I'm clearly not a forensic expert on document analysis. So I don't have any way of knowing or even coming up with a reasoned opinion about the authenticity of these documents published by CBS.
But one point of criticism doesn't seem as clear as many are presenting it. I'm talking about the suggestion that a superscripted "th" marks these as clearly the product of a word-processing program.
In an article today in Weekly Standard, for instance, Steve Hayes writes that ...
... in some references to Bush's unit--the 111th Fighter Interceptor Squadron--the "th" is a superscript in a smaller size than the other type. Again, this is typical (and often done automatically) in modern word processing programs. Although several experts allow that such a rendering might have been theoretically possible in the early 1970s, it would have been highly unlikely. Superscripts produced on typewriters--the numbers preceding footnotes in term papers, for example--were almost always in the same size as the regular type.
This AP article also quotes a person presented as a handwriting analyst making the same point.
But if you look at this document from <$Ad$>the official Bush records it shows a list of descriptions of various times Bush served. (See the paranthetical at the bottom of this post for specific notes on where to find this in the pdf I linked to.) Thus, we can assume that the same document was typed on by different people and different machines over time. This document has one entry with a superscripted "th" and another further down on the page with a non-superscripted "th" -- which of course suggests that both kinds of typewriters were being used in the Texas Air National Guard system at the time.
It doesn't look like the same script used in the Killian memos and it strikes me that in this case the typeface looks monospaced rather than proportional. But clearly some typewriter with a superscript was in use. I'll leave it to others to discern the meaning of all this.
This debate has quickly spiralled in so many different directions that I can't keep track of all the different points of suspicion folks have raised about these documents. But this suggestion about the superscripts at least seems not to add up.
(To find the reference in question, click here to see the document on the USAToday website. Then scroll down to page three of the .pdf document -- which is the first vertically-oriented page. If you look at the second entry on that document -- dated "4Sep68" you'll find a superscripted "th".)