Brent Nongbri - The ‘Objective’ Criteria for Dating Codex Sinaiticus

[Steven Avery]

The Date of Codex Sinaiticus
Brent Nongbri
https://academic.oup.com/jts/advance-article/doi/10.1093/jts/flac083/6652265?searchresult=1

2. The ‘Objective’ Criteria for Dating Codex Sinaiticus

To establish the ‘fourth century’ date, Roberts referred exclusively to the landmark study published by H. J. M. Milne and Theodore Skeat in 1938, Scribes and Correctors of the Codex Sinaiticus, which provides a detailed argument that Codex Sinaiticus was likely copied ‘before the middle of the [fourth] century.’ Here is how Roberts summarized their arguments in three points:

• A terminus post of c. A.D. 300–40 is provided by the Eusebian sections.
• Certain cursive notes, one of which can be seen in our plate (col. ii, l. 12), are in a distinctively fourth-century hand.
• The system of representing numerals points to a fourth-century date. In this century the practice of representing, for example, 1,000 by a stroke below the letter A (/A) replaces the old system of putting a curl above the letter (A͗). Milne and Skeat assign this change approximately to the years 338–60. As the codex was written to dictation and as it is certain that in some places in the exemplar the numerals were written out in full, the use of the old system is evidence of fourth-century date.22

Roberts thus provided a concise précis of the more ‘objective’ arguments for the date of the codex. In what follows, I will argue that point (a) is valid, point (c) is invalid, and point (b) is considerably more complicated than Roberts lets on. I will discuss the arguments in that order (a, c, b).

The terminus post quem mentioned by Roberts (the presence of the Eusebian canon and section numbers) is not controversial. The Eusebian apparatus as it appears in Sinaiticus has some anomalous features, but it seems almost certain that the Eusebian numbers were a part of the original production of the codex and not a later addition.23 The surviving evidence suggests that the Eusebian numbers were added after an early correction of the manuscript by scribe D but before the insertion of a replacement bifolium (again by scribe D) in the second quire of Matthew.24 The use of the canon and section numbers cannot predate their creation by Eusebius. The exact date that Eusebius developed and disseminated the system of canon and section numbers is not precisely known, but the terminus post quem of 300–340 offered by Roberts is reasonable.25

Less compelling is the argument that Roberts mentions in connection with changing customs of representing numerals. Roberts notes that over the course of the fourth century, one system of representing multiples of 1,000 with a curl (A͗) was replaced by a new system using a stroke (/A). Roberts concluded that ‘as the codex was written to dictation and as it is certain that in some places in the exemplar the numerals were written out in full, the use of the old system is evidence of fourth-century date’. There are at least two problems here. It will be useful to review what Milne and Skeat actually wrote in some detail:

The second point is the forms of certain numerals used in the text of 1 Maccabees. In the course of the fourth century the old method of representing the figures 1,000–9,000 by the ordinary cardinal numbers for 1–9 with a surmounting curl or crest (e.g. A͗ = 1,000, B͗ = 2,000, etc.) gradually went out of fashion, the curl being replaced by a simple slanting stroke to the left of the numeral (e.g. /A or /A = 1,000). … [Milne and Skeat then provide a table of dated papyri to show the window of dates for the shift.] From these data it can be seen that the change from the old to the new system took place about the years A.D. 338–60. In the Sinaiticus we still find the earlier system, B͑ in O.T. 47, col. 1, and Ͱ͑ in O.T. 43, col. 1, and 47, col. 1. All these occur in 1 Maccabees; elsewhere thousands are written out in words, as regularly in the Vaticanus. We may reasonably assume that in 1 Maccabees at least these numbers were represented by numerals in the exemplar, since this alone can explain the erroneous τρισχιλίους δέκα for ὀκτακισχιλίους in 1 Macc. v. 34 (i.e. Ͱ͑I for H͑), and the extraordinary series of numerals in 1 Macc. v. 20 quoted above (p. 57). It might in consequence be argued that the shapes of the numerals in the exemplar had influenced the copyist of the Sinaiticus. But now that we have shown that the manuscript was written from dictation, this possibility is all but excluded, and we can have confidence in the validity of the scribe’s own shapes as a criterion. If this is so, the Sinaiticus is not likely to be much later than about A.D. 360.26

First, as far as I can see, Milne and Skeat do not claim that ‘it is certain that in some places in the exemplar the numerals were written out in full’, as Roberts asserts. Rather, Milne and Skeat note that outside this small number of examples in 1 Maccabees, thousands are spelled out as words within Codex Sinaiticus itself. The fact that this older system using numerals with curls is present only in 1 Maccabees suggests that a copyist simply carried them over from an exemplar. Although Milne and Skeat mentioned this seemingly reasonable explanation, they rejected it because they believed that Codex Sinaiticus had been copied by dictation rather than sight.

This brings us to the second major problem. For the logic of Milne and Skeat’s argument about numerals to be convincing, it is necessary to assent that Sinaiticus was copied by dictation. But the argument of Milne and Skeat in favor of dictation has proven persuasive to almost nobody.27 Indeed, a recent article in the Journal of Biblical Literature has demonstrated that what Skeat regarded as ‘positive proof of dictation’ (the nonsense sequence of characters in 1 Macc. 5:20) was in fact based on a mistaken reading by Milne and Skeat.28 Barring some new and compelling evidence that Sinaiticus was copied by dictation, the argument about the orthography of numbers can carry no weight at all in the question of the date of the copying of the codex.

The other argument mentioned by Roberts, the presence of ‘certain cursive notes’ in ‘a distinctly fourth century hand’ also deserves more intensive scrutiny. Here is what Milne and Skeat say on the matter:

In the marginal additions made by scribe D while correcting the New Testament the directional signs are frequently supplemented with the words ανω and κατω, the former being placed in the lower margin and the latter opposite the place in the text (N.T. 2b, 66b, 73, 74, 80, 82, 92). These words are written in cursive script (no doubt to distinguish them from the text proper), and slender though the evidence of a few isolated words must be, they certainly belong to the fourth century, and probably the first half of it.29

 

[Steven Avery]

Footnotes​

1
Thanks to the organizers of the Text and Transmission Joint Research Seminar at KU Leuven and Ghent University for the opportunity to present portions of this material to a wide audience that provided helpful feedback. I am also indebted to Hugo Lundhaug, Gregg Schwendner, Mary Jane Cuyler, and Zachary J. Cole for comments on earlier drafts of this essay. Thanks also to the JTS production team for overcoming a variety of formatting challenges. This work was supported by the Research Council of Norway (project number 314240).
2
The online catalog entry for Codex Sinaiticus at the British Library (Add MS 43725) lists the date of the codex as ‘2nd quarter of the 4th century – 3rd quarter of the 4th century’ (<http://searcharchives.bl.uk/IAMS_VU2:IAMS032-002169711>). The ‘Reference Guide’ accompanying the 2010 photographic facsimile of Codex Sinaiticus comments only that ‘Codex Sinaiticus is generally dated to the fourth century, and sometimes more precisely to the middle of that century. This is based on a study of the handwriting’. A recent collection of essays produced by the British Library and dedicated to Codex Sinaiticus contained almost no discussion of the date of the codex, just a summary statement: ‘There is a strong consensus that Codex Sinaiticus belongs to the fourth century, and there are no good grounds to dispute that’. See Harry Gamble, ‘Codex Sinaiticus in its Fourth Century Setting’, in Scot McKendrick, David Parker, Amy Myshrall, and Cillian O’Hogan (eds.), Codex Sinaiticus: New Perspectives on the Ancient Biblical Manuscript (London: The British Library, 2015), pp. 3–18, at p. 6.
3
Constantinus Tischendorf, Codex Friderico-Augustanus sive fragmenta Veteris Testamenti (Leipzig: Koehler and Uckermann, 1846), p. 22: ‘cum magna veritatis specie medio fere seculo quarto eum adscripturus mihi videor’.
4
The number of copyists involved in the production of Codex Sinaiticus is a matter of ongoing discussion. See D. C. Parker, Codex Sinaiticus: The Story of the World’s Oldest Bible (London: British Library; Peabody, Mass.: Hendrickson, 2010), pp. 48–51.
5
On the various correctors, see most recently Parker, Codex Sinaiticus, 79–90.
6
See Constantin Tischendorf, Nachricht von der im Auftrage seiner kaiserlichen Maiestät Alexander II unternommenen herausgabe der Sinaitischen Bibelhandschrift (Leipzig: Giesecke & Devrient, 1860), p. 18: ‘Hierauf hab’ ich eine Mittheilung über das Alter der Handschrift verheissen; sie soll jedoch auf wenige Hauptstücke beschränkt bleiben, indem alles Ausführlichere für die Prolegomena aufzusparen ist. Vor Allem kann ich nur mit Nachdruck wiederholen was ich bereits bei der ersten ins Vaterland gegebenen Kunde von dem aufgefundenen Schatze von Cairo aus geschrieben: “Für diese Handschrift nun bedarf es wenigstens zur Feststellung des Jahrhunderts ihrer Entstehung kaum eines Datums; denn dass sie im vierten christlichen Jahrhundert geschrieben sei, das lässt sich mit allen Argumenten, die in der paläographischen Wissenschaft gelten, fast ausser allen Zweifel stellen.” Bei der genaueren Untersuchung kann nur das Eine zweifelhaft sein, ob die Handschrift schon vor der Mitte des vierten Jahrhunderts oder erst in der zweiten Hälfte desselben geschrieben sei’. For the more extended arguments in the Prolegomena to his facsimile edition, see Constantinus Tischendorf, Bibliorum Codex Sinaiticus Petropolitanus (4 vols.; St. Petersburg, 1862), pp. 1.11*–1.14*.
7
For instance, the British New Testament textual critic Samuel Tregelles, despite having a strained relationship with Tischendorf, agreed with the fourth century assignment. After examining the New Testament leaves in Leipzig in 1862, he wrote: ‘I believe I know something of Greek MSS and I am positively convinced that this is a manuscript of the fourth century’. See Timothy C. F. Stunt, ‘Some Unpublished Letters of S. P. Tregelles Relating to the Codex Sinaiticus’, The Evangelical Quarterly 48 (1976), pp. 15–26, at p. 19. See also Frederick H. Scrivener, A Full Collation of the Codex Sinaiticus with the Received Text of the New Testament (London: Bell and Daldy, 1864), pp. xiii-xl, at p. xxix: ‘Codex Sinaiticus is coeval with its rival in the Vatican, and consequently a record of the fourth century of the Christian era.’ Scrivener had not seen the manuscript himself. He based his judgements primarily on the 17 reproductions published in the first volume of Tischendorf, Bibliorum Codex Sinaiticus Petropolitanus.
8
On Simonides, see J. Keith Elliott, Codex Sinaiticus and the Simonides Affair (Thessaloniki: Patriarchal Institute for Patristic Studies, 1982).
9
Thompson, Handbook of Greek and Latin Palaeography (New York: D. Appleton, 1893), p. 150.
10
Thompson, An Introduction to Greek and Latin Palaeography (Oxford: Clarendon, 1912), p. 200.
11
Bell, ‘Early Codices from Egypt’, The Library 10 (1909), pp. 303–13, at p. 307.
12
See Lake, Codex Sinaiticus Petropolitanus: The New Testament, The Epistle of Barnabas, and the Shepherd of Hermas (Oxford: Clarendon, 1911), p. x.
13
See Milne and Skeat, Scribes and Correctors of the Codex Sinaiticus (London: The British Museum, 1938), pp. 60–65. The arguments of Milne and Skeat will be treated in detail below.
14
Colin H. Roberts, Greek Literary Hands 350 B.C. – A.D. 400 (Oxford: Clarendon, 1956), p. 24.
15
Guglielmo Cavallo, Ricerche sulla maiuscola biblica (Florence: Le Monnier, 1967), pp. 58–61: phrased variously, as ‘una data intorno al 360 ca. o solo di qualche anno più tarda’ (p. 58) or ‘intorno al 360 ca. o poco più tardi’ (p. 60). A similar date—‘IV2 (ca. 360)’—has been advocated on palaeographic grounds more recently by Pasquale Orsini, Manoscritti in maiuscola biblica: Materiali per un aggiornamento (Cassino: Edizioni dell’Università degli Studi di Cassino, 2005), p. 240.
16
Theodore C. Skeat, ‘The Codex Sinaiticus, the Codex Vaticanus, and Constantine’, JTS 50 (1999), pp. 583–625. This view is common, but to my knowledge nobody has argued the point as thoroughly as Skeat.
17
Timothy Janz, ‘Greek Paleography: From Antiquity to the Renaissance’ (<https://spotlight.vatlib.it/greek-paleography/feature/1-majuscule-bookhands>). See also the review of Cavallo by Peter J. Parsons in Gnomon 42 (1970), pp. 375–80. For general caution about the use of palaeographic evidence to generate narrow date ranges for Greek literary manuscripts of the Roman era, see Brent Nongbri, ‘Palaeographic Analysis of Codices from the Early Christian Period: A Point of Method’, JSNT 42 (2019), pp. 84–97; Brent Nongbri, God’s Library: The Archaeology of the Earliest Christian Manuscripts (New Haven: Yale University Press, 2018), pp. 47–82; and Christian Askeland, ‘Dating Early Greek and Coptic Literary Hands’, in Hugo Lundhaug and Lance Jenott (eds.), The Nag Hammadi Codices and Late Antique Egypt (Tübingen: Mohr Siebeck, 2018), pp. 457–89.
18
Milne and Skeat, Scribes and Correctors, p. 62, my italics.
19
See, for instance, Anthony Grafton and Megan Williams, Christianity and the Transformation of the Book: Origen, Eusebius, and the Library of Caesarea (Cambridge, Mass.; Belknap, 2006), pp. 215–21.
20
Gamble, ‘Codex Sinaiticus in its Fourth Century Setting’, p. 9. This point was recognized already by Brooke Foss Westcott and Fenton John Anthony Hort, The New Testament in the Original Greek: Introduction, Appendix (Cambridge: Macmillan, 1882), p. 74.
21
See the critical discussion in Parker, Codex Sinaiticus, pp. 19–24.
22
Roberts, Greek Literary Hands, p. 24.

23
The strange features include (1) the fact that no Eusebian canon tables survive in Sinaiticus, either at the mutilated beginning of the codex or at the start of the New Testament, (2) the section numbers are only partially present (they are missing for sections 107–242 in Luke), and (3) the first 52 sections in Matthew are more elaborately executed than the rest of the sections. Milne and Skeat have explained this situation by noting that, according to one sequence of quire signatures, there is a full quire missing between the last quire of the Old Testament and the first quire of the New Testament. They hypothesize that a quire containing a set of tables was planned for but never completed because the effort to add the section numbers was abandoned before it was finished, thus also explaining the abandonment of the extra decorations after section 53 in Matthew and the complete lack of Eusebian numbers in much of Luke (Scribes and Correctors, pp. 7–9 and 36–7). In the absence of other data, this solution seems the least implausible alternative (if the canon tables had been completed and contained in the codex, it is hard to explain why the missing section numbers in Luke were not added by any later users of the codex).

24
A correction in the lower margin at Matt. 10:39 carries a section number in identical red ink and made in sequence with the section numbers used in the main text, quire 74 (=New Testament quire 1), folio 6r, column 3. The bifolium consisting of New Testament folios 10 and 15 is part of a quire copied by scribe A, but this single bifolium is copied by scribe D and lacks the Eusebian numbers (the surrounding leaves copied by scribe A all have the Eusebian numbers). See the discussion in Milne and Skeat, Scribes and Correctors, p. 36.
25
Compare the phrasing of Kirsopp Lake: ‘It is unfortunate that we do not know the exact date when Eusebius made his apparatus, but it is at least plain that the first quarter of the fourth century is the earliest date which has any reasonable probability’ (Codex Sinaiticus, pp. ix–x). See further Matthew R. Crawford, The Eusebian Canon Tables: Ordering Textual Knowledge in Late Antiquity (Oxford: Oxford University Press, 2019), pp. 79–80, especially n. 73.
26
Milne and Skeat, Scribes and Correctors, pp. 62–4.
27
See, for example, Lake’s review of Milne and Skeat in Classical Philology 37 (1942), pp. 91–6, at pp. 94–5; Parker, Codex Sinaiticus, pp. 54–5; and especially Dirk Jongkind, Scribal Habits of Codex Sinaiticus (Piscataway: Gorgias Press, 2007), pp. 250–52.
28
Zachary J. Cole, ‘An Unseen Paleographical Problem with Milne and Skeat’s Dictation Theory of Codex Sinaiticus’, JBL 135 (2016), pp. 103–7. For the quotation from Skeat, see Theodore C. Skeat, ‘The Use of Dictation in Ancient Book Production’, PBA 42 (1956), pp. 179–208, reprinted in J. K. Elliott (ed.), The Collected Biblical Writings of T.C. Skeat (Leiden: Brill, 2004), pp. 3–32, at p. 17.
29
Milne and Skeat, Scribes and Correctors, p. 62.
30
Milne and Skeat, Scribes and Correctors, p. 62 n. 1. Lake seems to have assigned the corrections associated with these cursive notes to corrector A2, whom he regarded as ‘almost certainly identical with scribe D’. See Lake, Codex Sinaiticus, Plate II (for the assignment of one of these corrections to corrector A2) and p. xxii (for the identification of corrector A2 and scribe D).
31
I have not been able to inspect the relevant leaves in person to judge the question of the identity of the ink, so I must defer to the judgement of Milne and Skeat on this point at present. It should be noted that this is an area in which chemical analysis of the ink would be a very useful undertaking. I am grateful to Jesse Grenz for alerting me to the presence of similar notes on four pages in Codex Vaticanus, a fact of which I was unaware. These notes are discussed briefly in Pietro Versace, I marginalia del Codex Vaticanus (Vatican City: Biblioteca apostolica vaticana, 2018), pp. 14–18 (thanks to Grenz for the reference). Without first chemically confirming the identity of the ink of the ‘cursive’ notes in Sinaiticus and the certain corrections of Sinaiticus Scribe D, I hesitate to follow the chain of logic that leads Versace to favor the hypothesis that Sinaiticus and Vaticanus were produced in the same scriptorium.
32
In discussions of the palaeography of Greek writing of the Roman era, the descriptor ‘cursive’ carries considerable weight because the number of securely dated examples of cursive writing is much greater than the number of securely dated examples of ‘literary’ writing. Thus, it is generally assumed that it is easier to establish more precise and accurate palaeographic dates for samples of undated cursive writing by comparison to dated samples. Yet, Eric Turner has sounded a note of caution on this score, pointing out ‘how little truth there is in the facile, often repeated dictum that cursive, quickly written handwritings are easier to date than literary hands. Both types of activity are equally aleatory’. See Turner, ‘Writing Material for Businessmen’, BASP 15 (1978), pp. 163–9, at p. 164.
33
See Herbert C. Youtie, ‘P.Mich.inv. 6223: Transtigritani’, ZPE 21 (1976), pp. 25–6.
34
See Guido Bastianini and Guglielmo Cavallo, ‘Un nuovo frammento di lettera festale (PSI inv. 3779)’, in Guido Bastianini and Angelo Casanova (eds.), I papiri letterari cristiani: Atti del convegno internazionale di studi in memoria di Mario Naldini (Florence: Istituto Papirologico ‘G. Vitelli’, 2011), pp. 31–45. The script of PSI XVI 1576 is a classic example of the so-called Alexandrian Majuscule. I do not wish to say that any of the meager selection of letters from the notes in Codex Sinaiticus should be classified in this way; I am only pointing out that the use of similar isolated majuscule letter forms is attested in the early fifth century.
35
See Henrik Zilliacus, Vierzehn Berliner griechische Papyri: Urkunden und Briefe (Helsinki: Societas Scientiarum Fennica, 1941), pp. 39–46.
36
This does not mean that no such examples exist, just that if they do, they have not come across my desk.
37
For a technical introduction, see R. E. Taylor and Ofer Bar-Yosef, Radiocarbon Dating: An Archaeological Perspective (2nd edn., Walnut Creek, California: Left Coast Press, 2014). For the specific benefits and drawbacks of the use of radiocarbon analysis on papyrus and parchment manuscripts, see Nongbri, God’s Library, pp. 72–80.
38
For general discussion of the analysis of parchment and pre-treatment methods, see Fiona Brock, ‘Radiocarbon Dating of Historical Parchments’, Radiocarbon 55 (2013), pp. 353–63. For an excellent overview of the recent radiocarbon dating of early Islamic manuscripts copied on different media, see Eva Mira Youssef-Grob, ‘Radiocarbon (14C) Dating of Early Islamic Documents: Background and Prospects’, in Andreas Kaplony and Michael Marx (eds.), Qurʾān Quotations Preserved on Papyrus Documents, 7th – 10th Centuries and the Problem of Carbon Dating Early Qurʾāns (Leiden: Brill, 2019), pp. 138–87.
39
See, e.g., Youssef-Grob, ‘Radiocarbon (14C) Dating of Early Islamic Documents,’ p. 181.
40
See T. M. Kasso, M. J. Oinonen, K. Mizohata, J. K. Tahkokallio, and T. M. Heikkilä, ‘Volumes of Worth—Delimiting the Sample Size for Radiocarbon Dating of Parchment’, Radiocarbon 63 (2021), pp. 105–20.
41
Given the excellent quality of parchment used in Codex Sinaiticus, I assume that the skins were prepared specifically for the production of the codex and that the animals were killed at a time not very long before the codex was produced. For a discussion of the parchment in Sinaiticus, see Gavin Moorhead, Sara Mazzarino, Flavio Marzo, and Barry Knight, ‘A Physical Perspective of Codex Sinaiticus: An Overview from the British Library Folios’, in McKendrick, Parker, Myshrall, and O’Hogan, Codex Sinaiticus: New Perspectives, pp. 221–38.
42
I am grateful to Josephine Dru for help in accurately formulating this summary.
43
On IntCal20, see Paula J. Reimer et al., ‘The IntCal20 Northern Hemisphere Radiocarbon Age Calibration Curve (0–55 CAL kBP)’, Radiocarbon 62 (2020), pp. 725–57.
44
For a joint statement of the holding institutions regarding claims to the codex, see at the project website ‘History of Codex Sinaiticus’ (<https://codexsinaiticus.org/en/codex/history.aspx>). For a critical analysis of the modern history of the codex, see Christfried Böttrich, ‘One Story—Different Perspectives: The Discovery of Codex Sinaiticus’, in McKendrick, Parker, Myshrall, and O’Hogan, Codex Sinaiticus: New Perspectives, pp. 173–87.
45
Again, reliable analysis would require only about 1 cm2 (or less) of uninscribed parchment.



© The Author(s) 2022. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
© The Author(s) 2022. Published by Oxford University Press.

 

[Steven Avery]

He maintained a date for the production of the codex in the fourth century while allowing some leeway on whether it should be assigned to the first or second half of the century.6

6 See Constantin Tischendorf, Nachricht von der im Auftrage seiner kaiserlichen Maiestät Alexander II unternommenen herausgabe der Sinaitischen Bibelhandschrift (Leipzig: Giesecke & Devrient, 1860), p. 18:

‘Hierauf hab’ ich eine Mittheilung über das Alter der Handschrift verheissen; sie soll jedoch auf wenige Hauptstücke beschränkt bleiben, indem alles Ausführlichere für die Prolegomena aufzusparen ist. Vor Allem kann ich nur mit Nachdruck wiederholen was ich bereits bei der ersten ins Vaterland gegebenen Kunde von dem aufgefundenen Schatze von Cairo aus geschrieben: “Für diese Handschrift nun bedarf es wenigstens zur Feststellung des Jahrhunderts ihrer Entstehung kaum eines Datums; denn dass sie im vierten christlichen Jahrhundert geschrieben sei, das lässt sich mit allen Argumenten, die in der paläographischen Wissenschaft gelten, fast ausser allen Zweifel stellen.” Bei der genaueren Untersuchung kann nur das Eine zweifelhaft sein, ob die Handschrift schon vor der Mitte des vierten Jahrhunderts oder erst in der zweiten Hälfte desselben geschrieben sei’.

For the more extended arguments in the Prolegomena to his facsimile edition, see Constantinus Tischendorf, Bibliorum Codex Sinaiticus Petropolitanus (4 vols.; St. Petersburg, 1862), pp. 1.11*–1.14*.

‘After that’ I promised a communication about the age of the manuscript; however, it should be limited to a few main pieces, while everything that is more detailed is to be reserved for the Prolegomena. Above all, I can only emphatically repeat what I wrote from Cairo when I first received news of the treasure found in my fatherland: for that it was written in the fourth Christian century can be put almost beyond all doubt with all the arguments valid in paleographic science. On closer examination, only one thing can be doubtful, whether the manuscript was written before the middle of the fourth century or only in the second half of it.