Some thoughts on Craig Evans, “How Long Were Late Antique Books in Use?”

A few days ago I wrote about the statement of Peter of Alexandria (d.311) that the original manuscript of John’s gospel was still around and that readings could be obtained from it.

A few days ago I came across an interesting article by Craig Evans, “How Long Were Late Antique Books in Use? Possible Implications for New Testament Textual Criticism”, Bulletin for Biblical Research 25 1 (2015) 23-37.  The article makes some interesting points.

The first part of the article is a dossier of evidence that ancient papyrus books could be in use for considerable periods of time, perhaps even a couple of centuries.  Our papyrus data is from Egypt and the desert of Qumran, both locations where the climate may play a part.  But there are literary references as well, that refer to Rome.  The article might also have referenced Aulus Gellius for further examples of long-lived literary texts, such as Fabius Pictor.  The evidence does suggest that a papyrus book containing a literary text might be in use 200 years after it was written.

The article then digresses, before giving us the statement of Peter of Alexandria that the church at Ephesus preserved the autograph of the Gospel of John, and what a particular reading was said to be in that autograph.

These two things together lead to an interesting thought.

Firstly, we have clear evidence that papyrus books could remain in use in the ancient near east for a couple of centuries.

Secondly, we have clear evidence (from Peter of Alexandria) that this actually happened to the original manuscript of the Gospel of John.

These two points together seem rather interesting.  Neither point is exciting by itself.  Just because a few rare books did remain in use for centuries does not mean that the gospel autographs were among them.  But then we have an ancient literary testimony that one of them was among them.

I don’t see that the force of this evidence can be disregarded.  The autograph, the original manuscript of the Gospel of John, was around much later than we would expect; possibly as late as 300 AD.

This is a rather amazing statement, but it is what the evidence says.  What kind of contrary evidence do we have?  Nothing.

Having established this point, unfortunately the article introduces various distractions.

The first distraction is references to the parchment codices of late antiquity; but the long life of such items is irrelevant, because these are parchment codices of a higher technological standard than anything available in the first century AD.  I have myself handled British Library Additional 12150, which was written in 411 AD, and is none the worse for its fifteen centuries.

It may be guessed that the reason for introducing parchment books lies in 2 Tim. 4:13, where Paul refers to the books and the parchments.  This phrase was used as a title for a famous book by F. F. Bruce, and was likely enough in Dr.E’s mind.  But there is no evidence in our possession that the NT autographs were written on anything but papyrus.  They may have been in roll or codex form, but we can’t suppose a parchment codex without evidence.

The next distraction is an attempt to draw upon Tertullian, De praescriptione haereticorum 36:1-2:

Age iam, qui uoles curiositatem melius exercere in negotio salutis tuae, percurre ecclesias apostolicas apud quas ipsae adhuc cathedrae apostolorum suis locis praesident, apud quas ipsae authenticae litterae eorum recitantur sonantes uocem et repraesentantes faciem uniuscuiusque.  (CSEL 57)

Come now, if you are ready to exercise your curiosity better in the business of your own salvation, run through the apostolic churches, where the very thrones of the apostles preside to this day over their districts, where the authentic letters of the apostles are still recited, bringing the voice and face of each one of them to mind.[84] (Greenslade’s translation)

Greenslade indeed adds a footnote:

84. Eusebius, H.E., VII, 19, believed that the actual throne of James still existed at Jerusalem. Some think that Tertullian means by cathedrae here the physical objects. That is unnecessary, and on the whole unlikely, but not impossible. But “authentic” will scarcely mean autograph; he means unmutilated texts.

If we look to see how Tertullian uses “authenticae”, we find only one other reference, in De Monogamia 11:16:

Sciamus plane non sic esse in Graeco authentico, quomodo  in usum exiit per duarum syllabarum aut callidam aut simplicem   eversionem: Si autem dormierit vir eius, quasi de futuro  sonet ac per hoc videatur ad eam pertinere, quae iam in fide  virum amiserit. (CSEL 57, Bulhart)

Let us plainly know that, in the Greek original, it does not stand in the form which (through the either crafty or simple alteration of two syllables) has gone out into common use, “But if her husband shall have fallen asleep,” as if it were speaking of the future, and thereby seemed to pertain to her who has lost her husband when already in a believing state. (ANF, Thelwall)

There seems no reason to suppose a reference to the original manuscript of 1 Corinthians; but instead to the correct reading, the authentic reading of scripture, rather than the somewhat dodgy Old Latin translation then in circulation.

Dr. E. plainly consulted his dictionaries. The Oxford Latin Dictionary does indeed define “authenticum, -i, n.” as “an original document, autograph” and the adjective as “(of documents) Original”, referencing the Greek αὐθεντικός (Glare, p.220), but the examples given in both cases refer to documentary or legal texts, where the original means the actual thing itself.  But Souter’s Glossary of Later Latin has the adjective meaning “authoritative, genuine, true, original”, and even “authentica” used for Greek antigrapha, “copies” – presumably authentic ones.  Both of the Latin usages belong to the time of Tertullian, but here he is anticipating the later, and indeed the modern use of “authentic”.  A look at Liddell and Scott reveals both meanings hanging around the Greek word.

Tertullian, then, does not provide any support for the idea of autograph New Testament manuscripts; the subject is rather of honest copies instead of heretical forgeries or corrupt translations.

The article continues with some interesting general points about ancient letters, but this really only distracts from the key message.

Finally it ends with the very distracting claim that the NT texts were more textually stable than the gnostic writings, and hypothesises that the availability of the autographs may be the reason why.  This leads up to the tremendous final statement:

…. there really is no justification for supposing that the text of the NT writings underwent major changes in the first and second centuries.

Indeed not; but that does not follow from what has been said.

Such claims weaken the paper, because they have nothing to do with the evidence presented.  We can discuss them briefly, but as I said, they are a distraction.

The gnostic texts are known to us in tiny quantities of manuscripts, and, to the best of my knowledge, show a great deal of variation which is not a matter of textual transmission, but rather of editing changes.  In this they resemble the later hagiographical texts.  But we know from Irenaeus and especially Tertullian that the gnostics themselves did not consider that a rigid fidelity to what they heard was important.  As Tertullian pointed out – “what has Athens to do with Jerusalem” – they drew upon the pop-philosophical schools, where innovation was necessary in order to become a teacher.  Consequently every gnostic master taught something different.  The texts that they produced – we need hardly suppose any of them to be very old – will not be exempt from such an attitude.  Consequently they tell us nothing about the stability or otherwise of the New Testament texts.

Likewise the documented existence of the autograph of John by Peter of Alexandria does not force us to understand that it was consulted in such a way as to control the text.  Cathedrals preserve relics, in our own day, but getting access can be a test of patience.[1]  In fact the reading offered by Peter of Alexandria, with the authority of the autograph, looks very much the opposite, a lectio facilior.

This corrupt reading is most likely evidence that the Ephesus church was not doing what the article suggests.  If they were not doing it then we cannot suppose that others were doing so.

It is perfectly possible that Peter of Alexandria was an isolated instance.  It is quite likely that he was copying from an earlier, now lost source – he must have had access to lost works of Origen, for instance.  So the date of his anecdote may easily be fifty years earlier.

Nor does the existence of a “standard text” prevent the creation of “wild copies”.  Most ancient copies of books were probably produced in-house, by a literate slave.  Such copies are commonplace in the texts of the classics at Oxyrhynchus.  The medieval codices that transmit the Iliad and Odyssey to us do not derive from such trashy copies, but from the corrected texts of Alexandria, as the scholia show.  But the trashy copies still got produced.  No doubt it was exactly the same for Christian texts.  It is telling that our solitary text critical anecdote itself also comes from Alexandria.

These points should not distract us from what is a fine and interesting paper.

It may be relevant that it is the Gospel of John that is preserved.  Anybody who reads Eusebius’ Church History will be struck by the way in which stories about John are preserved.  Little is recorded about the other writers of the New Testament.  There were few Christians in the early days, and most likely they died before the Christian community was numerous enough to preserve very much.  How could those early believers in their house churches have managed to preserve autographs anyway?  But Irenaeus tells us that John lived on until the time of Trajan, 100 AD, and that Polycarp came to Rome ca. 155 and preached about what he had himself heard the apostle say.  Pliny the Younger tells Trajan that the temples are deserted.  This is a numerous group.  If John wrote his Gospel around 90 AD, it would appear in a world a generation later from that which the works of Mark and Luke had to face.  We are not obliged to believe in all or even most of the stories recorded, but the point is that they were recorded at all.  Such a world and such an environment might have been favourable to preservation.  But all this is speculation.

But regardless of its defects, this is a useful article.

Note: For some reason Dr. E.’s article greatly offended some readers.  Of the material that Google throws up, the hostile blogpost by Brice Jones is written in terms that would most certainly mean a duel in earlier days.  It also has comments below by various scholars.  Timothy N. Mitchell gives a critical but less excited reply.

  1. [1]This writer was once comprehensively prevented over a period of months from accessing an incunable by the apparently motiveless spite of an archivist at Canterbury Cathedral.