Musings of an American in Holland

On 14 December 2009, The Times of London trumpeted what it claimed were notes produced in early 2007 “from Iran’s most sensitive military nuclear project” documenting Iranian plans to build and test a neutron initiator, a key component of an atomic weapon. The article received widespread international attention and called into question the US intelligence community’s 2007 National Intelligence Estimate (135 kb PDF), which judged “with high confidence that in fall 2003, Tehran halted its nuclear weapons program.”

But Gareth Porter of the Inter Press Service News Agency today reports that “U.S. intelligence has concluded that the document published recently by the Times of London…is a fabrication” (emphasis added). Porter notes that while such a document would presumably be highly classified, the document (4.5 mb PDF) does not bear any security markings–a fact that The Times failed to disclose.

I have read The Times’ document for myself and note another significant irregularity that The Times failed to note: this allegedly “Iranian” document was composed on a computer that evidently lacked Persian (Farsi)–the Iranian national language–resources. It is readily apparent that the document was composed using Arabic, not Persian resources. For example, the Arabic letter corresponding to the English “y” is “ي”. In Persian, however, the dots are always omitted when the letter appears at the end of a word. (In fact, with a Persian input method selected, one has to press the shift key to obtain the Arabic letter “ي” with two dots underneath.) Throughout the document, the Arabic form of the letter, with dots underneath, appears:

Telltale Dots

In addition, the Persian word-initial long “a” (آ) consistently appears as a regular “a” (ا) in the document. The “little hat” above the long “a” is generally not considered to be optional.

In short, it is evident that an Arabic, and not a Persian, input method was used for composing this document. Why would a computer at Iran’s “most sensitive military nuclear project” be configured to use Arabic language resources instead of Persian ones?

Moreover, why would the Times fail to note the lack of security markings and the glaring typographical irregularities of this document?

Update: It should be borne in mind that while the use of Arabic input resources is noteworthy, it is not dispositive. Use of Arabic input methods for documents composed in Persian was fairly common in the early 1990s (before Persian input methods became widely available for computers running Microsoft Windows) and is still sometimes seen today. Still, the document’s lack of any date, classification stamp, or special handling instructions, combined with the typographical peculiarities discussed above, preclude me from taking it at face value.

Update 2: Scott Horton of Antiwar.com interviewed Philip Giraldi regarding The Times’ “nuclear trigger” document on 29 January 2009. Audio is available here.

Update 3: Scott Horton interviewed me about The Times’ document (as well as about polygraph policy) on 30 January 2009. That interview is now available on Antiwar.com. On 3 January 2010, Times leader writer Oliver Kamm, in response to comments that I left on his blog two hours before my interview with Scott Horton, wrote, among other things:

George Maschke, the whole of your comment is undermined by your mistake in assuming that the document that you read online was the document in its original form. It was in fact a retyped version of the relevant parts of that original document. The original document contained a lot of classified information. The Times did not publish the original document, because of the danger that it would alert the Iranian authorities to the source of the leak. The full version of the document is in the hands of the IAEA.

It should be noted that The Times represented the Persian version of its purported “nuclear trigger” document as “Iran’s nuclear trigger: document in full.” A running commentary on Oliver Kamm’s blog may be followed here.