Melvin Goodman, in his Dec. 19 Outlook article, "Who is the CIA Fooling? Only Itself," said that the CIA failed because it did not predict the collapse of the Soviet Union and the Warsaw Pact. From the perspective of the policymaker, I think he got it wrong.

Did the CIA tell President Bush on Inauguration Day that on Dec. 25, 1991, the Soviet Union was going to disappear? No. And if it had? The policymakers would have treated it as nothing more than an informed "guess," perhaps better than many, but nevertheless a guess. More important, policy choices themselves, and a myriad of other variables, modify the future and therefore the intelligence community's "estimate" of that future. Formal intelligence "estimates" of future events, apart from the intellectual value of the preparation process, may be the least valuable intelligence product for the policymaker.

The most difficult task the foreign affairs policymaker faces is making decisions in an environment of ambiguity and inadequate information. The role of intelligence is to narrow the range of uncertainty within which a decision must be made. What really matters is not how well the intelligence community predicts particular events but its ability to spot, track and interpret trends and patterns.

A principal policy goal of the Bush administration in the last days of the Cold War was to encourage liberalization in the Soviet Union, and especially in Eastern Europe, but at a rate that would not result in a crackdown by Soviet security forces. Our problem was, we did not know what rate of movement was sustainable. The CIA's analysis of the situation helped to keep our policy within sustainable bounds. Had there been an "intelligence failure" in this case, we might still have a hostile Soviet Union facing us.

BRENT SCOWCROFT

Washington

The writer was President Bush's national security adviser.