PRESIDENT CLINTON has promised to decide by this summer whether to move ahead with deployment of a national missile defense system. The promise was made mainly for political reasons; it was a way of buying time and maneuvering room on the issue a year ago.

It has become increasingly clear, however, that by summer Mr. Clinton won't have sufficient information to make a reliable decision. Tests of the elaborate technology have been inconclusive and likely will remain so. The decision should be deferred; the administration should begin to lay the groundwork for deferral now. The issue is too important to be driven by election-year considerations. This is not the right way to measure which party is strongest on national defense, and the president should muster the courage to say so.

As currently envisioned, the system would be a defense against a limited attack, not a massive one. The principal threat is no longer thought to lie with the former Soviet Union but with smaller states such as North Korea and Iraq. About 100 interceptors would be built at an estimated cost of $12.5 billion (and rising). Democrats have tended to be skeptical about constructing even a limited system, partly on cost-benefit grounds, partly because deployment would deviate from the 1972 Anti-Ballistic Missile Treaty and complicate further nuclear arms negotiations with Russia. But Republicans have urged deployment and made it into a totem of dedication to defense. They were helped in this by a North Korean test firing in 1998 that showed that country to have a greater missile capability than U.S. intelligence agencies had previously believed. A well-respected congressional advisory panel in 1998 also urged deployment. The president last year yielded, though with asterisks, signing legislation that said a system would be deployed if "technologically possible" and promising the decision this summer.

But based on what? After years of disappointment, the Pentagon finally succeeded in intercepting an incoming missile in a test last October, but critics say the test was eased, if not quite rigged, to produce the result. A further test over the Pacific last week resulted in a miss, though apparently not by much. A third test is scheduled this spring. But a Pentagon panel has meanwhile warned against a "rush to failure," and after last week's result, even some Republican supporters were urging that a decision be postponed. The sensible course is to put off deployment, continue research and development--and engage in serious further discussions with the Russians after their own elections this spring. Delay in this case is not a way of getting to no by another name. It's a way of getting to the right answer.