Typesetting might be important if it is alleged that a forgery is in a different type than the original.
That's part of it. Irey did not examine a copy that had been certified to be visually identical to the alleged original, to the extent that this type of analysis was legitimate. As you well know, a copy of a birth certificate may be certified as legally probative as long as it portrays legibly the same information that is contained in the original, regardless of how it is (re)produced. Hence Mitt Romney's legally acceptable birth certificate is a computer-generated abstract, while Barack Obama has produced not only the abstract but also, by special permission, a facsimile copy. The evidence of that certification in all cases is the seal of the issuing authority, which has appeared on various forms of Barack Obama's birth certificate presented to election authorities, to the secretaries of state of interested states, and to representatives of the media.
However there is no representation from anyone that the online PDF is, in all respects, a copy of the original record that can be considered so visually faithful to the original that typographical analysis of the copy would apply also to the separate paper version used to certify the candidate's eligibility and from which the PDF purportedly derives. Specifically, the informal PDF copy contains significant artifacts attributable to a common post-process optimization step designed to reduce the size of the digital file containing a scanned document. While this optimization does not materially affect the legibility of the document, it does mean the optimized PDF would not be considered visually identical to the original copy, or even to a photographically reproduced copy, even though it retains its legitimate
legal value as a true copy if presented under seal or otherwise authenticated (as has happened many, many times at the hand of the Hawaii registrar of vital records).
Paul Irey, examining this highly post-processed facsimile copy, notes differences in the appearance of typed characters that he asserts should appear identical, such as letters apparently misshaped, misplaced, or otherwise inconsistent in his opinion with the expected appearance of typed text. He proposes the theory that these inconsistencies could not arise in a legitimately typewritten document, and hence must show the document to be inauthentic and likely forged. He theorizes further that digitized samples of typewritten text were manipulated by computer programs to compose the copy of the birth certificate he examined.
However, the differences Irey notes (which, as Judge Reid notes, does not necessarily require an expert to testify to them) are attributed to forgery only in the plaintiffs' fertile imagination. They are equally and more easily attributed to the process by which the PDF was optimized. In technical terms, when a "deep" digital image is rasterized to a bitmap (i.e., a "shallow" 1-bit image), the resulting image suffers from quantization errors that materially alter the shape and position of the quantized characters.
There is
absolutely no presumption that typed or typeset characters scanned as a color or grayscale document, then quantized to a 1-bit bitmap will retain the same shape in the bitmap. Yet this is the premise that Irey relies upon in his analysis. He thinks he can look at an optimized PDF and attribute differences in the shapes of characters to some sort of digital cut-and-paste forgery because type should look identical if it were a real copy. That is simply preposterous.
The forger may have been careful to match type, either with an identical typewriter wheel/ball or by painting in the information by hand.
I proposed above that the easiest way to forge a typewritten document is simply to type it. Hence Irey's proposed method makes no sense. But the trier of fact is expected to draw his own conclusion regarding the credibility of the plaintiff's theory.
The original document might have been in more than one typeface. A government office is going to have a jumble of different machines purchased through the ages. There are any number of reasons why a clerk might have sat down at one machine, started typing, and then needed to switch to another.
This would account for some, but not all of Irey's alleged inconsistencies. Most notably, Irey argues that the "R" in "BARACK" appears in a different typeface than the rest of the word, hence could not have been typed on the same typewriter as the rest of the letters. There would be no reason for a typist to switch machines in the middle of a word.
But otherwise yes, the workflow in a normal office would mean, perhaps, that one junior typist would be assigned to fill in, say, the hospital name and address in a stack of birth certificates while another typist, working on a different machine, will transcribe the personal information from handwritten applications, doctors' notes, etc.
Other evidence is so overwhelming that one small detail simply is not enough to move a reasonable mind.
Well then you're getting into the whole absurdity of the Birther claim, which is the bulk of this thread. But in the narrow context of this court case, there was almost no other information presented.
Conspiracists often demand that the "truth" be 100% consistent.
And indeed want to attribute even the least inconsistency to mind-bogglingly absurd and complex theories involving shadowy powers and massive cover-ups.
In this case there is inconsistency, and the trial judge noted the inconsistency, weighed it appropriately, and concluded there was no cause of action. But even so in this case, there are legitimate explanations for the inconsistency. In this more than many, we
can give you an equation. The inconsistencies Irey notes in the typewritten text are more likely to be the result of considerable computer processing designed to reduce the size of a file not part of the candidate's chain of evidence and presented solely as a courtesy and convenience to the public.
