Sunday, May 19, 2013

Presuming the Authenticity of Texts

In a discussion over at Diglotting, Kevin Brown wrote "The default position is that Nazareth in Mark 1.9 is authentic; one has to prove otherwise."   When I asked him why this was the default position, he responded, "Really?! This shouldn’t be a controversial issue." 

It's hard for me to see how presuming the authenticity of the text of Mark wouldn't be controversial.  We lack any manuscript evidence for more than the first century of its existence.  In the manuscripts that we do have, the earlier ones show a higher rate of variants than the later ones from which we can surmise that the highest rate of variants would have occurred during the period for which we lack evidence. There was no church authority overseeing the copying of the texts, but there were plenty of men who were willing to forge texts in the name of Peter or Paul or any other figure that the forger thought would lend authority to the writing.

In short, there is every reason to think that there were any of number of opportunities for alterations by men with both the motivation and the willingness to do so and every reason to think that there are any number of alterations that left no evidence in the manuscripts. How could we possibly justify authenticity as a default position?

Kevin argued for the presumption on the basis of "tenacity," which refers to the idea that the original reading of the text is always somewhere in the manuscript tradition.   Tenacity is lovingly embraced by conservative apologists, but my impression is that the consensus among textual critics is that there likely are original readings that were lost in the early transmission.  The problem is that tenacity is inferred from manuscripts that were overwhelmingly produced by trained scribes after the time that Christianity had become the official state religion of the Roman Empire with established structure and doctrine.  We cannot assume that it prevailed in the earlier period when Christianity was a disfavored minority religion composed of competing sects with significant variations in belief.

The idea that something should be presumed until disproved is a familiar one to judges and lawyers.  We all know that a man is presumed innocent until proven guilty.  What we may overlook, however, is the fact that such presumptions are usually based on considerations of policy rather than probability.  The presumption of innocence isn't based on the likelihood that the police have arrested the wrong man.  In fact, we hope that the presumption will help assure that they usually get the right one.  A husband is presumed to be the father of a child conceived by his wife not because matrimony proves paternity, but because we are more interested in protecting kids than cuckolds.  

I believe that the presumption of authenticity is also based on policy rather than probability.   As Kevin points out "We could argue that every word or phrase in the NT was interpolated with such specious argumentation."  So what?  If in fact our evidence isn't sufficient to eliminate the possibility of tampering and alterations, why not simply have New Testament scholars qualify their conclusions to reflect the appropriate degree of uncertainty.  Isn't that more intellectually honest than presuming authenticity.  New Testament scholars want to talk about the original texts, but maybe the evidence doesn't justify it.

I think Kevin overstates the problem though.  The problem isn't that every word might be an interpolation but that any word might be.  That's a problem that historians are always faced with, but they deal with it through  corroboration rather than presumption.  If a passage in one of Paul's letters contains an idea that is found three times in other letters, then we have stronger reasons for thinking it genuine.  However, if an idea only appears once, we are necessarily less certain about the passage containing it.  If by positing that passage as an interpolation, we would radically change our understanding of Paul, then our original understanding is necessarily less secure than we might otherwise have thought it to be.


  1. At one level, I understand and generally agree with Kevin Brown. Biblical studies has a myriad of positions, we almost need a “default” position to start from to communicate! If we have a particular word or statement within all copies before (as an arbitrary point) the 10th Century, I would agree the word or statement is “default” as being the author’s original writing.

    However, (as you point out), we see modification in numerous situations, and it would be extreme hubris to presume we have found all modifications ever made. If there were modifications made where we can see—i.e. on the 4th, 5th, 12th copies we have—it is extremely likely there are modifications made where we cannot see—i.e. the 1st, 2nd copies. Simply put, we have a “default” position while recognizing the probability there is interpolation within the default.

    Perhaps I deal with presumptions, and preponderance and persuasion so much, I do not see the controversy, nor do I see monumental hurdle in overcoming such presumptions. Again, as you point out, in our legal circle, we have “presumed innocent until proven guilty” and it takes a far higher standard—beyond a reasonable doubt—to overcome the presumption. Yet people are found guilty—the “default” is overcome—all the time. Every day. Every court.

    On this particular issue, I would agree with Kevin Brown—“Nazareth” in Mark 1:9 is the “default” position for being in the original writing. But the arguments supporting the claim “Nazareth” is interpolated overcome (easily in my opinion) the presumption of the default position. To simply reiterate “this is the default position” would be like the Defendant claiming, “I am presumed innocent!” after the prosecutor has introduced the Defendant’s confession, the video of the crime, the fingerprints, the eyewitnesses, the DNA, etc.

    Now it is time for the Defendant to present their own evidence and arguments, rather than relying upon a presumption that has been overcome.

    My position would be the consistent copies are presumed “default” with a caveat we understand there are unfound interpolations. Much the same as my position is people are presumed innocent with the caveat there are unproven crimes.

    1. I must disagree my friend.

      Although I do not deal with presumptions and burden of proof as often as you do, I do remember the concepts from law school and I don’t see why they are necessary in academic inquiries. I have often said that “burden of proof” is only necessary in the legal process because fact finders don’t have the option of saying “I don’t know” or “too close to tell.” In the academy, it should be perfectly legitimate to declare the evidence insufficient to reach a conclusion.

      Although the law does contain many presumptions, I think there is usually a reason beyond mere convenience. We presume that a minor is incapable of consenting to sex with an adult. We presume the innocence of the accused. We presume the paternity of the husband. We presume impairment when blood alcohol level exceeds the legal limit. In these cases, the presumption reflects the kind of clear policy choice that doesn’t (or shouldn’t) arise in the academic context.

      Even if some legal presumptions may be primarily matters of convenience in order to facilitate the litigation process, I’m not sure that the same logic justifies presumptions in an academic inquiry. The question for me is this: How likely is it that a passage in the earliest manuscript goes back to the autograph? If it can be shown to be very likely, then that can be offered as a positive reason and we don’t need a presumption. If it cannot be shown to be very likely, then it is preferable to simply admit that the possibility of interpolation or alteration is a variable for which we cannot control. In that case, we should take as our starting point that the manuscripts reflect the understanding of the communities at the time they were created, and anyone who wants to argue that a passage does or does not go back farther should be prepared to offer positive reasons for thinking so.

      As a personal matter, I have a rebuttable presumption in favor of the consensus of peer reviewed scholarship in most fields. This is a matter of convenience as I cannot become an expert in every topic. The only areas where I do not apply the presumption are economics and New Testament studies because I have found that there is entirely too much ideologically driven scholarship in both fields.

  2. Hmmm…interesting discussion. Do we disagree because of genre classification? I attempt to derive a consistent methodology in determining autographs vs. interpolations, and the contemporary (to the gospels) works I tend to compare them to are histories and bios.

    You and I (I think) both agree the gospels were not written as histories, but my problem is there are few discussions regarding interpolations in bios of the time as historians simply don’t care! We recognize the possibility but equally recognize we have little means to determine it as such. The closer the gospels come to historiography, the more likely latter copyists would integrate additional material, or make modifications suited to intended recipients.

    For any lurkers—a modern example. If we were copying a list of numbers to do a bank deposit, we would expect the copyist to be 100% accurate due to the genre of bank deposits. No interpolations or modifications or even unintentional errors are allowed. On the other hand, when it comes to movies, we allow new writers and directors to completely re-manufacture a story—such as the re-boot of Batman or Star Trek--where the original story is maintained in some particulars but completely abandoned in others. No one is screaming about “interpolations” in the new Star Trek series. (Yet, curiously, within the story arcs, fans point out anachronisms, demonstrating a need for some consistency even within fiction.)

    So what is the gospels’ genre? Can we even make such a determination? (And, of course, genres are not clear-cut definitions. They could be somewhere between bios and historiography with a smattering of history.)


  3. Maybe the historians don't care about the possibility of interpolations in bios because they don't try to draw conclusions that would change much depending upon whether a particular passage was preserved as written by the original author or modified by someone later.

    That is why I suggest that our starting point should be that the manuscript reflects the understanding of the community for which it was made. If our earliest manuscript of Mark comes from 300 A.D., then our default position should be that it reflects what the story looked like then. We should be very wary about making claims about what the story looked like 100 years earlier. In general contours, it probably looks like what the original author wrote, but it is probably foolish to talk about whether the original author used particular words at particular points.

    New Testament scholars routinely make arguments that are highly dependent on particular words such as histerio. Do other historians do that?

  4. Here's the problem with using presumptions in an academic argument:

    In the law, an unrebutted presumption is treated as an established fact. It's res judicata and no doubts need be entertained throughout the rest of the proceedings. However, even if the presumption that a husband is the father of a child conceived by his wife is unrebutted in court, it would be a very bad idea for a doctor to use it to diagnose a hereditary illness or to order a blood transfusion.

    Let's see what happens when presumptions are used this way in New Testament studies:

    (1) All manuscripts include "brother of the Lord" in Galatians 1:19 so Bart Ehrman presumes that it is genuinely Pauline. However, he knows perfectly well that we cannot know what happened to the texts in the first 150 years of transmission and that is precisely the type of thing that any scribe along the line might have added with the best of intentions just to make it clear which James it was that Paul met.

    (2) Ehrman presumes that words are used in their most natural sense. Therefore Paul is referring to a biological relationship, but there is certainly a non-trivial probability that Paul was referring to spiritual brotherhood.

    (3) Ehrman presumes Paul is telling the truth and that Paul really met the biological brother of Jesus even though their is a non-trivial probability that Paul is mistaken or lying for reasons of his own.

    (4) Ehrman writes that Galatians 1:19 proves the existence of a historical Jesus "beyond a shadow of a reasonable doubt" even though it is logically impossible for a conclusion to be any more certain than any link in the chain of premises and conclusions upon which it is based. In fact, it is likely to be less certain than any of the individual links in the chain due to the way probabilities are multiplied.

  5. And of course, courts actually find people 'not guilty' rather than innocent.
    Andrew Ryan

  6. Vinny,

    [sorry for the delay…busy…]

    I am unaware of interpolations being argued in other bios of the era…but of course other bios are treated much differently than the gospels. We understand different authors had different agendas regarding their subjects, and would modify the stories according to those agendas. See for a good explanation regarding how bios would result in conflicting tales regarding the same subject without concern.

    Vinny: If our earliest manuscript of Mark comes from 300 A.D., then our default position should be that it reflects what the story looked like then.

    But it is not that clean. We also have manuscripts from Matthew and Luke, who copied Mark. If we have a 200 CE Matthean manuscript (for example) can we trace back at least the copied portions of Mark to that date? Additionally we have Second Century authors who are aware of Matthew, Mark, Luke and John, although their quotations create additional difficulties.

    Further, I (personally) think Mark 13 makes a strong argument the Gospel was written some time between 70 – 100 CE, so we know the Gospel at least existed in some form at that time.

  7. If we have reason to think that a passage dates back farther, then it is perfectly reasonable to hold that it dates back farther. The question for me is whether there is anything about the earliest manuscript containing a passage that gives rise to a presumption that it was originally part of the original writing. I don't think that it does and I think that thinking in terms of rebuttable presumptions gives rise to the kinds of errors that Ehrman makes.