A Scientist’s Verdict: The Film is a Fabrication

John P. Costella, Ph.D.

94 pages (pp. 145–238), incl. Appendix; 45 figures (84 photographs; 8 diagrams)

Introduction and motivation (2 pages)

I introduce myself to the reader, my background and interests, and explain how I got interested in research of the assassination, and the Zapruder film of it in particular.

The Gang’s response:

No dispute.

The MPI debacle (1 page)

I summarise MPI Teleproductions “reference digitisation” of the “camera original” Zapruder film in 1997, and the unbelievable number of elementary and inexcusable “mistakes” made in this important work.

The Gang’s response:

No dispute.

Researchers: genuine or not? (2 pages)

I describe my experiences when entering the “JFK research community”, and relate a number of incidents that give the reader some idea of environment that one is jumping into. In particular, the “Internet assistants” described on page 149 are none other than David Wimp and Joe Durnavich, the two principal “scientists” for The Gang. David Wimp’s refusal to accept simple arguments of physics and mathematics relating to pincushion distortion, as described on that page, should be kept in mind when we return to that issue below.

The Gang’s response:

No dispute.

Moorman in the Street (2 pages)

In this section I describe how I delved into the “Moorman in the Street” issue, how I made serious mistakes in my assumptions about some of the photographic materials I was using, and how I was considered a villain on both sides of the alteration debate after I conceded making these mistakes.

The Gang’s response:

Remarkably, The Gang include, as part of their “new” critique of The Great Zapruder Film Hoax, a link to their May 2002 website analysing the Moorman in the Street issue! They want to continue to debate this issue, even though I state explicitly in The Great Zapruder Film Hoax that I agree with their criticisms of my earlier analysis of this issue. Note carefully the differences between their May 2002 work on the Moorman issue, and their October 2003 work on the proofs provided in The Great Zapruder Film Hoax. For the former, they made use of the very methodology that I had developed, and showed explicitly why I had made unwarranted assumptions about the Zapruder pedestal region, which dramatically altered the results and conclusions of the analysis. For the latter, on the other hand, they have claimed (cf. postings to jfkresearch.com by James R. Gordon, a member of The Gang) that they have been unable to figure out the necessary mathematics, and as a consequence decided instead to apply an indirect method of analysis, by programming everything into a 3-d computer simulation. Carefully compare their two websites. The former represents serious research, and demonstrates that they are not as naïve and incapable as they claim. The latter is obviously the result of frustration at not being able to find fault with the methodologies employed.

The Zapruder Film Symposium, Duluth, Minnesota, May 2003 (11 pages)

In this section I provide an overview of the Zapruder Film Symposium held in May 2003, out of which the volume The Great Zapruder Film Hoax arose. This will be superseded by the DVD’s and videos of the Symposium itself, when they are released shortly, but I felt that many readers of the book may be interested in a brief overview of the Symposium, what it meant, and how it all ties together.

The Gang’s response:

No dispute.

The Stemmons Freeway sign (8 pages)

I provide a new analysis of the Stemmons Freeway sign that shows that it is internally inconsistent in the extant Zapruder film, demonstrating clearly and unequivocally that it was incorrectly inserted into the Zapruder film in the days following the assassination—and, once published in Life, was forever locked in place.

The Gang’s response:

The most valuable part of The Gang’s section on this issue is the identification of a typographic error. The frame numbers in The Great Zapruder Film Hoax are reversed in Figures 5 and 6, as is clear by comparing them with Figure 4. I thank them for picking this one up—it had slipped past me! The reason for the error is that page 167 was all over the place in the early proofs, with images in the wrong places and the frame number parts of the captions missing. We got the figures back into the right configurations, but the images got transposed. Of course, the error is typographic only: it has no bearing at all on the demonstration.

So let us turn to matters of substance. Before you read the following, you need to look below at the discussion of the Stemmons Freeway sign and the lamp pole, because The Gang’s section on that precedes the one we are discussing. After reading that section, return here.

The first warning flag is the admission that is made just before the start of this section:

Note that the Zapruder frames used here were corrected for aspect ratio and lens distortion by David Wimp.

The only problem is that David Wimp has no idea how the correct the pincushion distortion at all! This is actually documented in my chapter—he is the “internet assistant” described on page 149. (I made him anonymous to protect the guilty—but if he wants to now profess to tell me that I am wrong, I’m afraid I’m not going to surround him with cotton wool any longer.) Why did they not use the frames available on assassinationscience.com, that have had the distortion removed correctly?

But let us give Wimp the benefit of the doubt—he was a slow but plodding learner—and persevere with their section, to see what their procedure was:

Well, we added a right post to our model sign, assuming it to be perfectly parallel to the sign's right edge and rendered two images, matching the camera view for each frame. The camera positions differ by about 2.7 inches.

How wonderfully descriptive! They added a post to their animation engine: this part is fine. But then they simply tell us that the “match[ed] the camera view for each frame”. How? We’re back to their incredible scientific six-dimensional parametric fit, done by matching up a few edges until they “look about right”. (Cf. the section below.) Then they “rendered two images”. Sure—they can render two images. But how did they then overlay these images on the Wimp Zapruder frames? Another process of “yeah, looks about right”! And, surprise surprise, they find that the result is what they expected to find. Given that they were doing it “by eye”, what did they expect?!

We’re then shown a couple of images of their large rendered rectangles, superimposed on the edges of the sign. How wonderful. They managed to fiddle with the image so that the rectangle fit in the sign. My two young sons (ages 5 and 2) are doing something similar right at this very moment, on the Lion King Activity Centre CD-ROM game.

That The Gang put this drivel up against my careful and direct construction would be outright insulting, if I didn’t know how much they struggle with this sort of analysis. As it is, it is simply annoying that I have to take the time to explain why their fiddling is worse than useless.

My demonstration takes two frames of the film, removes the pincushion distortion, and then (for definiteness) transforms the camera axis directly to point in the same direction (although this last step actually makes a negligible difference; I perform it for completeness only). These steps were explained and demonstrated, slowly and carefully, in my Duluth Symposium—with fancy animated graphics to aid understanding by the audience—which I hope will appear on the DVD of my presentation when it is released shortly. The Gang would do well to study these steps. I am more than happy to supply them with the correct mathematical equations for removing distortion and changing the direction of view. They can then perform these direct transformations directly on the Zapruder frames themselves. They will not need to program a 3-d animation engine. They will not need to cut down the fidelity of the 3-d model so that the rendering can be “accelerated” (see section below) during their fiddling process. They will not need to ponder the mammoth task of mapping the results back into the imagery of the Zapruder film. And, best of all, they will not need to try to fathom how on earth they could perform an uncertainty calculation on this entire convoluted process—something they didn’t even think of in their first attempt. It’s simple, and direct.

I am also more than happy to explain how moving the camera around will change the results. Moving it sideways, or up and down, won’t help: that will shift the sign sideways, or up or down, but the images I present show that the bottom of the pole, at least, does not move sideways, and the top of the sign does not move up or down. (See how easy a verification is with the direct method? You can just look at the images with your own eyes. No sleight of hand, no incomprehensible “sticks” from some sort of graphics engine, no wondering how on earth they put the sticks onto the Zapruder film frames.) Moving the camera toward or away from the sign will changes its width and height (see the section below), but it won’t make it flip and flop. There are fundamental mathematical and physical explanations for why this must be so, which I am happy to explain to The Gang if they are still motivated. Finally, changing the three Euler angles of rotation of the camera doesn’t do anything at all, because we have shifted the optical axis of each frame to the same direction (that wipes out two angles), and we have simply rotated the images to match up the background precisely (which knocks out the final, third angle).

Which is probably more explanation than most of you readers of this page really wanted to know. That’s why I didn’t describe these things in The Great Zapruder Film Hoax—it’s not a physics textbook, nor an opportunity for me to show that I’m a physicist, but rather a book on the assassination of the 35th President of the United States. But rest assured that I have these explanations for you, should you want them.

And that offer goes for you too, Wimp. There’s hope for you yet.

Background magnification (1 page)

In the course of explaining the analyses of the Stemmons sign and the lamp pole, I show how the phenomenon of pincushion distortion, together with the panning of Zapruder’s camera, is responsible for previous researchers mistakenly believing that the background grassy area in the extant Zapruder film was enlarged. In fact, the background is exactly as would be expected for a genuine film.

The Gang’s response:

No dispute, of course!

Panoramic views of Dealey Plaza: July 2002, November 1963, and the Zapruder film (4 pages)

I describe how the mathematical tools used to analyse the Zapruder film were questioned by The Gang and their allies, and how the challenge was issued to create a seamless panoramic view of Dealey Plaza to demonstrate the mathematical and physical correctness of the tools employed. For the first time in the history in this case, such a seamless panorama was created, based on photographs taken by Jack White in July 2002. Comparisons with the topographic survey created for the House Select Committee on Assassinations confirmed that the panorama is accurate to around a tenth of one degree, throughout its almost 180-degree span. Similar panoramas based on photographs taken after the assassination, and from the frames of the Zapruder film itself, were also created. These answered a number of outstanding questions about the Zapruder film, including the issue of “background magnification” described above. Note carefully that these panoramas did not, in themselves, provide any proofs at all about the fabrication of the Zapruder film—to the contrary, they disproved the long-held beliefs about “background magnification”.

The Gang’s response:

No dispute, of course! To dispute these constructions would be to lose their only concrete proof against background magnification—something they have failed to provide, themselves, in the decade or more that this phenomenon has debated. But The Gang could not have it both ways. They cannot argue that the tools used to create these panoramas are flawed, because they are demonstrably correct, and indeed support their own theses about the background of the film. But, once they have accepted the correctness and appropriateness of the tools and methodologies employed, they must abide by this acceptance in their remaining attacks. We are thus left with the unbelievable argument that “they couldn’t figure out the mathematics”—but which they accept as being correct; and so they are forced to employ indirect constructions in their arguments.

The Stemmons Freeway sign, the lamp pole, and the panoramas (2 pages)

I now compare the panoramas, in which all 1963 features should, ideally, align identically. There are several obvious points of difference, and some not so obvious. The shift of the Stemmons Freeway sign sideways can be explained simply as the effect of “parallax”, if the camera is shifted sideways, and does not imply anything about the authenticity of the film. (I explain “parallax” with a simple example that you can perform with your own finger.) However, the incorrect size and angles of the edges of the sign are things that cannot be changed by parallax. These provide additional proof of the incorrectness of the Stemmons Freeway sign in the film, in addition to the internal inconsistencies separately proved above.

The Gang’s response:

In their page on this topic, The Gang curiously seize on parallax shifts as their “saviour”—even though I explicitly point out why I expected that there were parallax shifts, namely, because of the sideways movement of the sign and the lamp pole. You should carefully read my section of this chapter explaining parallax to understand what is occurring here. I explain clearly that the parallax shifts do not impugn the film, even though early researchers believed that they did. In this, The Gang are merely reflecting my own observations.

Even more curiously, The Gang seek to “prove” why there was a parallax shift, by providing photographs of Barnes on the pedestal—even though it was I who had insisted that such a shift was implied by the images themselves! Their technique is clever: they imply that I have denied that the two cameras were in different positions, and they are providing their “scoop” evidence that my (nonexistent) denial is wrong. Again, do not be deceived: read both accounts carefully, and note the points of agreement.

However, you can already start to understand the unravelling of their arguments when you consider their diagram showing where the Barnes and the Zapruder cameras were located. How did they determine these precise positions? They looked at the photographic evidence (itself in doubt), and made a guess! These locations were then fed into a 3-d computer engine. But have they programmed the uncertainties in these positions into the engine? Their presentation and their code suggests not.

David Mantik has recently summarised most concisely the inherent mistake in going down this path—in the context of the Dale Myers animations that the US ABC network are using to “prove that Oswald did it alone”:

[A] typical and very serious problem encountered by many computer modeling programs [is that] the errors in each step proliferate until the final error can become so large as to negate the entire process. Such errors are like snowballs from hell rolling downhill totally out of control.

Of course, it was on the very day that I started making similar comments that The Gang hastily released their website!

Saying that “the mathematics is too hard for us” is a foolish way for The Gang to proceed. If they are mathematically incompetent to perform the transformations of perspective directly on the photographic evidence, as I have done, then they are doubly incompetent to handle the calculation of uncertainties resulting from the compounding of individual uncertainties in a 3-d computer engine—it is an exponentially more difficult problem. Moreover, they have to then convince you, the reader, that they have some way of mapping the photographic evidence to their 3-d computer model! But if they can’t handle the mathematics that I have used to do this (in my panoramas, for instance), how can they make any connection back with the Zapruder film?

The simple answer is that they can’t. If they could, they wouldn’t have needed to program everything into a reconstruction engine.

Having said that, let me point out most clearly where The Gang have done something right. Yes, they did something right! (Crack out the champagne at The Sixth Floor Museum ….) And let me show you how they should have done it—directly. And clearly.

Their success is their prize section, “The Sign Width Discrepancy Explained”. What did they do right? They made the reasonable assumption that the Barnes position was taken from a position up to two feet further away from the Stemmons Freeway sign than the Zapruder film. Although the exact distance is uncertain, let us work on the basis that it was two feet, and then see what happens when we vary that distance. The Stemmons Freeway sign was something like 55 feet away from the pedestal, approximately. The size of objects for an ideal imaging system varies as the inverse of the distance that they are away. Moving two feet further away from the sign will therefore decrease its size (on the film) by around 3.6%. Is that the final answer?

Not quite. The “background” objects on the film, which determine the overall calibration of each image, are only around 180 feet away (if we take, say, the white wall for calibration, which is about the best object to use for the Zapruder film). Moving two feet further away from the white wall will make it shrink too. But two feet out of 180 feet is only 1.1%. So the sign shrinks by 3.6%, but the background shrinks by 1.1%. For these small changes, we can just subtract one from the other (or do it the long way if you like), and conclude that the size of the sign relative to the background should decrease by about 2.5%.

What if we were to now change that figure of two feet? The calculations are almost linear for such small changes, so the effect scales by direct proportion. A one-foot shift backwards would shrink the sign by about 1.3%. The feet backwards (whoops—he’s fallen off the pedestal!) would give you about 3.8%.

Is this enough to explain the discrepancy in width of the sign? It doesn’t seem to be quite enough, but it’s in the right ballpark. Moreover, we have to keep in mind the fact that the sign is inconsistent in the film itself, and so my panoramic view contains a sort of Frankenstein version of it. Given these uncertainties, I’d be completely willing to pay The Gang this one, and acknowledge that the width of the sign can be brought into consistency by means of having Barnes standing on the back of the pedestal.

Now, go back and read The Gang’s explanation of this for yourself. What was their approach? “We couldn’t do the mathematics, so we put it in the computer engine. It drew these rectangles for us. Look: it gets smaller!” That’s poor science. They have not identified the physics responsible for this effect. They know it has something to do with perspective, but do not appear to know the elementary formula that governs perspective foreshortening (which is the technical term for the “railway track” phenomenon they describe). The equation is so simple (inverse variation) that my Year 10 students have it in their Exam a few days from now!

If you’re of a scientific bent, you may have picked up something at this point. Since The Gang did not understand foreshortening, physically or mathematically, how did they know that they had to correct for the shrinking of the background? They didn’t! They didn’t do it. They just asked the computer for an image from the “forward” position, and an image from the “backward” position, and overlaid them. They probably didn’t even ask for enough detail of the background for them to realise that it, too, had changed! Their diagram is, in fact, wrong, for this very reason: it exaggerates the widening by around 30%.

Already, you might be gathering that this physics business is not as simple as it looks. And you’d be right. By my working directly with the photographic evidence, and calibrating each set of images with its corresponding panoramic mapping into the three-dimensional space of Dealey Plaza, many variables are eliminated from the outset. As David Mantik (and any other physicist) knows only too well, minimising the number of variables is absolutely crucial—unless you are prepared to do (and capable of doing) an enormous amount of algebraic correction.

The Gang have ignored this principle, and already they have made a 30% mistake in something that should have been an elementary calculation on the back of an envelope. David Mantik’s “snowball from hell” is no exaggeration. The errors compound out of sight from here.

So why am I prepared to “pay” their explanation of the sign width, if they have done their homework so awkwardly, and made such huge errors? Because I can see, immediately, from their assumption that the camera was some feet further away from the sign, that the perspective foreshortening is of the right magnitude. My incorrect assumption was that anyone that knew that Zapruder and Sitzman were up on that little pedestal, wouldn’t have tried to replicate their camera location by standing in Sitzman’s location. That’s been identified as wrong. Well done! I’ll give them a C+.

But you need to be cautious again here. Consider the final paragraph of their homework:

That Barnes took his photo from the rear of the pedestal explains the shift in sign position, the change in sign size, and the change in sign edge angles.

Whoa! Hold your horses! Did you see them slip that one in? They must think I’m a pretty poor teacher if they think I’m not used to having students sneak in conclusions that aren’t at all demonstrated by their working!

The Gang did not, at all, demonstrate anything about the angles. This is clear on several levels. Firstly, they didn’t say anything at all about it—it just sneaks in there, as the last phrase of the section. Secondly, there is a nice coloured diagram showing the sign width change (generated by the 3-d computer engine)—albeit wrong by that 30%—but no explanatory diagram of any angles. These are the obvious ones—the “I cheated on my homework” ones. But there are other reasons, that are equally obvious to me. Firstly, any angle in an image is an angle of a line relative to some other line. The camera is free to rotate about its optical axis, so that absolute angles are irrelevant. (For example, if you want the sign to be leaning by three more degrees, just rotate the image by three degrees!) All that is physically relevant are relative angles. Secondly, there is, of course, no comprehension by The Gang about how or why all of this works anyway. If they had been able to generate some angles, they would not have known how to calibrate for the changes to the rest of the image. And, indeed, they would have made the same mistake of forgetting to factorise out the effects on the background.

Let’s, therefore, back up just a tad, and see how they dealt with this whole problem. Here is the crucial paragraph:

To accelerate rendering, we used green “sticks” to represent the edges of prominent features such as buildings on Houston, etc. that we used for alignment. After matching the Barnes field of view, we panned, tilted, and rotated, the camera until it pointed in the same direction the Barnes camera pointed. To within a small degree of error, this provided the same perspective that Barnes had, as evidenced by the similar angles of the green lines and the diverging building edge lines:

GAK! We can see the wheels falling off the cart altogether here.

Firstly, The Gang needed to “accelerate rendering”. What does this mean? The “rendering” is the process of generating the view—the image. Why did they need to “accelerate” it? As someone experienced with three-dimensional graphics and virtual reality engines, I can tell you why: they would have found that it took them all day to re-render their images as they fiddled with all these parameters! Eventually, they realised that if they wanted to get anything out of it before Christmas, they needed to simplify their model. Eventually, all they were left with was a few “sticks”, and a rectangle for the sign.

Can you see how cack-handed it is to go through all this computer animation bunkum, when it can all be done directly and simply?

Secondly, we view with horror their “scientific” procedure:

After matching the Barnes field of view, we panned, tilted, and rotated, the camera until it pointed in the same direction the Barnes camera pointed.

As any first-year physics student will tell you, they had six degrees of freedom: the north-south position of the camera; the east-west position of the camera; the height of the camera above the pedestal; the panning angle of the camera (“left–right”); the tilt angle of the camera (“up–down”); and the rotation angle of the camera (twisting the image clockwise or anticlockwise). So how did they determine these six parameters? Did they do some sort of multi-dimensional maximum likelihood parametric fit? No, of course not. That would take some mathematical understanding! They just tweaked all six parameters until it looked about right. It’s exactly as if they had six knobs in front of them, each one adjusting one of these parameters, and three of The Gang sat there twiddling the knobs until Tink yells out, “Yeah, that looks about right! No worries, mate—she’ll be right!” (Sorry, that’s a standard Australian saying, which translates roughly as “close enough is good enough—and I’m buggered if I can be bothered to do any better than that!)

And what was the Tink-meister’s criterion for “getting it right”?

To within a small degree of error, this provided the same perspective that Barnes had, as evidenced by the similar angles of the green lines and the diverging building edge lines:

Uh-huh. A couple of angles look the same. That’s a six-dimensional parametric fit? I’m sure the statisticians will be overjoyed to hear about that simplification!

Let’s just highlight one aspect of this sentence on further time:

To within a small degree of error,

A “small” degree of error? How “small” is “small”? Well, if you want an answer to that question, I’m about to disappoint you:

We stress, though, that we do not offer this model as an exact model of features in Dealey Plaza. As such, our analysis here is qualitative rather than quantitative.

What’s that again?

our analysis here is qualitative rather than quantitative.

Uh, I’m sorry, guys. I’m not into “qualitative physics”. You either know how to calculate these things, or you don’t. And, clearly, you don’t.

So what does “a small degree of error” entail? Let me try and quantify it for them. Surely, a few degrees can’t be picked up. With six degrees of freedom, you can have two perspective effects that cancel in some aspects (such as those couple of angles that “look about right”), but not in others that aren’t measured.

So let’s return to The Gang’s piece of cheating:

That Barnes took his photo from the rear of the pedestal explains the shift in sign position, the change in sign size, and the change in sign edge angles.

The position was explained by parallax, in my chapter: Barnes is not just further away from the sign, he is to the left of the Zapruder position. I am surprised that The Gang were not even able to give a simple explanation of this point. The sign size I have conceded—when their proof is repaired. But the sign edge angles are completely unexplained. Even if The Gang had shown the angles changing by some degrees, they would not have been in any position to claim that this explains anything, because they have no control at all over the parameters at their disposal. Worse, they are making assumptions about the sign itself, which introduces yet more degrees of freedom; of the buildings; and so on. We return to David Mantik’s “snowball from hell”.

But even given all this, they don’t actually show any change in the angles at all!

Let me tell you why. It’s not possible! The shifts that they are fiddling around with are all small changes. They can’t push Zapruder or Barnes off the pedestal. And the perspective effects of such small changes—called, in mathematics, the generators of the large, arbitrary changes—are much simpler to analyse, mathematically, than the full perspective transformation. Basically, shifting the camera position in three dimensions gives you parallax shifts (vertical and horizontal) and foreshortening changes (the size change effect). The three angles of rotation (“pan, tilt, twist”), on the other hand, don’t give you anything at all! Why? Well, twisting the camera just rotates the entire image. Big deal—no relative angles are changed. Pan and tilt, on the other hand, basically specify the optical centre of the image. But in creating my panoramas, and then my comparison images, I have explicitly brought all optical centres to the same direction! (I’d be happy to give The Gang the formula for this, if they want—although the last time I gave David Wimp advice on image transformations, it took him two years to believe me!)

This is why the angles remain the “damning proof”. They cannot change,

I concede that the sign width should not be included in this description, and will ensure that this is corrected in the third printing of the book.

As a final piece of humour, however, consider The Gang’s attempt to analyse the lamp pole angle discrepancy. Some understanding of the history of this research will again be helpful. The discrepancies in the angles of the sign between the film and the Barnes photographs have been known about for some time. However, they have been “explained away” by suggesting that the sign itself had been moved, or re-erected, in the days between the assassination and the day that Barnes took his photos. The panorama work, however, allowed me to extend the analysis to the lamp pole to the right of the sign—the only other “foreground” object relatively close to the Zapruder pedestal.

The Gang tuck away their discussion of the lamp pole angle at the back of their piece, separated from the discussions above to which they logically belong. Why is this?

An examination of the section will reveal the reason very quickly. This is their complete analysis:

We modeled the light pole by first referring to the HSCA map and then adjusting it its position to match the lamppost in the Barnes photo. We had to move the lamppost east on Elm 5.7 inches and then tilt it westward about 0.8 degrees to get a good match. A view that matches Zapruder's frame 263 shows the model lamppost at the same angle as the lamppost in the Zapruder frame:

That’s it! Their image shows the overlay of a single “stick” from their computer animation engine. This represents the lamp pole. By fiddling the view, they found that they could make the lamp pole to appear on the desired angle. Wonderful! Why didn’t I think of that? Perhaps because it is the relative angle between the lamp pole and the background objects in Dealey Plaza that has any significance? If The Gang wanted to make the lamp pole lean, without reference to anything in the background, why didn’t they just rotate the image? That’s much easier than fiddling with an animation engine!

Understanding the motion blurring in Zapruder frames (2 pages)

One of the smoking guns of Murder in Dealey Plaza was illustrated by the apparent lack of blur in Frame 303 of the Zapruder film, as argued by Dr. Roderick Ryan in Noel Twyman’s Bloody Treason. Unfortunately, when I first started investigating the Zapruder alteration claims, I discovered that this particular one was in error. I start this section of my chapter with a thorough yet (I hope) understandable explanation of how the correct blurring is calculated, and show explicitly why Frames 302 and 303 are both consistent with the depicted motion of the presidential limousine.

The Gang’s response:

No dispute. Again, The Gang do not highlight the fact that, in the course of putting Zapruder alteration research onto a thorough and rigorous physical and mathematical footing, I actually showed that several previously made claims of anomalous phenomena were actually wrong. The reason for their silence is obvious: they do not want to admit that we displayed this degree of scientific honesty in our own work; and, moreover, they would be implicitly endorsing our mastery of the physics involved.

The Zapruder frame they forgot to blur (2 pages)

I show that Frame 232 of the film, as published in detail in colour in the Memorial Edition of Life, is not correctly blurred. The entire frame is impossible, if the film had been taken through the Zapruder camera. I explain how this implies the complete fabrication of the film, not just minor alterations.

The Gang’s response:

Unbelievably, NO DISPUTE! One of The Gang, James R. Gordon, opined on the jfkresearch.com forum that The Gang had “not realised the significance” of this proof, because it “came after” my discussion of the Stemmons Freeway sign. What a remarkable statement! The importance of the lack of blur in Frame 232 is listed as the primary evidence for fabrication by Jim Fetzer in the Prologue: he shows the image, and spends half a page summarising it. The image then appears again at the end of the colour section, in the highest possible quality of reproduction available to us. Finally, two pages of my chapter are dedicated to the frame, which is shown another two times—preceded, of course, by a two-page explanation of how to analyse the blur in Zapruder frames. Finally, I emphasise that the lack of blur in the entire frame is the first proof presented that demonstrates that the film is not merely an altered film, but rather is a completely fabricated construction. That The Gang attack sections of my chapter that follow this section is an obvious indication that they did not, in fact, “overlook” it. Rather, this proof is so remarkably simple that, once pointed out, anyone at all can understand it.

Changing his tune, Gordon’s final whimper was that I “claimed objects to be point-like”, and that they had to take my word for that claim. But the image is right there, for all to see! Even my five-year-old son can tell me what is there. Moreover, I show that these objects are not photographic blemishes, but rather were real objects in Dealey Plaza. Gordon’s apparent slip of the tongue demonstrates that The Gang realised why I chose point-like objects as especially important markers in this proof: there is a way to “weasel out” of the conclusions, mathematically, that is impossible to employ if the objects are point-like. I anticipated this argument of The Gang’s before it was even made (actually, it had been used on a previous occasion some years previously, and I had to admit on that occasion that it was, indeed, a possible way to “weasel out” of a similar conclusion), and Gordon has revealed the fact that I “snookered” them this time around.

The sprocket hole that is not a hole (3 pages)

I show that part of a sprocket hole shown in the Warren Report issue of Life magazine does not appear to be a hole at all, but rather a triple-exposure of light that looks white. This is the only sprocket hole edge “ever” shown in Life (the November 1966 issue showed the region between the sprocket holes, but the holes themselves were blacked out), excepting the 25 August 2003 issue showing Frame 313 (that was received too late for an amending note in the first printing; this is due to be noted in the second printing).

The Gang’s response:

No dispute.

The disappearing blood spray (3 pages)

The “spray” of red and white immediately during and after the head shot in the extant Zapruder film has always appeared fake to film and photographic experts. In this section, I perform an analysis of Frames 304–326, with the limousine stabilised and maintained in exactly the same position and orientation in each frame. Apart from illustrating the weird behaviour of “the blob” that appears on the crater of Kennedy’s right forehead (see David Lifton’s chapter for more details on what this told him already in the early 1970s), I perform an analysis of the red, green and blue components of each frame and display them on a graph. This provides confirmation of what the eye discerns, frame by frame: the red “spray” disappears after one frame, and the whitish “haze” within about six frames. This is impossible if the frames were showing real blood and brain matter, which would take some time to be ejected and dissipate—precisely as described by many early viewers of the film. Moreover, the “trunk” of the limo remains completely shiny and mirror-like, and the blood and brain matter that witnesses report splattering everywhere is nowhere to be seen. (Frame 323, published early in colour and mentioned for other reasons below, is particularly clean and clear, in contradiction with, for instance, Governor Connally’s testimony.)

The Gang’s response:

No dispute.

The missing object in the grass (5 pages)

In this section I build on an observation by David Mantik in Assassination Science, that the bright object that can be seen in the grass beginning in Frame 313 of the extant film, which moves across the field of view as Zapruder pans to the right, is missing from all three frames in which it should be visible in the issue of Life that was rolling off the presses on the weekend of the assassination. I note that this object is the only thing that ties down the speed of the limousine in the crucial period of time surrounding the head shot. Remarkably, there is a “puddle” on the largest of these first-weekend images, Frame 323, that is visible underneath the object in the MPI digitisation of this frame from the “camera-original” film. And even more remarkably, in the slides sent to him by the U.S. National Archives, this object is missing in Frame 323 of Stewart Galanor’s book Cover-Up, but is missing in no other frame—just like on that first weekend! As a bare minimum, this implies that there are two versions of Frame 323 in the National Archives. I make my best guesses as to what this means in the context of the fabrication of the film.

The Gang’s response:

No dispute.

The impossible head turns of the driver, Secret Service agent Greer (3 pages)

For over a decade it has been observed that Greer, the driver of the limousine, snaps his head from front to rear and rear to front faster than is humanly possible. Noel Twyman has performed the most extensive research into these anomalies in his book Bloody Treason. I provide a summary of this observation. I emphasise the fact that MPI’s “reference” digitisation of the film is inferior to previously available copies, excepting the region between the sprocket holes (which was previously unavailable except for what was shown in Volume VXIII of the Warren Commission Hearings) and a small region surrounding the President and Governor (which was shown on the DVD in full resolution). I then describe in detail how any interested person with a home video camcorder can perform their own experiment to try to replicate Greer’s feat. I provide an explanation of why the typical rotation times are all in the same general ballpark for real humans, based on some simple scaling properties of the physics variables involved. This proof is not as powerful as those previously presented, relying on the limitations of human physiology, rather than absolute laws of physics. Nevertheless, it is one of the most easily understood and verified demonstrations that something is seriously wrong with the extant Zapruder film.

The Gang’s response:

When I learnt of The Gang’s release of a website attacking The Great Zapruder Film Hoax, but before getting a chance to see it for myself, you could have laid bets with me as to which topics they would have had a go at. You would have been able to win any amount of money at all off me by suggesting that they would go for Greer’s head turns.

That a major portion of The Gang’s site is a re-analysis of this decade-old anomaly is simply beyond belief. This is research that was already well-known when it was extensively investigated and published by Noel Twyman in Bloody Treason (1997). I included a reference to it for several reasons. Firstly, for completeness: it is one of the few “early” pieces of Zapruder alteration research that remains relatively unscathed, and unqualified, so many years after it was first reported. Secondly, I wanted to provide a tutorial on how anyone at all can investigate this issue for themselves, using only a normal domestic camcorder and a VCR with frame-by-frame playback. Thirdly, I wanted to comment, from a physicist’s point of view, why it becomes increasingly difficult to turn one’s head increasingly rapidly: the g-forces increase like the square of the rotation speed, so that a turn that occurs twice as quickly implies four times the g-forces; a turn three times quicker implies nine times the g-forces.

Despite the relative insignificance of the Greer head turns compared to the new discoveries presented in The Great Zapruder Film Hoax, the topic is, apparently, James R. Gordon’s only field of “speciality”. His only other assigned task seems to be to annoy serious researchers on the jfkresearch.com forum.

But Gordon’s mathematical credibility gets off to a rather inauspicious start:

It is a short section, comprising of only four pages.

Unfortunately for James, it is less than three pages in length.

However, it would be prematurely unkind to conclude that Gordon is numerically challenged. If he followed the lead set by his Gang co-authors, it is highly likely that he has never read the book at all. So when Captain Tink writes to him, and explains, “It’s on pages 193, 194, 195 and 196,” Gordon counts: 1, 2, 3, 4. Four pages, right? Except for the fact that there are only eighteen words of it on page 196.

But let us ignore the fact that he apparently hasn’t read the book at all, and see what entertainment he has in store for us.

Gordon’s basic thesis is that Twyman’s figures are not quite correct, and that I, in turn, endorsed Twyman’s research without actually investigating it scientifically. Gordon then provides his own laborious, frame-by-frame comments on the Twyman analysis. After page upon page of descriptions of what Gordon believes he can see in each Zapruder film, he concludes that he does not agree with Twyman’s analysis. He complains that I did not re-perform Twyman’s analysis, and so, in the absence of such work, he will substitute his own. He then accuses me of being terribly unscientific in this omission.

Remarkably, Gordon makes use of the “Costella Combined Edit” Zapruder frames that Jim Fetzer has posted on assassinationscience.com—even though a major portion of my three-page description (sorry, James, I didn’t really mean to rub salt into the wound)—about one-third of it—is my complaint that the MPI images (on which my “Combined Edit” is based) are inferior to those available from other sources, unless we want to look between the sprocket holes, or at a small region surrounding the President, neither of which is applicable to the images of Greer! I further noted that even the halftone reproductions of Twyman’s images in his book, published before the MPI disc was available, are superior for analysing Greer. Of course, since Gordon has never actually read my chapter, he didn’t realise that I had made this criticism of MPI’s “reference digitisation”.

So we have the ridiculous situation where Gordon is using my images, which I have explained are inferior to Twyman’s, to argue against Twyman’s analysis, and then to complain to me that I didn’t re-perform the analysis using my own inferior images!

Just a word of advice, Jimmy Boy: as Clint reminds us, “A man’s got to know his limitations.”

But let’s ignore this major blunder of Gordon’s, and consider the question that seems to have him perpetually confused: why didn’t I actually do any mathematical analysis on the Greer head turns?

Anyone who has actually read the book is, by now, screaming out the answer: I was explaining how you can perform the experiment yourself, using nothing more than a camcorder and a VCR. Do you need to do all sorts of obscure mathematical guesswork, and write an essay on the merits of each frame, to perform such an experiment? Hell no! All you need to do is to set yourself up on a chair with the camera appropriately pointing towards you, and then snap your head around from front to rear as fast as you can possibly do it. Do it until you give yourself whiplash. Challenge your spouse or your kids or your neighbours to try it. It could well be the best party game since Pictionary or Trivial Pursuit.

I can just imagine Gordon now, complaining, “But how do you measure the time to snap your head around, if you don’t do any obscure mathematical guesswork?” Well, it’s quite simple, James—and it’s explained right there in the book, if you bothered to read it. You look at how Greer’s head looks in a frame of the film, and then use the frame-by-frame playback on the VCR until you find your own head in a matching pose. Then you look at Greer’s head in the next frame, and again move frame by frame until your own head is in a matching pose. That’s it! As the saying goes, Keep It Simple, Stupid!

All you then need to do is count how many frames it took you to match Greer, and multiply by the number of milliseconds per frame.

The only prerequisite is that you can count.

It is a short section, comprising of only four pages.

Oh. Sorry, Jimmy Boy.

Where could other cameramen be hidden? (3 pages)

Given that the Zapruder film is a fabrication, I provide my best speculations as to where the cameraman or cameramen that actually filmed the assassination could have been hidden. The Betzner photograph provides a tantalising clue of a tripod located in the pergola shelter; but the object is not shown in the Willis photo, and so one must determine whether either of these photographs has been retouched or altered. Certainly there was more than enough opportunity to do so, given that they were not actually published until Life rounded up some of the photographic evidence in 1967 (this, ironically—or perhaps not so ironically—being the project that the Leader of The Gang, Josiah Thompson, was connected with when employed by Life). It is difficult to know what extant photographic evidence to put one’s faith in—if any at all.

The Gang’s response:

No dispute.

The troublesome Mr. Altgens (8 pages)

In this section I offer an insight into how history can be subtly “revised” as the decades pass. James “Ike” Altgens was an Associated Press photographer who took a number of photographs of the presidential limousine in Dealey Plaza, including one during the shots themselves, and one as the limousine departed. The problem is that Richard Trask, in his important books Pictures of the Pain and That Day in Dallas dealing with the photographers the captured the assassination and its aftermath, embellishes Altgens’ story, to the point where he has Altgens taking two pictures that his own affidavits and Warren Commission testimony make impossible, and which Altgens himself denied to Trask himself. Trask ignored Altgens’ protestations as either modesty, or the confused mind of an old man. But not only does this provide an example of historical fraud: one of the two “extra” photographs, apparently first published by Josiah Thompson himself in Six Seconds in Dallas, appears to show Abraham Zapruder and his assistant Marilyn Sitzman departing from the pedestal from which is ostensibly took his film. Although only a view from behind, this photograph would be by far the clearest photographic evidence for the presence of Abraham Zapruder in Dealey Plaza that day: as Jack White has shown (one example of which is contained on the front cover of The Great Zapruder Film Hoax), every other grainy, blurry image of “Zapruder” and “Sitzman” shows a tiny “Zapruder” with a giant “Sitzman” in configurations that would simply not allow “Zapruder” any possibility of shooting the film that bears his name. But apart from his explicit denial, Altgens’ description of his own movements after the shooting, and photographic evidence of him being on the wrong side of the street, prevent this photograph from being genuine.

The Gang’s response:

No response.

Why the film had to have been completely fabricated (7 pages)

In this section I tie together a number of technical properties of the film that are actually correct (namely, pincushion distortion, “ghost” images, and motion blurring), to show why any “small alteration” theory of the Zapruder film can be unequivocally ruled out. Rather, the only possibility is that the film was carefully fabricated, frame by frame, out of raw footage of the actual assassination, together with background views of Dealey Plaza, and the bystanders as they waited for the motorcade to arrive.

The Gang’s response:

No dispute. This is remarkable, because in this section I have actually ruled out completely and unequivocally any possible “simple alteration” of a genuine camera-original Zapruder film. This includes many variations on the idea of simply using an optical printer to delete frames, whether individually or en masse. In proving this point, I am actually refuting a number of scenarios presented or implied in Assassination Science and Murder in Dealey Plaza.

I can only imagine that this proof was welcome only back in 2001, when I did not know much about claims that the film was a fabrication, and did not yet have the resources and data necessary to find those few mistakes that were made. (See the introduction to my chapter for a discussion of this.)

How the fraud was perpetrated (8 pages)

In this section I explain how such a fabricated film could easily have been created, despite the apparently tight restrictions presented by the fact that numerous people viewed the film in the days following the assassination. I explain how there would have to have been two parallel paths of development. In the first, the optical printing and editing techniques described by David Healy would have been used to perform quick (hours) but massive changes to the content of a genuine film of the assassination, that was passed off as “the Zapruder film”. As the days, weeks and months passed, this editing was progressively fine-tuned. (Of course, in those early days, no one had memorised the content of every frame of the film, as researchers of today have done—it was all new. Indeed, early viewers of the film described important and consistent differences from what is now in the extant film.) In the second path of development, high quality photographic editing was performed to create the frames that would be published in Life magazine and selectively released to other media outlets. This work would have been immensely more time-consuming, but only a few dozen such frames were published within even the first year. Most importantly, the task was completely possible with a very limited staff using 1963 technology. In the course of this work, some minor mistakes were made (as described in earlier sections)—but not many. Indeed, if it weren’t for those early issues of Life, those isolated mistakes would have been buried in the hundreds of technically perfect frames, like needles in a haystack. The most important point of this summary is not to provide a definitive explanation of what actually happened (only the perpetrators can tell us those details), but rather to show that it could easily have been done.

The Gang’s response:

Josiah Thompson’s recycled “proof” that the film is authentic is the only indirect response to my scenario. I encourage you, the reader, to carefully read through Thompson’s timeline, and then my own. I expect you to conclude that they are both completely self-consistent and plausible, if one ignores the proofs that the film’s authenticity is physically impossible. That is all that Thompson is presenting you with: a plausible, self-contained description of the film’s history. If I had a lot of time to spare, I could write you a thousand different plausible histories of the film—with, say, half of them having the film genuine, and half of them having it fake, if you would like. But not one of them would “prove” a thing.

If I had not explained how the film could have been fabricated, given what we know of its history, then Thompson could have rightly pointed to such an important omission as an acknowledgment of failure on my part.

But I did, and so he can’t.

Instead, we get the same recycled drivel that he posted to the Internet for Assassination Science, and recycled for Murder in Dealey Plaza, and now resurrected yet again from the dead for The Great Zapruder Film Hoax. All he changes is the book title.

Appendix: Strange experiences en route to Duluth (15 pages)

In this Appendix I provide what I hope to be an entertaining account for my experiences when visiting Dealey Plaza and in travelling to Duluth, Minnesota, for the Zapruder Film Symposium in May 2003. Included are a variety of eyebrow-raising experiences, individually humorous and curious, but collectively providing food for thought for anyone who has not lived the experience of being a known (feared?) JFK researcher first-hand. To fully appreciate this Appendix, one must read it in the context of the rest of the book, in particular David Lifton’s chapter about the shady characters and dealings that are rife in assassination research, and Jim Fetzer’s overviews of the curious “conversions” of Gary Mack and Josiah Thompson. Included amongst the experiences are: “rain sensors” that are hidden in such a way that they cannot collect rain, and separated by only fifteen yards (so that they can detect if it’s raining on one side of Main Street but not the other)—but whose symmetrical layout makes much more sense for audio surveillance purposes; “tourists” who acted remarkably like government agents; an “adjustable lamp pole”, propped up by about an inch on all sides by small washers, whose angle could be adjusted at will, overnight, simply by lifting the pole and moving a few of the washers, that could be used to discredit research investigating angles of objects (as mine has); another “tourist” following us to Minnesota and around the airport; my luggage having been gone through during a seven-hour wait at Minneapolis–St. Paul airport, with the “coincidental” results being the destruction of my electric shaver, holes being “clawed” in my shirts, and the memory card of a digital camera being destroyed, all being discovered on the morning I was to make my presentation; and the American-accented substitute teacher taking my place in Melbourne who had apparently lived in Australia for seventeen years, but couldn’t understand Australian accents, didn’t know who the Prime Minister of Australia was, and spent the entire four hours he remained in the job (before claiming he had found full-time work elsewhere) searching through the drawers of my desk in my classroom. Of course, none of these amusing events prove anything at all about the Zapruder film. Some intuitively-challenged researchers have offered the deadpan response, “Oh, that’s an interesting set of coincidences,” to which one feels like Holmes dealing with Watson on one of his particularly dull days. I leave this Appendix to the individual reader to form their own opinions; and all but those with an obvious agenda to pursue tend to arrive, not remarkably, at roughly the same conclusions. Read it yourself, and tell me what you think. Really! As it has no bearing on the authenticity of the Zapruder film, I really don’t mind where it takes you. (Consider it my contribution to the entertainment value of the book!)

The Gang’s response:

Given the importance of the rest of the book, it is remarkable that The Gang has yet another “specialist” whose only contribution is to argue about observations that have no bearing at all on the thesis of the book, namely, that the Zapruder film is a hoax. Barb Junkkarinen (known generally as “Junk”—I now know why) provides us a short section of scientific indignation and studious research on a burning issue in JFK assassination research: Wireless RainSensors.

Junk mounts her soapbox by complaining that someone with a Ph.D. in theoretical physics, an Honours degree in Engineering, an Honours degree in Science, and a teaching diploma should always investigate—scientifically, and until they drop of exhaustion—any object of scientific or engineering interest that is put before them. I guess she thinks that scientists and engineers are like Pavlov’s dogs: show us something technical, and we start salivating.

Junk’s good mate, Stu Wexler, has the same belief. He spent months of my time in 2001, trying to draw me in on research designed to debunk Ken Rahn’s absurd neutron activation analysis (NAA) claims. I spent some time on the issue, but told Stu very quickly that Rahn’s statistical arguments were hilarious. If Ken Rahn were counting sheep, he’d conclude that half of them were actually dogs, because he had a sheepdog rounding them up in one paddock. Mathematically speaking, that’s how ridiculous his claims were.

Unfortunately, Stewart Galanor, who also earns his keep these days teaching high school mathematics, did get sucked into this furphy. Galanor’s a good man, and it saddened me to learn that his talents had been absorbed by this sort of rubbish.

Anyway, I digress. Back to Wexler’s mate, Junk. She apparently believes that I should spend all of my energies investigating and researching these RainSensors. She is indignant that I “determined, to my satisfaction” what was going on in Dealey Plaza, too rapidly for her comprehension. She insists that it is a field of research that warrants a major investment of time and energy:

Did Costella do any research into the irrigation system installed in Dealey Plaza? He mentions none in his chapter, and failed to respond when I asked him this very question on alt.assassination.jfk.

Junk is being very cute here. A serious back-down, no less! What actually happened is that I broke with my own rule (see the introduction to my chapter) of not posting to the public newsgroups, and responded to a comment about The Great Zapruder Film Hoax on alt.assassination.jfk. What Junk fails to tell you is that this newsgroup is “moderated”, which means that it is censored by the man who “owns” it, John McAdams. There is more than enough on the Internet to tell you about McAdams and his modus operandi, if you are interested in the slimier side of assassination research.

My first posting to McAdamsLand on this topic appeared within a normal time frame, and Junk responded. After that, my postings seemed to disappear into a black hole—only to reappear many days later, attached to Junk threads, with second-hand responses from other members of The Gang.

I remembered why I stopped posting to McAdamsLand in the first place, and returned to that policy. Obviously Junk posted a challenge in some of her namesake, that I did not answer.

But let us put that missed opportunity to one side. Junk proceeds to tell us what she has learned from her Google search researches.

The devices have a range of 300 feet to the receiver, which must be housed indoors. Hang on a minute! Is Junk trying to attack me, or agree with me? Where are these “indoor” receivers in Dealey Plaza? In the former Texas School Book Despository? If the actual irrigation is to occur in the grassy area between Elm and Main, and Main and Commerce, how could a receiver indoors be of any help at all? And if the range is 300 feet, why was there a need for two of them within 50 feet of each other?

Junk then asks whether I tested the devices for audio transmission on an RF link. Yes, Barb, I pulled an all-band RF receiver out of my backside and tested the devices. Of course, if Junk were the electronics engineer she scolds me for being, she’d know that there’s no way of testing if an RF transmission is “actually audio”. If she has no understanding at all of digital devices and digital transmission, why is she bothering to write this section at all? Did she owe Tink a favour? Maybe she should have an Internet Phone conversation with him, and have someone pick up the other extension and listen to what it sounds like. “No, no audio conversation at all,” they would report. “All I heard was a noise that sounded like a fax machine or a modem.” DUH!

But let’s imagine that I really did want to spend all of my time in Dallas ignoring the reason I was there, and instead playing with RainSensors. Junk suggests that anyone with such an interest (which isn’t me, but let’s pretend it is) would “rent, borrow, beg or steal” the equipment necessary to analyse the devices. Well, let’s just walk through the logic of that one, Barb, real slowly for you. There are two logical possibilities. If the device is a rain sensor, it will still be one when I get back with all that gear. If it’s a listening device, then … someone’s listening, right? Oh YEAH! I can hear the penny dropping in Junk’s head from here.

And that’s it. All the Junk she could think of.

Acknowledgments (2 pages)

Given the large number of people who have provided me assistance and encouragement during this work (not always easy from this side of the planet), I took a couple of pages to gratuitously acknowledge them.

The Gang’s response:

No dispute. (Just as well. Tink doesn’t want to mess with my wife Robyn. J)