We publish on the opposite page today a ringing endorsement of the analytical quality and the political integrity of the CIA's intelligence assessments. It comes from one who is in a position to know a good deal more about this supremely important matter than the rest of us -- Robert Gates, the civil servant in charge of the process out there in Langley. He declares that the CIA has paid systematic attention to creating intellectual and bureaucratic conditions favorable to good intelligence judgments. The hallmarks of its system, Mr. Gates says, are rigorous review, an openness to criticism and a devotion to professionalism.

Except that we still can't really tell whether the CIA is doing it right. Mr. Gates, as a professional skeptic, might be among the first to acknowledge that he has not put before the public the materials for a trustworthy assessment -- he is not free to. For that we would have to know many other things about CIA: what questions of analysis or policy it had been asked to address, how it had handled the evidence in different kinds of cases, what had been done when a hard-charging analyst felt stymied by unworthy obstacles.

It seems unlikely to us that a simple lack of intellectual or physical resources would be a major difficulty now. Raw politicization can perhaps be dealt with. But there is a range of more subtle traps, bureaucratic, intellectual and cultural. The question of whether intelligence is designed not just to prevent surprises but also to predict the course of events is a continuing conundrum. Ferreting out or minimizing bias of various kinds is essential, but how do you ensure or maximize insight? Is the team approach institutionalized in the American intelligence community conducive to the play of minds -- to the play of a single mind -- which produces the major breakthroughs in other fields? We do not mean to trivialize Mr. Gates' account, but if we told you that that is the way we write editorials, you would be forgiven if you stopped reading.

Still, progress of sorts is visible. For the first three postwar decades in which intelligence was a central preoccupation of national policy, the work was secret and what discussion there was of it took place mostly behind closed doors. In the next decade, the debate went public and focused heavily on clandestine intelligence operations. The main subject, however, has always been intelligence collection and especially analysis, and in recent years enough questions about American performance have been raised to bring this subject to the fore, inside as well as outside the intelligence community. Mr. Gates carries forward this essential inquiry.