MUCH OF the writing on code-breaking during World War II suffers from an occupational hazard: the reader needs a code book in order to decrypt the prose. This is, thankfully, not the case in The American Magic, Ronald Lewin's study of the contribution of military intelligence to the defeat of Japan in the Pacific War. The technical aspects of signal intelligence, the complex world of codes and ciphers, are explained straightforwardly, so that the layman can appreciate the enormous difficulty of the American cryptanalysis effort without having to go through the experience himself. The human elements of code-breaking, the men, their triumphs and tragedies, are Lewin's main concern.

The need for military intelligence is not always appreciated in American society. When he was Hoover's secretary of state, Henry L. Stimson closed down the code- breaking shop, proclaiming that "gentlemen do not read each other's mail." Later, as Roosevelt's secretary of war, Stimson was rather less interested in such gentlemanly virtues and fully supported the Magic efforts.

The outlines of their success are well known. We broke the Japanese diplomatic code, known as Purple, well before Pearl Harbor. Unfortunately, as the Tojo government did not inform its own ambassadors of the Oahu plan, "listening in" to their communications did not help us prevent the surprise attack on December 7th. JN25, Tokyo's naval code, was a much tougher nut to crack. We began decoding their messages at sea in the spring of 1942, yet as the Imperial Navy routinely altered its code, the Magic cryptanalysts faced the nightmare of starting over periodically throughout the war. The Japanese army code finally yielded its secrets in 1943, so that by the end of the war Japanese signal traffic was largely transparent. It was no longer difficult to read their mail, but rather, as the military informed Roosevelt in 1944, "the problem is how to avoid being buried under the mass of information."

How vital was this code-breaking effort to our fighting forces? Very. A contemporary's desperate plea for secrecy is a more moving testimonial to Magic than any later historian's analysis.

In September 1944, Gen. George C. Marshall sent a secret letter to Republican presidential candidate Thomas E. Dewey, upon learning that Dewey had heard of the Magic success. Dewey, believing that Roosevelt must have known about the Pearl Harbor attack, was planning to turn code-breaking into a campaign issue. Marshall feared "tragic results" if "the least suspicion were aroused" in Japan and tried to convince Dewey of Magic's continued importance: "The battle of the Coral Sea was based on deciphered messages and therefore our few ships were in the right place at the right time. Further, we were able to concentrate our limited forces to meet their naval advance on Midway, when otherwise we almost certainly would have been some 3,000 miles out of place." The chief of staff reported that our island-hopping strategy depends on exact intelligence of enemy strength, and the destruction of Japanese shipping results "from the fact that we know the sailing dates and routes of the convoys and can notify our submarines to lie in wait at the proper points." In the end, Marshall was persuasive; Dewey kept the secret.

All of this has been known by scholars for some years and, thus, Lewin's book breaks very little new ground. Unlike the Ultra secret--the British breaking of the German ciphers which was kept under cover until 1974--the existence of Magic was made public in the harsh light of the Pearl Harbor Hearings immediately after the war. Yet, if the dust jacket's claim to present a "fundamental recasting of the history of the war against Japan" is quite disingenuous, Lewin's study is still valuable in that it provides an opportunity to examine the true nature of intelligence in war and peacetime.

For despite the remarkable technical achievement of Magic, this episode serves as a reminder of the inherent limits of the intelligence trade. Even if one possesses absolutely accurate intelligence on the enemy, the distribution of this information is always immensely difficult: the dangers of inter-service rivalry, the need to preserve secrecy, the reluctance of generals to believe their intelligence officers, all can be seen in Lewin's history of the Pacific War.

A more profound limitation of intelligence stems from the fact that, despite the apparent mechanical nature of the task, one is always dealing with an utterly human enemy. A human enemy reacts: if one prepares for an anticipated enemy action, "accurate" information can become false, as the enemy may change his plans seeing that surprise has been lost. A human enemy suspects: if one acts on signal intelligence too often, the enemy is likely to discover that his codes have been broken and change them or practice deception. Lastly, a human enemy decides: intelligence predictions often must be modest simply because one cannot know what the enemy does not know himself. In the days before the outbreak of the Pacific War and before the dropping of the Hiroshima bomb, American officials could "overhear" the Japanese debating their options, but could not know, as indeed the Japanese could not themselves know, exactly what would be Tokyo's final decision.

Most importantly, a historical awareness of the Magic success should not engender confidence that it can be easily repeated. Statesmen and soldiers have to make many decisions in a kind of twilight: the other side's words and actions are usually ambiguous or contradictory. Clear vision is required, and knowing what you want to do is still as important as knowing what the enemy plans to do. Diplomacy and war turn on the play of stragegy, sacrifice and chance. The legacy of the Magic triumph in the Pacific would be ill-served if it led us to believe that even the best intelligence information can make our defense decisions for us.