Ace of Spades HQ

« Even Zogby, Now: Bush Up by 4 | Main | AP Poll: 5 Points »

September 10, 2004

Dispensing (?) of Three Nits

There are three arguments made by those who wish to claim these documents are genuine:

1) There is a superscripted "th" in a contemporary document in Bush's record that we know for a fact is genuine.

This just isn't really true, even though Josh Marshall says it is.

The document in question is here. It's on page 3, a page that is vertically-aligned (i.e., up and down, the wrong way; you have to flip it to read it properly). The
"superscripted th" which Dan Rather will presumably be claiming proves the genuineness of his documents later tonight can be found on the second line, beginning "4Sep68." Scan along that line and you will find the alleged "superscripted th" in the ordinal "111th."

First of all, that character is not clearly anything. Look closely. It's a badly-distorted character. What it actually looks like is a square with a vertical line through it.

Could it be a "superscripted th"? Yes, it could, except independent experts keep telling us that such a character was not available on typewriters.

It could just be a folded-up bit of paper. It could be some sort of distortion in the photocopying process.

What else could it be? For crying out loud, it could be typo. The typist might have typed up the "111th" as "111t" or "111h," omitting one character, and leaving no room to squeeze it in without retyping the whole line; so he might have whited-out the character and drawn in a little "th" by hand.

Is that "reaching"? Well, it would be "reaching," I suppose, if we were just making this business up about the superscript th. But actual experts who deal with questioned-documents day in, day out, and who have entire bibles of available typefaces, tell us that this is either an extremely rare feature or something simply not available.

So, we can either decide to junk 50 or so years of document-verification science, or we can decide that maybe that unclear character is hand-drawn or some distortion caused by age and repeated photocopying.

2. When Little Green Footballs created his Word 97 version of one memo, he had to shrink it (or expand it) slightly to get it to overlay.

This one's pretty simple. The forged document has obviously been photocopied and re-photocopied a whole bunch of times to make it appear aged.

I'm not a photocopying expert, but I imagine that the platten (glass) on a photocopier is intended to lie a precise distance from whatever device is snapping a picture of the document. If that platten is just a little bit too far from the photocopying device, it will produce a slightly smaller reproduction. If it's a little too close, it will produce a slightly smaller reproduction.

You probably couldn't notice this if you took one photocopy, but if you repeatedly photocopied a document -- photocopying the new photocopy each time, as a forger would do -- that .1% difference would begin to multiply through successive runs through the copier. If a copy is only 99.9% the size of the original for a first-generation document, it becomes more than 1% smaller by the tenth generation (you just keep multiplying .999 by .999 ten times).

A .1% difference isn't noticeable; a 1% difference is quite noticeable, particularly on close (one laid over the other) inspection.

3. The White House did not declare the documents to be hoaxes when they released their copies (obtained from CBSNews) to the press.

Well, there are four reasons they might not have raised objections:

a) Because Bush knows the documents are real.

This is Dan Rather's interpretation, we must assume.

b) Because Bush knows the documents are forgeries, but the "story" the documents tell -- Bush was ordered to undergo a physical, which he refused -- is true. In other words, the forger framed a guilty man. Bush didn't object to the veracity of the documents because he knows he's actually guilty of the charges they suggest. He can't quite deny the validity of the documents without having to answer tricky questions about the "facts" they allege. So he stays mum.

This is Josh Marshall's hope.

c) Because Bush had no idea if the documents were authentic or not, particularly since most of them were memos to files that he would never have been shown. How the heck does the press expect Bush to know with certainty whether or not Killian's "secret personal files" contained derogatory information about him? They were, you know, "secret" and "personal."

The document ordering him to take a physical is trickier, but it's possible Bush thought this might just be some sort of miscommunication. He thought he'd been granted an exemption due to his non-flying status by one officer; now he was confronted with a document he didn't remember from Killian telling him to take a physical. He might be genuinely unsure if this document were real or forged, and yet still be sure he did what his commanding officer told him to do.

d) Karl Rove and the White House knew damn well they documents were forgeries -- how could they not? -- and they let CBS and Kerry swing on their own rope.

I don't say this is necessarily the most likely -- it's a risky maneuver; how do you know that the press will actually admit the docs are forgeries when evidence is presented -- but it does seem to be a live possibility. From the looks of things, these were extremely crude forgeries, so they would have a high degree of confidence they could prove their case after Rather hanged himself.

I honestly can't say which is true. Obviously, I don't want a) or b) to be true, but they could be, for all I know. My evidence that they're probably not true is my assumption that the documents are, in fact, forgeries-- can I direct your attention once again to the smoking gun?

I don't expect Dan Rather will even address the smoking gun tonight, except obliquely.

At any rate: we begin from the likelihood that these documents are forgeries. That's not some self-pleasing wild-eyed partisan surmise; that seems to be what all of the real experts are saying-- these documents are, to close to a 100% certainty, fake.

There's no need to go repealing years of document-authentication science and everything that's known about common 1970's typewriters in order to explain an unclear character on a single document. Yes, it could be that all of the experts are wrong and that superscripted-th's were common in typewriters.

Or it could just be a written-over white-out.

If the latter seems to easy, it also seems too hard to go abandoning all we know about document-authentication to explain one stray smudge.

And there are are many problems with the documents besides supescripted th's. There are smart quotes; there is proportional horizontal spacing;
there are proportional vertical spacing, again only available on word-processors; there is kerning; there is a dubious provenance, since the records did not come from the Pentagon or the man's family but some anonymous operative.

How exactly did that supposedly work, by the way? Did Killian take these papers from his base-files, where we'd imagine they're supposed to be, thus protecting Bush from exposure from them, but then give them to a secret friend (not his wife or son) with instructions, "Should I die, and, years later, should Lieutenant 'Sugar-Coat' Bush ever be up for re-election as President (not simple election; I'll give him four years-- gratis), then I want you, my good secret friend, to anonymously leak my private papers to the press and expose him for the chimpresident I expect he will be"?

And, oh yeah, there's the tiny little problem it's an identical copy of a document typed up in Windows 97 just yesterday.

posted by Ace at 05:06 PM

Comments

Has anyone contact Buck Staudt on these documents, as he is the one who is alledgedly putting on pressure?

Was he even still in the service in 1973 when the memos were written?

I believe he retired in 1972.

How could he still be pressuring Colonels a year after he retired?

Posted by: quasi on September 10, 2004 05:29 PM

As I understand it, there are four documents, two of which seem highly suspect.

So it's not out of the realm of possibility that not all the documents are fake--better to mix at least one authentic document in there to lead credence to the other two.

Which would make it all the more likely Bush and the White House would not deny them. That is, you show four docs to Bush, you say these are all secret docs from Killian's stash, and what does the President have to say for himself.

He might recognize the one and not the others, but certainly he's not going to assume a major news organization is trying to pull a Columbo-type trick on him.

We can be pretty sure that Bush's record wasn't perfect at this time. So it's not like he could be 100 percent sure his former superior would not have anything bad to say about him.

Posted by: Nicholas Kronos on September 10, 2004 05:30 PM

I've been trying to find out info about "th" on typewriters all day. The point isn't that "th" was or wasn't absolutely unavailable on typewriters in the late 60s. The point is that, to create these memos, it had to have been available on a typewriter capable of producing a font that looks like Times New Roman. The only machine I've heard of that might be able to do that is the IBM Selectric Composer, and for various reasons described on my blog, that just doesn't look likely. Bill at INDC also talked to his document expert this afternoon and he's hearing the same thing. It may be possible on the Composer, but it's so unlikely in this case as to be nearly impossible.

Posted by: Allah on September 10, 2004 05:36 PM

Sorry but the need to uniformly reduce or expand the LGF copy to match the "original" is bogus. As long as you can register four corners and the relationship between letters and words is the same that's all that matters. I do this kinda crap all the time in my work (tv graphics). Even a pixel off between letters and words when you're trying to match over something you did earlier and the difference jumps out at you. That's why originally the misplaced TH jumped out at me. But there ain't no way in hell you can tell me that you can resize a field of that many words and have them drop right on top of the others if they aren't from the same source. I fake crap all the time for TV ads (especially during politcal seasons with re making newspaper endorsements) believe me, you never get that freaking lucky.

Posted by: Paul B. on September 10, 2004 05:43 PM

Regard to the alleged superscript th referred to in the linked document, I would want to know if it is indeed superscripted in that one instance, why are not all other references to 111th in that document not superscripted. There are at least two other 111th written in that document that are not superscripted.

Posted by: Dman on September 10, 2004 05:48 PM

notice that there are several other instances of 111th where the th arent superscripted. Why the discrepancy?

Posted by: Big E on September 10, 2004 05:50 PM

PaulB,

Thanks, but I don't think I really know what you're talking about. You mean the "problem" LGF experienced was just an on-screen problem that wouldn't show up in a printed document?

Dman,

Well, they could have just NOT used the special key in those instances. Or they could have been using different typewriters at different times.

Look, a tiny bead of water on a photocopier platten can cause a character or two to contract and appear smaller or expand and appear larger, depending on what side of the water-lens you're looking through.

I just don't buy that any of these explanations are so much harder to buy than the one being offered by Dan the Document Man.

Posted by: ace on September 10, 2004 05:53 PM

If you go down two lines there's 111th again but without the superscript. And the next line and two lines down from that. You mean to tell me they had the mostest super fanciest typewriter in the USAF and they only weilded their mighty superscripting ability once?!

As I've said in previous posts, please. Something probably just went screwy in that line.

Posted by: Ken J on September 10, 2004 05:54 PM

And another big point here is that .pdf file is not written in Times New Roman. It's in courier or elite or one of whatever your two choices of typewriter ball were for that time period.

So even if they had the power to do a superscript in a standard Courier font...which that little blur doesn't prove...that means nothing about their ability to produce a superscript th in the then nonexistent-or at least exotic-Times New Roman font.

Next they'll be telling us that since DARPA invented the internet in 1969, then Scrappleface's e-mail could be for real, since that technology did in fact exist at the time.
But at some point all this evidence is just going to slice like an Occam's razor.

Posted by: See-Dubya on September 10, 2004 05:55 PM

The (th)'s are not matched at LGF because of the difference between how Word displays and and prints the superscripts (they print with a 3 point baseline shift it looks like).

The resize issue that the Kos man and others keep harping on is irrelevant because Charles at LGF only adjusted the total dimensions of all the words, he didn't resize indivual words and fiddle them around. The example I can think of is like facial recognition, regardless of how big or small one picture of your face is from another it's still you because the proportions between your nose and eyes remain the same. That's what I'm saying the proportions between the letters and words are all the same once you register the four corners of the text.

I'm saying, just from experience, if I have a printed document and I'm trying to match that printed document in Photoshop to "clean it up" even if I type the same words in the same font, my kerning and tracking (the proportions between the letters and words) are going to be different if the document I'm reproducing was not typeset in photoshop as well. So I can squeeze and stretch my copy all I like, its the relationship that's always going to be wrong. LGF shows the relationships are identical. No way that happens on accident.

Posted by: Paul B. on September 10, 2004 06:01 PM

The Lefties who are telling us that these nit-picky details prove the veracity of these documents are like the pet shop owner trying to tell us that the parrot is not dead. Sorry, but this is an ex-parrot.

Posted by: Raoul Ortega on September 10, 2004 06:03 PM

Dman and Big E:

Because, silly-- the person who typed this obviously replaced the typewriter ball to get the superscript "th" on the first reference, and then realized how stupid it was to switch out the ball each time they wanted to type 111th.

The other references are just the typist being lazy, that's all.

(Sarcasm is such fun. . .)

Cheers,
Dave
Garfield Ridge

Posted by: Dave on September 10, 2004 06:04 PM

Nit #3-- I prefer explanation e) Whatever.

Bush or Rove or Card or anyone else high up never saw these. Some low level staffer is under instructions to take any of these documents that come in and make them available immediately. Because to the Bush Admin/Campaign, they don't care about this subject and believe that since there's nothing incriminating or damaging in this stuff, they aren't going to worry about it.

Posted by: Raoul Ortega on September 10, 2004 06:11 PM

This resizing thing is even less an issue than y'all are making it out to be. Charles at LGF is using a *screenshot*. The documents are presumably scans of paper documents. If everybody's using the same screen resolution, both Word and Acrobat are set to scale the documents at 100%, and the scans were made at a whole multiple of Charle's monitor's DPI, then maybe you wouldn't have to resize it.

That wouldn't explain the kerning or centering, however. Whoever pulled that off with a Selectric should get a freakin' medal.

Posted by: Rube on September 10, 2004 06:49 PM

I'm with Raoul. Like a good defense lawyer, Rove knows where GWB is vulnerable. This AWOL from the Guard stuff didn't bite them in 2000, didn't bite them earlier this year, and they know it doesn't have the potential to bite them.

The Dems and the MSM are "crying wolf" on this. They look sillier every time they bring it up, and GWB's team is just helping them do so.

Posted by: Kirk on September 10, 2004 07:17 PM

Here's more on the little th and the selectric typewriter. Seems it can't make things smaller; the superscripting only goes up or down but stays the same size. So whatever that little th-looking blob is, it's not a selectric supercript th, like appeared in the forgery.

Via Allah: http://defeatjohnjohn.com/2004/09/10000-part-two-ibm-selectric.htm

Posted by: See-Dub on September 10, 2004 07:38 PM

"I'm not a photocopying expert, but I imagine that the platten (glass) on a photocopier is intended to lie a precise distance from whatever device is snapping a picture of the document. If that platten is just a little bit too far from the photocopying device, it will produce a slightly smaller reproduction. If it's a little too close, it will produce a slightly smaller reproduction." - Ace

I'm not an expert, and it's been a loooonnnnggg time since I worked professionally in printing, so I'm not going to swear that my recollection on this is correct. But, IIRC, at one time photocopiers didn't do an exact 1:1 reproduction of the copied documents - it was more like 1:0.998 or so. You would get error creep repeatedly copying the copies over and over, unlike a photostat which did do a 1:1 when calibrated properly.

I have no idea, without doing the research, if that's still true of modern photocopiers - I work exclusively with computers and printers these days, and have since '93 or so. Someone who works currently in the printing and photorepographics industry might be able to tell you.

Also, having no idea how they did the copying - wether it was copying one doc over and over, or copying the copies - that question is impossible to answer definitively for me.

Posted by: Ironbear on September 10, 2004 07:40 PM

Or they could have deliberately reproduced it at 99% on a modern copier. Not that they would all of a sudden start doing something clever to hide their tracks.

Posted by: See-Dubya on September 10, 2004 07:58 PM

I was amazed: NPR actually ran a story on this today. But worry not, the world as we know it is not ending. They ended the segment with someone--I can't remember if it was a Democrat or a CBS guy--questioning the timing of the debunking of the story. He noted that the blogs started picking apart the story almost the moment it aired. He then questioned how one could know enough about the documents to try to debunk them after only seeing a few breif clips on TV....

I guess the implication being that people knew these documents were fake, and this was all some sort of set-up to smear Bush's opposition. Not that he tried to refute any of the debunking, but he said it brought up questions.

Giving this guy the benefit of the doubt, the big question would be how CBS could MISS these forgeries?! It means either CBS is incompetant, or else they are in on the whole 'scheme.' And does anyone in their right mind believe that Dan Rather is part of the Vast Right Wing Conspiracy (TM) and is willing to sacrifice his credibility just to re-elect Bush? So the guy's only real implication is that this was a set-up, and CBS was too stupid to notice. Not exactly the line of reasoning I'd want to bring up if I were part of the loyal opposition...

Posted by: Charlie on September 10, 2004 08:18 PM

IronBear,

You're absolutely right about the sizing, and it seems that photocopiers haven't changed. I just photocopied my own printout of the memo ten times, and the text got noticeably bigger.

More at http://sobekpundit.blogspot.com/2004/09/forgeries-update.html

Posted by: Sobek on September 10, 2004 09:30 PM

I vote with Raoul on scenario E--that the Prexy and the evil wizard Rove aren't bother with whatever shit CBS faxes over.

I believe fotocopiers *must* not reproduce items exactly to their original dimensions, as a means of defeating quickie counterfeiting for vending machines. That's what I heard, but it may have been on CBS, so take it for what it's worth.

Cordially...

Posted by: Rick on September 11, 2004 12:39 AM

poker me up

Posted by: poker me up on December 29, 2004 02:22 PM