Greek Catenae and the “Western” Order of the Gospels

In: Novum Testamentum
View More View Less
  • 1 University of Birmingham, Birmingham, UK
Open Access


The “Western” order of the gospels—Matthew–John–Luke–Mark—is found in a few important ancient codices in both the Greek and the Latin tradition. Previous attempts to identify Greek minuscule manuscripts with this sequence have been inconclusive. This article presents five Greek minuscules which feature the gospels in the Western order. These five manuscripts, along with two Greek majuscules, contain the earliest form of the catena commentary on Matthew, John, and Luke. The analysis of these catenae reveals that the sequence of their composition is reflected in the codicology of these manuscripts, as well as non-standard orders of the gospels in other catena witnesses. It is therefore the presence of the commentary which explains the adoption of the Western order in seven of the eleven known occurrences in Greek.


The “Western” order of the gospels—Matthew–John–Luke–Mark—is found in a few important ancient codices in both the Greek and the Latin tradition. Previous attempts to identify Greek minuscule manuscripts with this sequence have been inconclusive. This article presents five Greek minuscules which feature the gospels in the Western order. These five manuscripts, along with two Greek majuscules, contain the earliest form of the catena commentary on Matthew, John, and Luke. The analysis of these catenae reveals that the sequence of their composition is reflected in the codicology of these manuscripts, as well as non-standard orders of the gospels in other catena witnesses. It is therefore the presence of the commentary which explains the adoption of the Western order in seven of the eleven known occurrences in Greek.

1 Introduction*

The anomaly of the so-called “Western” order of the gospels—when the gospels appear in the sequence Matthew–John–Luke–Mark—is a striking feature in significant early manuscripts such as 𝔓⁴⁵, Codex Bezae, and many Old Latin gospel books, in contrast to the early development and stability of the traditional order (Matthew–Mark–Luke–John).1 However, until now it has been unclear whether or not this non-traditional order is also found in Greek minuscule manuscripts. Bruce M. Metzger stated that several older minuscules could be found with the Western order of the gospels but did not provide specific details.2 The present article identifies five Greek minuscules with the Western order of the gospels, though none of them is included in the Kurzgefasste Liste.3 Significantly, each of these codices contains a catena—a commentary on the biblical text composed of a chain of scholia from various early Christian writers. The reason for the sequence of the gospels in these manuscripts is to be found in the characteristics of the catenae they contain.

2 Greek Manuscripts in the “Western” Order

The most recent list of manuscripts with the Western order is provided by Matthew Crawford, building on earlier studies and adding GA 073 + 084.4 All of the Greek witnesses are in majuscule script: 𝔓⁴⁵, 05/D, 032/W, 033/X, 055, and 073 + 084. Crawford’s evidence for GA 073 + 084—a sixth-century manuscript consisting of four fragments from Matthew—rests on a marginal tabular cross-reference system with the columns arranged in the sequence Matthew–John–Luke–Mark.5 While partially an argument from silence, the case that the marginal tables referred to the corresponding gospels in the same codex is persuasive, especially with the parallels to other versional codices.6 While Crawford’s article was in press, two further majuscule codices with this sequence were put forward by Daniel B. Wallace: GA 036/Γ and GA 0234.7 Neither of these, however, can be substantiated. In the case of GA 036, divided between the Bodleian Library and National Library of Russia, the present arrangement of the manuscript approximates the Western order. However, the folio currently numbered 52, which contains the kephalaia on John on its verso, bears extensive offset ink on the recto from the final page of Luke (now identified as 142v) due to water damage.8 Therefore, John followed Luke in the previous arrangement of this codex, and it cannot be taken as a representative of the Western order. GA 0234 remains a doubtful witness because, as only Matthew and John are extant, the full order cannot be determined. It is worth observing that the two latest majuscule witnesses, the tenth-century GA 033 and the tenth-century GA 055, are both catenae: although the commentary text is written in minuscule, they have been placed in the majuscule category because of the script used for the gospel text.9 This makes a total of six Greek papyrus or majuscule manuscripts which clearly attest the Western order, four from the third to the sixth century and two from a later period.

The existence of Greek minuscules with the Western order has been disputed. Although Metzger remarked that “several of the older Greek minuscule MSS” arranged the gospels in this order, he provided no specific details.10 Parker attempted to elaborate on Metzger’s “vague” comment by identifying three potential candidates.11 One was GA 055, which, as we have just noted, has been classified as a majuscule witness because of the script adopted for Mark despite the use of minuscule for the other biblical lemmata and commentary. Further examination discounts both Parker’s other examples. GA 594 is a fourteenth-century gospel manuscript held at the Biblioteca di San Lazaro in Venice (MS 1531). Parker voiced doubts about its sequence based on discrepancies in Caspar René Gregory’s Prolegomena.12 Hans Freiherr von Soden’s list of this manuscript’s contents indicates that the beginning of Matthew, the end of Luke, and the whole of John are missing, although he makes no comment on the order.13 An examination of the digitised microfilm in the NT.VMR confirms that this manuscript follows the traditional order of the gospels.14 Moscow State Historical Museum, Syn. gr. 138 is a twelfth-century gospel manuscript with catenae. Parker, once more relying on Gregory’s Prolegomena, proposed it as an example of the Western order, but Gregory again is inconsistent, listing its sequence as Matthew–John–Mark–Luke in his Textkritik.15 The latter order is found in Christian Friedrich Matthaei’s description of the codex, which has independently been confirmed by Georgi Parpulov.16 This codex offers evidence for another non-traditional sequence for the gospels in the context of a catena, which will be explored further below, but it does not attest the Western order.

While not directly addressing the lack of evidence for Metzger’s assertion, Pierre-Maurice Bogaert suggested that GA 2964 (Bibliothèque nationale de France, Gr. 200), a tenth-century minuscule manuscript, was also a witness to the Western order.17 This is yet another catena manuscript, and Bogaert asserted that it once held all four gospels in the Western order because it has the same types of catena in Matthew and John as GA 033 and 055. This manuscript is lacunose at the beginning and end; its biblical text is abridged, and only Matthew and John are extant. As in the case of GA 0234, it is impossible to be confident about its original order or contents given variations in manuscripts with these catena types, which will be described below.

In contrast to Parker’s endeavour to supply details that were lacking from Metzger’s comment, Wallace rejected it as a mistake, citing an email correspondence with J.K. Elliott.18 However, Wallace immediately went on to draw attention to GA 1411, an eleventh-century codex housed at the National Library of Greece (EBE 95), observing that “If this is a minuscule MS, it is the only one known with the Western order.”19 GA 1411 has an abridged form of the biblical text in an alternating catena format, currently extant in portions of Matthew, John, and Luke.20 Previously, Gregory and Wallace had identified this manuscript as only containing portions of John and Luke.21 My investigation of the manuscript, especially its catenae, resulted in the identification of thirty-eight leaves (fol. 1 and 126–163) covering Matt 8:19–13:3. During rebinding, these leaves had been placed after the leaves on Luke, except for folio one, which belongs between fol. 157 and 158. Wallace’s argument for the Western order was based on codicological features. Unaware of the surviving Matthew leaves, he observed that the codex must have originally contained Matthew because the first quire number of John is κβ (22), while asserting that Mark must have been lost after the missing leaves of Luke.22 The newly identified leaves of Matthew verify that it did indeed come first, but it remains unclear whether the codex included Mark. Gregory, who examined the manuscript in 1886, acknowledged that it was impossible to know how many leaves were missing from the end.23 What is more, Wallace claimed that, like GA 033 and 055, GA 1411 should in fact have been classified as a majuscule manuscript because of the script used for the biblical lemmata—if it qualified as a New Testament manuscript at all.24

This examination has shown that none of the manuscripts noted by Parker, Bogaert, and Wallace can confidently be classified as Greek minuscules with the Western order. In fact, in no case is this order likely to have been present, with the exception of GA 055 (treated as a majuscule). There is therefore no solid evidence to support Metzger’s claim that the Western order is found in some minuscules. Instead, this appears to be an error, perhaps based on misleading descriptions in Gregory’s Prolegomena.

3 Greek Minuscules in the “Western” Order

As part of CATENA, a project at the University of Birmingham, Georgi Parpulov has prepared a catalogue and database of all known Greek New Testament catena manuscripts.25 Among the catenae are five manuscripts in minuscule script with the Western order of the gospels, listed in Table 1.

Table 1
Table 1

Greek minuscules with the Western order

Citation: Novum Testamentum 64, 1 (2022) ; 10.1163/15685365-bja10012

As none of these five manuscripts are included in the official catalogue of Greek New Testament manuscripts, it is unlikely that Metzger would have been aware of them.26 Although the Liste does include some catena manuscripts with an abbreviated biblical text (such as GA 1411), this appears to be through inadvertence as such witnesses are normally excluded.27 All five of these manuscripts have an abridged biblical text, so are not currently eligible for inclusion in the Liste. Even so, they may be taken in support of Metzger’s claim if one is willing to look beyond complete copies of the Greek New Testament.

4 The “Western” Order in Catenae

The seven catena manuscripts with the Western order constitute almost two-thirds of the Greek manuscripts with this sequence and all of those after the sixth century. Therefore, the catena commentaries deserve closer examination as a common trait between these seven and an important difference from the other Western order manuscripts.28 Table 2 lists these seven codices with details about the type of commentary in each gospel according to Parpulov.29 The classification of catenae types follows the Clavis Patrum Graecorum.30

Table 2
Table 2

Types of catenae in each Western-order catena manuscript

Citation: Novum Testamentum 64, 1 (2022) ; 10.1163/15685365-bja10012

It is clear from that table that these are not simply all catenae, but they feature the same types of catenae in Matthew, John, and Luke—with greater variation in Mark. The types in Matthew (C110.1), Luke (C130), and John (C140.1) are the earliest forms of catenae on their respective gospel.31 While variety exists in the type of catena on Mark among these manuscripts, the two primary types, C125.1 and C125.2 are closely related to one another.32 The ninth- and tenth-century witnesses above are also among the oldest surviving New Testament catena manuscripts. Thus, these manuscripts account for both the earliest forms of gospel catenae and the earliest witnesses to those texts.

4.1 Internal Connections between the Gospels

The ways in which the catenae may have influenced the sequence of the gospels have hitherto not been studied in detail.33 In the case of the seven manuscripts in Table 2, the production of the catenae in each gospel and their relationship to one another explains why these manuscripts were copied in the Western order. Joseph Sickenberger and Joseph Reuss both posited that the same sixth-century compiler produced the earliest catenae on Matthew (C125.1), John (C140.1), and Luke (C130). They justified this claim by noting the use of common sources and the similar method of extracting comments from those sources. In addition, in these compilations, the editor did not usually name the source of each scholium.34

The common catenist for Matthew, John, and Luke is further corroborated by internal connections between the gospels in these catenae, which also accounts for the use of the Western order. The catenist left particularly noticeable links between the catenae on Matthew and Luke. The C130 type of catena on Luke contains forty scholia with some variation of the phrase προεγράφη εἰς τὸ κατὰ Ματθαῖον (has already been written in [the Gospel] According to Matthew).35 Immediately preceding this, an indication is given of passages in Luke which have parallels in Matthew. For example, one of the comments states: περὶ τοῦ λεπροῦ καὶ ἐπὶ τοῦ παραλυτικοῦ καὶ περὶ τοῦ τελώνου ἕως περὶ τῶν ἀσκῶν τῶν παλαιῶν καὶ τῶν νέων, προεγράφη εἰς τὸ κατὰ Ματθαῖον (“About the Leper” and “On the Paralytic” and “About the Tax Collector” until “About the Old Wineskins and the New” has already been written in [the Gospel] According to Matthew).36 In this case, the contents of Luke 5:12–39 had already been expounded in the catena on Matthew and are not discussed here. Accordingly, the scholia instruct readers to return to the catena on Matthew for the interpretation of the passages mentioned.37 The manuscript tradition supports this explanation: when both gospels are present, the C130 catena on Luke almost always follows C110.1 in Matthew, or one of the later recensions in the same class (C110.2–C110.4) that built on this early form. The connections between these catenae on Matthew and Luke are thus established both by these cross-references and their codicology.

The relationship between Matthew and John is less obvious; the catena on John does not include προεγράφη comments relating to parallel passages but it does incorporate occasional comments on gospel parallels. Instead, the basis for linking the catenist of Matthew C110.1 with the catenist of John C140.1 is the shared dependence on John Chrysostom and the limited use and attribution of other sources.38 This is confirmed, as in the case of Luke, by a strong codicological connection. Of the thirty-two manuscripts of John C140.1 that also include catenae on Matthew, thirty have the C110.1 type. The two other manuscripts contain one of the other C110 sub-groups.

Even though scholarship has not attributed the earliest catena on Mark (C125.1) to the same figure who wrote the catenae on the other gospels, the question of its authorship is pertinent to a discussion of the Western order.39 One of the primary reasons for not associating the catena on Mark with those on the other gospels is the fact that Mark largely depends on them. According to Reuss, one-third of the scholia on Mark derive from the catenae on the other gospels, especially the comments from Chrysostom included in C125.1.40 This is a striking departure from the compositional practice in the other catenae. What is more, there are no προεγράφη comments in Mark: instead, scholia are provided for more than eighty passages paralleled in the other gospels.41 It is therefore most likely that the earliest catena in Mark is the work of a different compiler. This dissociation of the earliest catenae on Matthew, John, and Luke from that on Mark explains the manuscripts with the Western order of the gospels that consistently contain the same catena type in the first three gospels but differ in Mark. As Reuss noted, the later addition of a catena in Mark “served to supplement the catenae on Matthew, Luke, and John so that one owned a complete commentary on the four gospels.”42 It also clarifies why GA 033 includes the full biblical text of Mark without a catena and why GA 055 treats Mark differently: omitting the kephalaia and titloi for Mark at the beginning of the manuscript, changing for Mark the type of script used to copy the other gospels, and giving full biblical lemmata for Mark only. Their Vorlagen may only have included the three gospel catenae without Mark, to which Mark was added, using a different exemplar, for the sake of completeness.

There are other parallels for the order of composition and exclusion of Mark in catena tradition. The sixth- or seventh-century catenist known as Peter of Laodicea compiled catenae on Matthew (C111), Luke (C132), and John (C141.1), but not Mark, although the order of composition has not been established.43 Nicetas of Heraclea in the twelfth century compiled his three gospel catenae in the order Matthew, John, and then Luke.44 The exclusion of Mark in the production of gospel catenae reflects the primacy of the other gospels, especially Matthew, among early Christian interpreters as seen by the dearth of commentaries and homilies on Mark.45 Additionally, the use of προεγράφη comments in Luke suggests that Mark was left out because it had little unique material after the other three gospels had been interpreted. While excluding Mark from the catena manuscript was more expedient, Mark’s canonical status led to the composition of catenae on Mark and its inclusion in later catena manuscripts. The composition practices of the catenists and the connections between the catenae found in these Western order manuscripts provides a logical explanation for this otherwise exceptional phenomenon.

4.2 The Order of the Gospels in Catenae

The CATENA project’s database of catena manuscripts facilitates detailed comparisons of Greek catenae based on many more manuscripts than was previously possible.46 This database was used to examine whether patterns existed in the sequences involving the types of catenae included in these Western order manuscripts.

First, the catenae types for Matthew, John, and Luke found in the seven Western order manuscripts appear more frequently in non-traditional orders than the other catenae types for these gospels.47 This is shown in Table 3, where the categories “Traditional Order” and “Non-Traditional Order” refer to any combination of gospels that follows the usual order Matthew– Mark–Luke–John or not. Accordingly, a manuscript with only Matthew followed by Mark reflects the traditional order whereas Matthew–John–Luke reflects a non-traditional order even though not all four gospels are present. The category “One Gospel” refers to copies only containing that gospel. These were isolated because they reflect no order.

Table 3
Table 3

Order of the gospels in Greek catenae

Citation: Novum Testamentum 64, 1 (2022) ; 10.1163/15685365-bja10012

The early gospel catenae on Matthew, John, and Luke usually have a non-traditional order, whereas all other catenae types overwhelmingly put the gospels in the traditional order. It is therefore not surprising to find the Western order among manuscripts of the earliest types.

The early catenae types in Matthew, John, and Luke also exhibit greater variety in how they arrange the gospels. Catenae on Luke represent this trend well. Table 4 shows the number of ways the gospels are arranged in each type of catena on Luke, and then indicates the number of manuscripts in a traditional or non-traditional order within that type.48 Here, a gospel arrangement refers to any combination of gospels and order, so a manuscript with Matthew– John–Luke counts as one arrangement and a manuscript with only Matthew– John would count as another arrangement even though they overlap.

Table 4
Table 4

Arrangements of catenae on Luke by type

Citation: Novum Testamentum 64, 1 (2022) ; 10.1163/15685365-bja10012

Manuscripts that include Luke C130 exhibit the most variety, with twelve different arrangements—almost double the next highest total. Among these manuscripts, the majority (35/53) are in a non-traditional order. The most frequently occurring arrangements include: (1) seven in the Western order (Matthew–John–Luke–Mark), consisting of the five minuscules and two majuscules described above, (2) three with John first: John–Matthew–Luke– Mark, (3) fifteen in the Western order without Mark: Matthew–John–Luke, including GA 1411 described above, (4) four that omit Mark from the traditional order: Matthew–Luke–John, and (5) four copies of only John–Luke. Among the eighty-seven multiple-gospel codices with types C131–137, seventy-five are copies of the four gospels in the traditional order. Scribes copying manuscripts of the later types of catenae on Luke clearly preferred a four-gospel manuscript in the traditional order. The catena tradition to which these Western order Greek minuscules belong is characterized by a variety of gospel arrangements, especially utilizing non-traditional orders.49

5 Conclusion

Though Metzger claimed to know of older minuscules with the Western order and other researchers attempted to supply the details he omitted, this article argues that the answer to whether he was correct is both no and yes. Greek continuous-text minuscules with the Western order that contain all four gospels cannot be found. In that sense, Metzger erred. What do exist are five minuscule catenae not catalogued in the Liste with the Western order, along with two majuscules which are also catenae. These seven manuscripts share the same catena types in Matthew, Luke, and John, and it is this commentary that explains their sequence. Matthew C110.1, John C140.1 and Luke C130 were compiled by the same catenist in the order Matthew, John, then Luke. Later manuscripts of these catenae on the three gospels sometimes added the gospel of Mark with a catena derived primarily from extracts on Matthew, in different places in the sequence, resulting in a variety of non-traditional gospel arrangements. Therefore, while these codices are unusual in relation to continuous-text gospel manuscripts and later catenae, they are characteristic of their catenae types. The use of the Western order in at least three of the four other Greek manuscripts (𝔓⁴⁵, GA 05, 032, and possibly 073 + 084) predates the creation of catenae, so a different explanation is needed for why those manuscripts were copied in a non-traditional order. Even so, the catenae may reinforce arguments about the logic which underlies such an arrangement of the gospels.


This article has been prepared and published as part of the CATENA project, which has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 770816). I am grateful to fellow CATENA project members H.A.G. Houghton, Amy Myshrall, and Georgi Parpulov for their insightful comments on this article.


M. Crawford, “The Eusebian Canon Tables as a Corpus-Organizing Paratext within the Multiple-Text Manuscript of the Fourfold Gospel,” in The Emergence of Multiple-Text Manuscripts (ed. A. Bausi, M. Friedrich, and M. Maniaci; Studies in Manuscript Cultures 17; Berlin/Boston: de Gruyter, 2019) 113; M. Hengel, The Four Gospels and the One Gospel of Jesus Christ: An Investigation of the Collection and Origin of the Canonical Gospels (Harrisburg, PA: Trinity Press International, 2000) 42–47.


B.M. Metzger, The Canon of the New Testament: Its Origin, Development, and Significance (New York: Oxford University Press, 1987) 296.


K. Aland et al., eds., Kurzgefasste Liste der griechischen Handschriften des Neuen Testaments (2nd, rev. and enlarged ed.; ANTF 1; Berlin/New York: de Gruyter, 1994), now updated online at


M. Crawford, “A New Witness to the ‘Western’ Ordering of the Gospels: GA 073 + 084,” JTS 69.2 (2018) 477–478. See also Metzger, Canon, 296–297, and D.C. Parker, Codex Bezae: An Early Christian Manuscript and its Text (Cambridge: Cambridge University Press, 1992) 116–118. These sources also give the non-Greek manuscripts including many Old Latin manuscripts and the Gothic manuscript Codex Argenteus. Another list, with the same Greek witnesses, is given in P.-M. Bogaert, “Ordres anciens des évangiles et tétraévangile en un seul codex,” Revue théologique de Louvain 30 (1999) 302–304.


Crawford, “New Witness,” 481–483.


There is another early example of a synoptic cross-reference table giving the gospels in an irregular order. In Codex Zacynthius the kephalaia and titloi are accompanied by a cross-reference table in the order Luke–John–Matthew–Mark. The manuscript, only extant in the first eleven chapters of Luke, likely was only a volume of Luke. In this case, the scribe put the manuscript’s gospel first, then reordered the other three gospels sequentially. See H.A.G. Houghton, P. Manafis, and A. Myshrall, The Palimpsest Catena of Codex Zacynthius: Text and Translation (TS [III] 22; Piscataway, NJ: Gorgias, 2020) 5–6, Fol. IV; H.A.G. Houghton and D.C. Parker, “The Gospel of Luke in the Palimpsest,” in Codex Zacynthius: Catena, Palimpsest, Lectionary (ed. H.A.G. Houghton and D.C. Parker; TS [III] 21; Piscataway, NJ: Gorgias, 2020) 33–36.


D.B. Wallace, “Medieval Manuscripts and Modern Evangelicals: Lessons from the Past, Guidance for the Future,” JETS 60.1 (2017) 9 n. 19.


I am grateful to H.A.G. Houghton for this identification.


The gospels in GA 055 were copied in different scripts within an alternating catena format. The biblical lemmata in Matthew, John, and Luke are abridged and written in minuscule whereas the complete text of Mark was written in majuscule. The biblical text present for the first three gospels would not qualify the manuscript for a GA number, but the full text of Mark warrants the designation as a majuscule manuscript.


Metzger, Canon, 296.


Parker, Codex Bezae, 116.


See further Crawford, “New Witness,” 478 n. 5.


H. Freiherr von Soden, Die Schriften des Neuen Testaments: In ihrer ältesten erreichbaren Textgestalt (2nd ed., 2 pts. in 4 vols.; Göttingen: Vandenhoeck & Ruprecht, 1911) 1:199.




C.R. Gregory, Prolegomena to C. von Tischendorf, Novum Testamentum Editio Octava Critica Maior (Leipzig: Hinrichs, 1890) 3:516; id., Textkritik des Neuen Testamentes (3 vols.; Leipzig: Hinrichs, 1900–1909) 1:172.


C.F. Matthaei, D. Pauli epistolae ad Thessalonicenses et ad Timotheum: Graece et Latine; Varias lectiones codd. Mosqq. scholia Graeca inedita et animadversiones criticas adiecit ac denvo recensvit (Riga: Hartknoch, 1785) 222–225; G. Parpulov, Catena Manuscripts of the Greek New Testament: A Catalogue (TS [III] 25; Piscataway, NJ: Gorgias, 2021) 40.


Bogaert, “Ordres anciens des évangiles et tétraévangile,” 302.


Elliott wrote: “Metzger was wrong (a rare occurrence!) re ‘several’ minuscules with the W order” (Wallace, “Medieval Manuscripts and Modern Evangelicals,” 9 n. 19).


Wallace, “Medieval Manuscripts and Modern Evangelicals,” 9.


D.C. Smith, “A Study of the Text and Paratext of the Catena on Luke in GA 1411” (ThM, Dallas Theological Seminary, 2019). In a master’s thesis focusing on Luke 1 in GA 1411, Smith found that the manuscript only included 502 of the 1200 words in Luke 1 according to the Robinson-Pierpont edition of the Greek New Testament and therefore suggested that its inclusion in the Liste should be reconsidered. Wallace, who supervised Smith’s thesis, also suggested GA 1411 might not belong in the Liste (“Medieval Manuscripts and Modern Evangelicals,” 9).


Gregory, Textkritik, 1:261; Wallace, “Medieval Manuscripts and Modern Evangelicals,” 9.


Wallace, “Medieval Manuscripts and Modern Evangelicals,” 9 n. 22.


Gregory, Textkritik, 1:261. I. Sakkelion made a similar observation: Κατάλογος των Χειρόγραφων της Εθνικής Βιβλιοθήκης της Ελλάδος [Catalogue of the manuscripts of the National Library of Greece] (Athens: Ethnikon Tupographeion, 1892) 18.


Wallace, “Medieval Manuscripts and Modern Evangelicals,” 9.


Parpulov, Catena Manuscripts, which was preceded by a working paper: G.R. Parpulov, “A Checklist of Greek New Testament Catena Manuscripts,” University of Birmingham, 2018,


The two Moscow manuscripts could be the exception as they were described by Matthaei (noted in Table 1), Gregory, and von Soden: Gregory, Textkritik, 2:855; von Soden, Schriften des Neuen Testaments, 1:539–540.


Parker discussed the inconsistencies and complexities in classifying catenae as Greek New Testament manuscripts in D.C. Parker, Textual Scholarship and the Making of the New Testament (Oxford: Oxford University Press, 2014) 40–52, and also in H.A.G. Houghton and D.C. Parker, “An Introduction to Greek New Testament Commentaries with a Preliminary Checklist of New Testament Catena Manuscripts,” in Commentaries, Catenae, and Biblical Tradition: Papers from the Ninth Birmingham Colloquium on the Textual Criticism of the New Testament in Conjunction with the COMPAUL Project (ed. H.A.G. Houghton; TS [III] 13; Piscataway, NJ: Gorgias, 2016) 2–4 and 25–27.


A connection between the Western order catena manuscripts and the earlier continuous-text Western order gospel manuscripts cannot be proven by their order alone. While the catenist may have used a Western order gospel book as the exemplar for the biblical text, it seems more likely that the compiler intentionally chose the Western order for the catena while using a traditional order manuscript. First, the traditional order was established and more popular by the sixth century. Second, the catenist used the Eusebian apparatus and kephalaia and titloi to navigate between the four gospels (see n. 37). Finally, reordering the gospels may have been an intentional decision to facilitate interpretation and pedagogy. See W.R.S. Lamb, The Catena in Marcum: A Byzantine Anthology of Early Commentary on Mark (TENTS 6; Leiden: Brill, 2012) 75–102, on the pedagogical function of catenae. Historical evidence for this purpose comes from Nicetas of Heraclea who compiled catenae on the gospels while he held the διδάασκαλος τοῦ εὐαγγελίου (teacher of the gospels) post in Constantinople; see C.Τ. Krikones, Συναγωγή Πατέρων εις το κατά Λουκάν Ευαγγέλιον υπό Νικήτα Ηρακλείας (κατά τον κώδικα Ιβήρων 371) [Compilation of the fathers in the Gospel according to Luke: By Nicetas of Heraclea (according to Codex Iveron 371)] (2nd ed., Βυζαντινά κείµενα και µελέτες 9; Thessaloniki: Κέντρο Βυζαντινών Ερευνών, 1976) 21; B. Roosen, “The Works of Nicetas Heracleensis () τo Σερρῶν,” Byzantion 69.1 (1999) 139–142.


Parpulov, Catena Manuscripts.


M. Geerard and J. Noret, eds., Clavis Patrum Graecorum: IV Concilia; Catenae (2nd ed.; CCSG 4; Turnhout: Brepols, 2018).


The CPG’s dating and classification of catenae types for Matthew, Mark, and John follows Joseph Reuss’s conclusions made in his three volumes on those gospel catenae: Matthäus-, Markus- und Johannes-Katenen nach den handschriftlichen Quellen untersucht (NTAbh 18/4–5; Münster: Aschendorff, 1941); Matthäus-Kommentare aus der griechischen Kirche aus Katenenhandschriften gesammelt und herausgegeben (TUGAL 61; Berlin: Akademie-Verlag, 1957); Johannes-Kommentare aus der griechischen Kirche (TUGAL 89; Berlin: Akademie-Verlag, 1966). The basis for Luke is M. Rauer, Origenes Werke, vol. 9: Die Homilien zu Lukas in der Übersetzung des Hieronymus und die griechischen Reste der Homilien und des Lukas-Kommentars (2nd ed.; GCS 49; Berlin: Akademie-Verlag, 1959), which was later supplemented with findings from Reuss’s volume on Luke: Lukas-Kommentare aus der griechischen Kirche (TUGAL 130; Berlin: Akademie-Verlag, 1984). For a concise English summary of catena types and the dating of catenae, see W.R.S. Lamb, “Conservation and Conversation: New Testament Catenae in Byzantium,” in The New Testament in Byzantium (ed. D. Krueger and R.S. Nelson; Dumbarton Oaks Byzantine Symposia and Colloquia; Washington D.C.: Dumbarton Oaks, 2016) 277–300.


Lamb, Catena in Marcum, 52–53.


Gregory perceptively proposed that the order of writing the commentary may have influenced the order of the gospels in GA 033 in the addendum to the third volume: Textkritik, 3:1356.


J. Sickenberger, Titus von Bostra: Studien zu dessen Lukashomilien (TU 6/1; Leipzig: Hinrichs, 1901) 40; Rauer, Origenes Werke, XXXVIII; Reuss, Matthäus-, Markus- und Johannes- Katenen, 20; id., Lukas-Kommentare, XII.


F.D. Duc, Bibliothecae Veterum Patrum seu Scriptorum Ecclesiasticorum: Tomus Secundus Graecolatinus (vol. 2; Paris: Compagnie de la Grand-Navire, 1624) 780. See also J.A. Cramer’s edition of the C131 type that includes all the προεγράφη comments from C130: Catenae in Evangelia S. Lucae et S. Joannis, ad Fidem Codd. Mss., vol. 2 of Catenae Graecorum Patrum in Novum Testamentum (Oxford: Typographeo Academico, 1844) 29.27.


See Paris, BnF, Grec. 188, fol. 156r.


Multiple earlier scholars have also recognized the link between catenae on Matthew and Luke via the προεγράφη comments: R. Devreesse, “Chaînes exégétiques grecques,” in Dictionnaire de la Bible, Supplément (Paris: Letouzey & Ané, 1928) 1:1182; Reuss, Lukas-Kommentare, XI; M. Rauer, Der dem Petrus von Laodicea zugeschriebene Lukaskommentar (NTAbh 8; Münster: Aschendorff, 1920) 71; id., Origenes Werke, XXXVIII; Sickenberger, Titus von Bostra, 24; Harold Smith, “The Sources of Victor of Antioch’s Commentary on Mark,” JTS 19.76 (1918) 352. Lamb demonstrated the catenist on Mark used the Eusebian apparatus to locate parallels to other gospels (Catena in Marcum, 145–149). The locations of the προεγράφη comments in these catenae on Luke suggest the catenist likewise used the Eusebian apparatus and the kephalaia to identify and label parallels. On the use of the Eusebian apparatus and kephalaia in gospels reading and exegesis see J. Coogan, “Mapping the Fourfold Gospel: Textual Geography in the Eusebian Apparatus,” JECS 25.3 (2017) 354–357; id., “Transmission and Transformation of the Eusebian Gospel Apparatus in Greek Medieval Manuscripts” in Canones: The Art of Harmony (ed. A. Bausi, B. Reudenbach, and H. Wimmer; Studies in Manuscript Cultures 18; Berlin/Boston: de Gruyter, 2020) 42–44; id., “Transforming Textuality: Porphyry, Eusebius, and Late Ancient Tables of Contents,” Studies in Late Antiquity 5.1 (2021) 21–27; M.R. Crawford, “Ammonius of Alexandria, Eusebius of Caesarea and the Origins of Gospels Scholarship,” NTS 61.1 (2015) 26–28; and id., The Eusebian Canon Tables: Ordering Textual Knowledge in Late Antiquity (Oxford: Oxford University Press, 2019) 96–121; J.R. Edwards, “The Hermeneutical Significance of Chapter Divisions in Ancient Gospel Manuscripts,” NTS 56.3 (2010) 413–426.


Reuss, Matthäus-, Markus- und Johannes-Katenen, 155; id., Johannes-Kommentare, XI.


See Lamb for a survey of views on the composition of the catena on Mark. Lamb himself concluded that an unknown figure began compiling comments on Mark sometime between 490 and 553 CE (Catena in Marcum, 32–58, 71–73).


Reuss, Matthäus-, Markus- und Johannes-Katenen, 138–141. See also Lamb, Catena in Marcum, 147; Smith, “Sources of Victor of Antioch’s Commentary on Mark,” 352.


Lamb, Catena in Marcum, 147.


Reuss, Matthäus-, Markus- und Johannes-Katenen, 141: “Er diente zur Ergänzung der Katenen zu Mt, Lk und Jo, um für die 4 Evv ein abgeschlossenes Kommentarwerk in einer Hs zu besitzen.”


Rauer, Petrus von Laodicea, 6–7; id., Origenes Werke, XXXIX; Sickenberger, Titus von Bostra, 128. Note that Reuss was not certain the same person composed the C132 catena on Luke, but he did not rule out the possibility either (Lukas-Kommentare, XIII).


Roosen, “Works of Nicetas Heracleensis,” 142. Likewise, the fourth-century Latin commentator Fortunatianus of Acquileia omitted Mark from his gospel commentary and only commented on sections of Luke and John which were not paralleled in Matthew: see H.A.G. Houghton, Fortunatianus of Aquileia, Commentary on the Gospels: English Translation and Introduction (in association with Lukas J. Dorfbauer; CSEL Extra Seriem; Berlin/Boston: de Gruyter, 2017) XVI, 25–26.


M. Bockmuehl, “The Making of Gospel Commentaries,” in The Written Gospel (ed. M. Bockmuehl and D.A. Hagner; Cambridge: Cambridge University Press, 2005) 286–289; Hengel, Four Gospels, 39; B.D. Schildgen, Power and Prejudice: The Reception of the Gospel of Mark (Detroit: Wayne State University Press, 1999) 38–41.


“Database of Greek New Testament Catena Manuscripts,” Access to catena manuscripts due to external circumstances and an incomplete catalogue of them is a significant limitation in earlier studies, as noted by D.C. Parker and J.N. Birdsall, “The Date of Codex Zacynthius (Ξ): A New Proposal,” JTS 55.1 (2004) 125–126: “Neither he [Rauer] nor Reuss studied more than a small proportion of the total number of [catena] manuscripts.”


Mark has not been considered in these comparisons for two reasons: first, the catenae on Mark in the Western order came in more types, meaning there was not a pattern among the manuscripts being studied in this article. Second, catenae in Mark were copied in fewer distinguishable types than the other gospels; the CPG lists only three types in Mark—two related recensions and a separate codices singuli category. The result of the lack of development in the Markan tradition is that catenae on Mark follow less discernable patterns and each of the main types appears in manuscripts with various forms of catenae on the other gospels.


See P. Manafis, “Catenae on Luke and the Catena of Codex Zacynthius,” in Codex Zacynthius: Catena, Palimpsest, Lectionary (ed. H.A.G. Houghton and D.C. Parker; TS [III] 21; Piscataway, NJ: Gorgias, 2020) 137–147, for a current survey of the catena types in Luke.


The dates of the manuscripts for all types in Luke range between the ninth and sixteenth centuries, and whether the older or newer manuscripts use traditional or non-traditional gospel arrangements holds throughout each century depending on which catena on Luke was being copied.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 758 758 37
PDF Views & Downloads 1265 1265 31