It recommends that the CIA consider disciplining then-Director George Tenet and other senior CIA officials. The recommendation was rejected by Porter Goss, who was director of the CIA in 2005, when the report was produced. That rejection was reaffirmed Tuesday by Michael Hayden, current head of the CIA, who also objected to the report's declassification.

The Office of the Inspector General is an organization internal to the CIA whose task it is to investigate failures and misconduct within the agency. It is the internal watchdog. It was rumored that the inspector general had issued a blistering report on the CIA's pre-9/11 performance, but it was not known (at least to us) that disciplinary action had been recommended. There is always tension between any federal agency and its internal investigations unit, but generally investigators are gentler than a congressional committee -- especially on senior management's execution of its job. To recommend discipline for the top officials of the agency is a pretty startling step.

The heart of the report levels this criticism at Tenet: that a strategy for fighting al Qaeda was needed but that it didn't exist, and that Tenet "bears ultimate responsibility for the fact that no such strategic plan was ever created." The report also criticizes the CIA for failing to maintain an effective watch list of potential terrorists, making it impossible to screen threats. It also says that Tenet diverted money allocated to counterterrorism programs for other intelligence uses unrelated to terrorism.

The most important criticism, of course, is the lack of a CIA strategy for combating terrorism. Over the years, the CIA had become driven by process. Obviously process is an important aid in achieving goals -- but in some organizations, and it would appear in the CIA, process stops being a tool and becomes an end in itself. What that means, in practical terms, is that getting the wrong answer became tolerable at the CIA, so long as the process was followed. Getting the right answer was unacceptable if it did not follow the process. One obvious problem is that gut insights do not map well to processes, but it is frequently those insights that get you where you need to go in intelligence.

The problem being raised here is the tension between process and strategy. Process is designed to serve as a template for recurring events -- so the same thing is done the same way each time. You can't generate a strategy via a process. Strategy, the broad approach to a problem, doesn't turn into a process because -- at least in intelligence -- every case is so different. Using the same process to mount an intelligence operation against the Soviet Union and to deal with al Qaeda makes little sense.

The CIA under George Tenet didn't search for a strategy for defeating al Qaeda. It didn't take apart al Qaeda, identify its weak point and systematically attack it. Rather it tried to create a process for dealing with terrorism. In trying to build a replicable, definable process, it failed to understand its enemy and therefore never created a strategy.

Strategy is to process as Clausewitz is to a PowerPoint. It is not clear whether the U.S. intelligence community or the military has learned this lesson. Understanding the nature of strategy is difficult, disorderly and can't be reduced to three bullet points. Process is easier, orderly and can be briefed in 15-minute sessions. Tenet rejected the charges in the inspector general's report. He had built sophisticated processes. But as the report said, he never built a strategy.