Doomsday scenarios are back. The most credible has Pakistan unleashing a pre-emptive strike against India. Or perhaps North Korea, a suspected nuclear power, attacks South Korea. Or maybe China hits Taiwan with a few nukes before a land invasion.

After a decade of civil wars in Africa, the Balkans and Russia, as well as the spread of the AIDS pandemic, the world's most serious threats seem decidedly non-nuclear. Not so. "Nuclear weapons have become usable again," says Sergei Karaganov, chairman of the Council on Foreign and Defense Policy, a Moscow think tank that's close to the Russian government.

Anxieties about nuclear weapons are rising, fed by the new Bush administration's avowed interest in building a missile "shield." Defense experts fear this could spawn a new arms race--or worse, launch an actual nuclear attack.

Even before the new president can decide whether a missile-defense system is technically feasible, political costs are mounting. European leaders are bewildered by the U.S. position, and China is angry. Meanwhile, Russian President Vladimir Putin warned in January that a missile shield would "irreparably damage" global stability and probably kill the possibility of an agreement between Russia and the United States to dramatically reduce weapons stockpiles.

What's pathetic about all this is that even if a missile defense system were to be developed, which is itself unlikely, everyone from the CIA to independent security experts say that a missile is the least likely form of nuclear attack on the United States--for the obvious reason that such a missile would carry a "return address," guaranteeing swift and certain U.S. retaliation. More plausible threats come from below-radar "cruise" missiles, ships and even hand-delivered bombs.

So why is missile defense so appealing to the new administration? To start with, U.S. political leaders have long sought to placate Americans by promising technological "fixes" for the inherent instability of being a world superpower. And then there is the need to find new rationales for huge increases in spending on military innovation. Bush clearly sees in a missile defense system a respectable cover for a giant defense boondoggle (official price tag $60 billion but likely far higher) of the sort not seen since Reagan launched "Star Wars."

There is a long history of the U.S. government setting fantastic goals for nuclear weapons builders to attract funding and public fascination. In the '50s, Edward Teller, then chief of one of the nation's two bomb labs, personally lobbied President Eisenhower to endorse work on a radiation-free nuclear bomb. No matter that Teller himself knew such a "clean" bomb was a fiction. The mere possibility convinced Eisenhower to oppose a ban on nuclear testing, which was Teller's real aim.

Graham Allison, a nuclear strategist at Harvard University, concedes that domestic political needs have always driven U.S. administrations to push weapons systems with little regard for the effects on international relations. Now it's déjà vu all over again. Allison warns that many countries are unlikely to understand that U.S. politicians aren't serious about missile defense. He worries that China, which he estimates has only 20 working nuclear weapons and none deliverable by missile, might choose to launch a pre-emptive strike against a U.S. ally (such as Taiwan) in order to reset the international balance of power prior to the deployment of a missile defense system.

The Bush administration, of course, wants to undermine any international backlash by insisting that U.S. allies would be covered by any defense system. But this offer is meaningless. "A defense against nuclear missiles is technologically possible, but it won't provide a real cover," Karaganov says.

And this is precisely the point. Reducing nuclear tensions must come the old-fashioned way, by patient negotiation, compromise and giving incentives (or imposing penalties) on nations intent on developing nuclear weapons. Until all countries agree to foreswear nukes and destroy their stockpiles, the risk of nuclear war--accidental or otherwise--remains. No technological fix can change this.