 and 2Clem 9 (10th cent.).