Tuesday, October 21, 2003


The fundamental theme of this blog is that government, the private sector, and civil society work better when they obtain and utilize knowledge in order to select the right course of action. I am one of those who believes that knowledge should be the result of analysis and distillation of information. I also believe that in many important instances, knowledge is socially constructed – the joint construct of a number of people rather than something held solely inside the head of an individual.

In this New Yorker piece, “The Stovepipe,” Seymour Hersh looks at the Congressional investigation of U.S. intelligence agency reports over the last decade. He notes that over half a century or more, the U.S. government had evolved complex procedures in which professionals assessed and evaluated information gathered from the field, providing the results to the politicians holding top elected offices. He suggests that under the Bush Administration, “stovepipes” were created to funnel raw intelligence, especially that supporting Administration positions, directly to very small numbers of ideologically homogenous elected and appointed officials in the State and Defense Departments and the White House. Fully vetted reports from the professional intelligence community eventually reaching the top levels of government were then challenged with conclusions drawn from the far less credible stovepiped information.

The result was, perhaps, that the Administration went to the American public with a rationale for the war in Iraq which they believed, but which was not sufficiently credible to justify so dramatic an effort. The Administration now seems to be reacting politically to charges that it misread intelligence – a response that is natural enough, but perhaps not likely to help unravel what actually happened.

I don’t know if Hersh is right. I would point out, however, that the social construction of knowledge from foreign intelligence information is likely to be quite different within political circles than within professional intelligence agencies. The U.S. has a history of political leaders distrusting career government personnel and governmental bureaucratic processes. It is quite believable that on obtaining power, newly elected and appointed officials might develop a second system to bypass the professional intelligence agencies and their procedures, in order to play a more active role in the interpretation of intelligence.

Having worked in the U.S. governmental bureaucracy for many years, I know bureaucracies can be wrong. I also know that bureaucratic systems should include stovepipes that get critical information on time sensitive decisions to the top level officials very quickly. Hersh seems to suggest that the Bush Administration has installed jury rigged stovepipes that are used more generally, and undermine the processes that have evolved over two generations, creating a crisis of morale in the intelligence community in the process. I hope Hersh is wrong, or that I am interpreting him wrongly. It seems better to work to improve the analysis and evaluation of the professionals, rather than bypassing the professional process and putting the unevaluated information directly in the hands of the politicos. If in fact there has been a deterioration in the process by which the U.S. government construes information and turns it into knowledge, that is a problem for the whole world.

No comments: