As the author of a book in the burgeoning fields of digital archaeological reconstruction and virtual reality, and as a member of the University of Iowa Digital Humanities Cohort, I know the importance of transparency when it comes to representing archaeological data in digital media. While most find it boring (and while some critics claimed chapters 3 and 4 in my book, Qumran through (Real) Time, which detailed each tedious technological step of my digital reconstruction methodology would have been better left out), I find it essential to the credibility of the practice of digital reconstruction to demonstrate at each point in the process precisely how digital reconstructions are made and exactly how the data are handled and represented. When dealing with digital representations of archaeological data, it is essential for the establishment of a researcher’s credibility to document all alterations of digital imagery and data. For as a colleague of mine (who happened to disagree with part of my digital Qumran reconstruction) once memorably stated to me, “If you give the public a picture of your interpretation of the data, they’ll believe it!”
Images are powerful. And because they are powerful, archaeologists must take great care in representing visual data properly in publication. For just as well-handled visual data can greatly inform the reader and provide new insights into archaeological research, so too can mishandled, or worse yet, deliberately manipulated visual images distort reality. Unfortunately, deliberately altered visual data have been used to support sensational claims throughout history like Bigfoot and the Loch Ness Monster, and this abuse has only worsened with the rise of digital photography and editing. It is now possible for unscrupulous individuals to manipulate photographs for the purposes of supporting sensational claims in the hopes that said claims can be published in print and on television to generate tremendous revenues for those making the claims.
Unfortunately, digital technology is often times also used by some to distract from a weak argument. That is, some will elaborate upon and highlight the use of technology in the hope that the mere presence of sophisticated technology, which may have no bearing whatsoever on the interpretation of an object in under examination, will distract from any subsequent fallacious claims being made about said object. And, in more disquieting cases, digital technology is sometimes used to “enhance” or even fabricate evidence outright that supports an otherwise untenable claim.
Or, to put it another way: if the evidence doesn’t fit, Photoshop it (especially if it looks fishy).
Unfortunately, there is reason to believe that an image that has been circulating in the press as part of the marketing campaign in support of the new book by filmmaker Simcha Jacobovici and University of North Carolina, Charlotte Professor, Dr. James Tabor, entitled, The Jesus Discovery, and Mr. Jacobovici’s forthcoming documentary, has been digitally manipulated in such a way as to lead the reader toward a desired conclusion. That is, the image making the rounds in the press and published by Dr. Tabor on the Bible and Interpretation website has been digitally altered and made to look like an engraving of a “great fish” on an ossuary discovered in Jerusalem, in order to support the authors’ rather sensational claim.
What’s more, in Dr. Tabor’s article, the manipulation of the photo of the “fish” on the ossuary has been made without any acknowledgment that the image has been manipulated.
In his recent publication in Bible and Interpretation entitled, “A Preliminary Report of an Exploration of a Sealed 1st Century Tomb in East Talpiot, Jerusalem,” Dr. James Tabor describes “Ossuary 6” (the “Jonah Ossuary”) on page 20, and references two images: Figs. 20 and 21.
Fig. 20 is pictured on page 41 along with a caption that reads: “20. Jonah image on front façade of ossuary 6.”
Fig. 21 is pictured in page 42 along with a caption that reads: “21. Blowup of Jonah image.”
Nowhere in the text of Dr. Tabor’s article or in the captions beneath the images is there any acknowledgment whatsoever that Fig. 21 has been altered other than being “blown up” or enlarged. This differs slightly from the caption of the same image in the Jesus Discovery book, where the caption for Fig 26 on p. 86 reads: “A composite representation of the ossuary image of Jonah and the big fish” (italics mine). On March 1, 2012, the team’s “Jesus Discovery” media website labeled the image as a “CGI enhanced image of ‘Jonah and the Whale’.”
However, whether the image in question is a “blown up” or “composite” image, as soon as one looks at original photograph and the composite/blown up image side-by-side, one immediately notices that Fig. 21 is no simple enlargement of Fig. 20, but rather a highly-doctored digital artist’s representation of Fig. 20.
Let us examine the examples of digital image manipulation.
1. Fig. 21 in Dr. Tabor’s article has been enlarged, rotated, and cropped. Simple changes to scale that retain fundamental aspects of digital data, such as shape, color, and features are generally not considered compromises of the digital image.
However in cases where alterations of scale and orientation are made to images, it is common practice to supply a centimeter measuring stick or relative scale (see above) to convey relative size, and a compass pointing north (see above) to indicate the orientation (especially for rotated images).
These size and direction indicators can even be added to hard to reach areas (like those accessible only through robotic arms) and digital reconstructions after the fact (see above). Both of these are absent in both of Tabor’s Figs. 20 and 21.
As the classic optical illusion of the frog and the horse or the fish and the girl (note that this fish actually has an eye) demonstrates, rotation and orientation make a huge difference when identifying an object. The psychological process of “cognitive priming” can be used to lead the brain to interpret certain objects in a desired manner. Michael Shermer’s book, The Believing Brain, examines this process in detail.
It is quite telling that it was this digitally altered photo, Fig. 21, that was first sent to the press by Jacobovici and Tabor, and that the image was rotated to the side in most press reports. That is, the absence of any indication of proper orientation on the photograph allowed Jacobovici and Tabor to depict the image on its side, that is, in a manner more consistent with the natural orientation of a fish, rather than in its proper orientation with the tapered end down, which would more resemble some kind of ceramic or glass vessel.
For instance, in the Yahoo News story by Eric Pfeiffer and the MSNBC Cosmic Log story by Alan Boyle on Feb. 28, 2012, the doctored image appears rotated onto its side with no indication in the caption or in the story that it has been digitally altered. Similarly, the Photoshopped image depicted in the Haaretz.com story by Nir Hasson on Feb. 28, 2012 describes the image as an “enhanced image” without noting that the image is out of context and rotated to better resemble a fish. Likewise, in the LiveScience story by Wynne Parry and the syndicated FoxNews.com story on Feb. 28, 2012, both the digitally altered photo and the photograph are included among the images. And while the doctored image under examination has a caption that reads, “A CGI-enhanced image” (italics mine), both the Photoshopped image and the photograph are rotated to resemble the natural orientation of fish, and no indication is given for either image that they have been rotated from their in situ orientation.
Of course, Jacobovici and Tabor can “blame the press” for rotating the images, but they wouldn’t have had the problem had the orientation indicators and size and scale indicators been digitally placed on the images in the first place. Likewise, the fact that their own YouTube promotional videos also have the image turned on its side while they prime the brain with discussions about a fish (see the 1:35 mark) suggest that the image rotations were not the product of the press, but were done by Mr. Jacobovici and Dr. Tabor from the outset.
It therefore appears that the repeated rotation of the image to its side and away from its proper orientation is not the result of trying to fit it on a page for publication, but may be a deliberate attempt to orient the image in such a way so that it better resembles the natural disposition of a fish. The image Jacobovici and Tabor released to the press appears to have been intentionally disoriented.
2. One then notices that Fig. 21 is a different color than Fig. 20. It appears more brown or sand colored, possessing much less green than Fig. 20. While this may be an attempt by the authors to make Fig. 21 look more like other known images engraved in limestone ossuaries, and while this is, in fact, an alteration of the digital data, we can excuse this acceptable manipulation of the image as a simple color correction from the original photo (Fig. 20). That said, many archaeologists do not like the practice of “color correcting,” as maintaining color data is the purpose of employing Munsell color charts in archaeological recording and photography. Such color information is lacking from the article and the figures.
Thus far, the image has been enlarged, rotated, cropped, and color corrected. Unfortunately, Fig. 21 appears to have undergone an additional digital perspective manipulation to correct the oblique angle of the camera, which apparently was not perpendicular to the image when the photograph was taken. This is a more serious digital manipulation of the image, as it involves correcting the lengths and angles of objects to make them appear as if they were photographed at a perpendicular angle to the object (straight on). This technique is also used to correct so-called “barreling” and “fish-eye” lens effects that result from certain cameras and angles, especially during close-up shots. This more severe correction to the image in Fig. 21 can be excused as a digital correction of a poor initial camera angle, but it is customary to acknowledge that the image has been altered to correct for perspective. This is nowhere indicated in Dr. Tabor’s article.
Unfortunately, the evidence of image manipulation with regard to Fig. 21 (the image that has been sent to the press) far exceeds simple rectification of orientation, size, color, and perspective. The evidence below details a number of digital alterations to Fig. 21 that simply cannot be described as a “Blowup of Jonah image.”
3. An object covering the right side of the supposed “tail fin” (marked as “Digitally Removed” in the upper right corner of Fig. 20 above) is present in Fig. 20, but suddenly absent in Tabor’s Fig. 21. On p. 83 of the Jesus Discovery book, this object is identified as another ossuary (#5) that is “jammed up against it so closely we were unable to see its full decorated façade.” In Fig. 21 above, Ossuary #5 been digitally removed and the right portion of the “tail fin” has been digitally generated using a Photoshop process called “clone stamping.” This is evidenced by the fact that it appears darker than the rest of the “fin.” Likewise, the dark shadow that appears down the right side of the “tail fin” in Fig. 21 may be explained as the unintended result of the process of cloning and creating that portion of the “fin,” as there is a dark spot present in Fig. 20 at the intersection of the right side of the image and Ossuary #5. The shadow is the result of cloning that dark spot up along right side of the “tail fin.”
4. Perhaps one the most egregious alterations to Fig. 21 appears in the so-called “tail.” The shape of the “tail” is altered to make it look more like the tail of a fish. The horizontal top of the “tail” in Fig. 20 is straight, but the corresponding line in Fig. 21 shows a tremendous bend on the right side. This is a deliberate result of the cloning process that produced the right side of the “tail fin” after Ossuary #5 covering part of the image in Fig. 20 was digitally removed from Fig. 21. The creation of this portion of the “tail” appears to have been deliberately drawn at an angle to further the illusion of a natural fish’s tail. Likewise, the left side of the “tail” in Fig. 21 also appears to have been gently rounded at the top, and then curved toward the bottom so that it better resembles a natural fish’s tail, while the corresponding area in Fig. 20 reveals a nearly L-shaped angle.
It is worth noting that the reproduced image on the “museum quality replica” exhibited at Jacobovici and Tabor’s Feb. 28, 2012 press conference in New York does not match Tabor’s Fig. 21 image above, which was the image released to the press. The image is in its proper orientation, and not on its side. The artist(s) who engraved the replica more faithfully followed the original photograph in Fig. 20 above than did the doctored “composite” photo that was distributed to the press, as the replica more accurately reconstructs the top of the image.
5. The zig-zag triangle and braided border design visible in Fig. 20 to the left side of the image suddenly disappears from Fig. 21. The full size image of Fig. 20 (above) shows that the supposed “fish” is surrounded by a double border consisting of a line zig-zagging in between two parallel lines forming a column of triangles, that sits inside of a beautiful braided or herringbone design, all of which passes just outside the left “fin” of the “fish” in Fig. 20. However, this design element, which appears to be engraved equally as deep as the “fish” in Fig. 20, suddenly disappears in Fig. 21, despite the fact that many of the lines in the double border design appear at many of the same angles present in the “fish” design. Given the lighting, many of the lines comprising the border should appear along with corresponding lines making up the “fish” image, but are peculiarly absent. The border design appears to have been Photoshopped out of Fig. 21 using a combination of clone stamping, feathering, and use of the “healing brush” tool, as the resulting brown texture remaining after the deletion of the border appears indicatively blurry and feathered.
Again, the motivation behind the deletion of the border may be understood as an attempt to remove the image further from its actual context. Combined with presenting the image on its side, eliminating the border gives the viewer the illusion that the “fish” is swimming freely in the ocean rather than bound and framed narrowly by an ornate border, which, along with its proper orientation, certainly detracts from its interpretation as a fish.
6. A segment of the “fish’s abdomen” near the supposed “tail” (that I have labeled Seg4 above) suddenly appears clearly in Fig. 21, but is barely recognizable if not completely absent in Fig. 20. Changes in lighting cannot account for this change, as the three segments (Seg1, Seg2, and Seg3 above) are visible in Fig. 20. Even if we allow for the presence of Seg4 in Fig. 20, it still does not match with the altered angle of the tail in Fig. 21.
The addition or “enhancement” of the Seg4 layer may explain the more tapered, narrowed look of the “abdomen” of the “fish” in Fig. 21, perhaps leading the viewer away from interpretations of Fig. 20 as a nephesh monument or some other architectural structure, and encouraging an interpretation as a fish.
7. Finally, there are marks around the outside of Fig. 21 that betray the telltale signs of digital alteration, specifically, digital cloning. In a number of boxes above, I highlight examples of digitally produced marks that are identical in shape and size. (They do differ in color/tint, as this effect can be applied after the cloning process is complete.) Each mark of a mechanically engraved ossuary is unique in reality. At the pixel level of Fig. 21, however, the attempts at reproducing artificial “engraver’s marks” that I have isolated demonstrate, I believe convincingly, that someone used a Photoshop “clone stamp” tool to add artificial scratches all around the “fish,” and give the illusion of a naturally engraved image.
Take for example the white boxes numbered 1-4 above. I have cut-and-pasted (a process similar to Photoshop cloning) these examples in the chart to the right. As you can see, the marks bear the telltale signs of being copied time and again around the outside of Fig. 21. Each cloned “engraver’s mark” is comprised of a main line attempting to represent an engraved gash. However, identical marks labeled as “low lines” appear just below each of the “main lines.” Likewise, a light spot appears above the left end of each of the “main lines.” Because such identical markings at the pixel level would never appear naturally on an ossuary, anyone familiar with Adobe Photoshop and digital imagery can attest that these are obviously the product of someone who has cloned gashes and attempts at “engraver’s marks” around the edge of the supposed “fish” in an attempt to make the altered image look more natural.
Likewise, if we examine the yellow boxes above, I have highlighted additional examples of identical, digitally cloned marks that were added in an attempt to disguise the fact that cloning had been done to the image. By altering the shape of the “engraver’s marks” added to the digital image, it was hoped by the digital artist that the additional variety of artificially produced “engraver’s marks” would conceal the artificial marks, making the evidence of cloning less noticeable.
Yellow boxes 5 and 6 above demonstrate a variation of the “engraver’s marks” digitally added to the image. Likewise, yellow boxes 7-9 show yet another deliberate variation. I have enlarged the identical marks in boxes 7, 8, and 9 in the chart to the right. In each example, there is a central horizontal “main line” curving upward to the left, a small notch just below the center of the main line, and a curved, almost vertical mark just above each “main line.” Again, these examples reveal definite evidence of digital manipulation to Fig. 21 in Dr. Tabor’s article.
Tabor and Jacobovici’s Response
The scholarly community has been nearly unanimous in their rejection of Jacobovici and Tabor’s claims. The American Schools of Oriental Research (ASOR) has posted on its blog a series of posts by reputable scholars critiquing the authors’ claims. And while these rebuttals vary in substance and style, from critiques of the inscription to critiques of the authors’ use of the Bible, one of the most intriguing rebuttals has been from a host of scholars critiquing the above altered image in question above, perhaps without the knowledge that it had been digitally altered. Specifically, perhaps based partially upon the demonstrable Photoshopped Fig. 21 above, several scholars (including Dr. Christopher Rollston, Dr. Eric Meyers, Dr. Jodi Magness, Dr. Stephen Fine, and myself) have argued that the image would be better interpreted as a nephesh monument.
However, I cannot help but wonder how many of these expert opinions may have been based upon the digitally altered and deliberately disoriented image described above. It now appears possible that at least some of the scholars interpreting the image in question as a nephesh monument may have been basing some of their arguments on a digitally altered image, removed from context, and rotated away from its original orientation. The interpretation as a nephesh monument may still be a possibility. But alternatively, given a knowledge of the image’s proper orientation, it may also be interpreted as a representation of an amphora as suggested by Italian scholar Antonio Lombatti, or some other kind of vessel like a krater as recently suggested by Warden and President of Trinity College at the University of Melbourne, Dr. Andrew McGowan, or perhaps even an unguentarium, as suggested by Kings College London Professor of Theology and Religious Studies, Dr. Joan E. Taylor.
The one theme shared by scholars interpreting the image both as a nephesh monument and as some kind of ceramic or glass vessel is this: they all agree it’s not a fish.
But scholars can only evaluate the claims and evidence that authors publish, and proper context and the integrity of the image data are essential to one’s interpretation. The caption under Fig. 21 from page 42 of the original version of Dr. Tabor’s Bible and Interpretation article, “A Preliminary Report of an Exploration of a Sealed 1st Century Tomb in East Talpiot, Jerusalem,” published on Feb 28, 2012 simply read: “21. Blowup of Jonah image.” Except, it obviously was not.
As I have shared the above evidence of image manipulation with my scholarly colleagues, including an exchange with Dr. Tabor on the ASOR Blog, it now appears that Dr. Tabor and Mr. Jacobovici are attempting to take steps to correct their descriptions about the misleading photo in question (Fig. 21 above), which has, in fact, been removed from its context, rotated away from its in situ orientation, and digitally altered, by updating the captions describing these images, and in some cases, correcting their orientation after the fact.
For instance, the editors at Bible and Interpretation confirmed to me that they have, in fact, taken down Dr. Tabor’s original article (which I have reposted here for purposes of comparison), and replaced it with a new, revised version. They also updated the title of the revised version with an editorial comment noting that Dr. Tabor’s present article is a different version from the one originally published.
This revision appeared on the Bible and Interpretation website on March 2, 2012, after Dr. Tabor and I had discussed the Photoshopped and rotated image on the ASOR blog. To his credit, Dr. Tabor confirmed that he uploaded a revised version of his article at Bible and Interpretation, noting on his Tabor Blog:
By the way, if you are one of the 18,000 that have downloaded that article in the last two days take a look again at the picture in the appendix–we have added the museum reproductions of the ossuaries.
However, Dr. Tabor makes no mention in his revised article of the fact that he has also corrected the orientation of the supposed “fish” in the image in Fig. 21 by rotating it 90-degrees clockwise into its actual in situ orientation. In the original version of the article, the image was horizontal, with the “head” of the “fish” pointing to the right, perhaps in an attempt to better resemble the natural disposition of a fish. Likewise, the caption for Fig. 21 has been changed from the original, “21. Blowup of Jonah image” to now reading, “21. Museum replica showing placement of image on front panel and closeup of image.”
The problem is, if one examines the two images in Fig. 21 of the new article closely, the images still don’t match! The bottom image is not a “close up” of the top image at all. Rather, the top image is an artist’s reconstruction on a replica, and the bottom image is a highly Photoshopped image. Look closely at the so-called “tails”: the bottom image has a bent “tail” on the top left (as the digital reproduction from the revised Bible and Interpretation article has now been flipped horizontally), while the image on the “tail” on the replica is flat and straight.
The caption of the same image in their Jesus Discovery book beneath Fig. 26 on p. 86 reads: “A composite representation of the ossuary image of Jonah and the big fish.” Again, there is no mention of the multiple digital alterations that the image has experienced, just the words “composite representation.”
Finally, the caption under the same image on the thejesusdiscovery.org website finally concedes it is a “computer enhanced” photo. Unfortunately, these photos were not made available on the website until after the book had been released, after the press conferences, after scholars had begun their initial critiques, and after I had begun to point out that the image in question (Tabor’s Fig. 21) had been digitally altered as described above.
The question is: which pictures should we believe? Should we accept the artist’s rendition on the “museum quality replica,” or the doctored, heavily Photoshopped, “computer enhanced” digital “composite representation”?
And this is the point: the image that was released to the public was a Photoshopped image. It was rotated. It had been altered in other ways, including having the “tail” reshaped to more closely resemble that of a fish. Yet, these were the images given to scholars to evaluate. And that’s what we evaluated.
I cannot speak for other scholars, so I shall only speak for myself: I admit that my original suggestion of some similarity between the image in question and Absalom’s Tomb stemmed from my analysis of the photo sent to the press (Fig. 21 above), which I have now demonstrated to be a doctored photo that was rotated to more resemble the natural disposition of a fish, and which lacked any indication of size, scale, or orientation like a cm stick or compass point. Once I realized that I had based my analysis upon a doctored photo, I publicly conceded that I no longer favor the interpretation as a nephesh memorial, but instead favor an interpretation of an amphora, krater, unguentarium, or some other form of vessel.
I also noted that as scholars, we must be willing to alter our conclusions based upon new (or in this case, accurate, in context) evidence following a consensus of our trained peers. The question is: are Mr. Jacobovici and Dr. Tabor willing to do the same? Now that they’ve had the opportunity to experience the peer-review of trained professionals, will they heed the nearly unanimous voice of the archaeologists and scholars stating that the image under investigation is not a fish?
Unfortunately, the visual evidence detailed above compels us to conclude that Fig. 21 from pg. 42 of Dr. James Tabor’s original Feb 28, 2012 Bible and Interpretation article entitled, “A Preliminary Report of an Exploration of a Sealed 1st Century Tomb in East Talpiot, Jerusalem,” has experienced a high degree of digital manipulation. Given the changes to the “tail fin” of the supposed “fish,” and given the deliberate rotation of the image’s orientation causing it to more resemble the natural orientation of a fish without offering a compass point or any indication on the image whatsoever that the image has been rotated, it can be argued that the motivation behind making these digital alterations to the image was the desire to create, or at least “enhance” the illusion of a “great fish” swimming freely in the ocean, while vomiting forth a human head.
We should not state that the image has been “faked,” as there is obviously an image on the ossuary. However, we are forced to conclude that the image was digitally manipulated and its orientation altered in such a way so as to encourage and enhance its interpretation as a fish over other possible interpretations. The fact that Dr. Tabor is still using the doctored photo as “evidence” upon which to base his recent rebuttals of other scholars’ critiques of his theory on his own jamestabor.com blog and in a new Bible and Interpretation article is quite telling.
What is more troubling is the prospect that other images published by Mr. Jacobovici and Dr. Tabor may be similarly digitally “enhanced” without proper acknowledgment. If such image manipulation is demonstrated in the “Jonah fish” image, which is central to their sensational and already highly spurious claim, how can we be sure that other images, such as those of the inscriptions, have not experienced similar amounts of digital alteration? Let us remember that the image distributed to the press and on the “museum quality replica” are, in fact, artist’s renditions of the image on the ossuary and not the image itself – a rendition that the authors desperately want viewers to interpret as a fish.
With the credibility of the visual evidence demonstrated above now highly suspect, and with the scholarly consensus nearly unanimously interpreting the image as something other than a fish, we should be all the more skeptical of any and all claims made by Mr. Jacobovici and Dr. Tabor regarding any claim of Jonah, a fish, or so-called “new evidence” of early Christianity obtained from these tombs.
Because if it doesn’t look like a fish, and doesn’t swim like a fish, it may very well be an ancient vessel cleverly Photoshopped to look like a fish.
Filed under: archaeology, christianity, digital humanities, judaism, robert cargill | Tagged: Adobe, Andrew McGowan, antonio lombatti, asor, barreling, bible and interpretation, christopher rollston, cloning, digital, digital humanities, eric meyers, fish, fish-eye, handles, healing brush, illusion, image, James Tabor, Jerusalem, Jesus Discovery, Joan E. Taylor, jodi magness, jonah, manipulation, optical, ossuary, Patio Tomb, Photoshop, Qumran through (Real) Time, rubber stamp, simcha jacobovici, Stephen Fine, Talpiot Tomb, Talpiyot, The Jesus Discovery, The Resurrection Tomb Mystery, whale |