Paul KerrNPR on Iran Intel

NPR had a good segment [Via Laura Rozen ] earlier today about, among other things, the Pentagon’s new Iran desk. It notes that several of its staffers worked for OSD’s Office of Special Plans.

[FYI, one of those staffers, Abram Shulsky, wrote this gem about intelligence gathering analysis a few years ago. Ick.]

Another highlight: Negroponte said that he would be “terribly surprised” if unvetted intelligence on Iran was reaching US policymakers.

Yeah, that’d be effing shocking.

Part of the segment also features a hilarious defense of the OSP from none other than D Feith. Priceless.

Laura’s post reminded me of a similar one from Josh Marshall a few months back.

Here’s the relevant excerpt:

Here’s a topic I’d like to know more about.

As you may know, Vice President Cheney’s daughter Elizabeth is the deputy assistant secretary of state for Near Eastern affairs. She also has the title of “Coordinator for Broader Middle East and North Africa Initiatives.” Basically that means she’s in charge of democratizing the Middle East.

She has a budget of, I believe $75 million, for bringing about ‘regime change’ in Iran.

I also noticed this recent aside in The Nelson Report in which Chris Nelson wrote that his sources “say [Undersecretary of State Nick] Burns has been fighting an apparently losing battle with Undersecretary for non-proliferation Bob Joseph on a variety of issues, and that Vice President Cheney’s office seems to be sponsoring the hiring of exceptionally large numbers of political appointees, not career FSO’s, to staff the to-be-created Iran democracy projects to be run out of State.”

Pass the glue…


  1. Andy (History)

    I worked in the intel community for years, and the sad fact is that technology has allowed policy people access to largely the same raw intel data that analysts have. Politicians (and their appointed lackeys) are not exactly known for their capacity to examine evidence in an unbiased manner. In my experience, they often latched on to specific pieces of intel that supported their own beliefs and policy objectives and downplayed or ignored contradictory evidence. This trend continues today, especially given the unhealthy focus on current intelligence by the IC today and the lack of adequately trained and educated analysts.

    I think I’ve mentioned it here before, but the analyst-policy relationship is severely broken today, and policy intruding into analysis is but one example.

  2. John Smith

    Hmm. I would take a different view. Politicians are biased, that’s a given. But the IC has simply not been coming up with the goods for a very long time. Great intel can only come from COMINT and HUMINT. The CIA, NSA and associated agencies have focused too much on IMINT and ELINT (and now even more obscure stuff like MASINT) since the start of the Cold War, which means they have good or average information on matters of interest, which is never enough.

  3. Andy (History)


    I have to disagree. HUMINT sources can provide access to information not available from any other means. However, the history of HUMINT deception is long and prosperous. I recommend J. Richelson’s “A Century of Spies” for dozens of examples.

    COMINT can be very good as well, but it’s often difficult to judge the context of an intercepted communication. This certainly played a role in Iraq where COMINT was misinterpreted (See the Iraqi Perspectives Project report page 93).

    IMINT is somewhat limited in a strategic intelligence role, but it can and does provide crucial information, as the Cuban Missile Crisis demonstrated. It’s increasingly moving to the open source world and we can now follow things like the reactor construction at Arak.

    ELINT is really a tactical intelligence discipline since its main focus is on radars.

    MASINT is hugely important, in non-proliferation especially, since it is used to determine many aspects of weapon development programs. Perhaps the most well known use of MASINT is COBRA JUDY data collected on Soviet reentry vehicles which allowed us to determine many aspects of their technical design.

    Intelligence collection on closed regimes like North Korea and Iran will always be a problem. Without a formal diplomatic presence it becomes extremely difficult to recruit and manage covert agents to say nothing of overt collection. Add to that the inherent difficulties of closed regime intelligence collection and it’s easy to see why our information sources lack both depth and diversity.

  4. ewk (History)

    My $.02 worth, mostly from being around folks who did NIE’s during the “team A/team B period in the mid 1970’s – I’m more inclined to agree with Andy, who seems to state a variant of a Mark Twain adage: “it’s not what we don’t know that gets us into trouble- it’s what we know for sure that ain’t so.”

  5. John Smith

    Andy, I’m afraid you’re just coming up with excuses.Most HUMINT is worthless, that is obvious, which is why you do whatever is necessary to get the small amount of good HUMINT which is out there. If you get it, it usually gives you all the information you need. The problem is that getting it requires committment and perseverance, which is something the CIA does not do.COMINT is simple, in that it always yields results as long as all necessary means are employed to gather it. If the NSA set up a large unit to monitor Iranian communications and another to analyse them, a lot of information would come up. Instead, they listen to a lot of things which are completely irrelevant. Without the above two methods of intelligence gathering, strategic breakthroughs cannot be made. With Cuba, for example, Kennedy should’ve fired the entire CIA for not giving advanced warning of the Soviet deployment through HUMINT or COMINT.The US could relatively easily locate C3I nodes on Cuba through ELINT and could have gained good telemetric data in case of a IRBM test. That, however, is not what the IC exists for.If the country’s intelligence agencies can’t provide information of strategic significance before threats emerge, they should not be there.

  6. Andy (History)


    We actually had a lot of intelligence on Soviet activities in Cuba. We closely monitored the buildup there, which first appeared to be primarily defensive. When the missiles were discovered it did not come as a complete shock. The CIA director, John McCone, had been warning of offensive missiles for at least six months prior to their confirmed discovery, but in the absence of definitive, confirming evidence, Kennedy, Rusk, McNamara and Bundy were not convinced. In fact it was McCone at the CIA that continuously pressed the issue and fought for more surveillance – surveillance that eventually did confirm the missile’s presence. If anything, Kennedy should have fired McNamara and Rusk. The majority of the intelligence community at the time (not just the CIA, but DoD and State as well) were wrong in their estimates that the Soviets were unlikely to put missiles in Cuba. We can thank McCone, along with a few other key analysts at the CIA, for correctly analyzing the secondary evidence available and pressing for the increased surveillance that did find the MRBM’s before they became operational.

    Getting good HUMINT requires much more than mere commitment and perseverance. Doing “whatever is necessary” does not recognize the very real practical limitations to developing human sources in a largely closed country like Iran. This isn’t an “excuse,” it’s the reality of developing HUMINT sources. The benefits of high-risk operations must also be weighed against the likelihood of capture and public disclosure. If the Iranian’s, for example, captured and paraded an American agent in front of the world press, the damage to our policy objectives would be extreme. Iran is a very difficult target, but just take a minute to think about the North Korea where isolation and a Stalinist security apparatus make running agents virtually impossible. Iraq under Saddam was similar in many ways which forced us to rely on undependable exile groups for HUMINT.

    As for the NSA, I’m sure they are listening to and analyzing SIGINT. The fact is, the vast majority of any country’s internal communications are via landlines, which can’t be monitored unless they are tapped, which necessitates operatives in the country with both the will and access to perform that mission. All the money, commitment, perseverance and resources in the world will not guarantee such access.

    Despite the problems with collection, historically, failures of strategic warning were not from a lack of data or evidence, but rather a failure of analysis or an effective enemy deception program (often it is both).

  7. hass (History)

    This debate between various sorts of intelligence is really besides the point in a country where “the intelligence is made to fit the policy”

  8. Andy (History)


    True to a point, but this is hardly unique to the United States or even the West.

  9. Andy (History)

    Not sure if you want to post this or not, but it’s an interesting example related to the discussion here:

    9/20/2006 – WASHINGTON (AFPN)—The hot dog stand in the Pentagon’s center courtyard, which long has been a source of Cold War intrigue, will be torn down in the coming months and replaced with a new eating facility.

    “Rumor has it that during the Cold War the Russians never had any less than two missiles aimed at this hot dog stand,” said Brett Eaton, an information and communications officer for Washington Headquarters Services. “They thought this was the Pentagon’s most top secret meeting room, and the entire Pentagon was a large fortress built around this hot dog stand.”

    Reportedly, by using satellite imagery, the Soviets could see groups of U.S. military officers entering and exiting the hot dog stand at about the same time every day. They concluded that the stand was the entrance to an underground bunker.

  10. Jason (History)

    Well, here are my $0.02…I don’t think this should be an either or type of debate. The fact is that only HUMINT can reveal intent. COMINT, IMINT, MASINT only tell you the evidence. HUMINT is the only thing that can put the others in perspective. HUMINT is also one of the only ways to access the most highly secret programs of other countries where probes and satellites cannot reach….just my 2 cents. Compared to some of the brains on this blog, I am probably going to have my hiney handed to me.

  11. Jason (History)

    One more comment, the first time I had ever heard of intel and policy (or military planning) mixing together was during WWII. I believe it was Winston Churchill who had some COMINT on Rommel asking Hitler for more tanks. Thinking that Rommell was on his last legs, Churchill ordered an attack. His troops were completely beaten and found out the hard way that Rommell was just asking in advance, knowing that Hitler was stingey with his hardware…so he was asking early.

  12. Damien (History)

    I read the Leo Straus and Intelligence document ed linked to. Seemed like a reasonable appeal for actual listening of what ones enemies actually have to say; exploring ways of understanding.

    What did you find to be “Ick” about the document?

  13. Paul (History)

    I don’t have enough time for a proper response, unfortunately. But it should suffice to say that the document strikes me as a recipe for coming up with non-falsifiable intelligence analyses. The heavy emphasis on the deceptive nature of regimes, the deriding of a social-scientific approach to intelligence analysis, and the approving nod to “textual interpretation,” all seem to justify my concern.

    The article’s plea for listening to one’s enemies is, I think, at odds with its emphasis on the deceptive nature of one’s enemies. It is obvious that one should beware of mirror-imaging and the assumption of universal decision-making styles, but I don’t see what the piece is offering as a credible alternative methodology. It seems to me that the implied alternative is a combination of deductive reasoning and intuition by those blessed with the ability to divine the true intentions of our enemies.

    For example, it seems that intelligence analysis based on the implied alternative could look like the following: deceptive regimes disguise their intentions, so one must assume that those intentions are nefarious. The deceptive nature of these regimes will render it impossible to find evidence that they do not possess worrisome capabilities or intentions.

    Anyway, take at look at these guys’ performance to decide if anyone should take them seriously.

  14. Andy (History)

    I don’t believe that the social science approach to intelligence founded by Sherman Kent is as incompatible with differences in regimes, cultures, etc. as the article indicates. In fact, it’s obvious that those factors should be taken into account. Social science methods for analysis are explicitly intended to limit bias and are designed to prevent assumptions about how specific actors will behave. These methods have progressed and improved much over the years, so while some of the criticisms in the article may have applied in the 1950’s and perhaps the 60’s, they certainly do not apply today. The foundational work begun by Kent has grown, expanded and adapted, but the principals of using scientific methods in the evaluation of intelligence information are as sound today as they were in the 50’s, if not more so. You can surf on over to the CIA’s Sherman Kent Center where a lot of the groundbreaking work in intelligence methodology is published for all to read.

    I also agree with what Paul said. The article puts forth many unfounded, in my view, criticisms of social science methodologies, but offers no solutions of its own besides trite suggestions like more language and cultural expertise. Mirror-imaging and other cognitive biases are human foibles well known by the intelligence community that all human beings are susceptible to. Sound social science methods used by good analysts will limit this human bias, though it is impossible to eliminate. The Strauss article offers no solution to account for cognitive biases – in fact the methods suggested seem to reinforce them.

  15. Andy (History)

    This article may save Paul some additional typing.

    Although I don’t agree with the conspiratorial tone of the piece, I think it highlights many problems with Shulsky’s and Schmitt’s ideas. I will say that I do agree with some of their specific recommendations, but am strongly against their overall philisophy regarding intelligence and its role in policy.

  16. Akash (History)

    Report: Iranian Science Teachers May Be Enriching StudentsSeptember 26, 2006 | Issue 42•39

    WASHINGTON, DC—A recently released Pentagon report is raising new worries that Iran has been operating several large facilities designed solely for the purpose of enriching mass quantities of high-grade students.

    Enlarge Image

    1. Low-grade students bond and expend energy outside core curriculum

    2. Facility-contamination expert’s office, stocked with bucket, trash bags, mop

    3. Cache of classified 2-page reports

    4. Buses transporting loads of bright, glowing new students to facility

    “We have reason to believe that specially trained Iranian science teachers are taking raw, unrefined brain power and bombarding it with knowledge at accelerated levels,” said U.S. Undersecretary Of Defense For Intelligence Stephen Cambone at a Tuesday press conference. “If current levels of student concentration remain this high, Iran could be a mere five to eight years away from developing an atomic scientist.”

    Leading analysts believe that the teachers are using a widely applied enrichment process in which students are isolated from such elements as family, play, and cartoons, and are rotated through seven separative work units over the course of each day. This cycle is repeated for months, until the students are made highly reactive to reading matter, which enables them to absorb large amounts of information in short periods of time.

    The students are then continually exposed to heavy material, taught to achieve critical thought, and finally graduate to a state of explosive productivity.

    Hard evidence that would support the Pentagon’s findings includes a top-secret syllabus, acquired by the CIA, which indicates that Iran may begin testing their students, possibly without warning, as early as next Friday. Reconnaissance-satellite images also reveal the presence of two Tehran–area facilities identified by intelligence sources only as “P.S. 235” and “H.S. 238.”

    Despite the Pentagon’s announcement in mid-June that Iran had halted its nuclear-science program, additional satellite photos taken in early September clearly show 40-foot-long buses transporting multiple loads of students to these facilities in the morning hours between 7 and 8 a.m. Some images also reveal a short, 20-foot-long bus thought to contain a smaller number of highly volatile, non-reactive, and extremely dense students.

    “While we believe that a majority of these students were developed within Iran’s borders anywhere from 13 to 17 years ago, there is also evidence that they are importing older students from former Soviet republics and Pakistan in what officials have dubbed an ‘exchange program,’” CIA Director Michael Hayden said.

    Although no one is sure exactly what is being conducted inside the accelerated core curriculum, a team of UNESCO inspectors who visited suspected Iranian enrichment facilities in 2004 found a number of microscopes, Bunsen burners, centrifuges, and reference materials, including a stockpile of instructional materials and textbooks covered in brown paper wrapping intended to obscure the material’s subject matter.

    In a nationally televised Oval Office address Tuesday, President Bush expressed the concern that if Iran is allowed to enrich its students unchecked, many of them could end up anywhere, with some potentially landing in major university centers in New York and Los Angeles.

    “The U.S. stopped enriching its students decades ago, and we call upon Iran to do the same,” Bush said. “If the Iranians do not put an end to this program by the middle of December, and impose final examinations, they could face further isolation from the international community.”

    As the U.S. awaits a response to the ultimatum, American intelligence continues to monitor a rumored late-afternoon summit, consisting of a series of secretive bilateral meetings between parents and a female science expert known as Mrs. Bakhtiari.

  17. Max (History)


    Thanks for that, Brilliant!