Curing Analytic Pathologies

A new study published by the CIA Center for the Study of Intelligence calls for a fundamental reconceptualization of the process of intelligence analysis in order to overcome the “pathologies” that have rendered it increasingly dysfunctional.

“Curing Analytic Pathologies” (pdf) by Jeffrey R. Cooper has been available up to now in limited circulation in hard copy only. Like several other recent studies critical of U.S. intelligence, it was withheld from the CIA web site. It has now been published on the Federation of American Scientists web site.

Author Cooper provides a thoughtful critique that notes the intrinsic difficulties of intelligence analysis and observes how current organizational practices have exacerbated them.

“The Intelligence Community presently lacks many of the scientific community’s self-correcting features,” he writes.

One major impediment to improving analysis is the hypertrophied secrecy practices that prevail in intelligence organizations.

“Unfortunately, the more that evidence and judgments are restricted in dissemination by compartmentation and distribution limitations, the more likely it is that questionable judgments will pass unchallenged.”

Fundamentally, the whole concept of the “intelligence cycle” — referring to the conventional sequence of collection, processing, analysis and dissemination — is misleading, Cooper argues, and should be jettisoned.

“With its industrial age antecedents, it usually conveys the notion of a self-contained ‘batch’ process rather than a continuous spiral of interactions.”

Taking Cooper’s thesis seriously, one could respectfully say that his new study embodies some of the defects in intelligence analysis that he writes about.

Thus, the study presents what is essentially one moment in an ongoing conversation and freezes it in a nicely produced but static document. This reflects the kind of spurious finality that Cooper dismisses as the “conceit of finished intelligence.”

And then the CIA, by disseminating the document in hardcopy only, sharply limited its audience and effectively precluded a “continuous spiral of interactions” regarding its contents.

That last part, at least, can be corrected.

See “Curing Analytic Pathologies: Pathways to Improved Intelligence Analysis” by Jeffrey R. Cooper, Center for the Study of Intelligence, December 2005 (5 MB PDF).

Author Cooper said he would welcome feedback from interested readers. Comments can be posted here on the Secrecy News blog. Alternatively, Cooper’s contact information can be obtained from Secrecy News.

No Responses to “Curing Analytic Pathologies”

  1. Preston May 3, 2006 at 2:00 PM #

    Perhaps a “CIA wiki” might help the analysis issue. That way, one author’s analysis can be posted as an article on the wiki, approved analysts can modify/correct/update the article (changes are tracked by the wiki), and a broader community can benefit from the “living” analysis.

    This is also an alternative method for publishing books as a series of wiki articles, which might be a better way to circulate Mr. Cooper’s book.

  2. Mark May 3, 2006 at 2:20 PM #

    Re the previous comment by Preston: The use of blogs and wikis by analysts at the CIA has exploded during the past couple of years, and their use has spread to other intel agencies. Their use as a means of info-sharing and improving analysis is, believe it or not, being encouraged in at least some quarters of the intel community.

  3. K. A. Taipale May 3, 2006 at 2:52 PM #

    Re wikis and blogs in intel:

    See Andrus, D. Calvin, “The Wiki and the Blog: Toward a Complex Adaptive Intelligence Community” . Studies in Intelligence, September 2005 Available at SSRN: http://ssrn.com/abstract=755904

    and

    Feds look to ants, wikis and blogs, FCW.com (Apr. 21, 2006)
    http://www.fcw.com/article94152-04-21-06

  4. Paul May 3, 2006 at 3:02 PM #

    The Intelligence Community presently lacks many of the scientific community’s self-correcting features,….” well, there is room for improvement and scientific community is also concerned about its review processes: “For Science’s Gatekeepers, a Credibility Gap” http://www.nytimes.com/2006/05/02/health/02docs.html

  5. robert schaefer May 3, 2006 at 4:53 PM #

    Think “systems”.
    Where the author writes “structure”, it appears the author means “structure” as a black box, not as a white box. The black box model cannot be controlled because the insides are too complex and too resistant to change. The internal decision making processes, the white box model of the organization needs to change. But it cannot because it is too difficult, too many people with vested interests – they succeed although the purpose of the organization does not.
    Possible solution alternatives are to start over from scratch – (not likely to happen because what will happen while there is nothing?) or create parallel alternative structures with the culture and model the author proposes. Experiment with alternative organizations. Slowly replace the equivalent functions of the failed organization with the new (one hopes better) organization.
    Another alternative – implement employee suggestion programs. The employees know the work the best, better than any planner of organizations. Ask them what they think should be done, then test it out.

  6. Robert L. Cerra May 5, 2006 at 7:32 AM #

    Thanks Mr Aftergood for making this report available on your website.
    I am digesting Mr Cooper’s 73 page report and my initial reaction is that he is mixing information analysis and the end product of intelligence. I would be more comfortable if Mr. Cooper’s report read Curing Analytical Pathologies: Pathways to Improved Information Analysis. In that parts that I have read he uses intelligence and information interchangably -that suggests to me a basic misunderstanding of the intelligence process and how one applies it. I want to read all of his report before I comment further.

    Regards
    Bob Cerra

  7. robert schaefer July 25, 2007 at 3:47 PM #

    In the end, does the existence of analytic pathologies really matter when it is the decider who makes the decisions?

    http://www.commondreams.org/archive/2007/07/25/2751/