Take names. The memos are not forgeries.

[girl gone mad]

I've spent the last two days poring over all available evidence, perhaps out of boredom, perhaps scientific curiousity.

Where I had been on the fence, I'm now convinced these documents were created on a typewriter, not a word processor (and certainly not using MicroSoft Word). Thanks to intrepid bloggers and DUers, I think we have essentially narrowed the list of possible typewriters, with the most likely candidate being the IBM Selectric Composer.

The images below, combined with comments I have read from people who profess to have used the Selectric Composer and/or similar typewriters have all but convinced me.





These images actually come from a web log attempting to debunk the CBS memos. The blogger claims that the fact the font does not line up evenly is proof enough that the Composer could not have been used.

To my naked eye, the font is a perfect match. The discrepencies on the second line can easily be accounted for by the fact that the typist left out a comma after the word 'May'.

An experienced typist posting on Slashdot suggested that the unevenness of the text along the baseline in the memos indicates the typewriter used was in need of maintenance. The composer needed to be serviced on a fairly regular basis to produce the kind of picture perfect results we are accustomed to with word processors. I believe that the discrepencies on the final two lines can be accounted for by considering the subtle variations that might result om machines that were not equally maintained.


I have attempted to reproduce these memos using 3 different versions of MicroSoft Word, but so far, none of my reproductions matches up as closely as what we see above. In my analysis, the image displayed on Little Green Footballs web site was probably created through a process of stretching the Word document, then cutting and pasting parts, like the dateline. MS Times New Roman font is not a match with the font in the memo.

Now, I think it's time for us to take names. Hold the people that are pushing this dishonest diatribe accountable. Not just the bloggers and Drudge, but also those within the mainstream media who chose to run with the forgery angle before all of the facts were in.

[RobertSeattle]

1. Was Killian a full time TANG member or did he have a civilian job?

Wouldn't be "lovely" if it turned out he had a civilian job as say a....IBM salesman!

[JI7]

2. you are doing a great job with this

we should be really proud of the hard work some of the DUers such as yourself and walt starr(with Bush medal investigation)are doing. a much better job than many who claim to be professionals do.

[Skittles]

3. doesn't matter, girl

the Bush corporate media has done its job and the word FORGERY is firmly planted in the minds of the sheep.

[girl gone mad]

5. You may be right..

But I just don't feel like letting them get away with it so easily.

[lasttrip]

45. great job.

[librechik]

9. only because every talking head makes sure to mention that too

in the lead. Even Steve on Washington Journal. "The documents, wwhich some experts say may be FORGERIES, are still pathetically defended by that socialist Dan Rather and his commie buddies at CBS news."

Just like that!

[knight_of_the_star]

17. So we play with that

That Bush used FORGERIES to get us into the war. Those documents are more real than the intell "report" on the uranium from Niger.

[girl gone mad]

4. I wanted to add something I wrote earlier:

Here is a list of all of the things that the forensic document "experts" have been wrong about so far:

Experts: Variable spacing fonts were not available on the commercial market until the second and third generation word processors of the mid 1980s.
Reality: Variable spaced fonts existed on typewriters as early as the 1940s and were remarkably common by the 1970s.

Experts: No typewriter from the early 70s could produce a superscripted 'th'.
Reality: At least 9 typewriter models available at the time were capable of producing that superscript. At least two documents in Bush's official records contain that superscript.

Experts: The font in the memos was invented by Microsoft in the 90s.
Reality: Times Roman font was created in the 30s, and the font in the memos is not even a precise match with MS Times New Roman.

Experts: The memo lines up perfectly with a document created in MS Word, proving that they were created on a word processor.
Reality: The images created line up only after a considerable amount of skewing (and likely cut and paste) is performed, and even then the match is not perfect.

Experts: The memos show evidence of 'kerning', not possible in a 1970s era typewriter, therefore they must have been created in Word.
Reality: Kerning is turned off in Word by default. Turning it on produces a document even less similar to the memos.

Experts: curly commas were not available before the invention of word processors.
Reality: Several typewriter fonts available in the 1970s used curly commas and curly apostrophes.

Experts: The memos used 13 point line spacing, not available on any typewriter.
Reality: The Selectric Composer was capable of producing 13 point line spacing. Also, it would be difficult to determine the exact line spacing of the memos based solely on the low quality images available on the internet.

[girl gone mad]

6. There are two new memos..

in this USA Today link.

I am not sure where they came from or when they surfaced. Someone posted the link in a thread last night.

http://www.usatoday.com/news/nation/2004-09-09bushdocs.pdf

[Mr_Spock]

7. Did you notice that the IBM Selectric Composer site person DOES NOT

believe that the memo's were likely made on that machine due to it's being a very specialized machine in it's day.

Just an FYI as you didn't mention that.

[meti57b]

8. IBM Selectric: After getting my first "paying" job the first thing I spent

money on was a good typewriter for my own use at home. As an engineer, typing was not something we did ourselves in those days and my typing skills were marginal. I think it was 1970 that I purchased an IBM electric typewriter and it had this "type-ball" sort of a thing with the letters on it. I really liked that since as a marginal typist, it eliminated that annoyance of getting the keys crossed and stuck from hitting two keys at the same time.

I think that typewriter was the IBM "Selectric" and it had proportional spacing.

With regards to the Killian family saying Killian never wrote notes to himself but kept the info in his head, ... we have always written those "CYA" memos to the file in my line of work. They are most often written when you are ordered to do something by upper management that you don't agree with ... but that you still carry out, or do something that may be considered controversial, or do something that upper management may later claim they never ordered you to do. These memos are especially written when you take or are preparing to take any kind of a personnel action against a subordinate who is not performing his job to required standards.

[girl gone mad]

14. I did read his comments after you pointed that out.

I'd like to see some hard evidence on this issue, showing how many IBM sold and who was buying. IBM leased typewriters, and institutions could have gotten good rates, making the cost not as prohibitive as people might think.

From first hand accounts of people in the military and military historians, I have gotten a picture of how day to day paperwork was handled. I can't say whether this process applies to the National Guard, however.

Memos would have been hand written or dictated, then given to a secretary or a typing pool. The military often used heavy duty typewriters because many document required several carbon copies. At least at the upper levels, and especially during times of war, carbon copies of almost all documents were archived.

While the author of the selectric site might consider the Composer to be a complicated machine, would a good typist of that day, someone who spent most of her 8 hour day using the machine, have felt the same? Would the ability to type an original and 3 or 4 carbon copies at the same time trump what might be a difficult learning curve?

I might argue that if we were to travel back in time with a modern computer and try to teach a 1970s typist how to use it, she would probably have a very difficult time of it for a while, though it's second nature to most of us by now.

[hexola]

19. but not before you posted it...

The page makes a pretty convincing case for it not being the Composer.

The whole approach to proving the authenticity of these documents has revolved around producing evidence that IBM typewriters could and would do everything that Microsoft Word does...

Isn't there some evidence of human-ness? Common typing practices/hacks?

[yardwork]

33. The major evidence of human-ness is that the characters jump around

on the horizontal line. Print out one of the CBS documents and hold any straight edge along the bottom of any line. You'll see that the letteres and numbers are unevenly spaced - slightly above or below the horizontal line.

Now hold a straight edge to any document produced in Microsoft Word and printed off the computer. The characters are perfectly alligned on the horizontal lines.

The so-called expert who claims she would testify in court that these documents were produced using Microsoft Word would have a fun time explaining that to any competent cross-examining attorney. I'd like to see that!

[Mr_Spock]

46. Yes, the horizonal uneveness proves beyond any doubt these were typed

There is no doubt in my mind that these memos were typed on a typewriter - a good one, but no word processor could reproduce that uneveness of the horizontal line.

[gollygee]

42. I see evidence of human-ness (m)

if you look at the whole documents - particularly the one that was posted on one of these threads yesterday - the "th" is SOMETIMES but not every time superscripted. On Word, the default is to have every th after a number superscripted every time - you don't select to just do it in one or two cases. Also, the "st" in 1st. Lt. George W. Bush is not superscripted. A superscript "st" wasn't available on the old Selectric typewriter I used to use, but the superscript "th" was. MS Word superscripts "st" in 1st.

[girl gone mad]

35. oops.. I'm reposting this below.

I put it in the wrong place.

[librechik]

10. the main thing is typewriters leave a pressure imprint on the paper

the assertion that the forgery was done in Word is either absurd or knowingly false. Without examining the orginals, no expert could say for sure that it was not done on a typewriter. Evidence from the text, such as not perfectly matched at the baseline and slightly crooked letters also point to a typewriter.

Also yesterday I read a remark from an expert typewriter repair person who said the standard IBM Executive Model D had all the capabilities necessary for that document. The Composer wasn't even needed.

[yardwork]

11. I've posted the same point myself

We can't prove that these documents are genuine. A real independent expert would have to examine the originals to reach that conclusion.

However, the instant assertions that the documents "must" be fake because of the superscript, blah blah blah, don't stand up to scrutiny.

The bottom line is that nobody has proven these documents to be real OR fake in my opinion. It is poor journalism on the part of NBC, ABC, Fox, CNN, and MSNBC to be reporting the "certainty" that these documents are fake. It reveals them to be biased in their reporting and overly credulous of anything the White House says.

[USA_1]

12. "fake"???

I was not aware that NBC + ABC categorically said they were fake. Didn't they say there is still some doubt, instead?

[girl gone mad]

13. They rushed to judgement en masse.

I'll try to compile a list of quotes and articles, but they relied largely on bloggers and didn't bother to present opposing viewpoints on the issue, nor did they mention that their experts were partisan.

There simply was no balance in their journalism. I know it isn't their job to verify the documents for CBS, but it is their job to peresent factually accurate information, which they did not do.

Edot: Media Matters has done an excellent job reporting on the media's false assertions in regards to this story.

http://mediamatters.org/items/200409100010

[yardwork]

15. Thank you for your research, girl gone mad

It's a sad day when ordinary citizens have to remind the major news media of the most fundamental precepts of journalism.

It's a good thing we have the internet to serve as a public square!

I thank our host Skinner for providing this venue, and the many dedicated posters who volunteer their time and expertise to analyze polls, research topics, and link important information here.

[Bucky]

16. Forget the "line up" analysis. Look at the words "Examination will"

The words "Examination will" clearly are the work of a typewriter. The 1st "a" in Examination and the word will are both raised above the line slightly, which is a common effect from manual typewriting.

However your fantasy about Little Green Footballs stretching and pasting the forgery "evidence" is baloney. There's nothing in that site's work to suggest that nor any evidence in your claims to back up that sort of conspiracy-level manipulation. As with Occam's Razor, we can find a much simpler explanation, that LGF and you have different settings on your versions of MS Word.

Since the line spacing and word spacing in Word is very easily manipulated, it's really quite believeable that one person could produce a near-exact match and another find the copies completely out of alignment.

[girl gone mad]

21. LGF needs to let people know..

what settings were used and what verion of Word was used.

I was only able to get a close fit by stretching the Word document 12% Horizontally and 1% vertically in the image below. You're right that they would not have needed to cut and paste, since it looks like they simply left off the dateline.

[JP Stormcrow]

18. A nice way to see the MS Word Times New Roman font not matching

I think this was also put up by someone looking to prove they
are the same - but it certainly shows that the fonts are
different.
It is an animated .gif movie - of an MS Word generated doc
over the one of the memos.
Best way to view it is to open Windows Media Player and then
go to File .. Open URL. Expand to Full Screen view for best
effect.

http://homepage.mac.com/cfj/.Pictures/aug1873-pdf-animate.gif

If you look at several instances of the same letter you will
see some clear patterns of differences. Start with capital
"H" , "e" is another easy one. Kind of fun
actually -like "Where's Waldo". 

I must say that I am leery of almost any absolute conclusion
drawn from us amateurs comparing these faxed/copied/scanned
images to a digital document - but I think "this is not
MS Word Times New Roman" looks pretty solid.


JP

 

[Heath.Hunnicutt]

28. Also, lower case 'x' does not match

The lower-case x doesn't match between the letter typeface and the Times New Roman font. The x in TNR has a lighter stroke for both the lower-left and upper-right strokes, while the typed output (possibly from a line printer) has a heavy stroke for all but the lower-left.

There are also multiple examples of the numerals 1, 4 and 7 to compare with redundant versions to sample.

[yardwork]

29. Welcome to DU, JP!

I noticed the same differences in the supposedly identical flashing documents!

[John_H]

20. I guess i'm goint to have to go buy an IBM executive model D to show it's

possble without the composer.

[karnac]

22. try this doc file

Allegedly it matches up perfectly with MSWORD2004. I don't have a good printer to see if it's true.

http://peterduncan.net/18-August-1973.doc

But what alarms me is that you are saying is that we may NEVER find an IBM composer that will match this one because of varying wear from one machine to another.

So we are in perpetual limbo on this. :(

Karnac

[girl gone mad]

24. You don't need a printer to see if it's true.

And I don't believe you need the exact typewriter to make a case.

If the typewriter font is a better match than any Word fonts (and it is), and there are obvious typewriter characteristics such as raised letters (and there are), you can say with a fair degree of certainty, say 99%, that the document was made on a typewriter.

We're not putting a guy on trial for murder.

[karnac]

27. quite true..

The fonts don't match that well. and it's hard to tell with so much photocoping.

What about a computer(runnning MSWORD) but matched with, say a daisywheel printer instead of laser? With the proper driver and acompanying computer fonts,is it more likely to be a better WYSIWYG match. particularly since the font that IBM licensed is quote likely to be also licensed by somebody else. Like Wang which made some printers too.

Karnac

[jab105]

23. I think its a go nowhere argument (now the masses think

its fake, so BUsh is ahead, as usual)...

But, my brotherinlaw goes and tells me that its proof that the media is liberal...what and the what what...

So I took the damn pdf, made a screen print, and made another of the Times New Roman from MS Word. They dont "match" and the proof is IN the "th"...this is close up (since the whole idea behind it is that the Times New Roman is MEANT to copy Times New Roman...ummm...it should be "close"...so I'm doing a closeup...if it was indeed an MS word "fake" it would match up completely...but it doesn't/

The Red is the MS word, the black is the PDF...PS-I lined it up at the 7...the "th" is obviouosly off...but so are the other letters...

[girl gone mad]

25. Thanks jab..

The i's aren't a match, the g's aren't a match and in other images I have seenm the v's aren't a match, either (the angle is wrong).

I am of the opinion that this is a good argument for us, if we become more vocal and united. These RW shills have, once again, lied to the public. I happen to think they should be humiliated for racing to push such tenuous claims.

Remember those Republicans that disrupted the count in Florida by standing in the hallway and shouting? HOw many people knew that they were payed GOP aides bused in from Washington? Not many, because the Democrats decided back then it wasn't worth making an issue out of it. Constantly rolling over in these situations has not gotten us very far.

[yardwork]

30. And how many know that Shrub buses his adoring crowds in to rallies?

I've seen a few scattered news reports let the cat out of the bag on this, usually by accident. One was a puff piece on W that interviewed groupies who travel with W from "local" rally to rally all over the country.

No wonder his crowds don't react when he says bizarre things about Ob/Gyns practicing their love on women. They've heard his canned speeches so many times they don't even listen, just applaud when the sign flashes.

[JP Stormcrow]

26. Actually the proof is is in several other letters , but not the th

I have been spending way too much time on this.
Therefor I know that the actual printout of the superscript in MS Word moves it up a bit - the mismatch you show is not conclusive if it is from the screen only. As you say some of the others are off - but you have to look at several instance to separate the patterns from artifacts.

Hopefully, this animated .gif will work for some as an embed:




JP

[gmaki]

31. I noticed the same thing with the "th" printed...

However, the most obvious difference, which you can see in that animation but is even more apparent in a larger format are the 8's.

MSWord's Times New Roman 8's have equal sice loops top and bottom. The 8's on the memo all have a smaller loop on top. I know that there are ither differences but this is one that I think there is absolutely no other explanation for other than they are different fonts.

Conceivably, distortions in copying could make an e here or there appear higher or make the serifs look thicker but I don't think aby distortion could cause the difference in the 8's.

My 2 cents. (I have been spending too much time on this too, but I will use any excuse to goof around in Photoshop for an hour vs. doing any real work)

[jab105]

34. Impressive...never thought I'd get caught up in this...

but he pissed me off with the swipe at liberal media...my take is that its the exact opposite...I couldn't let him get away with that!!

PS-Thanks, that was excellent!!

[girl gone mad]

32. I think this 'frontline' article is relevant

frontline covered the 60 Minutes Jeffrey Wigand (tobacco industry memos and cover-ups) story which was the subject of the film "The Insider".

This timeline from frontline gives an indication of what goes on behind the scenes at CBS/60 minutes. It's also interesting to note that ABC decided not to report on the tobacco story (and issued an on-air apology for a brief segment they had done) after being threatened with a $10 Billion dollar lawsuit.

I can only imagine what kind of pressure 60 minutes would be under now in regards to any story that is negative on Bush. The media has gotten more coprporate controlled, not less.

[girl gone mad]

36. I'm adding some pertinent new information to this thread:

I just found an interesting article from the May 1971 issue of Mother Earth News. It appears to be a lengthy IBM advertisement for the Composer, encouraging women to start their own home businesses by leasing a Composer for $150 a month (including maintenance and free training).

There is a sample document made on a composer, with a perfectly centered header. This is something the RW bloggers have been saying was not possible before word processors. I overlayed the same header in 12 point arial from Word onto the document, and here is my result:




A perfect match, which means the Composer was capable of centering as perfectly as Word.

[JP Stormcrow]

37. Good find but....

Unfortunately do not think it shows what you were hoping.

I went to the web site - and it is not a scanned image, but rather typed (with embedded scanned images such as the signature). They retyped the words in - so you are matching Word to another digital text representation - which is their interpretation of how the ad looked, but not the original image.

.. now if you can find a hardcopy of Mother Earth News you could complete the experiment.

Keep it up though.
JP

[Bridget Burke]

41. Typists could center text before the Composer.

It was a matter of measuring & counting characters.

The Composer was not "heavy duty" in that it did lots of typing with many carbons. It was, indeed, very specialized. When I used one in the 70's, I was producing copy for a small newspaper. This involved:

1) Set column width & font size, install correct "type ball".

2) Type the copy, one line at a time. Watch the text & listen for the bell--see where the line wrap should occur. At that point, you checked the gauge on the front of the machine--colored bars against a "ruler". Tab (to the tab stop you'd set) & type the "code". If the Green bar was on 3, type "G3"; white on 4, "W4", etc.

3) Proofread that sucker.

4) Check the codes & set the dial to the right of the keyboard, then type a line. It will be right justified. Proceed to the end of the copy. Don't leave out any lines!

5) Proofread again.

6) Cut (with an Exacto knife), paste (with rubber cement), add headlines (rub-on).

My current vote is for the IBM Executive. Proportional spacing (& we made up to 5 carbons). And the "superscript" may have been on some keyboards--not a special feature, just the character. It has been seen on other documents.

However, somebody could be using the Composer for some kind of newsletter & doing some regular typing on the side.

[gollygee]

43. A lot of typewriters had a "half space" key and even a "quarter space" key

to allow things to be centered better.

[Heath.Hunnicutt]

44. You accidentally grabbed a non-scan.

Sorry, Girl Gone Mad... That article you went to, if you "View Source" on the document, it turns out that they recreated that example with HTML. So the image above is unfortunately not from a scanned printed output.

I went to look at the centering in that image, which is actually less good than the memo, and saw that none of the pixels have aliasing on the nonbold letterehead. The artifacts are from jpg compression, probably.

You gave me a really great idea about the memo though. It is too centered for a computer program. It's antialiased below the 1 pixel granularity that a computer would use! I am more sure than ever that something electromechanical got that centering so precisely. So, thanks, you kick ass! :)

[Nederland]

38. Question

One of the most damning pieces of evidence for forgery I've heard is that in order to get a Composer to to match the superscripted 'th' you'd have to change type balls. I find it difficult to believe that someone writing a memo of this type would put that much effort for something simply intended for personal use. Do you have an explanation for this?

[Hepburn]

39. I have tried to duplicate the memos, too...

...and I have used MSW and WP and I have not been able to match them up with the Killian memos. My problem with why the could not have been word processed on a computer printer - the defects in the baseline. It just looks all wrong to have been done on a printer.

[BurtWorm]

40. Okay. I was a skeptic on Friday. This jerk turned me the other way!

I still don't know if the CBS documents are forgeries, but now I know that a Selectric could:

a) make lines and words of almost identical length as MS TNR's

b) make a superscript "th" that looks like MS TNR's.


Those were my two main issues with the non-forgery claims. I'm not 100% convinced of their authenticity, but I am no longer mostly doubtful of it.

[starroute]

47. Similar font on IBM Executive typing from 1962

This is from a 1964 issue of a fanzine called Niekas and would have been done on an IBM Executive. It's a scan of a mimeographed page, which is why so many of the letters are broken.

The font is not quite identical to that in the memoes, but it's pretty close. The capital letters and numerals are a particularly good match, though some of the small letters are less so (particularly y, g, and r.) Also the fanzine item is more loosely spaced than the memo. However, the overall appearance, with the somewhat irregular lines, is very similar. This is why I don't believe the memoes were done on a Composer.

We owned a Composer in the 80's, and as it got older, its spacing became erratic. But the problems were all with keeping an even spacing between letters, which was controlled by some fairly delicate technology. It never lost its ability to maintain a consistent baseline, because that was provided by the basic IBM typeball hardware, which was extremely rugged and dependable.