Monday, December 14, 2009

Tremors, or Just an Optical Illusion? A Further Evaluation of Carlson’s Handwriting Analysis (the shorter blog version) by Roger Viklund

Roger Viklund offers a summary of his new online article "Tremors, or Just an Optical Illusion? A Further Evaluation of Carlson’s Handwriting Analysis" below. My own comments on the article follow afterwards.


I am now fairly certain why Stephen C. Carlson saw all signs of forgery in the handwriting of Clement's letter to Theodoros. It seems all to be due to the basis of his work; printed low resolution images from Clement of Alexandria and a Secret Gospel of Mark by Morton Smith. This is a shorter version of my newly written article Tremors, or Just an Optical Illusion? A Further Evaluation of Carlson’s Handwriting Analysis.

In 2005 Stephen C. Carlson came out with The Gospel Hoax: Morton Smith’s Invention of Secret Mark. As the title says, Carlson argued that Morton Smith in fact invented Clement's letter to Theodoros containing extracts from a so-called Secret Gospel of Mark. Among the several indications for this presented by Carlson, he also pointed to the actual writing. We only have photographs of a now lost copy of the letter. This shows a text that appears to be written in a hurried cursive eighteenth century Greek handwriting. Carlson thinks that “it should be possible to tell if the hurried handwriting of Theodore is natural or simulated.” (Carlson p. 27) He suspects that if Morton Smith forged that handwriting in the twentieth century, imitating a hurried cursive eighteenth century handwriting, he would have had to write much more slowly in order to form the letters in a way which was not typical for his own way of writing. This careful imitation where the letters are almost drawn would, according to Carlson, lead to a number of characteristics which could reveal the counterfeiter (and according to Carlson, this counterfeiter, or hoaxer, was Morton Smith).

Carlson also claims to have found signs revealing this slow imitating. He refers to a number of “[b]lunt ends at the beginnings and ends of the lines” of the letters. This would be the effect of a very slow shaping of the letters and that the pen thereby came to a stop and the ink would accumulate. He also finds many instances where the pen was lifted in the middle of a stroke and this would indicate that Smith needed to prepare himself for writing the next letter before continuing. Occasionally Smith still had to go back and retouch certain letters. Further Carlson sees a lot of tremors in the writing, tremors which reasonably would not occur if a skilled scribe would have written the letter in a fast pace. Accordingly the tremors are also an effect of slow writing.

In my previous study, Reclaiming Clement's Letter to Theodoros – An Examination of Carlson’s Handwriting Analysis, I did not compare my colour images (which were scanned in 1200 dpi directly from the colour photos taken by Kallistos Dourvas in the late 70s) to the images Carlson used (which he took from the printed images in Morton Smith’s book showing the black and white (b/w) photos Smith took in 1958). Now that I have scanned also these images in 1200 dpi, I realise that the low resolution prints, done on a printing press in the early 70s, have a line screen at a 45 degrees angle. I had of course noticed this before and I had also previously been updated by Scott G. Brown, who informed me of this in an e-mail. Further, also Walter M. Shandruk in a blog post named Carlson’s Handwriting Analysis on Secret Mark, noticed how pixelated the letters were. But the images he presents do not seem to fully show all the details.

A line screen can be described as lines made out of separate dots of different sizes, yet organised in straight lines separated by the same distance and reproduced at a fixed angle. In this case the angle is 45 degrees. As can be seen below from the enlarged image to the left, which is scanned from the printed black and white images of the letter found in Smith’s book, the dots are arranged very symmetric. When you are printing, as in this case in black and white, there is only one colour, and that colour is black There will either be a black dot, or nothing at all, just the background colour of the paper, which normally is whitish. If you wish to produce different shades of grey, you still only have black dots at your disposal. You fool the eye into believing that the image is grey by letting white and black areas interact with each other. The more black dots or the larger the dots are, the darker the area appears to be, and of course, the fewer black dots, the lighter the area appears. The area to the left below is simply a grey background area of the letter enlarged so that the dots can be seen by the naked eye. In normal magnification the area will appear just grey, as is illustrated by the same magnified area reduces into the small grey rectangle in the lower right corner of that image.

When you print for instance letters, these letters are also made up of similar dots. However, because of the low resolution reproductions in the book, the letters will not appear smooth in high magnification. This is due to the low resolution line screen. When you enlarge a halftone image very much, like the one printed in Smith's book, you will not see an accurate representation of lines that are not both perfectly straight and at an angle that accords with the screen. Since the dots are arranged in straight lines with the same distance between them, you can produce straight lines only horizontally, vertically or at 45/135 degrees angles (as the screen in this case is in 45 degrees angle). This is illustrated in the figure above to the right. Whenever you reproduce a line at a different angle, it will appear stepped and this could easily be mistaken for a hesitation in drawing the letter. Since Carlson must have looked at images similar to the ones I shall reproduce below, I claim that the reason he saw all these signs of forgery is not due to the fact that there are any such signs, but due the poor images he used, in which the letters appear to be stepped, but in reality are not.

Because of the limited space on this blog template, there is some problem publishing lots of images. I will therefore settle for showing just a few examples of why, in my opinion, Carlson went wrong. For a more thorough comparison, I once again refer to my main article Tremors, or Just an Optical Illusion? A Further Evaluation of Carlson’s Handwriting Analysis.

Ink blobs as a result of pen stops
Carlson sees “blunt ends at the beginnings and ends of the lines”. This he says indicates “that the strokes were written so slowly that the pen had come to a complete stop at the ends of the strokes.” I will give just one example in order to demonstrate that the ink blobs Carlson saw seems to be the a result of the low resolution line screen of the b/w images and hardly can be seen in the more high resolution colour images. Above we have this iota, where we in the colour image can see that there is a hook upwards to the left, while the b/w image only shows a big blob. This is in all likelihood the “ink blob“ that Carlson saw. The circles below show the ending enlarged. When these low resolution b/w images are magnified to this size the shading at the ends of the letters cannot be shown, as they are constructed of just a few large dots. All shades between black and white are missing and this is a good example of how impossible it is to render small details with a faded surrounding in the b/w images which Carlson used. This discrepancy between the colour images and the b/w images used by Carlson is obvious in almost all the examples presented by him. When the shading is missing, all small and circular movements are distorted, leading to an impression of ink accumulation.

Pen lifts and retakes
Carlson claims that on several occasions, where a skilled scribe would have written in one stroke, this scribe stops, lifts the pen and begins anew. Here there are many obvious and good examples to choose between, where the 45 degrees screen twists the images into looking as if there were retakes. In for instance this stroke between the omicron-upsilon ligature and the circumflex accent, Carlson sees a pen lift. That is on the middle of the long diagonal line, which also is enlarged in the circle.

On the colour image one can clearly see that this line is drawn in one stroke, yet in a curve which is a characteristic of this scribe when he writes this letter. However, in the b/w image the curve appears as if there is a break in the line and one or two retakes beginning beside/below/above the line. Now, this is a consequence of the line screen not being able to reproduce the curve, and instead beginning on another line of dots along the 45 degrees angle, in order to follow the letter’s curve.

Another example where Carlson spots a pen lift is between the epsilon and gamma in the image presented below. The actual place where the break appears is cut out and enlarged in the ellipse to the left.

It really looks like the epsilon is done separately and the gamma is connected a little bit above the epsilon. As can be seen in the enlargement, the line that begins at the epsilon first follows the 45 degrees angle of the line screen, then turns left straight upwards in a 90 degree angle and the turns right to follow the 45 degrees angle again. The colour image shows that the real angle is about 60 degrees and that there is no break. It can only be seen in the b/w image.

These examples (and there are many) show that when lines are produced at angles which differ markedly from the angles possible to reproduce correctly, the lines will be stepped and it will appear as if the pen was lifted and a new stroke was done beginning on a different level. This is particularly obvious in curves as the line screen is not able to produce smooth bends. In order to follow lines done at perhaps 30 or 60 degrees angle, the dots have to shift position as they are following the 45 degrees angle and the lines will then appear to be stepped.

Tremors due to slow writing
Finally there are the forger’s tremors. Carlson writes: “The ‘forger’s tremor’ appears in the shaky quality of lines that should be smooth curves.” But the curves only appear not to be smooth from the low resolution line screen of Carlson’s images. A good example is the theta presented below, where especially the lower part looks shaky, while none of this can be seen in the high resolution colour image.

One can see that Carlson’s claim that in “the first line of Theodore, the shakiness is evident in the theta of Θεοδώρῳ” only holds true for the b/w image.

Other good examples are the so called squarish omicrons Carlson spotted. He claims that they “are so shakily written as to appear square rather than circular”. Carlson identified four omicrons which he claimed were square-shaped, but here I am picking just one omicron which is quite circular in the colour image, but appears to be a square in the b/w image. This rather heavily magnified omicron may demonstrate why the b/w images become distorted.

When a small circle (in reality this omicrons is less than 1 mm in diameter) is created with these dots, it is mainly following the lines of 45 and 135 degrees angles, forming a square standing on its corner. If one looks at the inner circle in the centre of the omicron, which obviously is quite round, the b/w image composes of only four white squares and the surrounding black dots can then only follow the lines of 45 and 135 degrees angles. The same holds true also for the outer circle.

No basis for judgement
This short survey still demonstrates that Carlson’s assertion that the handwriting of Clement’s letter to Theodoros shows signs of ink blobs, pen lifts, retouches and tremors, cannot withhold a critical examination. These signs lie rather in the poor images Carlson used than in the writing itself. The signs of forgery which Carlson claims could be detected in the handwriting cannot really be used as a basis for judgement of the letter’s authenticity, since the quality of the images he used is simply too poor. On the contrary, it can be said that the high resolution colour images do not show any conspicuously marks of ink blobs, pen lifts, retouches and tremors; thereby strengthening the opinion that the text was indeed written rather swiftly. This does in turn strengthen the opinion that the text was written by a skilled scribe in the eighteenth century.

Roger Viklund


A Comment on Tremors, or Just an Optical Illusion? by Timo S. Paananen

One of the first questions I had regarding Stephen C. Carlson's "The Gospel Hoax: Morton Smith's Invention of Secret Mark" (2005) concerned his choice of sources: why did he utilize the B/W photographs that were published in "Clement of Alexandria and a Secret Gospel of Mark" in 1973 instead of the colour ones Charles W. Hedrick and Nikolaos Olympiou published in 2000 - especially when we consider the difference in quality these two sets of photographs have. Furthermore, the latter were available as high-resolution scans, as Roger Viklund succeeded in obtaining a permission to use them in his previous online article "Reclaiming Clement's Letter to Theodoros - An Examination of Carlson’s Handwriting Analysis".

The very same question was asked by Scott G. Brown in his first response to the hoax hypothesis, published in "The Expository Times" in early 2006. (Scott Brown: Reply to Stephen Carlson. The Expository Times 117, 144-149.) Brown concludes that Carlson's choice to prefer the low resolution halftone reproductions "makes no sense". (p. 145 n. 4) Similar doubts were raised by Walter M. Shandruk in 2008: the resolution in Smith's photos - as they were published in 1973 - is not high enough to spot "that sort of tremor" a professional QDE (Questioned Document Examination) expert would like to find for establishing that the text is a forgery. ( Shandruk confessed: "I do not see a tremor" - an observation that Roger Viklund, in early 2009, echoed after he had studied the high-resolution scans of the colour photographs. He, too, could not see many of the "signs of forgery" Carlson affirmed were present in the text. (

Viklund's newest article, "Tremors, or Just an Optical Illusion? A Further Evaluation of Carlson’s Handwriting Analysis" summarized above, draws yet another bold question mark over Carlson's assessment of the handwriting of Clement's letter to Theodore, and offers a plausible explanation of the general trend of not seeing anything out of place in the quality of the letter's drawn line - there is nothing to be seen, unless one is looking at halftone reproductions of the B/W photographs. There the drawn line is filled with rugged edges, ink blots, and smudges. When the two sets of photographs are put side by side and compared, the difference is striking: it looks like most of the "signs of forgery" disappear when the gaze is moved from the B/W photographs to the colour ones.

A line screen at a 45 degrees angle is a powerful tool. The best examples of the bunch are the various pictures of omicrons, like the one above, where the round omicron of the colour photograph transforms into a clear tilted square in the B/W photograph. Upon my questioning Viklund wrote to me that the technical questions behind the transformation are actually much more complex than the simple line screen limitation, including issues of resolution, the size of the dots, the pattern of the dots and the intelligence of their distribution, even the quality of the paper and the relative humidity of the space where the printing took place. But the end result is clear enough: there are vast differences between the two sets of photographs, differences that seem to question the basis of the handwriting analysis Carlson made.

It will be interesting to see what conclusions the two handwriting experts, Agamemnon Tselikas and Venetia Anastasopoulou that were hired by Biblical Archaeology Review, will offer. Personally, I have begun to feel that anything less than the original MS will be insufficient for establishing anything about the handwriting. I have cited QDE expert Hannah McFarland of the dangers of not using originals (a similar sentiment is shared by the source Brown uses in his Expository Times reply, p. 145 n. 4). Recently, David Henige has offered a similar dire warning in "Historical Evidence and Argument" (University of Wisconsin Press 2005, 198-199, 271-272 n. 56-57). If the BAR experts arrive at some firm conclusion, either for or against the authenticity, I will be a bit suspicious as well as surprised. Instead of debating the undebatable, we should triple our efforts at locating the missing MS.

"Tremors, or Just an Optical Illusion? A Further Evaluation of Carlson’s Handwriting Analysis" does not answer the question I began with. We do not know why Stephen Carlson chose to use the B/W photographs instead of the colour ones. Viklund has, however, presented a strong case for why Carlson should have left the halftone reproductions with line screen at a 45 degrees angle out of his hoax hypothesis entirely.


  1. Those are very powerful arguments Roger but I for one am not fooled by logic. All you need to do is just look at the guy. He's all skinny, weird and goofy looking. I can tell a forger just by looking at him.

    I don't care about all your fancy tricks, Morton Smith JUST LOOKS like a forger. That's what my 'gut' tells me.

    Besides he's got 'criminal eyes.' That's why he wore those glasses. I hear he could see perfectly well but he was just trying to hide those creepy eyes from everyone. It would give him his 'criminal intent.'

    And you know what else? I've never seen a picture of him in shorts. My mother always said that you can tell if someone is a criminal if they don't wear shorts.

  2. A few samples from some recently published books:

    “In fact, recent analysis reveal the typical ‘forger’s tremor’, throughout the document, as well as well as well as characteristics of Morton Smith’s Greek handwriting style, convincing many that Smith himself was the forger.” (Craig S. Keener, “The Historical Jesus of the Gospels: Jesus in Historical Context”, 2009, p. 60)

    “Indeed, some scholars have suggested that Secret Mark may help us nuance the solution of the Synoptic problem, and, of course, some scholars have suggested that Secret Mark is older and more original than public Mark. The sad thing is that all this labor has been misspent; the Clementine letter and the quotations of Secret Mark embedded within it constitute a modern hoax, and Morton Smith almost certainly is the hoaxer. … Magnification of the handwritten text reveals the telltale presence of what handwriting experts call the “forger’s tremor”. That is, the handwriting in question is not really written; it is drawn, in the forger’s attempt to imitate a style of writing not his own. These telltale signs are everywhere present in the alleged Clementine letter.” (William Lane Craig, “Contending with Christianity's Critics: Answering New Atheists and Other Objectors, 2007, p. 143-4)

    “S. C. Carlson has recently written the most extensive examination of the authenticity of the letter of Clement and has produced compelling evidence supporting the suspicion that M. Smith created the text as part of a scholarly hoax. … Second, certain characteristics of the script of the Clementine letter suggest that it is a twentieth century forgery. (1) The script bears signs of unnatural hesitations in the penstrokes that are consistent with the forger’s tremor as well as pen lifts and signs of retouching that suggest that the characters were drawn rather than written. The characters also are inconsistently formed. (2) The script is dissimilar in several ways from that of other eighteenth century manuscripts produced at Mar Saba … (3) The script is identical to that of another Mar Saba manuscript, manuscript 22, which Smith himself claimed was produced by a twentieth century person whom he identified as M. Madiotes.” (Andreas J. Köstenberger, “The Cradle, the Cross, and the Crown: An Introduction to the New Testament”, 2009, p. 134)

    “Carlson, a well-regarded patent attorney and amateur biblical scholar, thoroughly investigated the case, bringing in handwriting experts, and writing The Gospel Hoax: Morton Smith’s Invention of Secret Mark in 2005.
    “What’s your opinion about the authenticity of the letter?” I asked.
    Evans’s answer was dramatic: “I think the clues really lead to the conclusion that the letter is a hoax and that Smith is almost certainly the hoaxer.” – – –
    When experts examined the magnified photos of the text, they could see what they call ‘forger’s tremor,’ where the text isn’t really written, but instead it’s being drawn by a forger in an attempt to deceive. There are shaky lines, pen-lifts in the middle of strokes—all kinds of indications that this was forged. Lee Strobel, interviewing Craig A. Evans in “The Case for the Real Jesus: A Journalist Investigates Current Attacks on the Identity of Christ”, 2007, p. 50)

  3. Too many "as well as well"; sorry ...

  4. Yeah well the handwriting looks to me like it was made with a guy with long, skinny fingers. AND MORTON SMITH HAD LONG SKINNY FINGERS. Wiggle your way out that Einstein ...

  5. Three years ago I wrote a large paper with the headline: "Weder gefälscht noch authentisch? Überlegungen zum Status des geheimen Markusevangeliums als Quelle des antiken Christentums". It will be published next year in a volume about apocryphal gospels in WUNT (Mohr Siebeck, Tübingen) together with the first volume of the new Hennecke-Schneemelcher. There I show that Carlson's "great story" against Smith the forger's tremor only a part of is nothing more than a story of an author playing Sherlock Holmes who read too much fantasy literatur. Therefore he is not able to distinguish between his fantasies and the reality of texts.

    Eckhard Rau, Hamburg /Marburg