The Bibliographical Test Updated

For more than forty years Christians have appealed to what is called the “bibliographical test” as a means of establishing the New Testament’s (NT) transmissional accuracy. The bibliographical test examines the overall number of extant manuscripts (sometimes abbreviated to MSS or MS for the singular) and the difference between the date of the original writing, called the autograph, and the date of the earliest surviving, or extant, manuscript. Since we do not possess the autograph of even one ancient document, this test best determines transmissional accuracy for any ancient document.

Christians argue that if historians will consider an ancient document to have been accurately transmitted whose manuscripts are few and far between the date the autograph was penned and its earliest extant copy, then they should accept documents as accurately transmitted whose manuscripts are comparatively many and comparatively near their autographs. For many years Christian apologists have employed the bibliographical test to argue that since the NT surpasses all other ancient documents in sheer number of manuscripts and the nearness of the date between the autographs and extant manuscripts that the NT has been accurately transmitted.

However, some of the numbers that apologists appeal to are woefully out of date. For this post I will provide just on example. Those who wish to see the updated numbers for others ancient authors and the updated numbers for the New Testament’s translations into other languages can find the fuller research in my article in the Christian Research Journal. You can listen to a radio interview here.

Homer authored the Iliad and the Odyssey and for years apologists have claimed that while there are over 5,500 manuscripts of the Greek NT, there are, by comparison, only 643 manuscripts of Homer’s Iliad. For example, a recent Google query of Homer 643 manuscripts turned up 143,000 results, and a review of the first one hundred of those results revealed that all of them related to whether the NT has been accurately transmitted.1

But things have changed: the more recent number of Iliad manuscripts is 1,757.

Martin L. West, senior research fellow, All Souls College, Oxford, cataloged a total of 1,569 papyri.2 But this is a papyri only count and not a total manuscript count.3 West said that he didn’t believe there was a more recent nonpapyri count than that found in T. W. Allen’s Homeri Ilias, which contains 188 items.4 Thus 1,569 papyri, plus 188 parchment manuscripts, comes to a total of 1,757.5

Although there has been an increase in the number of non–NT ancient manuscripts, nothing has changed regarding the applicability of the bibliographical test. Even Homer’s Iliad, which has seen the greatest manuscript increase, is still dwarfed by the NT, which has more than three times the Greek manuscripts as the Iliad. When one adds the fifteen thousand manuscripts in other languages, and then considers that almost the entire NT could be reproduced by the quotations of the early church fathers, one must maintain that, despite the increase of non-NT ancient manuscripts, the NT remains in a class by itself: it is by far the most attested ancient work.

This troubles skeptics because if they reject the transmissional reliability of the NT, then they must also consider unreliable all other manuscripts of antiquity. As John Warwick Montgomery has often related: “Some years ago, when I debated philosophy professor Avrum Stroll of the University of British Columbia on this point, he responded: ‘All right. I’ll throw out my knowledge of the classical world.’ At which the chairman of the classics department cried: ‘Good Lord, Avrum, not that!’”6

  1. Conducted on August 26, 2011. []
  2. Martin L. West, Studies in the Text and Transmission of the Iliad (München: K. G. Saur, 2001), 86. []
  3. Even though manuscript literally means handwrite, scholars like West sometimes use manuscript to refer only to nonpapyri manuscripts. Papyri manuscripts they only call papyri. []
  4. Thomas W. Allen, Homeri Ilias (1931; repr., New York: Arno, 1979), 11–55. Personal correspondence with West on October 30, 2010. []
  5. West also lists 142 Homeric papyri (glossaries, commentaries, scholia minora) and 47 witness papyri (“miscellaneous papyri and inscriptions in which verses of the Iliad are quoted”), ibid., 130. []
  6. John Warwick Montgomery, Human Rights and Human Dignity (Grand Rapids: Zondervan, 1986), 139. []
This entry was posted in Apologetics and tagged , , , , , . Bookmark the permalink.

23 Responses to The Bibliographical Test Updated

  1. hiero5ant says:

    I have no reason to dispute the claim that we have “over 5,500 manuscripts of the Greek NT,” but I’m having trouble determining what this claim consists in. Does this mean that we have 5,500 copies of “all the canonical books assembled in one package”? Or 5,500 manuscripts with at least a sentence or three quoting some NT book? Or something different?

    Moreover, don’t we know independently that e.g. Mark has not been transmitted completely accurately? Every manuscript of Luke and Matthew (transmitters of Markan material) is a manuscript where subsequent authors clearly altered the text for identifiably partisan political and theological reasons.

    • tkjaros says:

      hiero5ant,

      Could you point me to spots where the text is “clearly altered … for identifiably partisan political and theological reasons.”

      Thanks.

      • hiero5ant says:

        Wikipedia is the friend of all:

        Regarding verses where Mark differs from Matthew and/or Luke, it is often easier to see why Matthew or Luke would alter Mark than the reverse. For example, the pericope starting at Matthew 20:20 lacks a criticism of the disciples found in Mark 10:35 and later verses.[2] Matthew 8:25 and Luke 8:24 both lack disrespect towards Jesus from the disciples, portrayed in Mark 4:38.[2] Henry Wansbrough writes: “Mark is highly, even shockingly, critical of the disciples’ lack of faith and understanding; Matthew and Luke both weaken this criticism, in a way that might be expected to have occurred at a time when reverence for the first leaders of Christianity was increasing.”[8]

        Mark’s Jesus often seems more human than Matthew’s. Davies and Allison[6] list a number of passages where Mark but not Matthew portrays Jesus as emotional (e.g. Mark 1:41, cf. Matthew 8:3), ignorant of some fact (e.g. Mark 6:37-38, cf. Matthew 14:16-17), or incapable of some action (e.g. Mark 6:5, cf. Matthew 13:58).

        http://en.wikipedia.org/wiki/Markan_priority#Alterations

        • tkjaros says:

          Thanks for the quick reply.

          Perhaps I’m not seeing your point, but I still don’t see how there would be any alteration of text. What the wikipedia quote does tell me is that perhaps Matthew or Luke recalled the event slightly differently or (for a more liberal approach) they altered the story for their own purposes.

          But even altering a story for your own purposes is different than altering a pre-existing text. The transmission argument that Christians point to is not the fact that there may be discrepancies or differences among the NT books, but that the text has remained pure from textual alterations (when an author deletes a previously written text from a document or adds something that wasn’t in the original). So even if Matthew and Mark disagree on something (which is what you seem to be arguing), that is saying something different than, what Matthew or Mark wrote in mid-1st century is what we have today (the transmission argument).

          • hiero5ant says:

            Your first two sentences baffle me. You can see how they altered the story, but you can’t see how that means they altered the story?

            One must also keep in mind that to recall something, one must first have experienced it, and so that cannot be the case here, as none of the gospels were written by eyewitnesses.

            I find the difficulties believers have in tracing the common descent of stories to often be exactly parallel to their frequent difficulties tracing the common descent of organisms. The argument is not “they look similar”. It is that there is a very specific pattern of a specific kind of similarities that matches exactly what one would predict as the result of known mechanisms (in this case, psychological mechanisms).

            There is a story with no miraculous birth and no resurrection appearances, which treats the disciples as putzes, and this story is amended by later writers to include more miracles and to elevate the authorities of the nascent catholic church. Just human beings doing what humans do, spinning things to suit their own agendas.

            • tkjaros says:

              “Your first two sentences baffle me. You can see how they altered the story, but you can’t see how that means they altered the story?”

              You’re missing the distinction that I’ve been making: text versus story.

              I can alter the story of the way the Titantic went down by writing my version on my own paper, or I can make literal edits on a story about the Titantic that’s already been written. The bibliographic test is regarding the former.

              You objected that the “text” is “clearly altered … for identifiably partisan political and theological reasons,” but then you went on to support that claim by arguing that the stories have changed (like how Matthew wrote something differently than Mark).

              For the “text” to be altered, you’d have to argue something like, ‘A redactor placed these three words in here because they aren’t in the earlier manuscripts.’

              Hope that clarifies.

              Oh, and John was definitely an eyewitness. Swinburne’s written a book called “Jesus and the Eyewitnesses.” I’d highly recommend it. In the meantime, don’t be put off by non-eyewitness accounts. We read non-eyewitness accounts of historical events all the time. Why the double standard?

  2. hiero5ant says:

    Look again at the examples of “editorial fatigue” in that link. The pseudonymous authors of Luke and Matthew were not editing Mark’s fashion sense. They were not editing his musical score. They were editing his text. Therefore, every copy of Luke and Matthew is an example of transmissional inaccuracy. (Not to mention the original short ending of Mark etc.) Moreover, they are examples of transmissional inaccuracy performed for precisely the reasons I claimed: political and theological posturing. QED.

    Really, not much turns on whether a gospel author tendentiously rewrote a “text” or tendentiously rewrote a “story”; in either case, the fingerprints of the gospels’ purely human authors are all over them. This is not some bizarre, fringe claim I picked up from some unemployed dental assistant’s vanity myther-blog. This is as close to mainstream consensus in the field as one is likely to find. For example, Mike Licona has pointed out that the story of the zombie saints was added for allegorical reasons, and that the author of John just straight-up changed the date of Jesus’ crucifixion to make a theological point.

    It’s not much of an apologetic to say, “look, we can trust Luke and Matthew because even though they were obviously photoshopped, we have lots of copies where no further photoshopping was performed.”

    Of course, accuracy is not veracity. If the number of texts of the Iliad quintupled overnight, that would not tend to increase the probability of the existence of Ares and Zeus. The OP is reasonably good about this, until the very end with that contextless quote about “throwing out all knowledge of the ancient world”. Please. As though the only options are %100 credulity or %100 skepticism.

    • tkjaros says:

      I think we are in agreement on lots of points, in reality. However, the wikipedia excerpt you pasted here is strictly referring to how the story was changed, not how the text itself was changed. The transmission argument is strictly about the text and not the story. That’s all I’m saying.

      Even in the ‘fatigue’ section from the wikipedia page, it is about how Luke and/or Matthew altered a story from Mark, while they (or whatever authors) were writing their own accounts down. So it doesn’t deal with how the text of Mark changed over time, just how the story changed.

      You wrote, “They were editing his text.”
      No, they were editing his story. Here’s the difference:

      Mark-o (original), to Mark-1 (first copy), to Mark 2-, etc.
      vs.
      Mark-0, to Matthew-0
      Mark-0, to Luke-o

      Or even
      Mark-3, to Matthew-0

      The transmission argument deals with: Mark-o (original), to Mark-1 (first copy), to Mark 2-, etc.

      You’re arguing Mark-0, to Matthew-0
      Mark-0, to Luke-o or even Mark-3, to Matthew-0. That’s something I don’t care to argue about (right now).

      • hiero5ant says:

        However, the wikipedia excerpt you pasted here is strictly referring to how the story was changed, not how the text itself was changed.

        So Luke and Matthew were not copying a text? If you agree they used Mark, what is it you think they “used”? Were they attuning their vibrations to his entry in the Akashic Records?

        I hate to do this, but look again at the entry on editorial fatigue. It lists several examples of copied elements of text. You can’t slip into Markan verbiage unless you, er, know the verbiage.

        You wrote, “They were editing his text.”
        No, they were editing his story. Here’s the difference:

        Mark-o (original), to Mark-1 (first copy), to Mark 2-, etc.
        vs.
        Mark-0, to Matthew-0
        Mark-0, to Luke-o

        Or even
        Mark-3, to Matthew-0

        The transmission argument deals with: Mark-o (original), to Mark-1 (first copy), to Mark 2-, etc.

        You’re arguing Mark-0, to Matthew-0
        Mark-0, to Luke-o or even Mark-3, to Matthew-0. That’s something I don’t care to argue about (right now).

        Yes, both of those are what I’m arguing. If Matthew (0) copied Mark (n), then from the standpoint of transmission, Matthew (0) is Mark (n+1).

        The most the argument in the OP gets you (and it’s not nothing) is that it lets you say with somewhat greater accuracy that there was a time after which this one specific text became too widespread to tolerate any more tampering, infection with legend, copyist errors etc. Beyond that, as an apologetic it becomes a worthless tautology: “the transmission of the texts was accurate all the way back to the point before which the transmission was inaccurate.”

        • tkjaros says:

          “So Luke and Matthew were not copying a text? If you agree they used Mark, what is it you think they “used”? ”

          That’s contingent upon what theory you think best explains how the gospels came about. For example, the McGrews, known for the Bayesian probability argument for the resurrection, take a hardcore stance that Matthew came before Mark, since that what history explicitly tells us.

          So, if you’re trying to argue “Matthew (0) is Mark (n+1)” then I think you’re trying to do something more than what the transmission argument does here. All it seeks to do is to get us to Matthew (o), not pre-Matthew material.

          • hiero5ant says:

            I note that you left Dagoods’ question on the synoptic problem unanswered. Consequently I am left taking shots in the dark at what you may or may not believe.

            I explained already that the Transmission Apologetic is useless for getting beyond something like Matthew (0), on pain of tautology. I am not “trying” to do anything with it, other than point out that it is being misleadingly deployed as a proxy for a lack of alterations in the Gospel accounts generally, or worse, for their trustworthiness.

            Incidentally, I would advise against relying on McGrew for analysis of New Testament scholarship. In one breath, he endorses Habermas’s silly “the majority agrees on these four ‘facts'” apologetic, and in the next breath he is making claims that the consensus of mainstream scholarship on the Synoptic issue is wrong, and that every gospel is written by the people it is legendarily attributed to! The hypocrisy and cynicism of that kind of “whatever argument works” apologetics is exactly the kind of thing that drives a door-knocking, pamphlet-out-handing born again christian to start doubting the integrity of the whole enterprise.

  3. hiero5ant says:

    p.s. did you perhaps mean Bauckham, not Swinburne? The book where he argues that because Don, Peggy, Roger, Joan, Pete, and Dick were common names in 1960s New York, that Mad Men was written by eyewitnesses?

  4. DagoodS says:

    tkjaros: I can alter the story of the way the Titantic went down by writing my version on my own paper, or I can make literal edits on a story about the Titantic that’s already been written. The bibliographic test is regarding the former.

    This was confusing to me. Do you mean, “The bibliographic test is regarding the latter”? My understanding of the numbers involved in this bibliographical test was to support the claim, “Mark is extremely close, if not exact to MarkCopy1, which is extremely close, if not exact to MarkCopy2, which is extremely close, if not exact to MarkCopy3….” and so on. That by virtue of the numerous examples, we can winnow and determine as close as possible the “original” writing.

    Whereas the synoptic altering is Matthew and Luke using the written Mark to make their own alterations in the stead of Mark.

    I would think (perhaps I am wrong) the copying transmission is closer to literal edits on a previously-written Titanic Story, as compared to writing one’s own version. Although this is not clear, because both the scribe attempting (and failing) to copy written Mark word for word could introduce edits (even intentional ones) in a similar manner Matthew and Luke DID introduce modifications from a written Mark.

    Out of curiosity, what is your solution to the synoptic problem? That may help clarify my confusion

  5. Clay, thanks for the good work. Near the end of your article you repeated an oft heard claim: almost all of the New Testament could be reconstructed from Church father quotes. I recently read somewhere, but cannot find where, that that claim is factually incorrect, that it is a claim made in the past by one scholar, and has been quoted over and over so many times that everyone takes it to be true. However, when looking at the details, much of the New Testament is missing from quotes. Sorry, I can’t find the source to this, but have you heard it?

    • clayjones says:

      The following quote comes from Bruce Metzger and Bart Ehrman:

      Besides textual evidence derived from New Testament Greek manuscripts and from early versions, the textual critic has available the numerous scriptural quotations included in the commentaries, sermons, and other treatises written by the early Church fathers. Indeed, so extensive are these citations that if all other sources for our knowledge of the text of the New Testament were destroyed, they would be sufficient alone for the reconstruction of practically the entire New Testament.*

      I don’t know of anyone who has proven them wrong, Thinker.

      Clay

      *Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament: Its Transmission, Corruption, and Restoration, 4th ed. (New York: Oxford, 2005), 126.

  6. Pingback: Really Recommended Posts 8/3/12 « J.W. Wartick -"Always Have a Reason"

  7. Thanks for this post. I’m updating my training materials. Any word on the most recent figure for Julius Caesar’s Gallic Wars? Do we still only have 10 ancient copies; the earliest one made 1,000 years after the original?

    • clayjones says:

      Hi Mikel,

      Instead of 10 there are 251 manuscripts (a 2,500-percent increase!) beginning from the ninth century (the majority belong to the fifteenth century). This is from Virginia Brown, “Latin Manuscripts of Caesar’s Gallic War,” Palaeographia Diplomatica et Archivestica: Studi in onore di Giulio Battelli (Rome: University of Rome, 1979), 105–7.

      Clay

Comments are closed.