HOW curious that the best recent book on American secret intelligence should be by a British scholar. Prof. Christopher Andrew of Cambridge University, who has previously studied the British secret service and the Soviet KGB, has now produced an absorbing and comprehensive narrative of American history in espionage back to George Washington.
His device is to tell the story in terms of how presidents interacted with intelligence agencies, often with profound effects on both. Thus, a distracted President Wilson was deceived into believing British disinformation about a German-Mexican conspiracy that hastened the United State's entry into World War I.
On the other hand, presidents like Eisenhower and Bush were wiser about the uses of intelligence, mainly because their previous experience enabled them to appreciate the marvels of technical intelligence.
It should be noted that throughout ''For the President's Eyes Only: Secret Intelligence and the American Presidency From Washington to Bush,'' the author leaves little doubt of his predilection for technical intelligence, on which Britain and the US cooperated fruitfully, over human intelligence, important though it was. He cites countless examples of the exploits of SIGINT (signals intelligence, which is basically code-breaking) and IMINT (imagery intelligence, which is pictures from airplanes and, recently, orbiting satellites).
Andrew suggests that presidents who relied more on HUMINT (human intelligence) than SIGINT sometimes made disastrously wrong judgments. For example, had President Roosevelt heeded Japanese naval communications more and human reports less in 1941, he could have anticipated the attack on Pearl Harbor.
If Andrew prizes HUMINT less than SIGINT, he appears to have utter scorn for the cloak-and-dagger covert operations side of intelligence.
Unlike the older British secret service, functioning mainly in peacetime, the Central Intelligence Agency came into being with an emphasis on clandestine action inherited from Allen Dulles and the wartime Office of Strategic Services.
Some presidents were more interested in what the CIA could do than what it could learn. President Eisenhower relished the agency's ability to perform such feats as overthrowing left-wing regimes in Iran and Guatemala. President Kennedy was dazzled by counterinsurgency and guerrilla action until the Bay of Pigs fiasco soured him on covert action.
Meanwhile, technical intelligence was winning the cold war -- the U-2 spy planes deflated the fear of a ''Soviet missile gap'' and spotted the missiles deployed in Cuba in 1962. The shooting down of a U-2 over the Soviet Union in 1960, Andrew says, ''marked only a temporary setback in one of the most successful intelligence operations of the twentieth century.''
In his antipathy to covert action, the author stacks the cards a little. He describes the Watergate coverup and the Iran-Contra scandal as failed covert actions. That was true only in a general sense; these were not intelligence-agency actions, but actions spawned by the White House.
But this is only a minor flaw in this sweeping account of how presidents have used -- and misused -- the intelligence resources. With the cold war behind us and a disorderly world facing us, one can only say amen to the author's closing line, ''The presidents of the twenty-first century, like their Cold War predecessors, will continue to find an enormously expensive global intelligence system both fallible and indispensable.''