From your link: “Here are the two documents superimposed—and please note that when I cut and pasted the images, I just eyeballed it with no special effort to match sizes:”
The graphic is too small to tell much. However, where is the smudge where the second m in ‘Memo’ of the original document goes below the baseline? Where is the mess where the e in ‘period’ touches both the p and the r? Where is the messed up, squarish f in ‘from Grp’? If the overlay was created as the website author claims, those defects should all be visible. They aren’t.
What sort of font package would your supposed forger have to use to create all the typewriteresque failings in the CBS originals?
If they’re going to cry forgery, they have to do a hell of a lot better than 72 (or 86 or whatever) dpi overlays to convince me. Here’s a link to the CBS originals, pdf’s listed in the box on the left of the page, if anyone else wants to look at them.
I’m curious Sam, why you think Mr. Bush and the white house have gone along with these memos, if they know them to be forgeries? Surely Mr. Bush knows what happened back then, and would recognize lies if he saw them. Why hasn’t the president denounced the memos?
Squink: I took your link, opened the memo in PDF format, took a screenshot, and copied it into Adobe Photoshop.
Then I opened word, and typed the memo in. I left Word at the default settings, and just typed, letting word wrap the text where it saw fit.
Then I took a screenshot of that, and pasted it as a new layer in Photoshop. Then I scaled and moved the two layers so that the CYA matched perfectly.
I reproduced the results as shown on LGF. It was a bit of a hassle to get the scale exactly right (I expected the scale to be off since I was looking at a PDF). The ‘th’ doesn’t match exactly, and a couple of other letters look a tiny bit off, but the margins and everything are exact. The date is EXACTLY 8 tab stops in word. There’s not even a thousandth of an inch difference.
The guy on LGF says that the slight differences in the ‘th’ and a couple of other characters are due to the difference between screen and printer fonts. He says that the differences vanish completely when you compare the printed documents.
Well, until you pull out an IBM typewriter of the same vintage as the one from the document’s provenance, type a document, and then perform the same test and show that they don’t match, you’ve proven pretty much nothing.
Sam, shouldn’t you expect that the standard margins, font styles, character size and tab size in Word are modelled on the standard typewriter styles? Given this, shouldn’t just about any document produced using the standard settings on modern-ish typewriter match a document produced using the standard settings on Word?
The reason the superscript “th” in Killian’s version diverges from the Word version could be because typewriters often had a superscript “th” key, as well as other handy characters like ®, ™ etc.
Why would you fake something like this by typing it in Word, if that is so easy to expose as a forgery? If I were John Kerry’s garbologist, and I had a devious plan to take down George Bush, I’d go find an old typewriter in the attic and produce the memos on that, making sure to make a lot of typewriter-style errors.
Conversely, if the other tinfoil-hat version is true, if I were George Bush’s garbologist trying to frame John Kerry for framing George Bush, I’d make sure that the forgery of the memos was beyond debate. I’d accidentally include things that were definitely unable to be produced by typewriters of the time, perhaps a certain character or two that definitely weren’t on 70s typewriters.
It’s most likely that these documents are exactly what they purport to be - documents typed by Killian, or his secretary, on a 70s typewriter.
And, no, I’m not at all mystified that the font metrics of Times New Roman and an IBM typewriter are very similar. Font metrics of some sorts are all very similar and inherit from classic designs. Something you, Sam, should be more than a little familiar with, and a bit less credulous about.
I’m also not very impressed that tab stops in Word and a typewriter are the same. They’re both based on a mysterious measure called the “inch” perhaps you’ve heard of it?
In fact, the more I consider this, the more I’m just…astonished at this line of attack on the documents.
“Oh my gosh, computer output looks just like a printed document!”
Um… wasn’t that exactly the aim of the vaunted “Desktop Publishing” revolution 10-15 years ago? To produce documents that look exactly like ones produced on a typewriter and better equipment?
I dunno man. There’s a lot of different typewriters and a lot of different fonts out there. Just about everybody sets there tabs up differently and formats differently… or they did until word processers came around. Even if you were trying to follow a pattern, it’s difficult to set the tab on a manual typewriter so that it’s in exactly the same spot.
Try it sometime. Type a paper in a typewriter and make a note of where you set your tabs and how many lines you spaced. Then, clear all your tabs and try to duplicate the paper you just typed. Now put the two papers together, square them and see if they match exactly.
No, the pitch was “at least as good” and “much better than those awful dot-matrix printers”.
Scylla, I don’t own a typewriter, but I do know that the IBM types held to be the authored machine was held up as the typeface/formatting standard for a typewriter of that era, and for a computer program to emulate those metrics seems not at all improbable.
Well, this half-brained guy has done his research on dates that fit the current records, who was in Bush’s chain of command, Air National Guard regulations, abbreviations etc, practices such as different types of memos (to file, to self etc). After this meticulous research, I find it hard to believe that he’d screw up by typing the thing in Word!
No, I don’t think it’s at all likely that these fonts would match exactly. We’re not just talking about font shapes, but kerning styles, leading, etc. Times New Roman did not even exist as a font 30 years ago. Times Roman did, but Times New Roman is slightly different. Even among different computers you often find differences in exactly how Times New Roman is rendered.
Now, I’m not saying it’s flatly impossible. I’m saying it seems to me to be highly unlikely, for several reasons. For one, every other military document I’ve seen from that era was printed on a plain old monospace typewriter. Proportional typewriters were hard to use and very expensive. Second, while those old Selectrics could do superscript and subscript, I don’t know that I’ve ever seen a general use typewriter that had a “th” key. Maybe they had them, but I’ve never seen one.
Finally, given that the old typewriters were mechanical, I would expect some slight differences in character spacing just from mechanical slop. These images line up perfectly.
As for tab stops, I don’t know the answer. I don’t know if there was a standard for Tab stops that might still be in play today.
All in all, it doesn’t seem very likely to me that these documents were made in the early 1970’s. But I’m open to new evidence. Can anyone find a similar typed document anywhere from that time that exhibits similar features?
Finally, the Weekly Standard consulted three of what they claim to be the most highly respected experts in typography and typewriters, and all three felt that these documents were forgeries. INDC Journal also contacted an expert in forensic document examination, and here’s what he said:
Squink, thank you for the links to the CBS originals. You’re correct about the typewriteresque mistakes. There are numerous instances of vertical misalignments, pieces of letters missing (probably due to dirt on the ribbon or page or dry spots on the ribbon), etc. I suppose one could hypothesize some diabolical font designed to emulate typewrite failings, but even that wouldn’t explain these documents. If that was the case then a given letter would show the same mistakes consistently and that isn’t what I see on those PDFs. I suppose a very clever programmer could write a printer driver that would somehow randomly emulate such mistakes but if you’re going to go to all that trouble then why not just use a friggin typewriter? If these are forgeries, then it sure as hell seems like they must have been forged on a typewriter.
The question is, does CBS have originals or not? If they do have originals then I’m sure that there are very easy ways to determine for certain whether they were created on a modern printer or on a typewriter. Chemical analyses of the ink for example.
I noticed that on that INDC Journal page they make a big thing about the superscript “th” which they say occurs at the end of dates. In the CBS documents I could only find two instances of a superscripted “th”. Once, in the May 4 physcial examination order where it is used in “111[sup]th[/sup]”; and once in the August 18 CYA memo where it’s used in “187[sup]th[/sup]”. I can find other instances of “th” being used in the same manner, but not superscripted, in those and other documents. I cannot find even one incidence of “th” superscripted or not, being used in a date. What documents are they looking at?
Personally, I think that if matches up perfectly with Word, that’s too much of a coincidence to be ignored. Is it “possible” that it could have come from a mechanical device and still match? Sure. But let’s face it, what are the chances?
I would hope that logically, most people would accept that.
It does not absolutely destroy the possibility that the documents might be genuine though.
A lot of times when people scan old documents, they run them through photoshop or Word to clean up the noise and make them clearer and more legible. Sometimes they have their software set up that way so it does it by default. That might alter or standardize fonts and formatting.
I’m not sure how well he’s done his homework. The date is listed as ‘18 August 1973’. Standard military format was more like "18 AUG 1973’. But then, this is a personal memo, so he wouldn’t have to use any sort of format, I suppose. I’m just curious what makes you think all the terminology is accurate? As for knowing the names of the people involved, wouldn’t anyone who’s read one of the anti-Bush Guard books know those kinds of details?
Anyway, I assume we’ll find out in a day or two. Maybe they aren’t forgeries, but I’m very surprised that a 30 year old document would match a modern word processed document so exactly. Maybe Microsoft is a real stickler for details and one of their design parameters for word is that it had to match the Selectric exactly. I suppose that could be the case.