Biblical Archaeology Review has continued their recent assault on the (nearly airtight) case that the “Secret Gospel of Mark” which Morton Smith “discovered” in the Mar Saba monastery is a clever hoax. This time, they’ve hired a handwriting expert to analyze Smith’s handwriting and the handwriting of the letter of Clement in which the Secret Gospel was found. Her conclusion? Based on the evidence presented to her, it is unlikely that Smith forged the letter.
On the surface, this looks quite impressive and would seem to be quite a blow to the camp claiming the Secret Gospel is a hoax—a professional handwriting analyst claiming Smith didn’t do it is pretty persuasive. Persuasive, that is, until one reads the actual report. The truth is that handwriting experts can only make judgments based on the evidence at their disposal, and, in this case, the evidence is especially spotty. Basically BAR‘s tests entailed an examination of the hand of the Clementine letter, an examination Morton Smith’s English and Greek handwriting from a number of his marginal notes, etc., and then a comparison of the two.
Ms. Anastasopoulou concludes, after noting that 1) the hand in the Mar Saba letter is a good example of an 18th Century formal calligraphic Greek hand, “written spontaneously and with an excellent rhythm. The letters and their combinations are curved fluently while at the same time the grammatical rules are followed,” while 2) Smith’s English hand is spontaneous and shows the typical fluidity, speed, and fluency expected of a native English speaker, and 3) Smith’s Greek shows grade-school level fluency and formulation, written much more slowly (letter-by-letter), occasionally getting accents wrong, and not demonstrating the ability to actively and constructively operate within the language. In the expert’s words: “His writing is like that of a school student. It is obvious that his hand is not familiarised in Greek writing so as to be able to use it freely and with ease and be able to express thoughts and beliefs.” On the basis of these three observations, Anastasopoulou concludes that it is highly unlikely that Morton Smith could have forged the Mar Saba letter, given that he showed no signs of ability to work fluidly with Greek in the writing of his own notes.
There are several problems that make the analysis questionable. First, it asks the opposite question this kind of handwriting analysis is designed to answer. That is, this test does not involve comparing an authentic exemplar with a suspected forgery to determine whether the hand is the same or to detect any discrepancies that would suggest an imitative hand. What is being asked here is much harder to answer, as the Mar Saba letter does not attempt to mimic a specific 18th Century hand, only a particular calligraphic style (that is, it is not tacked onto a manuscript purportedly using the same hand). It is one thing to have an original by which to judge a fake. It is a wholly other thing to have something original and try to determine whether a given person is capable of having produced it. So, unlike most forgery investigations, the question here involves trying to prove a negative rather than a positive. In other words, the only way this kind of test would have any meaning is if it was able to determine that Smith’s handwriting indeed bears some resemblance to the Mar Saba letter—positive results. But if Smith’s handwriting doesn’t look like the Mar Saba letter, that negative result doesn’t mean he couldn’t have altered his own hand (especially in a highly specialized script) to look different. That is, after all, the goal of a forger.
Secondly, the materials provided (samples of Smith’s hand in English and block Greek where he is copying specific phrases) are in no way comparable to the Mar Saba letter. No samples of Smith’s attempts to produce calligraphic style Greek are provided, nor any other materials that might show the ability to learn and copy something in a specific script. It is one thing to say that Smith’s block Greek handwriting shows a lack of fluency and speed in writing the letters; it a wholly other thing to suggest that this indicates he could not have produced the Mar Saba document. This is akin to suggesting that someone whose print handwriting looks like a third grader’s when jotting notes to himself couldn’t teach himself to copy a letter in flowing cursive. They’re two different types of writing and entirely different skills.
Thirdly, the question of fluency and grammar is really quite out of place here. It’s not as though Smith would have simply sat down at the Mar Saba monastery and written the letter extemporaneously. If those arguing for forgery are correct, Smith put a great deal of time into crafting a masterpiece forgery, complete with carefully-crafted in-jokes, relying heavily upon specific Clementine vocabulary, thoroughly vetting the grammar, and validating arguments about Mark that Smith himself had proposed some time earlier. Are we truly to believe that Smith would have simply gone to Mar Saba and just written this letter with no calligraphy practice? Most certainly not!
Put simply, BAR‘s handwriting analysis concludes that, all things being equal, based on the handwriting excerpts in his notes, etc., Smith couldn’t have simply sat down and copied out the Mar Saba letter, But this isn’t the proposed scenario at all! Rather, if Smith indeed forged Secret Mark, he would have worked through the Greek carefully for quite some time, checking and re-checking everything to make sure he had it perfect. Then he would have chosen a specific Greek scribal style to learn—preferably something that would be difficult (but doable) and distinct and traceable to the right timeframe. Then he would spend however long it took to learn this special style, making sure to use the right tools for the job (no ball point pens or #2 pencils for this kind of practice), ultimately practicing the Mar Saba letter in its entirety numerous times, likely memorizing the Mar Saba letter in the process, down to each accent, to make sure he could produce it without having to copy it from another sheet. Then, once he had mastered this style and sufficiently memorized the letter, he would have gone to the monastery and done the deed (or—more likely—done the deed and then gone to the monastery to plant the book), by then a practiced scribe at copying out his own creation in exactly the right hand with a smooth rhythm and comfortable speed. (This is where the fluency with the language argument breaks down; I’m pretty convinced that, given a few months, I could learn to write a script and copy down a very smooth letter in a language with which I am entirely unfamiliar, simply by practicing the letter enough. I wouldn’t even need to actually know what it said as long as I had memorized the movements adequately.)
In the process, of course, his overall Greek fluency wouldn’t have improved much, as he would be operating within such a small language base. Nor would his Greek block writing (with a pencil or ball-point, no less) have improved or changed much, since the two scripts are so extremely different. (For example, my Hebrew block characters haven’t changed a whole lot since I started writing in Hebrew cursive some years back; if anything, they’ve gotten clumsier.) If anything, the accents would be the only things to look somewhat similar (in placement and relative size), given that the accents aren’t all that different from our block script, though they’re certainly more fluid and the writing implement used would have some impact. (Incidentally, the accents are sized, placed, and angled similarly in Smith’s writing samples when compared to the Mar Saba letter, as the handwriting analyst notes—tending towards big accent marks in the 11 o’clock to 1 o’clock positions.) This would mean that his Greek block wouldn’t look much different before or after the exercise.
Unfortunately, the BAR analysis doesn’t consider this (what I would suggest is about the only feasible) scenario at all, content with a rudimentary comparison of Smith’s completely unrelated handwriting to a script and style entirely unlike his own standard hand—and then arriving at the dubious conclusion that, since his block Greek and English handwriting looks nothing like the 18th Century calligraphy in the Mar Saba letter, he couldn’t have pulled it off. At the end of the day, BAR‘s expert witness is only able to tell us that we have no material from Morton Smith comparable to the Mar Saba letter. But surely we didn’t need a professional handwriting analyst to tell us this, did we?
HT: Jim Davila
17 Comments. Leave new
On top of what you say, if you take a look at all the Greek letters that don’t rise above and below the lines, there is a remarkable similarity between Smith’s own hand and that of the letter found at Mar Saba. They’re so similar that the report should have tried to explain it.
I have one disagreement with you, however: you seem to imply that Smith wrote this letter while at Mar Saba. I think it more likely that he wrote it in his den at home and smuggled the book into Mar Saba. (That’s surely why the book was missing its front cover: Smith had bought the book, but he had to tear off the cover because it had been marked as to its former owner or seller. I’m not sure if those who think the letter is authentic address the missing front cover.)
Excellent point, and one which I had forgotten.
[…] This post was mentioned on Twitter by Stephen Carlson, Jason Staples. Jason Staples said: [blog] Biblical Archaeology Review's “Secret Mark” Handwriting Analysis Unimpressive http://bit.ly/awmmcX […]
Excellent post! As you said, the case against it seems virtually airtight, so I’m always interested in/puzzled by the defenses that come up.
[…] Staples on BAR’s assessment of Smith’s handwriting Jason Staples has written a blog post on BAR’s recent analysis of the handwriting in the Mar Saba document. If you’re not familiar with Jason’s blog, you should read it! Jason is a Ph.D student […]
Dedications are usually (though not always) written on the first page of a book, not on the inside flap of the cover. Naturally there are any number of counterexamples to this, but I don’t see that the missing cover to the Voss book is relevant.
If the point of this post is just “Well, Smith could have forged it anyway,” sure, I suppose he could have. The handwriting analysis doesn’t prohibit that, nor do I think anyone claims it does. But it surely does count as evidence *against* the case for forgery, any way you spin it.
(FWIW I am not typically much impressed by handwriting analyses one way or the other, though I think they tend to have some weak merits.)
Essentially the point of the post is indeed “he certainly could have forged it”—directly responding to BAR’s conclusion that he “probably couldn’t have.” I also don’t think handwriting analysis (especially this kind of report) is very weighty in this kind of hoax/forgery situation, especially on the negative side.
My point is just that when using probabilistic methods, you can always say the research hypothesis could be true. “He certainly could have forged it” is just as true whether or not BAR is right that he “probably couldn’t have”.
Pile on the ad hoc premises, and you can make any hypothesis plausible. (FWIW this is Popper’s point when he talks about pseudoscience.)
You write “the only way this kind of test would have any meaning” is if it tried to prove a positive, i.e. that Smith’s handwriting resembled the manuscript’s. But this is false–it still has meaning when it shows his handwriting does not resemble the manuscript’s. It just doesn’t have as much meaning.
the_cave, You’re right in that there is perhaps some meaning to negative results based on incomplete and stacked data, but when added into all the other (much more impressive) data, it is next to meaningless. I do think a comparison of Smith attempting a comparable script would have more meaning in this case; it’s just that measuring his handwriting in English and block Greek doesn’t have a whole lot of meaning when comparing to a wholly different script. It’s like looking at a Hebrew document and determining Smith couldn’t have written it because our samples don’t include any examples of Hebrew writing. My guess is that even an 18th Century scribe trained in calligraphy would be slow and hesitating when using a 20th Century block Greek script. A more productive sample for comparison might have been the M. Madiotes sample referenced in Carlson’s book, as at least it is in a somewhat comparable script (though again handwriting analyses are not especially weighty as evidence either way).
And again, we’re not talking about a series of ad hoc premises, nor is any argument for it being a hoax resting largely on handwriting.
Okay, let me get this straight: Carlson says it is a forgery because the handwriting is not fluent, shows hesitation, patching, ink blobs, etc. You say it is a forgery since it appears authentic with good rhythm, comfortable speed, etc.
I’m tempted to ask which one of you is correct, but I guess it doesn’t matter since you both look at the same evidence, interpret it in diammetrically opposite ways, and yet still arrive at the same conclusion.
Mary, I didn’t conclude that it “appears authentic with good rhythm, comfortable speed, etc.” Those were the conclusions of the handwriting specialist hired by BAR — conclusions that I would disagree with, incidentally. The point of my post, however, was to show that, as her conclusions rest on stacked data, they aren’t persuasive in the least in demonstrating that Smith didn’t do it. Given that there is a very strong case for a hoax even without handwriting analysis, BAR’s handwriting case doesn’t do any damage to those claiming it’s a hoax.
1) More than the cover was missing from the Voss book: also the title pages and front pages that could have been used as practice sheets.
2) Has anyone mentioned Clement of Alexandria page 293 Appendix on Palaeographic Peculiarities? The examples there were drawn by whom?
But wait, Smith already warned us off (Secret Gospel p. 23): “I doubt that its palaeographic peculiarities will yield much more information.”
The BAR report in fact did consider the handwriting of the Clement Appendix on Paleographic Peculiarities.
By saying the data is “stacked” are you suggesting that BAR’s analysis was deliberately dishonest?
Though you criticize the comparison of Smith’s handwriting to that of the document in question as meaningless, maybe you will at least agree that the BAR analysis of the hand of the Mar Saba document (in and of itself) shows that Stephan Carlson’s same analysis was amateurish and wrong? Perhaps the lack of available examples of Smith writing in cursive Greek means he never did write in cursive Greek?
In using the term “stacked,” I am suggesting that given the methods employed for this kind of test, anyone could have predicted the outcome before the analysis was done, so the samples were “stacked” in that BAR had a very good idea of the outcome before ever doing the analysis.
I wouldn’t exactly agree with your conclusion that Carlson’s analysis was “amateurish and wrong”; rather, I would simply say that I don’t see that the professional analysis was all that much more “professional and right,” leading me to conclude that—in the absence of better samples or more persuasive handwriting analysis, this kind of analysis is not a smoking gun for either side. This is why handwriting analysis is handled with care by the legal system these days—it is far from an exact science, especially given disparate samples from the start.
Sure, it could mean that he never wrote in cursive Greek. But it also doesn’t mean that he never did.
For my own views, see http://www.music.princeton.edu/~jeffery/VAnastasopoulou.pdf
Further comments on handwriting, in two parts:
http://salainenevankelista.blogspot.com/2010/06/further-comments-on-recent-handwriting.html
http://salainenevankelista.blogspot.com/2010/06/further-comments-on-recent-handwriting_14.html
The latter deals with Anastasopoulou’s report, and offers a suggestion that in my mind would be possible for everyone, the defenders of the authenticity and the proponents of the forgery hypothesis, to accept, namely that “On the basis of the handwriting in Clement’s letter to Theodore Morton Smith cannot be accused of forgery.”
Josh McManaway above has it right: I’m puzzled, too, for as far as I can see the hoax hypothesis as laid out in The Gospel Hoax is a cathedral completely crumbled, yet it is still perceived to be “nearly” or “virtually” “airtight” by others. How can we have such divergent assessments of the evidence?
I have been sitting on an interview I arranged between Professor Charlie Hedrick and Agamemnon Tselikas a respected Greek paleographer on the question of the authenticity of the Mar Saba document. It is now published at my blog http://stephanhuller.blogspot.com/
Hope you and your readers might want to check it out. Dr. Tselikas will be publishing an article on the same subject for BAR next year referencing the same material.
Sincerely
Stephan Huller