Baker Academic

Thursday, May 1, 2014

The Pericope Adulterae Conference—Chris Keith

I have just returned to London from having spent a week in the States, the first several days of which were in North Carolina for the pericope adulterae conference at Southeastern Baptist Theological Seminary.  The speakers were yours truly, Tommy Wasserman, Jenny Knust, John David Punch, and Maurice Robinson.  I feel confident in saying that a good time was had by all.  SEBTS treated us very well.  Tommy Wasserman and I were in the same house and opened each day with breakfast, coffee, and any variety of discussion topics while sitting on the front porch.  I had BBQ at least three times on the trip and they even brought in Krispy Kreme donuts and Chic-Fil-A chicken biscuits.  I was honored to introduce Tommy to these staples of American cuisine.  He tells me he loved the Chic-Fil-A but found the Krispy Kreme too sweet for a breakfast food.

As to the actual academic issues, Tommy, Jenny, and I argued against the Johannine authenticity of the story of the adulteress while John David and Maurice argued for it.  Tommy argued that the state of the manuscript tradition in the second century makes it unlikely for PA to have been interpolated then and suggested ca. 250 for its interpolation, agreeing (I'm happy to say) with the broad suggestion that I made in my 2009 book on PA, although I argued that point on other grounds.  Jenny ran down a long list of how the Fathers treated passages involving sexual sin and also how Alexandrian textual scholarship, and the variant of it practiced by Origen, makes it highly unlikely that early Christians would have taken the story out of John's Gospel.  In short, she showed that Christians did not shy away from stories about sexual sin and that scribes typically marked spurious passages found in their exemplar but nonetheless copied them; that is, even when bothered, they did not typically take texts out.  I argued that linguistic style cannot be a decisive criterion for authorial origin and reiterated my argument about PA's insertion by an attentive interpolator.  John David Punch argued that PA's linguistic style supports Johannine authenticity or, in the least, does not speak against it, and advocated a theory of ecclesiastical suppression.  Punch went first, so my argument about style came after his argument, as did Knust's argument about how Christians' treated sexually explicit passages.  Maurice Robinson went last and he argued that PA was most likely authentic based on linguistic style.  In a nice mixture of a variety of the theories on offer, Maurice argued against the theory of ecclesiastical suppression but for a (very) early removal of the text by the lectionary system.  He also argued against an argument in my book that the reading of katagrapho/grapho at John 8.6, 8 is the preferred reading.  He says it was most likely grapho/grapho (so also Holmes's SBL edition), and that a scribe then later changed it to katagrapho/grapho, and bases this argument on my earlier argument about katagrapho/grapho in 8.6, 8 being an intentional allusion to Exod 32.15.

Maurice's argument deserves a little more attention, if nothing else because he made it a point to come after me, silver-haired assassin that he is.  (I should add that Maurice spoke very highly of my work in general, for which I was grateful.)  At John 8.6, 8, the manuscript tradition includes katagrapho/grapho, grapho/katagrapho, katagrapho/katagrapho, and grapho/grapho.  In my monograph, I argued for katagrapho/grapho, agreeing with the critical editions of Nestle Aland (at least the last two; I haven't checked every one) and, e.g., von Soden.  I think it's far more likely, especially in light of the synonymous actions of the verbs in the narrative of PA, that scribes harmonized the verbs so that they match (thus producing the readings of katagrapho/katagrapho and grapho/grapho) than that a scribe purposefully changed one but not the other and, under Maurice's argument (that grapho/grapho was original) in the midst of this, introduced a hapax legomenon (katagrapho).  I think it more likely that an original hapax legomenon was either harmonized with (katagrapho/katagrapho) or removed (grapho/grapho) than that one was introduced.  (For what it's worth, no theory on offer, to my knowledge, explains adequately the wildcard reading of grapho/katagrapho in MS 28.)

Why am I telling you all this?  Well, in my original argument in my 2009 book, I described the katagrapho/grapho reading as the "majority" reading several times.  I was wrong on this point (I was soon to find out), and for my life I cannot remember what made me think this was the case.  For some reason along the way, I assumed that it was correct, despite my father having told me very early on what happens when you ass-u-me things.  I can't account for this lapse on my part, but I can describe the consequence.  During his paper, and with no small amount of goodnatured smugness, Maurice presented me personally with a full collation of all 1427 MSS including PA, showing that, contrary to what I'd said, katagrapho/grapho was not the majority reading.  A full 1220 manuscripts out of 1427 read grapho at 8.6.  Maurice was even kind enough to do the math for us--82.7%.  So here, folks, below is photographic evidence of precisely what it looks like the moment one is handed a document in front of a full conference showing his claim to be irrefutably wrong (collation in hand under the folder).  I'm so very happy that Tommy Wasserman was willing to take this picture so that I can preserve the moment.

During the question and answer, I kindly thanked Maurice for showing me the error of my ways.  I also defended myself a bit.  I think the collation is, strictly speaking, irrelevant for answering which reading is to be preferred.  I still think katagrapho/grapho is to be preferred for the reasons I just mentioned.  I'll add also that I can't go with Maurice's argument for the lectionary system.  He dates John's Gospel to before 70 CE and argues that the lectionary system was in place already before P66 (ca. 200 CE) and P75 (ca. third century), the earliest copies of John's Gospel that contain the relevant section of text, and both of which omit PA in that section.  So, in essence, his argument requires a lectionary system that had developed and widely removed PA by the end of the first century or in the second century.  There's just no evidence for a lectionary system this early.  Maurice is aware of this, however, and would respond (I suppose) by stating that there's no clear evidence against it either.  He appealed to Justin Martyr's descriptions of the public reading of the Gospels as his supporting evidence for a lectionary system (of course, as I suppose he would acknowledge, Justin says nothing about a lectionary system or removing texts or PA, etc.).  Most scholars date the existence of a full lectionary system closer to the eighth century and, to my knowledge, the earliest evidence for the liturgical reading of PA in John's Gospel is Apol. Dav. Alt. 1.1 and 2.5, dateable to the late fourth century if written by Ambrose, though this is debated.  Maurice's argument for Johannine authenticity also depends heavily on linguistic style, and in my paper I argued that this cannot be a conclusive factor because later authors were fully capable of copying the style of earlier authors (I provided examples, including the Longer Ending of Mark and Septuagintalisms in the NT).  So, I still think I'm right about which reading is to be preferred and the fact that PA is not original to John's Gospel, but . . . on whether katagrapho is the majority reading in John 8.6, there can be no doubt:  I was wrong and Maurice was right.  I have the photo, and a personalized collation, to prove it.  I couldn't have been proven wrong by a better man.

I hope readers of the blog will understand that I write all of this with a smile on my face.  That was the general tenor of the entire conference and it was absolutely wonderful.  We all had agreements and disagreements with each other, but we all thoroughly enjoyed the experience as far as I could tell.  I thank Maurice wholeheartedly for his hospitality, along with David Alan Black and the rest of the SEBTS crew.  I also thank John David Punch, Tommy Wasserman, and Jenny Knust for allowing me to squeeze in with them and be a part of the conference.  It was an honor to be involved.  Publication of the proceedings is forthcoming, but I'm not sure where yet.  I will put it on the blog when I know.  For other write-ups of the conference, see here, here, herehere, and here.


  1. Thanks for the overview, Chris. The conference sounds like it was great fun, as well as very productive. And since you confessed that were caught in the very act of misstating the facts regarding the majority reading, I trust your colleagues were gracious and did not condemn you. scott caulley

    1. They did not, but I must sin no more.

  2. Iiuc, Dr. Robinson's proposal does not require the existence of a "full lectionary system" in the second century -- just enough of a lectionary system to assign specific passages to a smattering of particularly prominent feast-days, such as Easter-time, Christmas-time, and Pentecost.

    C. R. Gregory proposed, "It seems to me very likely that at an extremely early date the lessons were chosen for the Sundays," and nobody seems to have thought that this collided with the available evidence, or that such a view should be set aside on the grounds that a "full lectionary system" cannot be shown to be so early.

    *If* Luke 22:43-44 is original, and its movement from its original location between Luke 22:42 and 22:45 is an effect of lectionary-based adaptation, would this not prove the existence of an early primitive lectionary? -- not a fully developed cycle of annual readings for every day, but a basic cycle of readings for major feast-days such as Pentecost and Eastertime?

    Does Luke 22:43-44 have to /not/ be original, if one is to deny the plausibility of the theory that the PA was dropped accidentally due to a scribe's misunderstanding of lectionary-related instructions (expressed, perhaps, via marks which he misinterpreted) to skip the PA?

  3. James,

    What does "liuc" mean?

    Neither I nor anyone else said that Maurice's view should be set aside simply on the grounds that a "full lectionary system" cannot be shown to be so early. There are many other reasons to set aside his theory and I referenced a "full" lectionary system only in my comment about what is dateable to ca. the 8th century. I said that Maurice's proposal requires "a lectionary system" insofar as it must be at least widespread enough to have already removed PA from multiple copies of John's Gospel from the third and fourth centuries. Let's be clear about this, however--there is no evidence *at all* for lectionary readings of the Gospels this early. Justin Martyr says they read the memoirs of the apostles, not that there is a system, no matter how nascent it might be, or that the memoirs had been broken into particular readings, or that this was done for particular holy days, or that John's Gospel was a part of the reading, or even that PA had been part of the reading and was removed. Justin claiming that they read the Gospels liturgically does not demonstrate a lectional system (again, no matter how nascent it might be) any more than Second Temple discussions of reading the law and prophets in synagogue demonstrate a lectional system or references to the public reading of Paul's epistles demonstrate a lectional system. Clearly, lectional practices developed from liturgical reading, but exactly how and when that occurs and how it impacted individual manuscripts is not clear just from statements about public reading of texts. In my honest opinion, and with all due respect to my friend and colleague, Maurice, I think that asserting Justin as support here is simply wishful thinking. Again, there's no clear evidence of liturgical or lectionary reading of PA in John's Gospel until the late fourth century at the earliest. That is when the evidence permits us to start talking about the impact of liturgical reading on PA with confidence.

    From another perspective, even *if* there was a lectionary system that early, this is not the same as proving that PA was part of it. One cannot simply assert the possibility of lectionary reading and then--deus ex machina--assert that this took PA out and that's why the manuscript evidence looks like it does. The second part does not follow from the first.

    Similarly, in my opinion, your theory for Luke 22:43-44 simply requires too many assumptions. I'm open to being proven wrong here, but I think that Drs. Wasserman and Knust have shown rather clearly that texts tended to be expanded, not contracted. I remain convinced of the lateness of Luke 22:43-44 in any event.

  4. Since it's unlikely that "Iiuc" means "International Islamic University Chittagong", I'd go with "If I understand correctly".

  5. Chris,

    As Ape said, “Iiuc” = if I understand correctly.

    If I understand correctly, you are affirming the following four points:

    (1) A full lectionary system is datable to about the 700’s. But this is a tangential remark that doesn’t really address Robinson’s theory, inasmuch as he does not posit a full lectionary system in the 100’s.

    (2) There is no evidence at all for lectionary readings of the Gospels in the 100’s.

    (3) Justin’s description, c. 160, of the liturgical reading of the memoirs of the apostles does not require or imply a lectionary-system at all, let alone a lectionary-system in which the reading for Pentecost was the same as what it is in the normal Byzantine lectionary (in which Jn. 7:53-8:11 is skipped).

    (4) The earliest clear evidence of the PA’s use as a lection is in the late 300’s.

    It’s regarding that second point that Luke 22:43-44 might be significant. Granting that you, today, are convinced that Luke 22:43-44 is late (defining its “late” production as some time before the production of Justin’s Gospels-text, well before the production of the earliest extant manuscript of Luke 22), *if* Luke 22:43-44 is original, then considering that its displacement in f-13 is lectionary-related, *then* wouldn’t this indicate that someone was adjusting the Gospels-text to conform to a rudimentary lectionary-system in the 100’s? I’m not arguing today for the correctness of the premise; I just want to get your take on what the premise, if correct, could imply about lectionary-influence in the 100’s.

    1. James, thanks for your response. By "late" I mean not initially in Luke's Gospel, but I think it was probably a very ancient story, much like PA. Also, I don't think my point about the eighth century is tangetial. The point is that this is when we can start saying things about a "system" with surety and thus how we must hold any conjectures about early periods with a loose grip because we simply don't know so much.

      In your scenario here, I would say that the implications of what you propose *could* indicate that lectionary-related reading influenced Luke 22:43-44's displacement *if* you assume its originality, but is far from proving it with certainty. I don't think we can automatically assume that a clear lectional impact in f13 was also exactly what would have created the scenario in the second century. For what it's worth, I don't doubt for a second that there was liturgical reading in the churches in the second century and that these practices c/would have impacted the manuscripts. But the problem is that we cannot here talk about a system, even a small one. In my mind, too, even if you're right about Luke 22:43-44's originality, and right about lectional influence that's clear in f13 also being active in the second century, we still haven't demonstrated anything at all about PA. The most we could say is that it *might* be something similar. For me, there's just too many assumptions in all this to stand on it.

    2. . . . but I appreciate the dialogue.

  6. Sounds like a great conference. Thanks for the summary.

  7. In relation to the lectionary issue (in general, not with specific reference to the PA), I have some reflections in Goulder and the Gospels, Chapter 19, which is on Early Christian Worship. I argue that there is no clear evidence of the use of the Gospels as lectionaries this early, but that one should take seriously the development of Gospel traditions in the liturgy, including the Passion narrative at Pascha, etc. One real difficulty is that Justin Martyr speaks about reading the memoirs of the gospels "as long as time permits", which does not sound like a regular lectionary portion.

    1. Right, Mark. That's exactly why I can't bring myself to say that what we have in Justin is lectionary reading. I think it's clearly part of a trajectory toward it, but it's just not enough to stand on securely.

  8. By the way, thanks for the great summary and discussion -- wish I could have been there.