How a Sketchy, Unsourced Claim in a Government Report Became Part of the Case Against Iran
Stay up to date with the latest headlines via email.
Several media outlets reported this month on an alarming finding from a new U.S. government study: Iran’s intelligence ministry, as CNN put it, constitutes “a terror and assassination force 30,000 strong.”
The claim that the intelligence ministry has a whopping 30,000 employees, first reported by a conservative website, spread to other outlets including Wired and the public radio show the Takeaway and landedelsewhere online, even on the intelligence ministry’s Wikipedia page. All cited the new government study, put out by an arm of the Library of Congress called the Federal Research Division.
So how did the government researchers come up with the number? They searched the Internet — and ended up citing an obscure, anonymous website that was simply citing another source.
The trail on the 30,000 figure eventually ends with a Swedish terrorism researcher quoted in a 2008 Christian Science Monitor article. But the researcher, Magnus Ranstorp, said he isn’t sure where the number came from. “I think obviously that it would be an inflated number” of formal employees, said Ranstorp.
We inquired with six Iran experts, and none knew of any evidence for the figure. Some said it might be in the ballpark while others questioned its plausibility.
“Whether the figures emanate from Iran or from western reporting, they are generally exaggerated and either meant as self-aggrandizing propaganda, if self-reported by Iran, or just approximations based on usually scant data or evidence,” said Afshon Ostovar, a senior Middle East analyst at the Center for Naval Analyses who writes frequently on Iran. The number “could be more or less accurate, but there's no way to know.”
Gary Sick, a longtime Iran specialist in and out of government, said the entire Federal Research Division study “has all the appearance of a very cheap piece of propaganda and should not be trusted."
Sick pointed to the study’s use of questionable Internet sources as well as flat-out errors. In one section, for example, the study lays out in detail how “Iran’s constitution defines” the intelligence ministry’s official functions. The problem, as Sick notes: Iran’sconstitution doesn’t mention an intelligence ministry, let alone define its functions.
Federal Research Division Chief David Osborne said in an email the report “was leaked to the media without authorization” and declined to comment further “because it is proprietary to the agency for which it was written.”
This is that we know about the 30,000 figure and its provenance:
On the morning of Jan. 3, the conservative Washington Free Beacon ran astory headlined, “Iran Spy Network 30,000 Strong.” The outlet said it had obtained a “64-page unclassified report” on the Ministry of Intelligence and Security, and published it with the story.
The Federal Research Service of the Library of Congress, which produced the study,provides “fee for service” research to other government agencies using the resources of the library. The study’s title page names no author but says it was produced under an agreement with an arm of the Pentagon called the Combating Terrorism Technical Support Office. (That office did not respond to requests for comment.)
The study flatly states that Iran’s intelligence ministry has “more than 30,000 officers and support personnel.”
But it also hedges. It notes Iranian intelligence is “a difficult subject to study because so little information about it is publicly available.” The study does not claim to feature any original intelligence or reporting. It says its main sources are news websites and Iranian blogs.
“The reliability of blog-based information may be questionable at times,” says the report. “But it seems prudent to evaluate and present it in the absence of alternatives.”
The evening after the report was first published, CNN ran a segment on what it called “troubling new details on a new report of Iran's intelligence service.” The story compared the 30,000 figure to the roughly 100,000 employees in the 17 U.S. intelligence agencies and offices, and went through various attacks over the years attributed to Iranian intelligence.
A CNN spokeswoman said the network “checked the number with sources that led us to feel comfortable that the report was in line with the national security community's understanding."
That post, from 2010, turns out to merely excerpt another study from yet another source.
That study, titled “Shariah: The Threat to America,” was put out by the hawkish Center for Security Policy. As the title suggests, it doesn’t focus on Iran but rather the purported threat of Islamic law.
That piece refers to Iran’s intelligence ministry having “some 30,000 on the payroll by one count,” which came from Ranstorp, the Swedish terrorism researcher.
Ranstorp told us that while he did not recall citing the figure to the Monitor, it might have originated with Kenneth Katzman, a Mideast specialist with the Congressional Research Service who often writes on Iran.
Katzman told us that the figure did not come from him. He added that 30,000 did not seem “inordinately unreasonable” but that he does not know of evidence supporting it.
Bill Gertz, the Washington Free Beacon reporter who obtained and published the Federal Research Service study, told ProPublica he stands by his story.
"In my 30-plus years in reporting on national security issues, I have found that such unclassified reports often use press reporting of such numbers to avoid having to use classified information,” Gertz said. “I also know that most of the people who write such reports have access to classified information about the subjects they write about and I doubt they would publish a figure that would be contradicted by classified assessments of the number of personnel in the [intelligence ministry]."
Gertz also pointed to another report on Iran, this one produced in 2010 by private intelligence firm Stratfor. But that report says that, as of 2006, Iran’s intelligence ministry had just 15,000 employees. It does not cite a source for the figure.