How do we explain the bizarre recent National Intelligence Estimate (NIE) on Iran, which stated in its opening sentence that the ayatollahs had halted their nuclear weapons program in 2003, even as, tucked away in a footnote, the same document noted that the most critical component of such a weapons program--uranium enrichment--was proceeding at full tilt?

Even Mike McConnell, the director of national intelligence, the man who presides over the 16 agencies that comprise the U.S. "intelligence community," was compelled to back away from the assessment issued by his own subordinates. In retrospect, he said in testimony before Congress on February 5, "I probably would have changed a thing or two" in the way the intelligence was presented to the public. The "halt" referred to in the NIE, he conceded, involved the "least significant part" of the program, which was "the only thing" in the Iranian nuclear effort that actually may have stopped.

McConnell's repudiation of the work of his own staff raises some obvious questions about the organization of his office. One such question: Is anyone in charge of quality control in the cockpit of the most pivotal intelligence position in the United States?

There is in fact a bureaucratic unit designed for such a task. Back in September, McConnell appointed an "assistant deputy director of national intelligence for analytic integrity and standards." The occupant of this position is to ensure that intelligence products--reports like the NIE--are created according to accepted norms and are vetted properly for accuracy and lack of bias.

Who holds this critical job? The present incumbent is one Richard H. Immerman, who up until his appointment was a professor at Temple University and the author of a number of books, including one about CIA depredations in Guatemala. Immerman has an essay in the current issue of Diplomatic History, the scholarly journal of the Society for Historians of American Foreign Relations, and it makes for interesting reading.

Titled "Intelligence and Strategy," Immerman's article is a survey of the various ways that intelligence has been distorted for political purposes across the history of the CIA. It ranges over a great many issues of historical importance, including Vietnam and the Soviet-American arms race, and also more recent controversies, like the Bush administration's decision to depose Saddam Hussein.

Immerman, it emerges quite rapidly in the article, is not a dispassionate student of these matters, but a combatant in the political intelligence wars himself. He traces a fair amount of the CIA's present troubles back to George H.W. Bush, who served as agency director under Gerald Ford. Once ensconced at Langley, writes Immerman, the future president "kowtowed to apoplectic conservatives," who were accusing the CIA of minimizing the pace and scale of the Soviet nuclear arms build-up. This led him to establish a group of outside analysts--dubbed Team B and headed by the "rabid anti-Soviet ideologue Richard Pipes"--to examine critically CIA findings. Team B "predictably ravaged" the existing CIA estimates and "undermined the agency's credibility" even though all of its findings, according to Immerman, were themselves just plain "wrong."