TANG Typewriter Follies; Wingnuts Wrong

(From the diaries -- kos)

Against my own better judgment, but because I believe that the more rapidly charges are countered, the better, I spend a goodly portion of the last day researching -- shudder -- typewriters of the '60s and '70s.  As everyone on the planet no doubt knows by now, the hard-right of the freeper contingent -- specifically, LittleGreenFootballs, a site which frequently is cited for eliminationist rhetoric and veiled racism, and PowerLine, a site linked to with admiration by such luminaries as Michelle Malkin and Hugh Hewitt -- discovered that if you used the same typeface, you could make documents that looked almost -- but not exactly -- like the TANG documents discovered by CBS News.  This qualifies as big news, of course, so from those two sites, the story has spread into the mainstream media through the usual channels, most notably Drudge, NRO, etc.

I do not believe there is any truly "new" information here, but I hope to condense it in one easy-to-digest reference.

So here are some point-by-point findings re: the "forgeries".

First Claim (LittleGreenFootballs): "The documents can be recreated in Microsoft Word".

What the LGFer did to "prove" this was to type a Microsoft Word document in Times New Roman font, and overlay it with the original document.  As he says:

Notice that the date lines up perfectly, all the line breaks are in the same places, all letters line up with the same letters above and below, and the kerning is exactly the same. And I did not change a single thing from Word's defaults; margins, type size, tab stops, etc. are all using the default settings.

We're going to make this simple.

First, of course, in order to do this, he first had to reduce the document so that the margins were the same, since the original PDF distributed by CBS is quite a bit larger.  Then he superimposed the two documents, such that the margins on all sides lined up.

What he then discovered is that Times New Roman typeface is, when viewed on a computer monitor, really, really similar to Times New Roman typeface.  Or rather, really really similar to a typeface that is similar to Times New Roman typeface.

Um, OK then.

You see, a "typeface" doesn't just consist of the shape of the letters.  It also is a set of rules about the size of the letters in different point sizes, the width of those letters, and the spacing between them.  These are all designed in as part of the font, by the designer.  Since Microsoft Word was designed to include popular and very-long-used typefaces, it is hardly a surprise that those typefaces, in Microsoft Word, would look similar to, er, themselves, on a typewriter or other publishing device.  That's the point of typefaces; to have a uniform look across all publishing devices.  To look the same.  You could use the same typeface in, for example, OpenOffice, and if it's the same font, surprise-surprise, it will look the same.

So kudos on discovering fonts, freeper guy.

Next, however: do they really match up?  Well, no.  They don't.

If you shrink each document to be approximately 400-500 pixels across, they do indeed look strikingly similar.  But that is because you are compressing the information they contain to 400-500 pixels across.  At that size, subtle differences in typeface or letter placement simply cannot be detected; the "pixels" are too big.  If you compare the two documents at a larger size, the differences between them are much more striking.

For instance:  In the original CBS document, some letters "float" above or below the baseline.  For example, in the original document, lowercase 'e' is very frequently -- but not always -- above the baseline.  Look at the word "interference", or even "me".  Typewriters do this; computers don't.  Granted, if you are comparing a lowercase 'e' that is only 10 or 12 pixels high with another lowercase 'e' that is only 10 or 12 pixels high, you're not going to see such subtleties.  That doesn't prove the differences aren't there; it just proves you're an idiot, for making them each 12 pixels high and then saying "see, they almost match!"

"This typeface -- Times New Roman -- didn't exist in the early 1970s."

There are several problems with this theory.  First, Times New Roman, as a typeface, was invented in 1931.  Second, typewriters were indeed available with Times New Roman typefaces.

And third, this isn't Times New Roman, at least not the Microsoft version.  It's close.  But it's not a match.

For example, the '8' characters are decidedly different.  The '4's, as viewable on other memos, are completely different; one has an open top, the other is closed.

So yes, we have proven that two typefaces that look similar to each other are indeed, um, similar.  At least when each document is shrunk to 400-500 pixels wide... and you ignore some of the characters.

"Documents back then didn't have superscripted 'th' characters"

That one was easy.  Yes, many typewriter models had shift-combinations to create 'th', 'nd', and 'rd'.  This is most easily proven by looking at known-good documents in the Bush records, which indeed have superscripted 'th' characters interspersed throughout.

"This document uses proportional spacing, which didn't exist in the early 1970s."

Turns out, it did.  The IBM Executive electric typewriter was manufactured in four models, A, B, C, and D, starting in 1947, and featured proportional spacing.  An example of its output is here.  It was an extremely popular model, and was marketed to government agencies.

"OK, fine, but no single machine had proportional spacing, 'th' characters, and a font like that one."

No, again.  The IBM Executive is probably the most likely candidate for this particular memo.  There is some confusion about this, so to clear up:  the IBM Selectric, while very popular, did not have proportional spacing.  The Selectric Composer, introduced in 1966, did, and in fact could easily have produced these memos, but it was a very expensive machine, and not likely to be used for light typing duties.  The proportional-spacing Executive, on the other hand, had been produced in various configurations since the 1940's, and was quite popular.

(Note: However, it is not immediately clear that the Selectrics and Selectric IIs could not in fact emulate "proportional" spacing.  There is skepticism in some circles that these memos really show "proportional" spacing.  Looking at the blowups, it appears pretty obvious to me that there is, but still researching.)

Did they have a font that looked like Times New Roman?  Unclear; they apparently were manufactured in a range of configurations, and with different available typefaces.  Note that these were not "typeball" machines, like the Selectrics; they had a normal row of keys.  But it is worth noting that IBM had what we will call a "close" relationship with Times New Roman:

Courier was originally designed in 1956 by Howard Kettler for the revolutionary "golfball" typing head technology IBM was then developing for its electric typewriters. (The first typewriter to use the technology was the IBM Selectric Typewriter that debuted in 1961.) Adrian Frutiger had nothing to do with the design, though IBM hired him in the late 1960s to design a version of his Univers typeface for the Selectric. In the 1960s and 1970s Courier became a mainstay in offices. Consequently, when Apple introduced its first Macintosh computer in 1984 it anachronistically included Courier among its core fonts. In the early 1990s Microsoft, locked in a font format battle with Adobe, hired Monotype Typography to design a series of core fonts for Windows 3.1, many of which were intended to mirror those in the Apple core font group. Thus, New Courier--lighter and crisper than Courier--was born. (In alphabetized screen menus font names are often rearranged for easier access so now we have Courier New MT in which the MT stands for Monotype Typography.)

Courier's vanquisher was Times New Roman, designed in 1931 by Stanley Morison, Typographical Advisor to the Monotype Corporation, with the assistance of draughtsman Victor Lardent. The Times of London first used it the following year. Linotype and Intertype quickly licensed the design, changing its name for their marketing purposes to Times Roman. Times Roman became an original core font for Apple in the 1980s and Times New Roman MT became one for Windows in the 1990s. (Ironically, at the same time IBM invited Frutiger to adapt Univers for the Selectric Typewriter, they asked Morison to do the same with Times New Roman.)

So, as you can see, both IBM and Microsoft specifically obtained the typeface "Times New Roman" from the designers of that font; neither was the creator of it.  And, as we said before, typeface includes not just the "shape" of the letters, but the size and spacing between those letters.

One of the differences between the Times New Roman as implemented on the IBM machines, as opposed to Microsoft Word?  The IBM machines apparently had the alternative '4' character that matched these memos, while Microsoft Word's TNR does not.

Oops.

Now, would the 111th Fighter Interceptor Squadron have extravagantly purchased typewriters that contained the th superscript key?  Would the military want or require typewriters with the 'th', 'nd', and 'rd' characters?  Hmm.  Ponder, Ponder.  What would the 111th need with a th character... I'll leave that to the enterprising among you to deduce.

This is not the final word on this, and it is certainly possible that any documents are forgeries.  But the principle argument of the freepers -- that it would be impossible for a TANG office in 1972 to produce documents that look like these -- is simply false.  Within a few days, however, we should know for sure either way; these typewriters still have a following, and type samples should be forthcoming.

Update [2004-9-10 4:26:25 by Hunter]: Also see kj's diary just after this one, for evidence on the IBM Selectric Composer, first marketed in 1966. This machine definitively had all the features necessary to produce these documents. Because it was apparently very expensive and difficult to use, the argument is that a TANG office would never have had one. Unclear. Nonetheless, it strikes down the theory that a 60s-70s era machine could not have produced these docs.

Update [2004-9-10 5:48:19 by Hunter]: Here is an excellent article explaining the recent history of Times New Roman in particular. Note that Adobe, Microsoft, Apple, and other firms redesigned their "Times [New] Roman" typefaces in the 80s-90s specifically to more accurately match the original design of Times New Roman:

When Microsoft produced its version of Times New Roman, licensed from Monotype, in TrueType format, and when Apple produced its version of Times Roman, licensed from Linotype, in TrueType format, the subtle competition took on a new aspect, because both Microsoft and Apple expended a great deal of time and effort to make the TrueType versions as good as, or better than, the PostScript version. During the same period, Adobe released ATM along with upgraded versions of its core set of fonts, for improved rasterization on screen. Also, firms like Imagen, now part of QMS, and Sun developed rival font scaling technologies, and labored to make sure that their renderings of Times, licensed from Linotype in both cases, were equal to those of their competitors. Hence, the perceived quality of the Times design became a litmus for the quality of several font formats. Never before, and probably never again, would the precise placement of pixels in the serifs or 's' curves etc. of Times Roman occupy the attention of so many engineers and computer scientists. It was perhaps the supreme era of the Digital Fontologist.

So as you can see, it has indeed been a primary design goal of Microsoft and other firms to make their Times New Roman font match the original 1930's typeface design as closely as possible.

Update [2004-9-10 6:47:49 by Hunter]: Here is an actual manual for an IBM Selectric Composer, circa 1966, itself created using a Composer.

Update [2004-9-10 14:26:41 by Hunter]: This is from a commenter at Kevin Drum's Washington Monthly site:

Kevin, I worked in the IBM Office Products Division field service area fixing typewriters in NYC for over 13 years in the 70s. I can tell you that the Model D can produce those documents, not only did it do proportional spacing, you could order any font that IBM produced AND order keys that had the aftmentioned superscripted "th." Also you could order the platen, thats the roller that grabs the paper, in a 54 tooth configuration that produced space, space and a half and double spacing on the line indexing, this BTW was popular in legal offices. The Model D had to be ordered from a IBM salesmen and was not something that was a off the shelf item, typical delivery time were 4-6 weeks. Also, typewriter keys were changed in the field all the time, its not that hard to do. I wish I had saved my service and parts replacement manuals to backup this claim but I'm guessing a call to IBM with a request for a copy of their font and parts replacement manuals would put this to rest ASAP. Posted by: BillG NYC on September 10, 2004 at 12:24 PM | PERMALINK

FYI, but I have found nothing that contradicts this information. It would appear you could order the humble IBM Executive with a wide variety of typestyles and characters, especially if you were a large, important client.

Update [2004-9-10 22:43:41 by Hunter]: An update to this post can be found here; it contains summaries of the new information, and a little about the media reactions in all of this.