Wednesday, February 10, 2016

Debunking Silly Statements about the Bible

There is a post about the textual transmission of the Bible at the (US) Gospel Coalition website: Debunking Silly Statements About the Bible. Unfortunately all around (for the author, Greg Gilbert, his book, the Gospel Coalition) this post contains a number of Silly Statements about the Bible. Christians who want to defend the Bible have a responsibility to know what they are talking about. Here are some quotations from the first half, with brief comments.
1. Whatever pieces of paper Luke, John, or Paul used to write Luke, John, or Romans have been lost to history, and it’s unlikely we’ll ever find a biblical manuscript about which we can say, “We are 100 percent certain this is the original piece of paper on which the author wrote.”
I think if we find a biblical manuscript on paper we can be 100 percent certain it is not the original.
2. even though we don’t have the originals, we do have thousands of other pieces of paper that contain original-language text from each book of the Bible—about 5,400 when it comes to the New Testament. These go back to the third, or second, or even (perhaps?) to the first century.
(just stepping over the ‘pieces of paper’ thing going on here) I think that it is seriously misleading to leap from a figure like 5,400 manuscripts to then say ‘these go back to the third ... century’. These 5,400 manuscripts do not go back that far. Less than 1% of that 5,400 figure goes back to to anything like the third century (or earlier, although clearly at this point none of them go back to the first century).
3. After all, the New Testament was written in the mid-to-late first century, and the earliest copies we have are from about the years 125 to 200. At best, then, there’s a gap of some 45 to 75 years between the originals and our earliest copies.
This is also potentially misleading. If we date P46 to around AD200 then we are looking at more like 140-150 years for the Pauline letters. For Mark we might have a gap of 200 years. For John perhaps 45-75 years works, but not for any other portion of the NT. Generalisations are not helpful.
4. One fascinating example is what’s called the “Codex Vaticanus,” a copy of the New Testament originally made in the fourth century, but which was re-inked in the tenth century so it could continue to be used. Do you see what that means? Codex Vaticanus was still in use 600 years after it was originally made! Therefore the claim that all we have are “copies of copies of copies of copies” of the originals is far overwrought. Indeed, it’s well within the realm of possibility that we have in our museums today copies of the originals, full stop.
A better example would be a manuscript (e.g. Sinaiticus) which shows more evidence of continuous usage than does Vaticanus. Nothing in this suggests that we have immediate copies of the originals in our museums, and as far as I am aware no one has ever argued such a case in any scholarly publication. Wishing doesn’t make it so.

5. Also, when you consider the gap between the originals and first copies of other ancient works, you can see just how small this “gap” for the New Testament really is. For example, for Thucydides’s History of the Peloponnesian War, we have exactly eight surviving manuscripts, the earliest of which is 1,300 years removed from the original!
I really wish people wouldn’t keep saying this sort of thing. It is nonsensically ill informed. In fact there are 99 early mansucripts of Thucydides, mostly papyrus, some of which go back to the third century BC, dozens from the first and second centuries (LDAB).
6. One scholar has asserted there are, astonishingly, up to 400,000 variants in the New Testament! There are several things to say about this charge. First, the manuscripts are not in fact riddled with variants, and that 400,000 number isn’t nearly as scary as it seems, even if it’s accurate. The scholar who used that number wasn’t just looking at the 5,000 pre-printing-press, original-Greek manuscripts we have, but also at 10,000 other manuscripts in other languages, and then on top of that another 10,000 or so instances where people quoted the New Testament during the first 600 years of church history! Put it all together, and what you’re really talking about is 400,000-ish variants across some 25,000 manuscripts and quotations covering 600 years. But at the far upper end, this comes out to . . . only about 16 variants per manuscript. To put it nicely, that’s really not many.
I’m really not sure where to start on this. It is nonsense from beginning to end.
Up-date: I don't know who this "one scholar" is, there are several possible candidates. It is not particularly astonishing, but without any careful definition of "variant" it is not even clear what we are talking about. It is not the case that these relate to 10,000 other manuscripts in other languages (although being manuscripts they will have variations) since the languages tend to be cited as a unit in critical discussions. The purported logic of dividing 400,000 by 25,000 and coming up with only 16 variants per manuscript is completely vacuous. It suggests a lack of understanding of how the standard critical editions actually work, and no one who works with manuscripts would think like this. Do we forget that one single manuscript, Codex Sinaiticus, has 23,000 corrections reflecting within itself at least 23,000 "variants"? Of course we should define that term whenever we use it in that context. For the only reputable published discussion of the subject see Peter Gurry, “The Number of Variants in the Greek New Testament.” New Testament Studies 62.1 (2016): 97-121. He certainly was not concerned with variants in versions and quotations in church fathers.  
7. Finally, it’s not as if the variants in all those 25,000 manuscripts just show up everywhere; rather, they tend to cluster around the same few places in the text over and over again, which means the number of actual places in the New Testament really at issue is surprisingly small. The point is that when you think about it beyond the soundbites, you don’t get a picture here of a mountain of copies with so many variants that we can’t make heads or tails of it. Not even close. On the contrary, you get a picture of a remarkably stable transmission history for the vast majority of the New Testament, and a few isolated places where some genuine doubt about the original text has given rise to a relatively large number of variations.
This is so wrong. I doubt this person has ever read a manuscript.
Up-date: Remember that there aren't actually 25,000 manuscripts (since even on the figures provided there are only 15,000 manuscripts and another 10,000 references in church fathers [where these figures come from I do not know]). The claim that textual variants within the New Testament "tend to cluster around the same few places in the text over and over again" is not supported by any sort of reality. Open the standard critical edition, some pages have more variants than others, some passages have more variants than others, but by and large variants are spread over the whole text of the Greek New Testament, from Matt 1.3 (whether to read ZARA or ZARE) to Rev 22.21 (whether to read AMHN at the end or not), there is no real clustering in the sense that only a few places within the New Testament text have variants - every page normally has dozens of variant readings cited with manuscript and versional support. How many passages are really "at issue" in the sense of textually uncertain is an interesting question and could be answered in different ways depending on the scholar or group of scholars (in NA28 there are diamond readings indicating that the editors are uncertain; there are many differences between different editions). Uncertainties exist even in relation to some questions of theological significance. And finally, to say that there are "places where some genuine doubt about the original text has given rise to a relatively large number of variations" has things the wrong way around: it is the variant readings in the manuscripts that give rise to our genuine doubt about the original text.

Surely we can do better than this.

The Overall Quality of Byzantine Manuscripts

Following up on my follow up on the Byzantine text, it’s instructive to see just how close the text found in the majority of our manuscripts is to the initial text in the Catholic Epistles. From the ECM supplement, here are the number of readings where the “undivided witnesses of the Byzantine text” differ from the ECM/NA28 initial text.*

Book Differences Var. Units
James 61 761
1 Peter 46 700
2 Peter 23 417
1 John 43 765
2-3 John, Jude 33 403
TOTAL 206 3,046

To put the matter positively, the Byzantine text as defined by the ECM agrees with the editors’ reconstructed text in over 93% of variant passages in the Catholic Epistles.

GA 35, a key representative
of the Byz text in the ECM
That’s quite a lot. You could even say that your NA28 is 93% Byzantine in the Catholic Epistles if you were so inclined. In any case, this high level of agreement is what led the editors to conclude that the Byzantine text is “an important witness to the early text.” Hence the re-evaluation for the ECM2 of all places where the Byzantine reading differed from the initial text of ECM1 (see ECM2, p. 34*).

*The stats are from the ECM2 supplement, pp. 10, 13, 15, 17 and Gerd Mink, “Contamination, Coherence, and Coincidence,” p. 147 n. 15.

Sunday, February 07, 2016

Codex Gigas Images Online

The Devil
Codex Gigas, otherwise known as the “Devil’s Bible” is considered the largest Medieval Latin codex. Its pages are 35 inches tall and 19 inches wide and contains the complete Bible along with several other works from Josephus and others meant to illuminate the Bible. The name “Devil’s Bible” comes from the picture of the Devil found on f. 290r. As the official website explains:
The portrait was intended to remind the viewer of sin and evil. It is opposite a page with a representation of the Heavenly City and the two pages were deliberately planned to show the advantages of a good life and the disadvantages of a bad one.
The website has lots of useful information and includes all the photos here. There are some impressive initial letters (e.g., Mark’s Gospel) and other decorations throughout. Interestingly, the only human represented appears to be Josephus. You can download high resolution images from Wikimedia Commons here.

Update: there’s more good info linked in Tommy’s post from 2007.

The beginning of Mark’s Gospel

Thursday, February 04, 2016

Defining the Byzantine Text

As a follow-up to my post about text-types from last week, here is an example of how our definitions of texts can influence our thinking about textual criticism. It matters particularly when it comes to comparing one “text” to another.

In their new book on NT textual criticism, Stanley Porter and Andrew Pitts make a common argument against the Majority Text position which could apply to the Byzantine Priority position as well. Porter and Pitts write:
Several other questions are left unanswered by the Majority text approach as well. Statistical probability of documents simply cannot explain why no distinctively Byzantine readings are identifiable in the Greek manuscripts, church fathers, or version from the first several centuries—certainly some remains would have been left, even if the manuscripts were in constant use. If the Majority text most accurately reflects the original, we would expect some traces of it chronologically close to the original. These significant obstacles to the Majority text approach still have not been convincingly overcome by its adherents (pp. 91–92 n. 3).
Now I admit that I find a form of the chronological argument against the Majority Text and the Byzantine text persuasive. But what I want to point out here is that this particular form of the argument hinges entirely on something that Porter and Pitts do not provide us with, namely, a definition of “distinctively Byzantine readings.”

To illuminate the problem, let’s consider how the Editio Critica Maior (ECM) defines the “Byzantine text.” You have to read it very carefully, but the ECM actually has two definitions. In the introduction (§3.1) we read that “the term Byzantine or Koine text refers to the form of text defined by those readings which are attested by the majority of the manuscripts and differ from the established text” (emphasis original). The supplementary volume gives a complete lists of just these readings for each book of the Catholic Epistles (pp. 10ff).

But the ECM also has a second, more implicit definition. On the same page (p. 10) of the supplementary volume we read about “the undivided (ungespaltene) Byzantine text” which differs from the initial text in only 61 of about 700 places in James. In the editors’ words, this undivided Byzantine text is, “apart from these 61 passages, an important witness to the early text” (p. 10; emphasis mine).

So we have two definitions: (1) the Byzantine text as all the shared readings found in the majority of Greek witnesses; and (2) the Byzantine text as that set of readings found in these same witnesses which differ from the initial text. The first entity is an important witness to the early text whereas the second is defined in distinction from essentially that same early text.

Now the problem should be obvious. An argument like that of Porter and Pitts risks circularity on this second definition and it is simply false on the first. There are hundreds of readings found in Byzantine manuscripts that are also found “chronologically close to the original.” In fact, according to the ECM, there are over 630 readings just in James that are the original or at least very close to it (depending on how you identify the initial text).

Now Perhaps Porter and Pitts have a definition of “distinctively Byzantine reading” that isn’t circular. From what they’ve written we simply don’t know because they don’t define their key term.

And that’s just the problem. Good arguments about the history of the text can only be made where we have clarity about what constitutes the textual entities—whether text-types or otherwise—that we are trying to relate to one another. Porter and Pitts have two texts they relate (“original” and “Majority/Byzantine”), but since they don’t define them, the relationship they propose is meaningless.

This isn’t to pick on Porter and Pitts. They just provide a good illustration of problem of not defining the texts we claim to relate to one another.

Wednesday, February 03, 2016

Fragmentarium Fellowship (Biblical Fragments) Grant for Postdocs

A grant for postdocs to study biblical fragments at St. Gallen has been announced by the Fragmentarium Fellowship (HT: Christina Kreinecker):

Fragmentarium fellowship 2016 Stiftsbibliothek St. Gallen 

The Fragmentarium Fellowship has been established to support research on medieval fragments.
Fragmentarium is an international digital research laboratory for medieval manuscript fragments, which collaborates with about 15 major European and American libraries. This international scholarly network will enable libraries, collectors, researchers and students to upload medieval manuscript fragments and to describe, transcribe and assemble them online. The grant will allow a scholar (postdoctoral candidate) to spend a period of minimum two to three months in St. Gallen, with the goal of carrying out a case study on the biblical fragments preserved at the Stiftsbibliothek St. Gallen.

The fellow will be responsible for this case study. He/she will have to document his/her research and make it available to Fragmentarium through the publication of an article. The fellow will also have to use the digital application which is now being developed by Fragmentarium in order to test its possibilities. The fragments on which the research will be carried out will be uploaded on the Fragmentarium website in close collaboration with its team.

The fellows of Fragmentarium will participate in two workshops. The first one will bring together the 2016 Fragmentarium fellows and will take place between June 6th and June 8th, 2016 in Fribourg. The second, during which the results will be presented, will take place in 2017. The research must be done between June 2016 and June 2017. The article will have to be submitted after the second workshop.

The candidate must be affiliated to a non-Swiss institution. He/she must be employed as a researcher in his/her home country before, during and after the fellowship. He/she may be DPhil or D.Theol. as well as a specialist in at least one of the following fields: edition, palaeography, codicology, latin, biblical sciences.

Depending on the destination, varying lump sums will be given to cover travel expenses. For room and board expenses, the support will range from CHF 3000 to CHF 3500 per month. Travel and living expenses for the workshops will be added to this sum.

Applications should be submitted (with curriculum vitae, a description of the research project with 10’000-12’000 signs, and planning) by February 28th, 2016, via email to:, Sylviane Messerli, Université de Fribourg, rue de l’Hôpital 4, 1700 Fribourg.


Sylviane Messerli
Université de Fribourg
Rue de l’Hôpital 4, CH-1700 Fribourg

Tuesday, February 02, 2016

Painful blunders

Find the missing word in this verse:

Luke 14:27 And whoever does carry their cross and follow me cannot be my disciple.

This is the text produced by the first hand of Vaticanus, though I am glad that eventually the missing 'ου' was added above the line.

οστις ουν \ου/ βασταζει
τον σταυρον εαυτου
και ερχεται οπισω μου
ου δυναται ειναι μου

Incidentally, the correction is not noted (yet) in the transcription on the NT.VMR.

Images of P.Bodmer VIII (P72) On-line

6 Vatican Library has just released images of P.Bodmer VIII online! (HT: Brent Nongbri)