by Lubomyr Prytulak
First posted on  www.xoxol.org/traw/stewart.html  17-Jun-2010 03:50pm PST, last revised 14-Oct-2010 09:45m PST

Can The Stamp-Imprint Curves Be Made To Fit Perfectly?

Larry Stewart testifying as expert witness for the prosecution at the 2004 trial of Martha Stewart (no relation).

Larry Stewart, one time Laboratory Director and Chief Forensic Scientist for the United States Secret Service, has attempted to make stamp-imprints coincide on the John Demjanjuk Trawniki ID Card 1393


Among the insufficiencies indicating that Demjanjuk Trawniki Card 1393 is forged is that the two stamp-imprints which overlap its photograph are supposed to have been made by the same circular stamp tool and therefore should have produced two identical circular stamp-imprints, and yet the two stamp-imprints deviate significantly both from each other and from circularity.  Irregularities in the circular perimeter of the stamp-imprints have already been described in the second and fourth of the nine articles listed below, pointing to the broad conclusion that the stamp-imprints are counterfeit, and to the more particular conclusion that arcs within the stamp-imprints do not fit each other, and cannot be made to fit by repositioning the photograph.

Nevertheless, prosecution witness Larry Stewart currently testifying in Munich concludes generally that Card 1393 is genuine, and particularly that its stamp-imprint arcs can be made to fit, and that the fit achieved is "perfect".  Yes, Larry Stewart describes the fit that he achieves as "perfect".  Not "reasonable" or "satisfactory" or "adequate", and not "nearly perfect" or "practically perfect" but an unqualified "perfect", a claim which cannot be taken lightly, as Larry Stewart originally made it while he was "laboratory director for the United States Secret Service" (p. 154), "in charge of all the forensic operations of the Secret Service" (p. 154), "the chief forensic examiner in the area of document authentication" (p. 154), "a fellow of the American Academy of Forensic Sciences" (p. 194), and "on the board of directors of the American Society of Crime Laboratory Directors" (p. 197) as is testified in the transcript of the John Demjanjuk denaturalization proceedings on 30 May 2001 before Judge Paul R. Matia in Cleveland.  The American College of Forensic Examiners International adds the following detail to Larry Stewart's high standing at the time of his Demjanjuk testimony:

Mr. Stewart most recently held the position of Laboratory Director and Chief Forensic Scientist for the United States Secret Service.  In that role, he managed up to 120 scientists, technicians, and support staff in the areas of document analysis, handwriting, fingerprints, trace evidence, audio and video analysis, photography, toolmarks, computer evidence and counterfeit analysis.

American College of Forensic Examiners International (ACFEI), Member Links  www.acfei.com/links/

In light of the innovative methodology invented by Larry Stewart for his analysis of Card 1393 stamp-imprints, and in view of his high standing in the field of forensic document examination, it behooves us to study his claim, paying particular attention to his method of curve alignment which, for reasons that will become evident shortly, is called here the "Larry Stewart Counterclockwise Solution".

What Is At Stake?

As an accumulation of evidence already attests to Trawniki Card 1393 being counterfeit, Larry Stewart's success in demonstrating that all stamp-imprint arcs can be made to align would remove only one piece of evidence of forgery from that accumulation, whose highlights can be read in the sources below:

    The "Demjanjuk" signature on the Trawniki ID Card is forged, to conceal which John Demjanjuk persecutors have rendered it illegible
    18 Feb 2010

    It was not the German military but the Russian KGB that glued the photograph of John Demjanjuk to the Trawniki ID Card
    19 Mar 2010

    Trawniki Card 1393 breaks the unwritten rule that outside and inside stamps must match
    14 Apr 2010

    Beaded and unbeaded lines within a single stamp-imprint indicate forgery, as does hand-inscription overtop of a lightly-inked template
    26 May 2010

    Kremlin forgery factories compromise the Russian archives
    03 Jun 2010

    By the time the Demjanjuk photograph was attached to the Trawniki ID Card, the photograph was already old and worn
    15 Jun 2010

    Irregularities in the duty-roster area of the Trawniki ID Card could be manifestations of Bazilevskaya sabotage
    10 Jul 2010

    Two gratuitous and ostentatious patches on Trawniki ID Card 1393
    04 Aug 2010

    The Trawniki Id Card was never tightly folded, and shows erosion where there should be none
    19 Aug 2010

Furthermore, if it were discovered that Larry Stewart's Counterclockwise Solution were in some way tainted, this would add to a different accumulation of evidence — the evidence that John Demjanjuk's persecutors are unethical and unscrupulous.

What Is A Reasonable Fit?

In order to recognize whether Larry Stewart has achieved a perfect fit, it might help to first see what a reasonable fit looks like, as for example in Display 1 for the stamp-imprint lying on the outside surface of Card 1393, to the left of the Streibel signature.  The fit whose goodness we are interested in is the fit between a software-created white BestCircle and the stamp-imprint perimeter over which the BestCircle has been digitally superimposed.  Material in pure black inside the stamp-imprint perimeter (consisting of the word "Dienstsiegel", meaning "Stamp", and the pair of concentric circles of lesser diameter) are pre-printed on the card, and are not part of the stamp-imprint added later — they merely serve as targets at which the stamp-tool handler is supposed to aim when creating the stamp-imprint.

As the name implies, the reasonable fit in Display 1 is not absolutely perfect.  The stamp-imprint perimeter is not of absolutely uniform thickness, and it vascillates somewhat around what should be its white BestCircle midline.  Nevertheless, this is as close a fit as is typically achieved, and so is considered here to be a reasonable fit.

DISPLAY 1.  Reasonable fit between a digitally-superimposed white BestCircle and the underlying ink perimeter of the stamp-imprint.

Reasonable fit of a BestCircle on a stamp-imprint on the John Demjanjuk Trawniki ID Card 1393

What Is The Full Depth of KGB-Forging Incompetence?

For the reader who might be inclined to doubt that the KGB is capable of stamp-imprint forgery as incompetent as it is accused of in connection with Demjanjuk Trawniki Card 1393, it may be eye-opening to examine Display 2 which shows the stamp-imprint on the inside surface of Juchnowskij Trawniki Card 847 whose perimeter curve

  1. travelling clockwise (CW) enters the left edge of the photo and when it arrives at the bottom edge, simply stops, failing to connect with anything beyond, and

  2. travelling counterclockwise (CCW) enters the bottom edge of the photo, and as it approaches the left edge of the photo, veers spasmodically upward before making a futile attempt to connect with the card-borne stamp lying left of the photo.
The two arcs entering the photo from opposite directions miss each other entirely, merely lying parallel to each other where they should coincide, clearly two independent lines which could not have been produced by any imaginable application of a stamp-tool.

DISPLAY 2.  Stamp imprint from Juchnowskij Trawniki Card 847 whose photo-overlapping segment (the "on-photo" segment) shows deviations from normalcy which are difficult to explain without hypothesizing severe incompetence and irresponsibility on the part of forgers.

KGB forging incompetence evident in stamp-imprint on Juchnowskij Trawniki ID Card 847

Display 3 shows the same Juchnowskij Card 847 stamp-imprint, but with a white BestCircle fitted to the portion of the stamp-imprint lying on the card itself (the "on-card" segment), which BestCircle serves to create a second instance of reasonable fit on-card, and serves also to highlight how wildly the stamp-imprint lying on the photo (the "on-photo" segment) deviates from that reasonable fit.  Despite reasonable fit having been achieved on-card, on-photo can be said to exhibit a failure to fit, meaning that it is impossible to accept that the stamp tool that created the on-card stamp-imprint could have created the on-photo stamp-imprint at the same time.

DISPLAY 3.  The BestCircle which achieves a reasonable fit over the on-card stamp-imprint fails to fit the on-photo stamp-imprint.

KGB forging incompetence made even more evident in stamp-imprint on Juchnowskij Trawniki ID Card 847 by the addition of a white BestCircle

To the question, "What is the full depth of KGB-forging incompetence?", the answer must be that that abysmal level has not yet been plumbed, but that in the case of stamp-imprints on Trawniki Cards, the example of Juchnowskij Card 847 may be scraping bottom.

The two instances above of reasonable fit (Display 1 and on-card Display 3), and the one instance of failed fit (on-photo Display 3), may be viewed as extremes defining a continuum on which the reader will have opportunity throughout the instant article of deciding for himself where to place a particular fit under consideration — sometimes toward the reasonable-fit end of the continuum, and at other times toward the failed-fit end.

A First Step In Evaluating Card 1393 Stamp-Imprint Curves

In evaluating Card 1393 stamp-imprint curves, an optimistic procedure based on the assumption that the card is genuine, and on the understanding that the two stamp-imprints are made by the same stamp-tool, is outlined, and whose steps can be seen put into effect in Display 4.

  1. Determine the diameter of the BestCircle by locating the largest-available single-surface stamp-imprint arc, and fitting a BestCircle to it digitally.  In our analysis, the BestCircle is always white, and the largest single-surface arc is the on-card segment of the stamp-imprint which overlaps the photo's lower-left, and its optimal diameter is 868 pixels, leading us to describe our BestCircle as BestCircle868.

  2. Fit another BestCircle868 to the largest single-surface arc in the other stamp-imprint, the one on the photo's upper-right.  In our analysis, the largest single-surface arc in this region is the on-card arc AB against the right-hand edge of Display 4.

  3. Evaluate the goodness of fit of the two BestCircles to the stamp-imprints on which they have been superimposed.

What we see in Display 4, then, within the lower-left stamp-imprint is a reasonable fit on-card and a failure to fit on-photo, reminiscent of that same fit pattern which we saw just above in Juchnowskij 847.  And within the upper-right stamp imprint, we see the on-card arc AB fitting perhaps reasonably, but the on-photo fit failing utterly.  We disregard arc CD for the moment, but will return to it shortly.

Our initial analysis, then, fails to confirm that the stamp-imprint arcs readily correspond to the requisite BestCircle868 paths.  A single stamp-tool applied twice seems incapable of coinciding with the stamp-imprints that we find overlapping the photo.  Larry Stewart, however, claims to have found an extrication from this quandary, one whose details will be discussed further below, but which for the time being can be characterized as his hypothesizing that the Demjanjuk photo is not in its original position on the card, and that if properly relocated will recapture the "perfect" fit that it had enjoyed in the distant past when the stamp-imprints were deposited on it.

Larry Stewart's solution, however, has no answer to the observation of failed Single-Surface Fits, as described in the following section.

DISPLAY 4.  Result of a quick first-evaluation of the stamp-imprint curves found in proximity to the photograh on Demjanjuk Trawniki Card 1393.  The white BestCircle868 which best fits the largest available single-surface arc (the on-card arc in the stamp-imprint on the lower-left) has a diameter of 868 pixels.  The same white BestCircle868 is also fitted to the largest single-surface arc AB within the stamp-imprint on the upper-right.

Gross fit failure of white BestCircles on both photo-overlapping stamp-imprints on John Demjanjuk Trawniki ID Card 1393

Four Obligatory Single-Surface Fits

In view of the overall lack of fit of the two BestCircle868s, a more detailed analysis can be performed to determine its cause, with the question in mind of whether the lack of fit results simply from the photo either having been deliberately removed, or else having fallen off, and then having been reglued in a slightly different position or orientation.  The more detailed analysis starts by fitting BestCircle868 to stamp-imprint segments lying entirely on a single surface, which is to say, lying either entirely on-card, or else entirely on-photo.  There exist a total of four such Single-Surface Fits, identified as SSF-1 to SSF-4 as explained in Displays 5 and 6.

Display 5.  Location of four Single-Surface Fits that need to be evaluated.

Location of four Single-Surface Fit areas on John Demjanjuk Trawniki ID Card 1393

Display 6.  Definition of the four Single-Surface Fits that need to be evaluated, along with the goodness-of-fit conclusion that results from each fit.     
Stamp Segment
Stamp Segment



We have already seen in Display 4 above that SSF-1 is reasonable.  It is the fit that defined BestCircle868.

Display 4 also serves to evaluate SSF-2 — that is, moving CCW along arc AB, we find BestCircle868 missing arc CD by a wide margin.  As this failed fit lies entirely on-card, no tilting or shifting of the photo can improve it, and therefore Larry Stewart's claim to be able to achieve "perfect" fit by repositioning the photo, even at this earliest stage of our analysis, must be recognized as an impossibility.

Display 7 below deals with the remaining two Single-Surface Fits, both of these lying on-photo, namely SSF-3 and SSF-4.

To evaluate SSF-3 in Display 7, then, BestCircle868 was fitted to arc EF, as can be seen quite tightly, with the result, however, of producing severe fit failure at FG.  No reasonable Single-Surface Fit of BestCircle868 in region SSF-3 is possible because BestCircle868 is a circle, and EFG does not follow the path of a circle.  Because this fit failure occurs entirely on-photo, relocating the photo is incapable of improving the fit.

To evaluate SSF-4 in Display 7, BestCircle868 was fitted to arc HI, which, however, resulted in fit failure at JK.  As JK is faint in Display 7, it is shown enhanced in Display 8.  And to repeat the obvious, fit failure here too takes place entirely on-photo, such that no tilting or shifting of the photo is able to improve fit.

DISPLAY 7.  The two remaining Single-Surface Fits are both on-photo.  Fitting BestCircle868 nicely to arc EF produces fit failure in the continuation of that arc at FG.  And, optimal fit of BestCircle868 to arc HI produces fit failure in the continuation of that arc at JK.

Failure of both on-photo, Single-Surface fits on John Demjanjuk Trawniki ID Card 1393

DISPLAY 8.  Ultra-high contrast improves the visibility of arc JK whose existence was relied upon in Display 7.

Ultra-high contrast reveals fit failure on John Demjanjuk Trawniki ID Card 1393

To sum up, whereas a convincing demonstration of overall fit requires that all four Single-Surface Fits be reasonable, in fact only one of them is (SSF-1), and the remaining three exhibit failure to fit (SSF-2 to SSF-4), in view of which, Larry Stewart's claim to have achieved a "perfect" fit seems implausible because it depends on photo repositioning, whereas none of these three Single-Surface Fit failures undergoes any change as a result of photo repositioning.

We turn next to examine Larry Stewart's Counterclockwise Solution, with the question in mind of how he was able to convince the Cleveland Court of its worth when its success seems so clearly impossible.

Larry Stewart's Counterclockwise Solution

As of this moment, my information on Larry Stewart's Counterclockwise Solution does not include all the images that he presented on his 30 May 2001 appearance at the Cleveland denaturalization hearing before Judge Paul R. Matia.  However, we can be confident that what information we do have allows us to render a fair evaluation.  What we do have is the images in Displays 12 and 14, taken from a pdf file of Larry Stewart exhibits that I believe originates from Larry Stewart, and just below that we have the relevant pp. 173-177 from the transcript of Larry Stewart's Cleveland testimony taken from a different pdf file.

DISPLAY 9.  Stamp Type C taken from the outside surface of Bondarenko Trawniki Card 1926 is the stamp-imprint that appears twice overlapping the Demjanjuk photograph.

Stamp Type C seen on outside surface of Bondarenko Trawniki ID Card 1926 appears twice overlapping the photo glued to John Demjanjuk Trawniki ID Card 1393  

DISPLAY 10.  The Reichsadler insignia, symbol of Nazi Germany, appears in the middle of the Type C stamp.

Reichsadler insignia, symbol of Nazi Germany, appears in stamp Type C, both photo-overlapping stamps on the John Demjanjuk Trawniki ID Card 1393  

Display 11.  Larry Stewart imagines that the oak-wreath-with-swastika is a separate stamp-imprint created by means of its own dedicated stamp-tool.

Larry Stewart imagines that the oak-wreath-with-swastika is a separate stamp-imprint on John Demjanjuk Trawniki ID Card 1393


One striking feature of Stewart's testimony reveals that he was unaware that the two stamp-imprints that he was analyzing are two instances of Stamp Type C (a designation proposed in FOUR STAMP TYPES, FIVE STAMP PATTERNS), and that Type C looks like Display 9, and that it contains the Reichsadler insignia shown in Display 10, consisting of an eagle perched on an oak wreath which encircles a swastika.  So, when during his analysis Stewart saw this oak-wreath-with-swastika overlapping the base of the Demjanjuk photo, and with the perching eagle lost except for a wingtip higher above (look at Display 7 above, or Display 12 below), he imagined this oak-wreath-with-swastika to be a separate stamp-imprint applied by means of its own stamp-tool dedicated to producing nothing more than an oak-wreath-with-swastika, like that in Display 11.

This Larry Stewart confusion can be read either in the Cleveland testimony of 30 May 2001 pdf file, or in the extract reproduced below, in either case on p. 175, lines 4-7 and p. 176, lines 19-22.  However, on p. 222, lines 4-6 Stewart interjects a retraction in the guise of a clarification: "And just for clarification, the swastika imagine in the lower stamp is part of that original stamp".  "Imagine" is in the transcript, though "image" is more likely intended.  As this retraction-posing-as-a-clarification comes without explanation of how Stewart came to recognize his error, one may wonder if some member of the prosecution team didn't bring it to his attention during the recess in court proceedings mentioned on p. 188, lines 20-24.

Turning now to the first of Larry Stewart's two relevant images, Display 12 shows him fitting a red BestCircle to the on-card portion of the stamp-imprint lying on the photo's LOWER-LEFT, and discovering that although the on-card fit is reasonable, the on-photo fit fails.  This is not yet Larry Stewart's Counterclockwise Solution, or any solution, it is merely a recognition of part of the problem for which a solution is sought, the same recognition that we came to when we inspected Display 4.

Without explanation, though, Stewart avoids the obligatory next step of attempting to fit another BestCircle of the same diameter to the stamp-imprint on the UPPER-RIGHT, where we have noticed in Displays 4 and 7 above that Single-Surface Fit failure is most remarkable.

DISPLAY 12.  Larry Stewart fits a red BestCircle to the stamp-imprint within the on-card area SSF-1, only to discover that it fails to fit the stamp-imprint within the on-photo area SSF-3.  The opportunity to fit the same BestCricle to the stamp-imprints within SSF-2 and SSF-4 is missed.

Larry Stewart fits red BestCircle to stamp-imprint within on-card area SSF-1 on the John Demjanjuk Trawniki ID Card 1393

Display 14 below shows Stewart's second relevant image, which begins to make an attempt at resolving some of the lack of fit, if only in the lower-left stamp imprint.

What Stewart needs to show, of course, is two BestCircles of equal diameter, one fitted on the lower-left and the other fitted on the upper-right.  And he needs to show each of these BestCircles fitted to the stamp-imprint portion lying on-card and at the same time to the stamp-imprint portion lying on-photo.  In other words, he needs to draw two BestCircles of equal diameter to cover all stamp-imprint arcs in all four regions, SSF-1 to SSF-4.  If he needs to relocate the photo to get a reasonable fit, then so be it, but the requirement of showing two equal BestCircles covering four regions cannot be waived.  But of these four indispensible locations, to how many does Larry Stewart show a BestCircle fit?  Well, only one is visible in Display 14; however, another can be inferred.  The fit actually visible in Display 14 is SSF-3; the fit that can be inferred is the SSF-1 fit shown in Display-12.  The red circles in Display 14 are the same as the red BestCircles in Display 12, and the larger of the Stewart red circles is pretty much the same as the corresponding BestCircle in our Display 4.  It is from his testimony below that it becomes evident that the red circles are on-card fits, even though the card is not shown, and the green circles are on-photo fits.  Imagine the photo being unglued from the card, and able to be tilted and shifted while carrying its green circles along with it, while the red circles remain fixed to the card.  By means of such tilting and shifting of a moveable photo with its attached green circles, these green circles can be made to coincide perfectly with the red circles.  This is the "perfect" fit that Larry Stewart takes credit for having achieved.  This is the "perfect" fit that he mentions in his Display 14 instructions, "Move vertically, horizontally and 1o counterclockwise and you get perfect alignment."

Larry Stewart's Counterclockwise Solution, however, can be faulted on several counts:

  1. Move — what?  The card with its best-fitted red circles or the photo with its best-fitted green circles?  It is reasonable to assume that Stewart means "move the photo" back to its original position, the position that he claims it had occupied when the two stamp-imprints were created.  After all, if it's the photograph that has been uprooted from its original position, then Stewart's analysis might be expected to replant the photograph back where it started.  Thus, it might be the photograph with its attached green circles that Stewart shifts and rotates 1o CCW until its green circles match the red circles which are attached to the card.  However, maybe he moves the card with its attached red circles upward until they match the green circles.  But if it's the card that Stewart rotates 1o CCW, that's equivalent to rotating the photo 1o CW.  The ambiguity forces us to test both rotations.

  2. In his perfect-fit Display 14, Larry Stewart crops so as to show the photo without the card, which makes it impossible to inspect goodness of fit within the two on-card regions, SSF-1 and SSF-2.  No justification is offered, or can be imagined, for the exclusion of information that is so highly pertinent.  No excuse is given, or is conceivable, for beginning to introduce the problem-to-be-solved in the full Display 12, but then showing the putative solution to that problem in the cropped Display 14.

  3. Larry Stewart fails to fit a comparable set of red and green BestCircles to the stamp on the upper-right of the photo, in region SSF-4.

  4. Comparing, in Display 12, how little of the oak wreath lies on the photo with how much lies on the card, and noticing also that the upper perimeter of the oak wreath is at best indistinct, partly because it vanishes and partly because it merges into the eagle talons, leads to the conclusion that it is impossible to fit any small green circle to the on-photo wreath with confidence.  Larry Stewart's claim to have not only fitted, but to have perfectly fitted, a small green circle to the on-photo segment of the wreath lacks plausibility.

  5. Larry Stewart's red and green BestCircles are thick and are near each other, the combined effect of which is to largely conceal the on-photo stamp-imprint perimeter, such that goodness of fit between BestCircles and stamp-imprints is obscured.

  6. Larry Stewart fails to follow the procedure for determining single-surface goodness of fit that was illustrated in Display 7 above, consisting of first achieving a reasonable fit to the largest-available arc EF, and which can then be seen to result in severe fit failure in arc FG.  Instead, Larry Stewart opts for fitting over the entire on-photo stamp-imprint, which though achieving a reasonable fit nowhere, yet reduces to a concealable level the maximal fit failure that is visible anywhere, which principle is illustrated in Display 13.

DISPLAY 13.  On the left, seizing a reasonable fit somewhere in order to highlight fit failure elsewhere.  On the right, using the same dot pattern and same circle size, seizing a reasonable fit nowhere in order to reduce the greatest fit failure that can be seen anywhere.

LEFT: seizing reasonable fit somewhere to highlight fit failure elsewhere. RIGHT: seizing reasonable fit nowhere to reduce greatest fit failure anywhere.

Despite such digressions from acceptable methodology, however, the stamp-imprint perimeter that habitually sticks too high above John Demjanjuk's right shoulder continues to peek out from underneath its green BestCircle, which it would not if the fit were truly perfect, or even if it were reasonable.

And this "perfect" fit of Stewart's, applies only to the stamp-imprint on the lower-left; the impossible-to-fit stamp-imprint on the upper-right is ignored.

Although we expect deception this primitive to issue from Kremlin forgery factories, to see it issuing also from the US Secret Service comes as a bit of a surprise.

DISPLAY 14.  Larry Stewart's fitting of red and green BestCircles is his Counterclockwise Solution of the problem that the Trawniki Card 1393 stamp-imprint perimeters don't align.

Larry Stewart's Counterclockwise Solution to problem of fit failure on John Demjanjuk Trawniki ID Card 1393

And below is Larry Stewart explaining his Counterclockwise Solution of the puzzle of stamp-imprints failing to align, taken from the transcript of the Cleveland proceedings of 30 May 2001.  The explanation is the same as the synopsis of it offered above, though perhaps not as clear.


13      MR. DRIMMER:  Your Honor, I am once again
14      giving you the original of Plaintiff's Exhibit 3.
15      Q.  Mr. Stewart, can you describe the examination you
16      performed on the photograph on this document?
17      A.  Yes, sir. During the fiber-optic light analysis, it
18      was apparent that the photograph didn't line up properly.
19      It had either — it was either a false document or had been
20      moved sometime during the years.  Those were the two
21      options.  That was apparent by looking at the rubber stamp
22      impressions that were over the top of the photograph.
23      So I came up with the idea of trying to show
24      whether or not that is, in fact, the case, whether or not
25      the photograph was originally on the document or had it


 1      been altered in some way or moved.  So I began using a
 2      computer and a scanner to assist me in making that
 3      determination.
 4      Q.  Okay.  Did you prepare a chart to help illustrate
 5      your testimony about the examination you performed?
 6      A.  Yes, sir, I did.
 7      Q.  Is that chart somewhere in this courtroom?
 8      A.  It's with me right here.
 9      Q.  Okay.  Using the chart, can you please illustrate the
10      examination that you did on this photograph?
11      A.  Certainly.
12      THE WITNESS:  May I stand, Your Honor?
13      THE COURT:  Yes.
14      MR. DRIMMER:  I believe we also have an easel
15      set up if that will make life easier.
16      THE WITNESS:  To help illustrate this as
17      well, I created handouts for Your Honor and the lawyers, if
18      that will assist.
19      Q.  Please go ahead.
20      A.  The first observation is simply that there was poor
21      alignment.  If we look at the photograph, you will see that
22      there is a series of seals.  There's a rubber stamp
23      impression on the right-hand side and one that's in the
24      lower portion of the photograph (drawing).  That's done as
25      a security measure when a document like this is created.


 1      If you're going to do rubber stamps, you do it over
 2      something like a photograph.  That way you can show whether
 3      or not a document has been altered later on.
 4           This lower seal actually is the product of
 5      two seals.  There's a smaller seal in the middle of it that
 6      also transverses across the photograph and the card stock
 7      document.  When you look at the image as it appears right
 8      now or the photograph and cards as they appear right now,
 9      you'll see that there's a slight break in the location of
10      where the seal meets the photograph in all of the circles,
11      all three of the circles.  So I needed to determine whether
12      or not that was the original photograph.
13           So what I did is, I began using a computer
14      and a scanner.  What I started with is the first image that
15      you have in the upper left-hand side, and after scanning in
16      the document I enlarged it rather large, and I drew a
17      circle just on the area of the seal that went across the
18      photograph at the bottom, the two parts that have the seal,
19      the small inner circle, and the outer large circle.
20           I then continued that and made it a
21      continuous circle as if it had been on the card.  I did the
22      same thing in reverse by taking the portion of the
23      photograph — the portion of the circle that appears on the
24      card and draw that in, drew that in, and then continued as
25      if it had gone across the photograph.  That's what the red


 1      and the green circles are on the image.
 2           You'll see the green is the alignment of the
 3      stamp on the photograph and the red is the alignment of the
 4      stamp on the actual document.  And you'll see that when you
 5      make it larger like that and you make the complete circle,
 6      you'll see truly that they do not line up.
 7           The next image is where I took that green and
 8      that red seal and I physically moved them using the
 9      computer vertically to see if I could get alignment that
10      way.  You'll see that there is better alignment, but still
11      not perfect alignment.
12           The next picture, Exhibit 4 there, shows the
13      movement of the seals horizontally.  After I moved them
14      vertically, I then began moving them horizontally.  Still
15      you get better alignment, but not perfect alignment.
16           The next image shows what happens if I take
17      those seals that I've moved both vertically and
18      horizontally and then rotate them one degree
19      counterclockwise.  At that point, if you remove the red and
20      the green seal, you get perfect alignment of all of these
21      three seals, these three rubber stamps, to include the area
22      that transverses across the photograph and is on the card.
23      All three of the circles are in alignment now.
24      Q.  And so based on the examination you performed and
25      your illustration here, what are your conclusions regarding


 1      the stamps in the photograph?
 2      A.  My conclusions are not only that this is a card from
 3      the 1940 era, that this photograph is the original
 4      photograph that was put on the card.  I make that
 5      conclusion based on the lack of possibility of someone
 6      taking a photograph, putting seals on them, and have them
 7      perfectly line up as this does.

Fits Following One-Degree Tilting

If it is impossible to obtain Single-Surface Fits for BestCircle868 in three of the four same-surface regions, then it follows that it will be impossible to fit two BestCircle868s to all stamp-imprint arcs simultaneously, and this no matter how the photo may be repositioned beforehand.  Despite the logical certainty of fit failure, the attempt is nevertheless made below on the possibility that some insight might ensue.

In the first three images below, the photo was tilted one degree CCW, then slid up-down and left-right until a best fit at some particular spot or spots was obtained.  The moveable photograph, the one being tilted and slid, is imagined as a copy of the one glued to the card, and as being placed on top of the Trawniki Card as it exists today, with the Demjanjuk photograph glued to the card as it is glued today, so that the glued photograph can always be seen peeking out from underneath the moveable photograph.

In Display 15, then, the tilted photo has been slid vertically and horizontally so as to achieve best-possible fit of stamp-imprint arcs at points L and M, trying to make the on-card arcs flow smoothly into the on-photo arcs at these two points.  Result: substantial fit failure within regions SSF-2, SSF-3, and SS-4.

DISPLAY 15.  Moveable photo is tilted one degree CCW.  Alignment is optimized at points L and M.

Moveable photo tilted one degree CCW, alignment optimized at points L and M on John Demjanjuk Trawniki ID Card 1393  

Display 16 shows fit optimized at N so that the on-photo stamp-imprint arc moves as smoothly as possible into the on-card stamp-imprint arc.  Result: massive fit failure within regions SSF-2, SSF-3, and SSF-4.

DISPLAY 16.  Moveable photo is tilted one degree CCW.  Alignment is optimized at point N.

Moveable photo tilted one degree CCW, alignment optimized at point N on John Demjanjuk Trawniki ID Card 1393  

Display 17 optimizes fit between on-card and on-photo oak-wreath-with-swastika, the wreath being pointed to by arrow P, and with a small white circle superimposed near the edge of the wreath to facilitate estimation of the reasonableness of the wreath-fit that was achieved.  Result: massive fit failure in regions SSF-2 and SSF-4.

SSF-3, however, shows the best fit achieved so far in this region, in fact possibly identical to the fit achieved by Larry Stewart's green circle in his Display 14.  So maybe this is the "perfect" fit that Larry Stewart boasts of, and that if hidden under a thicker BestCircle than is conventionally preferred here, in fact if hidden under two thick BestCircles, can be passed off as a "perfect fit".  But to cover up the confusion that continues to reign on the upper right was to perpetrate fraud on the 2001 Cleveland court, as it is to perpetrate fraud on the 2010 Munich court.

And even the unexpected success enjoyed in region SSF-3 may be viewed skeptically.  It falls short of the reasonable fits that we once admired in Displays 1 and 3 above, and falls short even of the shakier fit that we consider to be reasonable in region SSF-1 of Display 17.  Does not the impulse to accept this improved SSF-3 fit as reasonable result from our adaptation to the avalanche of catastrophic fit failures under which we have been buried?  Does it not result from our implicit comparison to dime-a-dozen cataclysmic fit failures, but not from comparison to the reasonable fits in Displays 1 and 3?  And does it not result from a deviation from proper methodology, already noted above, the improper methodology consisting of abandoning a reasonable fit everywhere in order to minimize the maximum deviation visible anywhere?

In any case, however warmly prosecutors might greet the improved fit at SSF-3, Larry Stewart's Counterclockwise solution still gets them nowhere.  The massive fit failures at SSF-2 and SSF-4 continue to play the role of spoilers.

Display 17.  Moveable photo is tilted one degree CCW.  Fit is optimized for the wreath-with-swastika pointed to by arrow P.

Moveable photo tilted one degree CCW, fit optimized for wreath-with-swastika pointed to by arrow P on John Demjanjuk Trawniki ID Card 1393

And on the possibility discussed above that Larry Stewart intended the photograph to be tilted one degree CW rather than CCW, several CW-tilt displays were prepared, but each arrived bearing the same sort of massive fit failures that we have seen resulting from CCW tilts above.  The best of the CW-tilt fits turned out to be, as in the case of CCW fits, the one that concentrated on optimizing the fit of the on-card and on-photo halves of the oak wreath, and that fit is reproduced in Display 18.

Notice one remarkable thing about the Display 18 fit — that while arc QRS in region SSF-3 continues improved, not much different from the fit that the region achieved in Display 17, the stamp-imprint arc TU in region SSF-4 is beginning to fit the white BestCircle868 that reigns on the upper-right.  For Larry Stewart, this might seem like a great victory.  If this promising fit in region SSF-4 were to be blanketed with thick red and green BestCircles, it might pass muster as a "perfect" fit almost as readily as Larry Stewart's "perfect" fit did in region SSF-3.

However, there would remain one very large shortfall.  Arc V in Display 18 would be sitting way up high, unconnected to anything, and the letters "tu" above arrow W that are companion to arc V would be floating way too high as well.  For prosecutors to seize upon the improved fits to arcs QRS and TU in Display 18, then, would be for them to admit forgery of arc V and companion letters "tu".  John Demjanjuk does not have to show that every curve in the stamp-imprints is forged, or even most of the curves, or even several of the curves.  For these stamp-imprint curves to be discredited, it is sufficient for John Demjanjuk to show that one of them is forged, and for that reason Display 18 shows Card 1393 stamp-imprints continuing to avoid fit, and therefore continuing to point to forgery, and therefore continuing to beg the Munich court to remove Card 1393 from its list of exhibits.

Display 18.  Moveable photo is now tilted one degree CW — note that the tilt is now clockwise.  Fit is optimized for the wreath-with-swastika, as it had been in Display 17, which fit the white circle superimposed over the wreath helps evaluate.

Moveable photo tilted one degree CW, fit optimized for wreath-with-swastika on John Demjanjuk Trawniki ID Card 1393

Did Larry Stewart Really Believe His Counterclockwise Solution?

Display 19.  According to Larry Stewart's Counterclockwise Solution, the Demjanjuk photo was stamped while in its original position, which is rotated one degree CCW from its current erroneous position, and also shifted somewhat.  If the Counterclockwise Solution is true, then underneath the photo in its current erroneous location, on the surface of the card in the area covered by the blue rectangle, Larry Stewart must expect to find an arc which is part of the perimeter of the stamp-imprint.  If in peeking under the upper-right corner of the photo Larry Stewart sees no arc, then his Counterclockwise Solution is disconfirmed.

Larry Stewart able to test his Counterclockwise Solution by looking for an arc under the photo in the region of the blue rectangle

Larry Stewart had at his disposal a sure way of verifying his Counterclockwise Solution.  The verification required no equipment, could have been performed at no cost, would have taken about one second of time, and would have provided a definitive answer as to whether his Counterclockwise Solution was true or false.  And the method of verification was obvious, and would have occurred to anybody who had the card in hand, and the Counterclockwise Solution in mind.

All Larry Stewart had to do was to peek under the upper-right corner of the photo to see what lay underneath, what lay on the card in the area which in Display 19 is marked by a blue rectangle with a question mark.

Larry Stewart's hypothesis is that to get alignment, he needed to relocate the photo back to its original position, the position it was in when the stamp-imprint came down on it.  Well, when the photo was in that original position, the blue-rectangle area was naked card, and would have received the stamp-imprint, and so the perimeter arc of the stamp-imprint should still be there on the card, right under that blue rectangle, although covered up by the photo in its current erroneous position.

Larry Stewart does not report what he saw upon peeking under that corner of the photo, nor does he report making the attempt.  In his testimony at the Munich trial, he did not prop up his repetition of his Counterclockwise Solution by inviting the court to view the arc which lay under the upper-right corner of the photo.  His neglect invites two alternative inferences.

One inference is that such a verification never occurred to Larry Stewart.  It never dawned on him that he could learn something from looking under the photo.  He was seized by no curiousity to lift up a corner of the photo and take a peek.  He overlooked the fact that his Counterclockwise Solution had this particular implication which could be readily tested.  He overlooked it during his initial research and right through his presentation of his findings in Cleveland in 2001, and then he continued to overlook it for the following nine years, and still hadn't thought of it when he testified in Munich in 2010.

The alternative inference is that Larry Stewart did take that peek, saw no arc, recognized that its absence disconfirmed his Counterclockwise Solution, but said nothing.

In other words, the choice seems to be between (a) Larry Stewart believing his Counterclockwise Solution but not peeking under the upper-right corner of the photo because handicapped by a deficit of competence, or (b) Larry Stewart disbelieving his own Counterclockwise Solution but not disclosing its disconfirmation by the absence of arc under the upper-right corner of the photo because handicapped by a deficit of integrity.

But can we be sure that there really is no arc hidden underneath the upper-right corner of the photo?  We can be fairly sure.  Prosecution expert witness Gideon Epstein testified that in 1981 "the photograph was loose enough to move with your hand, so that you could get some sort of play into the photograph", whereas in 1987 in Israel, "I found that the photograph was very firmly affixed so that it could not be moved" (Trial Transcript, 11-May-1987, pp. 5873-5874).  In other words, the photograph had been re-glued, to accomplish which it would have had to be removed.  If prosecutors responsible for the re-glueing had noticed any previously-concealed arc, they would have recognized its inculpatory value and would have brought it to the attention of the court, which they did not.  Reason to believe that at least one ungluing of the photograph was accomplished by means of a glue solvent is presented in WHO GLUED THE DEMJANJUK PHOTOGRAPH TO TRAWNIKI CARD 1393?

And even if there had never been any removal of the photograph by anybody, it is impossible to imagine that not a single one of the many prosecutors or of their numerous staff who have been laboring over the Demjanjuk case for over three decades, have been curious enough to peek under the photo.  Defense document examiner William Flynn took a forbidden peek under one corner of the photo, when Israeli monitors momentarily turned their backs on what he was doing, but doesn't report any arc, though he does not say which corner it was that he took his clandestine peek under.

Although I was not permitted to perform any destructive testing (under an agreement between the Soviets and Israelis), I did manage to lift up one corner of the photo enough to see that it had been glued to the card at least twice.  This fact is significant because of the ease with which a composite photo can be made.
William J. Flynn, Demjanjuk: a victim of Soviet forgery, Ukrainian Weekly, 18-Sep-1988, originally published in the 10-July-1998 Arizona Republic.

Although the above is the only publicly-acknowledged peek under the photo, there have been others, one at least motivated by a search for the arc which found none.  And if Larry Stewart really believed in his Counterclockwise Solution, he would this very day be demanding that this missing arc be searched for.

Yes, we can be fairly sure that such an arc does not exist.

Larry Stewart Gets The Last Word

In the course of the 2001 Cleveland hearings, Larry Stewart was pointedly asked why his Counterclockwise Solution failed to include the stamp-imprint material on the upper-right.  Defense attorney Michael E. Tigar had apparently noticed the omission and had been puzzled by it, and asked Stewart what it meant, and so we do have Stewart's explanation on record:


22      MR. TIGAR: Thank you, Your Honor.
23      Q.  I'd like you now to look at Government's Exhibit 3
24      and your charts.  Now, you did a computer movement in order
25      to look at these stamps, correct?


 1      A.  You could call it that, yes.
 2      Q.  Now, did you do the computer movement with respect to
 3      one stamp or both stamps?
 4      A.  Both stamps.  And just for clarification, the
 5      swastika imagine in the lower stamp is part of that
 6      original stamp.
 7      Q.  When I say two stamps, you notice that on the front
 8      of the card there are two big rubber stamps, one in the
 9      lower left and one in the upper right; do you see that?
10      A.  That's correct.
11      Q.  Now, did you do your computer examination with
12      respect to both of those big ones?
13      A.  No, sir, I did one, and I did that for a reason.
14      Q.  What was your reason?
15      A.  If you take the computer and ask it to draw a circle
16      on the one rubber stamp, it was important to determine what
17      effect moving that rubber stamp would have on the other
18      image, and so all of that was useful in obtaining my final
19      conclusion.

However, Larry Stewart's statement — contained within Lines 15-19, p. 222 — fails to weaken the perception that he avoided discussing the stamp-imprint fragments on the upper-right because he was aware that they refused to be brought into any semblance of alignment.  When Michael Tigar invited Larry Stewart to explain his exclusion of the upper-right stamp-fragments from his Counterclockwise Solution, it is on record that all Larry Stewart was able to come up with was double-talk.

Refutation Demo

The persecutors of John Demjanjuk do not abandon arguments just because they are meritless and have been publicly refuted.  Rather, they repackage and rebroadcast them, hoping to win over popular opinion by mere repetition.  Even an astute observer may not have the technical background, or the time, to evaluate these repackaged repetitions, and so will be left with the impression that because the persecutors' views seem to predominate, they must be correct, or at least must contain some merit.

In order to fortify the public against being won over by unmeritorious arguments, at least on the question of unaligned stamp-imprint curves, below is presented a demonstration of how one such repackaging can be refuted.

Display 20 presents a repackaging of the Larry Stewart CCW Solution, incorporating what may be a new twist, where the left-hand panel defines the problem to be solved — BestCircles fitted reasonably on-card, both on the lower-left and on the upper-right, each produces fit failure on-photo.  The middle panel shows a 1o CCW tilt plus repositioning of the photo which seems to produce a reasonable fit overall, but where the author of this analysis claims that the 1.5o CCW tilt plus repositioning shown in the right-hand panel fits even better.

Five characteristics of this presentation are salient.

  1. The images are of such poor quality that the stamp-imprint sometimes disappears, which suggests an attempt to conceal absence of fit.
  2. The BestCircle lines are thick, which is reminiscent of an attempt to conceal the stamp-imprints underneath as part of an attempt to conceal absence of fit.
  3. Whereas it is the photograph that is obviously being tilted from left panel to right, yet in the right-most panel, the card has been tilted as well, which is purposeless and distracting, and which suggests carelessness.
  4. The Single-Surface on-card fit of the BestCircle on the upper-right (region SSF-2) seems reasonable in all three panels, which contrasts stunningly with the blatant failure of this same single-surface fit that we have been seeing consistently above.
  5. In every panel, the upper-right BestCircle seems larger than the lower-left one (which impression is quickly confirmed by holding up a ruler to the computer screen to compare diameters), but which difference of diameters violates our assumption that either the same stamp-tool, or else two identical stamp-tools, created both stamp-imprints.

DISPLAY 20.  Larry Stewart Counterclockwise Solution repackaged.
Larry Stewart Counterclockwise Solution repackaged.  Original with BestCircles of different sizes added on John Demjanjuk Trawniki ID Card 1393
Original with BestCircles added
Larry Stewart Counterclockwise Solution repackaged.  Tilt photo 1.0 degree CCW and reposition, superimpose BestCircles of different sizes on John Demjanjuk Trawniki ID Card 1393
Tilt photo 1o CCW and reposition
Larry Stewart Counterclockwise Solution repackaged.  Tilt photo 1.5 degrees CCW and reposition, superimpose BestCircles of different sizes on John Demjanjuk Trawniki ID Card 1393
Tilt photo 1.5o CCW and reposition

High-definition versions of the three above images, and with thinner BestCircles, are shown in Displays 21, 22, and 23.  The lower-left BestCircle continues to have a diameter of 868 pixels; to get the same on-card fit on the upper-right as has been presented to us in Display 20, BestCircle diameter had to be expanded to 896 pixels.  To improve the visibility of the bottom of the on-photo stamp-imprint on the upper-right (region SSF-4), it has been subjected to hyper-contrast.  Within each display, the three points of poorest fit are numbered.

DISPLAY 21.  Original with BestCircle868 added on the lower-left and BestCircle896 added on the upper-right still leaves unacceptable fits at points 1, 2, and 3.

BestCircles 868 and 896 superimposed on John Demjanjuk Trawniki ID Card 1393 leaves fit failure at points 1, 2, and 3

DISPLAY 22.  Tilting photo 1o CCW ssand repositioning, as in the middle panel in Display 20, still leaves unacceptable fits at points 4, 5, and 6.

Tilting photo 1.0 degree CCW and repostioning, then fitting BestCircles 868 and 896 on John Demjanjuk Trawniki ID Card 1393 leaves fit failure at points 4, 5, and 6

DISPLAY 23.  Tilting photo 1.5o CCW and repositioning, as in the right-hand panel in Display 20, still leaves unacceptable fits at points 7, 8, and 9.

Tilting photo 1.5 degrees CCW and repostioning, then fitting BestCircles 868 and 896 on John Demjanjuk Trawniki ID Card 1393 leaves fit failure at points 7, 8, and 9

The repackaged Larry Stewart Counterclockwise Solution above was able to achieve improved fit in on-card region SSF-2 by allowing a higher BestCircle diameter than was evident in SSF-1, which the instant article never allowed up until the instant demo because of the assumption that the two stamp-imprints overlapping the Demjanjuk photo had been produced either by the very same stamp-tool, or else by duplicate stamp-tools.  Employing BestCircles of different diameters would only be justified if it were shown that in fact Stamp Type C had been manufactured and employed simultaneously in two sizes during WWII, one with a diameter 3.23 percent larger than the other.

But even if, for the sake of argument, two sizes of stamp-tool Type C are imagined to have existed, so that BestCircles of two diameters are permitted, fit continues to fail, as is demonstrated in Displays 21-23 above.  Admittedly, poorness of fit can be reduced by such stratagems as permitting disparate BestCircle diameters, and tilting and shifting the photograph, but the resulting fit never reaches reasonable.

And in any case, single-surface fit is never reasonable within on-photo region SSF-3, as has been demonstrated above in Display 7 using arc EFG.  Fit only appears reasonable to the unwary who greet uncritically methodologically-defective analyses that distribute fit failure over as large an arc as possible so as to reduce the maximum deviation from fit that is visible at any particular spot, as we have seen illustrated in Display 13.

And in any case, again, the failure of this new variation of the Larry Stewart Counterclockwise Solution, as of the original, is proved by the absence of a perimeter arc under the upper-right corner of the photo, as has been discussed above under the heading Did Larry Stewart Really Believe His Counterclockwise Solution?

And in any case, yet again, the tilts which John Demjanjuk persecutors imagine the photo having originally possessed are so extreme as to be implausible, no known Trawniki card showing a photo that crookedly glued.

And in any case, yet again, the presence of two stamp-imprints overlapping the photograph is by itself indicative of forgery (the norm being only a single stamp-imprint), as has been argued at THE FORGER'S DILEMMA THEOREM EXPLAINS DOUBLE-STAMPING within the article WHO GLUED THE DEMJANJUK PHOTOGRAPH TO TRAWNIKI CARD 1393?.

And in any case, yet again, the presence of both beaded and unbeaded lines within a single stamp-imprint by itself indicates forgery, as does hand-inscription over a lightly-inked template, as is discussed at A CLOSER LOOK AT A STAMP-IMPRINT.

And in any case, finally, the disparity between the photograph-overlapping stamp-imprints on the inside surface of the card and the stamp-imprint on the outside surface of the card is also by itself indicative of forgery (the norm being that the inside and outside stamp-imprints match), as has been noted at FOUR STAMP TYPES, FIVE STAMP PATTERNS.

In short, the stamp-imprints which overlap the Demjanjuk photograph fail to pass muster in so many respects as to undermine their credibility many times over.

What Is The Explanation Behind Larry Stewart's Bogus Counterclockwise Solution?


Larry Stewart is on indefinite suspension, without pay, from his job as laboratory director for the Secret Service. If convicted of both perjury charges, he could get as much as ten years in prison.  (AP)  21 MAY 2004  www.cbsnews.com/~

Although Larry Stewart's unconscionable testimony in Cleveland and in Munich is not what the public expects from the director of the Secret Service forensic laboratory, it is not out of line with news reports concerning his testimony at the 2004 trial of Martha Stewart.

U.S. Attorney David Kelley announces perjury charges against Secret Service lab director Larry Stewart, but not in connection with John Demjanjuk Trawniki ID Card 1393  

21 MAY 2004: U.S. Attorney David Kelley discusses perjury charges brought against a national ink expert who testified in the criminal case against Martha Stewart.
MSNBC  The video transcribed here can be viewed at


Lest anyone doubt our committment to ensuring the integrity of all official proceedings, I am here today to announce that we have charged Larry Stewart, the director of the Laboratory for the United States Secret Service, who was called as a government witness at the Martha Stewart trial, for perjuring his testimony in that trial.  [00:22]

Larry Stewart has been employed as an examiner, not a special agent, with the Secret Service since 1982.  [00:30]

In sum, it is alleged, in a complaint we filed today, that Larry Stewart made false statements in his testimony on two subjects.  [00:40]

First, that he personally participated in the forensic ink examination of the disputed notation "@60" which was on a paper that was maintained by Peter Bacanovic concerning positions of Martha Stewart's stock.  [00:59]

Secondly, he lied, or it is alleged that he lied, whether or not Larry Stewart was familiar with a book, a book proposal on ink analysis, that was asked of him by Bacanovic's lawyer during cross examination.  [01:14]

While our investigation has revealed that Larry Stewart testified falsely about the nature and extent of his participation in the examination of the disputed document, we have uncovered nothing to raise questions about the accuracy or validity of the results of that examination.  [01:36]

In fact, the central conclusion drawn by Larry Stewart in his testimony — that is, that the same ink was used to create all of the entries on the document, except the "@60" notation, as well as another small mark on the paper — was confirmed by the examiner who in fact had conducted the examination, and it was not disputed by Bacanovic's own lawyer, who himself had also examined the document.  [02:07]

As set forth in today's complaint, Larry Stewart falsely testified that he worked on the case together with the examiner who did in fact conduct the examination, that he observed and participated in parts of the examination, that he was involved in the preparation of a specific laboratory test, and that he performed a series of specific tests when the document was re-tested, in January of this year, right before trial.  [02:42]

That today's charges are indeed troubling, very troubling, because a trusted and accomplished lab examiner and public servant violated a public trust, as well as a trust that so many of his colleagues in law enforcement had in him.  [03:00]

However, we are quite confident that the false testimony will have no impact on the convictions of Martha Stewart and Peter Bacanovic for both factual and legal reasons, including the fact that the jury in that case acquitted Bacanovic of the charge specifically relating to the disputed document and found that we, the Government, had not proven each of the alleged false statements and perjury specifications relating to the document, as well as to the the sixty dollar price agreement that was between, as re-alleged, between Bacanovic and Martha Stewart.  [03:42]

Ultimately, however, that is an issue that will rest with the court to decide.  [03:49]

While there is no doubt that Larry Stewart's conduct cannot and will not be tolerated, the results of those ink tests are not in doubt.  [04:00]

But by allegedly lying about whether he was present when the tests were done, or that he participated in the examination, the defendant has hurt a tremendous organization of Federal Agents, the United States Secret Service, whose agents put themselves in the line of fire, literally, each and every day to protect our nation's leaders and its citizens.  [04:22]

I also owe a debt of praise and gratitude to the Secret Service Inspection Division for detecting the corrupted testimony, and handling the matter, and this investigation of perjury, in the most professional and expeditious of ways.  [04:40]

In fact, this whole episode unfolded very recently, and in a very short period of time.  [04:48]

Stewart, I expect, will be presented before a United States magistrate judge this afternoon on the charges.  [04:58]

And as we have said before, when we announced both the charges as well as the convictions in this case — in the Martha Stewart case, I should say — we have to hold all people to the same standard of integrity, and today we hold true to that committment.  [05:15]

Thank you.  [05:16]

The complaint of perjury was made by John Jordan, Special Agent with the United States Secret Service, and can be read in full here (originally at news.findlaw.com/cnn/docs/mstewart/uslstewart504cmp.pdf).  And the indictment can be read here (originally at www.perjury.us/~).

From the above, it would appear that everyone believed that Larry Stewart had perjured himself: Martha Stewart's attorneys, government attorneys prosecuting Martha Stewart who had relied on Larry Stewart's testimony, the Secret Service Inspection Division which complained of Larry Stewart's false testimony, five employees of the Secret Service whose testimony is cited in the indictment, and the Secret Service itself which suspended him indefinitely without pay.  And Larry Stewart knows he lied too — paragraph 17 on pp. 12-14 of the complaint describes Larry Stewart admitting the allegations against him.

Nevertheless, prosecutors subsequently failed to prove Larry Stewart's guilt to a jury, which the press is fond of misconstruing as proof of innocence.

NEW YORK, Oct. 5, 2004
Martha Ink Expert Didn't Lie
Jury Clears Scientist Larry Stewart In Perjury Case
By Bootie Cosgrove-Mather

(AP)  An ink expert was acquitted Tuesday of repeatedly lying on the stand at Martha Stewart's trial — a case the homemaking expert has seized on in her bid to have her conviction thrown out.

A federal jury deliberated about seven hours over two days before acquitting Secret Service scientist Larry Stewart — no relation to the millionaire homemaker — of two counts of perjury.  Stewart had faced 10 years in prison if convicted.

"Mr. Stewart, good luck to you," U.S. District Judge Denny Chin said.  The scientist appeared to lose color in his face as the verdict was read, and hugged his lawyers when it was complete.

With tears in his eyes, Stewart, 48, told reporters outside court that the ordeal had been "long, expensive and painful" and said it was difficult being on the other side of "a system I've worked for diligently for 25 years."

Federal prosecutors accused Stewart of exaggerating the role he played in ink analysis testing of a stock worksheet that was used as a piece of evidence against Martha Stewart and her stockbroker, Peter Bacanovic.

The pair were convicted in March of lying to investigators about why she sold stock in 2001.

Regardless of the verdict, the mere allegation that a prosecution expert witness was accused of lying on the stand is expected to figure prominently in appeals by both Martha Stewart and Bacanovic.

Both were allowed to remain free while they appealed.  Stewart decided instead to begin serving her five-month term, and must report by Friday to the minimum-security women's prison in rural Alderson, W.Va.

Federal prosecutors said all along they did not believe the perjury charges against Larry Stewart put the Martha Stewart verdict on shaky ground.

"I think we're on solid footing on that point," U.S. Attorney David Kelley said Tuesday after the verdict.

Brooke Morganstein, a spokeswoman for Martha Stewart, declined comment on the verdict.

In the trial, the government contended Susan Fortunato, who worked for Larry Stewart at the Secret Service, had actually conducted the tests in the summer of 2002 and January 2004.

Fortunato also testified that Stewart once confronted her after an office meeting and kissed her.  She complained to her supervisors but later withdrew the claim.

Jurors who spoke to reporters after delivering the verdict said they doubted Fortunato's credibility as a witness.

The trial offered a curious mix of scientific testimony about arcane ink testing methods and a detailed description of what the defense called a three-year feud between Fortunato and Stewart.

The defense noted that the pair, as well [as] at least one other Secret Service employee, regularly went to lunches where the conversation veered into sexually charged material.

Reflecting on the sordid testimony about Secret Service office politics, juror Faustino Galan said: "I hope that the division that protects the president runs much better than the document section."

Stewart has been on suspension from the Secret Service since he was indicted this year.  Asked whether he wanted his job back, he said: "Not that job.  I don't want it."


It is a disturbing disparity that of the two Larry Stewart misleading testimonies that we know of, he was arrested and tried for only the lesser.  His lesser misrepresentation merely took credit for laboratory work done by another.  But Martha Stewart was not on trial for her life, and Larry Stewart's misrepresentation did not increase her probability of conviction.  In contrast, John Demjanjuk had earlier been on trial for his life, and had come perilously close to being hanged.  No one could tell whether the 2001 Cleveland proceedings in which Larry Stewart submitted his bogus Counterclockwise Solution would lead to John Demjanjuk being extradited to a jurisdiction intent on executing him.  And Larry Stewart's falsehood in Cleveland was not the trivial one concerning who had actually conducted the research, it was the prejudicial one of misrepresenting exculpatory evidence as inculpatory.  And yet for the latter deception, the more serious deception, the deception that pushed John Demjanjuk toward the gallows, Larry Stewart has never been arrested or tried.

Perhaps it should not be surprising that Larry Stewart has come under complaint of perjury again, this time for his 2010 Munich testimony, though not for continuing to flog his sham Counterclockwise Solution, but for performing an about-face concerning the two holes in the Demjanjuk photograph (holes which have been discussed under the heading THE PHOTOGRAPH IS PUNCTURED BY A PAIR OF STAPLE HOLES in WHO GLUED THE DEMJANJUK PHOTOGRAPH TO TRAWNIKI CARD 1393?)

Investigation opened into Demjanjuk witness
By ANDREA M. JARACH (AP) – 17 Jun 2010

MUNICH — Munich prosecutors said Thursday they have opened an investigation into a former U.S. Secret Service forensics expert who testified at the John Demjanjuk trial following a motion from the defense accusing the witness of perjury.

Trial prosecutor Hans-Joachim Lutz told The Associated Press that his office was obliged to open the investigation against former agent Larry Stewart, who testified in Munich last week, after defense attorney Ulrich Busch filed a complaint with the court accusing him of perjury.

Busch argued Stewart's Munich testimony contradicted statements he had made in U.S. District Court in Ohio in 2001 — the year after he examined documents being used as evidence against Demjanjuk.

Lutz refused to comment on the possibility of charges being filed, saying the evidence first had to be examined.

Stewart rejected the allegation, telling the AP in a phone call from California that his two testimonies did not contradict one another.

"I was asked different questions this time, so I answered the questions I was asked," he said.

Demjanjuk, a retired Ohio autoworker who turned 90 in April, is standing trial on some 28,060 counts of accessory to murder on allegations he was a guard at the Nazi's Sobibor death camp in occupied Poland. He denies ever being at any camp, claiming he is the victim of mistaken identity.

But the prosecution argues, among other things, that a Nazi-era identity card has Demjanjuk's picture on it and indicates he was a guard at Sobibor.

Stewart, who analyzed the identity card and 21 other documents being used in the case in 2000 by the U.S. Department of Justice's Office of Special Investigations, told the court last week he found staple holes through the photograph but not through the card itself.

He told the Munich state court that indicates the photograph — which is now glued to the document — was once stapled to another piece of paper.

Stewart said, however, that did not necessarily mean it was a forgery, testifying that in his experience photographs often fell off wartime documents as they aged and were then stapled to separate pieces of paper in postwar archives.  He said the fact that there was no rust around the staple holes indicated that they were made after the war, when iron staples were no longer used.

But according to the 2001 testimony from Ohio, which Busch provided the court, Stewart said at that time he had not determined that the holes were from a staple, did not know whether there were similar holes in other service passes, and that he had not looked for holes in the identity card itself.

"I was looking at the ink, the paper, and the photograph, and the holes didn't have anything to do with that in my opinion," Stewart said, according to a transcript.

But Stewart said in the telephone interview that in preparation for the Munich case he went over his original findings again, and was also able to examine the original ID card again when he was asked about it in court.

"In this case they had the original card, and they had me come up to the judge and they asked me specific questions about the holes," he said.

Stewart added that he would be happy to return to Munich and answer any additional questions they had.

He testified in Munich over the objections of Demjanjuk's defense team because he had been charged with perjury in 2004 in the United States when he was an expert witness in an unrelated trial.  Busch argued that even though he was acquitted by a jury of those charges, it made his testimony suspect.

Stefan Schuenemann, an attorney for a Sobibor survivor who joined the trial as co-plaintiff as allowed under German law, said that if Busch found Stewart's testimony contradictory, he should have questioned him about it in court last week.

"He had the chance to ask him directly here about it, and he didn't," Schuenemann told AP.

Perjury can carry up to 5 years in prison in Germany, depending upon the severity of the case.

Associated Press Writer David Rising contributed to this story from Berlin.


As the allegation of perjury concerning holes in the Demjanjuk photograph arises from a possible discrepancy between his 2001 Cleveland testimony and his 2010 Munich testimony, we can begin to evaluate this latest perjury allegation by reading Stewart's 2001 statements:


20      Q.  All right.  I'll go on to something else.  Another
21      thing that you examined, sir, was the Government's Exhibit
22      3, correct?
23      A.  Yes, sir.
24      Q.  And you prepared a chart for us.  Do you remember
25      that?


 1      A.  That's correct.
 2      Q.  Now, when you examined it, did you notice that it had
 3      staple holes in it?
 4      A.  I noticed it had holes.  I did not determine they
 5      were from a stapler, no.
 6      Q.  Were there holes of that character in any other
 7      service passes that you examined?
 8      A.  I don't know.
 9      Q.  You never looked for that?
10      A.  It was not material to my examination, so I did not
11      choose to look for holes in the document, no.
12      Q.  Why wasn't it material to your examination?
13      A.  I was looking at the ink, the paper, and the
14      photograph, and the holes didn't have anything to do with
15      that in my opinion.
16      Q.  Didn't you tell us a little while ago, sir, that you
17      were free to conduct any tests that you thought were
18      necessary?
19      A.  Yes, sir.
20      Q.  Would you agree with me that over time there's been
21      some question about that photograph on that card?  Right?
22      A.  I had a question about the photograph on the card, so
23      yes.
24      Q.  Would you agree with me that whenever it is suspected
25      that pages have been substituted in a document bound with


 1      staples a careful examination of all staple holes in each
 2      sheet should be made?  Would you agree with that statement?
 3      A.  If it's pertinent to the examination, yes.  It is not
 4      pertinent to determining whether or not that photograph was
 5      originally on that card, and they both date back to the
 6      1940s.
 7      Q.  May I ask you, sir, what I just said, the beginning
 8      is "Whenever," that's what your expert, Mr. Hilton says,
 9      "Whenever it is suspected"; that's what I read, right?
10      A.  That's what your expert has said in this book.
11      Q.  Right.  And not only is he my expert, he's your
12      expert, isn't he, sir?
13      A.  He is a person that is widely regarded for his
14      ability in that field, but I believe we are taking some
15      things out of context here.  Again, if I can direct you
16      to —
17      Q.  I'm going to give you the book and I'm going to —
18      excuse me, Your Honor?
19      THE COURT:  We are getting way far afield
20      here.
21      MR. DRIMMER:  Thank you, Your Honor.
22      MR. TIGAR:  Thank you, Your Honor.

And the question of staple holes is reverted to only to underline the conclusion that no one seemed interested in having staple holes examined, either in the Demajanjuk Card 1393 or in any other card:


19      Q.  Did the defense ask you to examine any staple holes
20      on Government's Exhibit 3?
21      A.  No, they did not.
22      Q.  Did the defense ask you to examine any staple holes
23      in the photograph of any other service passes?
24      A.  No, they did not.

To summarize, Larry Stewart's 2001 Cleveland position was that no one had asked him to examine holes, and he did not do so on his own initiative because he considered holes to be irrelevant, and he never considered the question of whether the holes in the Demjanjuk photo were staple holes, and so could not say that they were, and it goes without saying that he never examined the holes for rust that might have come from a staple, and he was so uninterested in holes generally that he did not notice whether any other documents that he examined had holes, and so that he had nothing to say about holes one way or the other.

Now if in his 2010 Munich testimony Larry Stewart speaks of the holes as if he knows them to be staple holes, and if he is ready to talk about them because he sees their relevance, and if he now remembers that the holes through the photograph did not extend through the card as well, and if he now understands that the photograph had been stapled to some piece of paper because it had fallen off the card, and if he now speaks as if he had examined the holes for rust but found none — then this does seem like radically different testimony.

If Larry Stewart led the Munich court to believe that his revised 2010 conclusions concerning holes resulted from laboratory analyses that he had performed in preparation for his 2001 testimony, then perhaps he is guilty of perjury.  If, however, he had made it clear in Munich that his revised views on holes resulted not from laboratory analyses that he had performed, or that anybody anwhere had performed, but rather from watching CNN and from having glanced at Card 1393 in the Munich courtroom, then perhaps the perjury allegation should be withdrawn and he should be commended for his frankness instead — his frankness in admitting that he had abandoned giving expert testimony and had relapsed into passing along hunches.

Of pivotal significance to the question of this latest accusation of perjury might be the 17 Jun 2010 sentence written by reporter Jarach which seems to impute to Stewart the claim that he examined the holes for rust, but found none: "He said the fact that there was no rust around the staple holes indicated that they were made after the war, when iron staples were no longer used."  But does a forensic expert say "there was no rust" unless he has performed a laboratory test for rust?  Does not rust sometimes come in minute quantities that may be invisible to the naked eye and that require laboratory analysis to detect?  By analogy, if a forensic expert says "there were no fingerprints", is it not fair to conclude that he intends us to understand that he performed laboratory tests which discovered no fingerprints?