The debate over strategic defense.

The United States is in the midst of a wrenching debate over how to prevent nuclear war. Questions about deterrence and weapons of mass destruction - for years the nearly exclusive domain of technicians and theoreticians - are now openly argued by scientists, strategists, politicians, theologians, and the public. It is a sharp and often bitter exercise unseen since the first relatively puny and unsophisticated atomic bombs exploded 40 years ago.

One event propelled the issue onto the front pages and into dinner-table conversation: President Reagan's speech 18 months ago in which he called for new efforts to render nuclear weapons ''impotent and obsolete.''

The speech - which startled the Pentagon's research establishment - has accelerated Defense Department activity in military space systems. It has set the aerospace industry scrambling for billions of new dollars. It has galvanized defenders of arms control treaties. And it is the reason Washington and Moscow toyed with negotiations on space weaponry that might have started tomorrow in Vienna, but won't.

The strategic defense initiative (SDI) - dubbed ''star wars'' by critics - holds out the possibility of defending the US and perhaps its allies against nuclear attack by the Soviet Union. It relies for its promise on the striking advances in science and technology made in recent years. Ultimately, it would have to include space-based systems to detect and destroy enemy warheads before they reach their targets.

During most of the last 20 years - when the Soviet nuclear arsenal began to catch up with that of the US - the threat of ''mutual assured destruction'' (MAD) has reigned. In sum, MAD says: There is nothing you can do - even launching a first strike - that can prevent us from retaliating with unacceptably devastating force.

It has never been clear, however, that the Soviets accepted this theoretical notion, let alone its reality. Survival - whether it be from an attack by Mongol horsemen, Napoleon's troops, Hitler's tanks, or nuclear weapons - has always been paramount in Russian history. And Soviet military planners see their country's survival in the nuclear age dependent on their ability to launch a preemptive strike.

As nuclear weapons on both sides became more numerous and - most important - very accurate, many strategists in this country grew doubtful that MAD could remain a credible US strategic doctrine.

It is now possible for intercontinental ballistic missile warheads to destroy hardened missile silos and command facilities. Next-generation strategic missiles - like the US Trident II submarine missile - will have warheads that can maneuver even closer to their targets. Warheads on the US-built Pershing II intermediate-range nuclear missile, now being deployed in West Germany, can already do this.

Does this mean that a nuclear attack could wipe out another country's retaliatory capability in a ''counterforce'' strike? If so, the basis for stable deterrence could be undercut, especially if advances are made in antisubmarine warfare (since most US strategic warheads are on relatively invulnerable subs).

Beyond questions of force vulnerability, many experts as well as laymen also wonder if there isn't something better than the ''balance of terror'' to deter nuclear war, or at least reduce the terrible results of a nuclear holocaust. That is why a clear majority of Americans (more than 80 percent, according to several polls) favor some sort of active defense against nuclear attack.

Two groups influenced Mr. Reagan's quest for such defenses, according to those who took part in the early discussions leading up to his speech:

The first included key members of his pre-presidential ''kitchen cabinet'' - especially businessman Adolph Coors and physicist Edward Teller. They reinforced Reagan's natural suspicion of arms control treaties by urging him to consider ways of countering nuclear weapons instead of simply assuming that Soviet ballistic missiles, once launched, would reach US soil.

The second group consisted of senior military leaders, who have always worried about the military utility of strategic nuclear weapons. They became increasingly concerned that expensive and politically controversial strategic modernization programs such as the MX missile could undercut improvements in conventional forces, which is where the services would prefer to concentrate.

When the Joint Chiefs of Staff expressed their hopes about ballistic missile defense to the President just a month before the speech, their concerns ''came through loud and clear,'' says one Air Force officer.

Since then, despite the fuss over the proposal, not much has happened to advance the cause of strategic defense that would not have happened in any case.

The Strategic Defense Initiative Organization (SDIO) at the Pentagon is still ''in the throes of getting set up,'' as one officer involved puts it. This effort includes only about 60 people, fewer than the 80 authorized for this fiscal year. Funding for strategic defense this year (and next) is much less than 1 percent of the Defense Department budget.

Research on related technologies continues as it has for some years, and most of this is not on exotic weapons but more challenging sensors and computer software.

But beginning next year, and assuming Reagan is reelected, the push for strategic defense will intensify greatly. The SDI budget request will double to nearly $4 billion for fiscal year 1986, totaling some $26 billion over five years. This is about $10 billion more than was being planned for that period before the President's speech last year.

According to Pentagon officials, the biggest increases will come in space-based sensors (to be tested in the late 1980s or early 1990s) and kinetic-energy weapons - very high-speed projectiles that home in on their targets thousands of kilometers away.

It is thought that these will be essential to protect battle-management satellites (''battlesats'') in space.

''We've accelerated those efforts a lot,'' says a Pentagon official involved in strategic defense.

''Big increases'' are also planned for computer software development, says this official, where ''the problem is orders of magnitude higher than anything we've ever handled before.''

Since the issue is so touchy, however, there is a conscious effort to lie low until after the November election. The SDIO director, Lt. Gen. James A. Abrahamson, has declined recent public appearance requests.

''Everybody, in a way, has wanted to play it down because of its potential for getting caught up in election-year politics,'' said Colin Gray, a member of the Reagan administration's General Advisory Committee on Arms Control and Disarmament.

''You really want to discourage people from making speeches on it at the moment,'' he says.

Officers working on strategic defense expected the Democratic platform to blast the program.

But they weren't much happier with the publicity that accompanied a glowing Republican endorsement. ''We didn't like either of the platforms,'' one officer observes wryly.

Big questions may remain about whether a space-based missile defense system would work.

But this has not stopped defense contractors from lining up for the billions of dollars to be spent on beefed-up research.

''All of the major aerospace industries are deeply and heavily into this now, and all of them view this as an opportunity they can't pass up,'' says one company executive. ''Twenty-five billion dollars is an awful lot of money. That's a strong incentive for any company to pursue it, even if Congress cuts it in half and we only get 10 percent.''

Meanwhile, interservice rivalry and bureaucratic foot-dragging - not unknown in a massive organization in which supporters of programs threatened by something new fashion their own strategic defense - continue to play a part. ''There are a lot of wars going on,'' one closely involved officer observes. This source says the Navy, in particular, ''is just not playing in this thing except in a one-half-of-one-percent way.''

Defense Secretary Caspar W. Weinberger and other administration spokesmen have suggested that strategic defense technology might one day be shared with the Soviets so that both superpowers could feel protected and therefore less likely to initiate nuclear war. But little serious consideration is being given to this idea, except in the sense (as one officer puts it) that ''it makes good press.''

SDI promoters also find it difficult to explain how nuclear weapons could be ''rendered impotent and obsolete'' without creating an ''astrodome'' of leakproof protection over the US.

Reagan may have been speaking rhetorically with a sense of vision, but no one professionally involved with strategic defense believes that this is possible or even necessary.

General Abrahamson and others explain that to inject enough uncertainty into Soviet planning to prevent Moscow from launching a first strike, the US only needs to field a ''robust'' - not a leakproof - defensive system.

''With sufficient defense, the military utility of nuclear weapons would be reduced to the point where they are, in effect, 'obsolete,' '' says Edward T. Gerry, a member of the scientific panel that explored the possibilities of nuclear defense for the Defense Department last year.

''If deterrence of nuclear war by whatever means is successful, then the population is 100 percent protected,'' he says.