Bedrock Evidence in the Kennedy Assassination

by Josiah Thompson

Part 1    Part 2    Part 3

Note: A somewhat shorter version of this paper was delivered in Bethesda, Maryland on November 19, 2005 at a conference sponsored jointly by the Assassination Archives and Research Center and the Cyril H. Wecht Institute of Forensic Science and Law.

If the history of human inquiry tells us anything, it tells us there is no natural and necessary progression for research in any field. Inquiry ... historical, scientific, theoretical, practical ... proceeds in fits and starts, sometimes veering far off course before discovering a new and important line of advance. Sometimes, when inquiry veers off line, it is simply a waste of time and energy. At other times, however, an apparently fruitless and frivolous line of inquiry proves to be the most important. My talk today illustrates this point. It illustrates the importance of unexpected consequences in the study of the Kennedy assassination.

Over the last decade, the assassination research community has been riven by a raucous, high-decibel argument. The claim has been made by Harrison Livingstone, James Fetzer, Jack White, John Costella, David Mantik, David Lifton and others that the Zapruder film and other films/photos taken of the assassination have been altered. The alteration claimed is gigantic in scope although the particular films/photos purported to be fake vary from week-to-week. In spite of an enormous effort spread over a decade, the exponents of alteration have failed to produce a single persuasive instance of fakery.

One way of looking at this continuing argument is to see it as a gigantic waste of time, as a prime example of junk science from educated people who ought to know better. It may have amusement value in some chronicle of "silly science," but, in terms of knowledge about the Kennedy assassination, it has produced literally nothing. If I took that view of what has happened, I certainly would not be making this "dry hole," so to speak, the subject of my talk to you today. That view would motivate passing the silliness by in silence, leaving it, without comment, to molder in some mausoleum of bankrupt claims.

However, I don't take this view. I believe that this decade-long failed attempt has great significance for future inquiry into the Kennedy assassination. The failure of this attempt to impeach the authenticity of the Zapruder and other films permits us to focus on these films and photos as a kind of evidentiary "bedrock." I mean to argue today that the alteration argument was not a worthless detour but rather an important step along the way towards distinguishing a logically prior kind of evidence in the case. Most simply, I mean to argue that the films and photos taken in Dealey Plaza form a self-authenticating whole which can be used to judge other evidence. In this way, this evidence can become a foundational structure upon which a reconstruction of the crime can be built.

But first we have to come to some understanding of the failure of the alteration argument. Debunking of the alteration argument has been carried out at various conferences and on a number of web sites over a number of years. As soon as one alleged "anomaly" showing alteration is explained, others are put forward. Not a single one has survived inspection and criticism. For our purposes, I am going to choose a few to illustrate the nature of claims made and how they've been debunked. I would point out that I haven't cherry-picked the claims to select "sitting ducks" as illustrations. The claims selected are those that continue to draw support from a dwindling number of enthusiasts. And let me warn you in advance, many of the claims of film fakery are couched in obscure technical language. It takes patience and extreme attention to detail first to understand the argument and then to grasp its weakness.

* * * * * * * * * * * * * *

In December 1998, Martin Shackelford began a brief article entitled, "A Brief Overview of the Zapruder Film Alteration Argument" by stating:

In the early 1990s, researchers whose theories weren't fitting well with the available evidence began to argue that the evidence that didn't fit was faked. Unfortunately, my colleague, Harrison Livingstone, opened the floodgates with a concept he called "the smorgasbord of evidence," suggesting there were two of everything and that the Zapruder film had been altered. It seems that two people who had worked with him picked up on the idea: Dr. David Mantik and Professor James Fetzer.

It is now six and one-half years later. It is fitting that Harrison Livingstone, who was there at the beginning, has just produced a book which marks the reductio ad absurdum of the whole argument. The book, The Hoax of the Century: Decoding the Forgery of the Zapruder Film, is a fat volume, 495 pages long with 29 pages of badly reproduced color plates and two pages of black-and-white. It is privately printed by a firm in Victoria, B.C. and costs $80.00.

In his book, Livingstone intends to lay out an independent attack on the authenticity of the Zapruder film by arguing with the conclusions of Roland Zavada, the retired Kodak expert who studied the Zapruder film for the ARRB. Here Livingstone has a problem concerning his own credibility.

Roland Zavada has a towering reputation in the field and no conceivable reason for cooking his conclusions. Forty some years ago, Zavada was part of the Kodak team that developed the very film Zapruder used in his camera on November 22nd. Over an eighteen-month period in the late 1990s, Zavada personally examined the Zapruder film and its first generation copies, interviewed the technicians who worked on the film on November 22nd and performed experiments with cameras of the same make and model as Zapruder's. Although not tasked specifically by the ARRB with authenticating the film, the results of his exhaustive examinations, interviews and experiments led Zavada to conclude that the original of the film and its first generation copies were genuine and unaltered.

Since the arguments are intensely hyper-technical, why should we believe Livingstone when he and Zavada come to contrary conclusions? In a curious "apology" at the beginning of his book, Livingstone admits that "a great deal of what I was dealing with often enough was over my head." However, even if many of the technical issues were over Livingstone's head, if he could show a glaring logical inconsistency in Zavada's approach or offer clear reasons why a conclusion does not follow from the observed evidence, he could have made a real contribution to our knowledge. Sadly, Livingstone does neither.

This may become clear if we attempt to follow Livingstone's most obvious claim — the point he showcases on the front cover of his book and emphasizes throughout the book. Livingstone prints three Zapruder frames in living color across the front of his book. Then he hangs a banner question in red across them: "Do you see first-frame over-exposure between Z-132 and Z-133?"

The arcane nature of the question speaks volumes as to how technical the arguments have become. Frame Z-132 is the last frame of a sequence of motorcade cyclists shot by Zapruder. He then stopped his camera so that he'd have sufficient film and wind-up spring left to photograph the limousine. Frame Z-133 is the first frame of the next sequence started when Zapruder saw the limousine and began to photograph its glide down Elm Street. When the camera is stopped and started, the first frame after the restart often shows some over-exposure. This overexposure can come either from light leaking into the first frame or from startup inertia in the camera mechanism.

Livingstone's claim is that the transition between Z-132 and Z-133 shows no such over-exposure and that therefore the camera was not stopped at all. Rather, says Livingstone, a number of frames were simply excised.

What does Zavada have to say about this? I called him on the phone and asked.

Zavada points out that he was aware of the challenge from Livingstone and did some further research in 2004. His research using Photoshop histograms for luminosity showed there were two examples when the camera was stopped and started which showed significant overexposure on the first frame after restart. These two restart frames showed the characteristic overexposure due either to light leaking into the camera or mechanical inertia at startup. However, other instances of stopping/starting appeared to show no significant overexposure on the first frame after restart. These instances, however, did show a tell-tale 10% decrease in exposure as one moved from the first frame after restart to the second. One instance is the transition from Z-133 to Z-134. Other examples are found in the Zapruder home movie part of the film: (1) the stop/start between a toddler at various zoom angles and a toddler walking in the grass; (2) the stop/start between a toddler outlined against grass and shadow and a young boy standing by a lawn chair. These sequences are reproduced onpages 136 through 138 of Livingstone's book. To the naked eye, both the stop/start between Z132 and Z133 and the other stop/start sequences in the backyard portion of the film show no overexposure.

The counter-examples to Livingstone's thesis can be found reproduced in his own book as well as in segments from Zavada's reported research. Both are included on pages 136 through 138 in a topsy-turvy, fifty-page section of the book which is virtually unreadable. If one remains focused on the central argument throughout and works through this maze, an appalling fact becomes apparent. Although Livingstone never admits it and spares no effort to obfuscate the point, his own book reproduces clear stop/start sequences from Zapruder's film which show the same characteristics as Z-132/Z-133.

Only because I had already talked to Roland Zavada could I penetrate through the mass of verbiage to discern the shell game that was being played. Awash in a flood of technical terms and details, the reader is swept along by Livingstone's rhetoric. When analyzed, however, it becomes readily apparent that there is nothing unusual or amiss about the stop/start sequence at Z-132/Z-133. It looks exactly like two other stop/start sequences on the same film.

There is however something very much amiss about the captions Livingstone applies to the illustrations above. They are shown below:

"What's going on here?" I asked myself. Is Livingstone claiming that the backyard Zapruder frames he publishes in his book really show the opposite of what they show? I reached him by email to ask what the captions meant. In between him (1) calling me several names, (2) threatening to sue me if I published anything negative about him or his book and (3) wishing me an early heart attack, he replied as follows:

"Each time the camera stopped and started during the backyard sequence, it was quite apparent that there was first frame over-exposure (we were looking at the alleged original S.S. copies). I'll stake my life on it. I had police officers with me, and they tried to photograph it but the pits did not turn out. I also had two other persons in my team in the NARA lab." (Email from Livingstone dated 4/17/05)

Since it's difficult to believe anyone would claim this, I went further. I checked with people who worked with Livingstone on the book. "Harry said that when he saw the Z film unrolled on a light table at the National Archives," I was told, "there were flash frames in that particular backyard sequence. Now, they aren't there." This was supposed to have happened in the mid-1990s.

In short, Livingstone publishes in his book Zapruder frames which destroy the basic claim he's making. When this is pointed out to him, he claims these very frames have been altered at the National Archives to show the opposite of what they showed ten years ago.

Meanwhile, the solution to the alleged mystery of first frame over-exposure is quite simple. Most often, it is caused by the inertia of the camera as it starts up. This leads to the first frame in a sequence picking up more light than later frames. In cases where the camera mechanism has been idle for some time, the first frame over-exposure is quite obvious. In cases, where the mechanism has been idle for only a few seconds, the over-exposure is minimal. As with the short time gap between Z-132 and Z-133, the backyard sequences shown below indicate that the camera has only been stopped for a few seconds or minutes.

The failures, prejudicial omissions and outright misstatements of fact in this book are legion. There are interminable discussions of the odd septum line on Zapruder film copies but no mention of the critical fact that Zavada found a septum line on a film copied earlier on the same Jamieson "J" printer which copied the Zapruder film. In a discussion of the "National Photographic Interpretation Center (NPIC)," there is no mention of the critical observation made by technician Ben Hunter that the film he and McMahon worked on at NPIC had no intersprocket images and thus had to be a copy. It is a book which seeks to persuade by overwhelming the reader with arcane detail. In the final analysis, however, it is parasitical upon the work of a genuine expert who came to diametrically opposed conclusions. For Roland Zavada, it must be like being pecked to death by geese!

Nor was this the first time that Zavada has been attacked by zealots. In James Fetzer's latest book, David Lifton mounted the claim that Zavada himself had discovered unambiguous evidence of Zapruder film fakery but would not admit it. Relying upon hearsay from ARRB staffer Doug Home, Lifton claimed that Zavada had experimented with cameras of the same make and model as Zapruder's but had been unable to produce images that matched Zapruder frames. What was at issue here was what Lifton called "full flush-left penetration of the image in the intersprocket area." Lifton claimed that Zavada had never been able to achieve this full flush-left penetration when using cameras of the same make and model as Zapruder's.

I asked Zavada about this claim. He said he could not understand why Lifton "needs to revert to hearsay to support his arguments" when he had produced and published in his report instances of "full flush-left penetration" that Lifton said he had been unable to achieve. Here are two instances of "full flush-left penetration" from Zavada's report:

In a later edition of Fetzer's book, Lifton dealt with Zavada's claim by publishing a muddy black and white illustration and then wrongly claiming in the text that "in none of the tests (shown here) could he [Zavada] replicate the continuous `full flush left' phenomenon seen on the previous two pages."

Let's compare the muddy illustration from Fetzer's book with the photo in Zavada's report from which it was copied. The photo from Zavada's report gives the lie to Lifton's claim. It shows the full flush-left interpenetration that Lifton said Zavada could not achieve with cameras similar to Zapruder's.

Zavada has been vehement from the beginning that his study of the film and its authenticity is independent of any content analysis. His examination and investigation is simply of the artifacts presented to him (the camera-original film and its three first generation copies) and their provenance. This was the limit of his investigation and his conclusions. Given these limits he is willing to conclude as follows:

There is no detectable evidence of manipulation or image alteration on the Zapruder in-camera-original and all supporting evidence precludes any forgery thereto.

The film that exists at NARA was received from Time/Life, has all the characteristics of an original film per my report... It has NO evidence of optical effects or matte work including granularity, edge effects or fringing, contrast buildup, etc.

In the world of paintings or antiques, authentication of artifacts is a job best left to experts. The same applies here. In the specialized realm of "questioned document" or "questioned photograph" examination, amateurs venture at their peril. Long before Lifton and Livingstone ventured forth, David Mantik offered his own mistaken theory concerning the Zapruder film as artifact. In an early article in Fetzer's volume, Assassination Science, he opined that the "ghost images" found at times between the sprocket holes signaled alteration of the original film. This thesis evaporated as soon as Anthony Marsh began circularizing snippets of eight millimeter film taken with a similar camera that showed similar ghost images. Later, Zavada showed how the "ghost images" were produced by a simple double-exposure of the primary image. Other amateurish efforts over the years that sought to undermine the authenticity of the film via technical criticism have met with similar fates.

Continue on to Part 2