Monday, January 2, 2023

About That "Embarrassment of Riches" (and Quantities of Manuscripts)

“We have an embarrassment of riches when it comes to the evidence for the New Testament.”   Ever hear that one?  Such a claim is routinely made by Christians who fill, or appear to fill, two roles as apologists and researchers.  And they are mostly right:  the quantity of manuscripts of the Greek New Testament is staggeringly superior to the evidence for any other literary work of a comparable age.  But they are partly wrong, for at least three reasons. 

First, the relative poverty of textual support for specific readings in the works of Suetonius (to pick one ancient author) does not make other authors (such as the authors of the books of the New Testament) rich. 

Second, the New Testament did not initially circulate as a single book, but as 27 books - which were not copied and distributed evenly.  (There are over 1,600 Greek Gospels-manuscripts; there are fewer than 400 Greek MSS of Revelation.) 

Third, quantity is not necessarily quality.  Kurt & Barbara Aland (as in "Nestle-Aland compilation of Novum Testamentum Graece," the primary base-text of the New Testament in the ESV, NIV, CSB, NRSV, NLT, and NASB), after listing numerous Greek manuscripts, candidly stated in their 1981 handbook The Text of the New Testament (translated into English by Erroll F. Rhodes), “All of these minuscules exhibit a purely or predominantly Byzantine text.  And this is not a peculiarity of the minuscules, but a characteristic they share with a considerable number  of uncials.  They are all irrelevant for textual criticism, at least for establishing the original form of the text and its development in the early centuries.”

How many Greek manuscripts did Kurt and Barbara Aland consider “irrelevant” to the task of reconstructing the original New Testament text?  Looking over their list on pages 140-142 (“Table 7”), I count 887 manuscripts.  Aland & Aland, though, seem willing to put “more than 1,175 minuscules” (p. 138) into the category which they dismiss as “irrelevant.”

The number of minuscules that they did not consider “irrelevant” is given on page 138:  “a little more than 175.” 

What about uncials, a.k.a. majuscule MSS?  The total number of majuscules is easy to calculate, since each is identified by a number preceded by a zero, and we saw the addition of 0315 in 2015 – so the current total number of majuscules is just a bit higher than 315, right?  Wrong.  Some majuscule manuscripts were obtained by researchers, and were given identification numbers, after the manuscripts had been torn up.  Only later did researchers discern that they had portions of the same manuscript, with a different identification-number given to each portion. 

029 is same manuscript catalogued (in portions) as 0113, 0125, and 0139. 

070 is the same manuscript catalogued (in portions) as 0110, 0124, 0178, 0179, 0180, 0190, 0191, 0193, 0194, 0124, and 0202. 

063, according to Aland & Aland, “belongs with 0117.”

064 is the same manuscript (in portions) as 074 and 090.  (Part of this manuscript was found among the New Finds at St. Catherine’s monastery.)

073 is the same manuscript as 084.

083 is the same manuscript (in portions) as 0112 and 0235.  (Take note NET-readers; this manuscript is erroneously double-counted in the NET’s notes.)

087 is the same manuscript as 092b.

089 is the same manuscript as 092a and 0293.

0100 is the same manuscript as 0195, and neither one merits an identification-number among continuous-text uncial MSS, because each one is part of lectionary 963.

0102 is probably (according to Aland & Aland) the same manuscript as 0138.       

0129 is the same manuscript as 0203, and neither one merits an identification-number among continuous-text uncial MSS, because each one is part of lectionary 1575.

0137 is the same manuscript as 0138.

0152 is a talisman, technically not a continuous-text uncial manuscript.

0153 is an ostracon, technically not a continuous-text uncial manuscript.

0192 is lectionary 1604.

0212 is not a continuous-text uncial manuscript, and thus does not merit inclusion in the list.

A simple (perhaps too simple) count brings the total number of continuous-text majuscule (uncial) manuscripts down from 315 (in 2015) to 285.    

Uncial manuscripts that display a Byzantine text (according to Aland & Aland), and which are thus “irrelevant,” include 07, 09, 011, 013, 014, 017, 018, 021, 022, 023, 024, 026, 027, 028, 030, 031, 033, 034, 036, 039 (the same manuscript as 566 (Codex L; the text of Matthew and Mark is written in minuscule; Luke and John are written in majuscule – but it is all a single manuscript), 041, 042, 043, 045, 046, 047, 049, 052, 056, 0104, 0116, 0133, 0135, 0197, 0211, 0248, 0253, 0255, and 0257.  These forty majuscules are in the same “irrelevant” category in which Aland & Aland placed about 1,175 minuscules.

So the maximum number of continuous-text majuscule parchments that were used for the compilation of the Nestle-Aland NTG is . . . (let’s see:  285 – 40 . . . ) 245.

Most of these are not complete.  It should be kept in mind than even a tiny fragment, if it is not part of another manuscript, receives its own identification-number, and is counted as one manuscript.  A complete New Testament = one manuscript, and a fragment of a single page = one manuscript.  Without this factor constantly in mind, people who hear about the “embarrassment of riches” might tend to imagine that we have 245 relevant majuscule continuous-text copies of the New Testament.  But in real life, as I mentioned, many of the majuscules are fragmentary. 

Instead of referring to “New Testament manuscripts,” majuscule or minuscule, it would be more accurate to refer to “Gospels-manuscripts,” (about 1,800) and to manuscripts of Acts and the Epistles,  and to manuscripts of Revelation, and to manuscripts that contain the entire New Testament, whether majuscule or minuscule, are anomalies.  (I think about 70 such copies exist.)   (Manuscripts with other combinations also exist.)

Minuscules are not immune from the same (or similar) kind of double-counting that slices off the number of real continuous-text majuscule manuscripts by ten percent.  Georgi Parpulov, of the Akademie der Wissenschaften zu Göttingen, reported in 2022 in the open-access journal Fragmentology that GA 674 and GA 1284 are portions of the same manuscript.  And one minuscule, GA 2427, which was featured prominently in the apparatus of the Nestle-Aland NTG, but was proven to be a nineteenth-century forgery, has to go.  Another, GA 2795, is lectionary 2198.

Finally we come to lectionaries.  Minuscule 2795 is part of the same manuscript as lectionary 2198.  Parpulov also reported that lectionary 849 and lectionary 309 are portions of the same lectionary.   There are over 2,300 lectionaries to consider (and here again one should differentiate between Gospel-lectionaries, and lectionaries of the remaining New Testament books).  But although lectionaries have been the focus of considerable research, one would think from the apparatus in the Nestle-Aland NTG and the UBS GNT that hardly anyone is considering them.  Almost all of them display (with expansions and modifications) the Byzantine Text that Aland & Aland dismissed as irrelevant.
As Maurice Robinson has observed – as noticed by Peter Gurry in 2017 at the Evangelical Textual Criticism blogThe resources of the pre-fourth century era unfortunately remain meager, restricted to a limited body of witnesses. Even if the text-critical evidence is extended through the eighth century, there would be only 424 documents, mostly fragmentary.”  

 Am I disturbed by individuals who, in one breath, give soothing assurance about the “embarrassment of riches,” and in the next breath endorse the Nestle-Aland compilation that was made with the working premise that over 1,175 minuscule manuscripts, and 40 majuscule manuscripts, are irrelevant?  Well, to answer that question, I must diverge from today’s main topic. 

It is disturbing that anyone would brag about our “embarrassment of riches” and then proceed to dismiss 85% of the coins in the royal treasury as counterfeit.  (Meaning:  Wallace & Co. talk about our “embarrassment of riches” but at the same time habitually reject the reading found in the vast majority of manuscripts (not just 85%, but sometimes 95% – keeping in mind that MSS should be generally divided into Gospels/Acts-Epistles/Revelation categories) when that reading disagrees with a favored reading in the Alexandrian Text.)

But this is essentially a point against bad rhetoric, bad apologetics, and bloviations (or combinations of all three), not a point against the evidence for the New Testament text, about which I am not disturbed.  I disagree with the idea that the Byzantine text, and the manuscripts supporting it, are irrelevant.  Aland & Aland’s anti-Byzantine bias is obsolete. 

The approach used to compile the New Testament base-text of the ESV, NIV, NLT, CSB, and NRSV is basically the same obsolete, never-was-valid approach that was used for the 1881 edition of Westcott & Hort.  (NA27 and WH1881 fully disagree in only 661 readings; I use “fully” to modify “disagree” because the editors of NA27 made non-decisions at multiple points and put some readings in brackets and double-brackets (a feature which I guarantee was not in the original text).  The number of tenuous disagreements is higher:  1,372, as I have explained here.) 

One doesn’t have to think about that long and hard to discern that the Nestle-Aland compilation is unstable at 711 points – not counting the 34 readings introduced in NA28, which included a conjectural emendation (based on zero Greek manuscripts).  (Many of which are trivial as far as meaning is concerned.)

That doesn’t make the Byzantine Text synonymous with the original text of the New Testament.  But it should make it a lot more than “irrelevant.”  English translations that take the Byzantine Text seriously (not the similar Textus Receptus) are already on the market.  More are coming, and I hope some major Bible-publishers will see this as an opportunity to amend the mistakes of publishers in the past 142 years.  So should English Bible-readers who desire the text in their English Bibles (not just the footnotes!) to reflect the text found in the rich manuscript-evidence that is available. 

This should not be interpreted to mean minority readings cannot be original.  Sometimes they are original (as I have repeatedly insisted), and in such cases the reading found in the majority of MSS must give way, lest scribal inventions, no matter how popular, usurp the original text.  But today’s main point should not be diminished:  talk about the “embarrassment of riches” by advocates of a New Testament compilation that is 99% Alexandrian (at points where the Alexandrian Text and Byzantine Text disagree) should stop.


Mm said...

We have less than 60 complete NT manuscripts. And many were copied after the 10th century.

Joel Metzger said...

Thanks James for your detailed presentation of the facts, giving us an accurate picture of the situation.

Peter Gurry said...

A 99% Alexandrian compilation is Alexandrian in how many points exactly? And what is it in the other places, and are those places based on the embarrassment of riches or on something else?