Causing What It's Meant to Prevent

In a chauffeur-driven limo in what was then still communist Moscow, a pioneer of the Soviet missile program was describing to me the unique danger of systems so complex they might create the very situation they're meant to guard against, and so powerful they can't safely be triggered even once. His concept remains at least as relevant today as in 1986.

Visiting Moscow as a citizen diplomat right after the summit meeting in Reykjavik, I heard my eminent acquaintance start to argue against the "star wars" anti-missile hopes of President Reagan. As police waved the limo through traffic lights, I expected a defiant statement that, though the U.S. might try to bankrupt its rival, the Soviets would do whatever was necessary to match the build-up. Or perhaps the argument (which Gorbachev made at the summit, as I later discovered) that any anti-missile system could easily and cheaply be overwhelmed by a profusion of dummy warheads. Or that under guise of launching an anti-missile system, the U.S. would devise a space-based weapon with which to attack the "socialist" countries.

But no, this Soviet scientist voiced a concern that had never occurred to me. Any space-based system, he said, would have to defend itself instantly, without time for second thoughts by humans on earth. It would also require computer code far more complicated than written for any existing system. With the sides rushing to get their systems up quickly, the code would surely contain even more bugs than usual, he said.

One of the bugs in system A might falsely signal an attack by system B, causing an actual attack on the other space-based system. This would trigger a retaliatory response of nuclear rockets by the country that had put up system B, in which case side A, watching its radars, would launch its own rockets. In short, the scientist argued, the successful launching of the two systems would probably lead to the war it was intended to help guard against.

Whether or not the Soviets had the funds to launch such a system and would choose to do so rather than bewilder a U.S. anti-missile system with a scatter of decoys, my host had envisioned an example of a special class of dangers: a system so complicated that it could lead quite unintentionally, quite accidentally, to disaster. (The system could be financial, for example, as well as military.)

BC Crandall of SpaceWealth calls these dangers "surprising inevitabilities," an oxymoronic formula for catastrophes that cause officials to say with chagrin, "this was not supposed to happen." The financial writer Nassim Nicholas Talleb calls them "black swans." The terms used by these writers do not specify disaster, so we still need a general phrase that describes dangers that are not only supposedly low in probability, but also very high in negative impact.

Insurance firms operate on the basis of actuarial tables, based on many people dying at known ages, many houses burning down or cars crashing at known rates, allowing statistical predictions. While the death of an individual is a loss to his or her loved ones and friends, it does not threaten the continuation of the species or the collapse of civilization.

In the financial world, risk is routinely estimated for events that are negative but less than catastrophic. In a "very high negative impact, supposedly low probability" situation, one occurrence is by definition too many. Estimates of low probability are thus less than comforting, especially when based on very little (or no) data.

For roughly a half century our species relied on nuclear weapons to keep the peace, through a "doctrine" called "mutual assured destruction." (Satirists had sport with the acronym, MAD.) Though some U.S. warriors wanted to attack while the other superpower was relatively weak, and more recently to lord it over the whole planet, nuclear weapons are so terrifying that, with some exceptions such as the Cuban missile crisis, the two super-powers avoided direct military confrontation. Historians even attributed the "long peace" to the system of deterrence.

Imagine an epitaph for humanity that reads: "Oops: the system worked really well for decades, until suddenly it didn't." Among progressives, the goal of abolishing nuclear weapons has long been articulated, never more eloquently than by Jonathan Schell in The Fate of the Earth (1982) and subsequent books.

After only four or five years Schell was joined by some unlikely allies. At the Reykjavik summit in October 1986, the negotiations worked around to a surprising plan for reducing nuclear warheads to half within five years and then to zero within a decade. (We have the memoranda of these remarkable and once secret conversations, as prepared by both Soviet and U.S. note takers and now posted on line.) The most dramatic exclamation came when the U.S. Secretary of State burst out, with the proposal for reducing nuclear arms to zero on the table, "let's do it."

However, the proposal was the victim of Reagan's insistence on having the freedom to develop a space-based military system outside the laboratory, as he claimed was permissible under a broad interpretation of the Anti-Ballistic Missile Treaty. As he presented it, this system would be solely for defensive purposes and would be shared with the Soviets.

Gorbachev said that he could not take seriously the offer to share the system, given recent U.S. refusals to share technology for drilling oil wells and for processing milk. Unspoken but hanging like a fog in the Icelandic air was the Soviet apprehension that a U.S. space-based system would allow a first-strike on the USSR.

It seems that Reagan had a moral objection to a system that allowed no defense other than the threat to kill hundreds of millions of people in a retaliatory strike ordered by the President, by him. For this reason, he said, he was willing to abandon all nuclear warheads, provided that others did so. Dismantling of the Soviet nuclear establishment would have to be thoroughly verified at every stage. Don't worry, replied Gorbachev, it would now be the Soviets who would be pressing for (and agreeing to) careful verification.

Blame has fallen on Reagan for scuttling an historic abandonment of nuclear weapons just so the U.S. could be free to develop a defense that (a) would fail realistic tests, (b) could easily be overwhelmed in an exchange of missiles, and (c) wouldn't be needed if the nuclear powers disarmed. However, from the Soviet point of view, articulated in vague terms in Reykjavik, launching a space-based system might allow the U.S. to dominate the Soviets. Reagan was reduced to giving assurances of a kind he would never have accepted if the situations were reversed (it's solely defensive, we will share).

But the idea of going to zero never died. In January of 2007, the Wall Street Journal published an op-ed signed by four high former officials (two U.S. Secretaries of State, including the one who had been at Reykjavik, a Senator who had chaired the Armed Services Committee, plus a Secretary of Defense). They were concerned primarily about proliferation and about "non-state terrorists" getting "their hands on nuclear weaponry." They spoke about the need for nuclear weapons in the past tense: they "were" essential during the Cold War as "a means of deterrence."

The former officials foresaw the danger of "a new nuclear era that will be more precarious, psychologically disorienting, and economically even more costly than was Cold War deterrence." For this reason, they called for "a solid consensus for reversing reliance on nuclear weapons globally... ultimately ending them as a threat to the world."

Without quite using the word "luck," they asked, "will new nuclear nations and the world be as fortunate in the next 50 years as we were during the Cold War?" Their goal was to "rekindle" the vision shared by Gorbachev and Reagan at Reykjavik.

These authors of this historic op-ed joined in at least two subsequent appeals (Schultz, Perry, Kisisnger, Nunn) and one of them, former Senator Nunn has directed an initiative for securing warheads and fissile materials, in part so terrorists cannot obtain them. However, the Reykjavik idea of reducing nuclear arms to zero remains on hold, as if the answering system of the big powers were saying, "your call is important to us, please stay on the line," and so far failing to pick up the phone.

At President Obama's April 2010 Nuclear Security Summit, leaders focused on keeping nuclear weapons out of the hands of governments that do not yet have them and of non-state terrorists, and on cuts in the vast overkill in the hands of the two former rivals. The summit touched on many helpful means but not on abolition (which, like single-payer health care, was apparently "off the table").

Which takes us back to the Moscow limo and the example of a system so complex it unintentionally causes the disaster its builders seek to avert. Can we continue to focus on one crisis at a time (oil in the gulf, trouble in Afghanistan, reform of the financial industry) while assuming that a fragile and complex nuclear system will continue to protect us? The former U.S. officials ended their op-ed with these words:

"We endorse setting the goal of a world free of nuclear weapons and working energetically on the actions required to achieve that goal..."

If we adopted this goal, what questions would arise? We would ask:

• What are nuclear weapons good for? They provided deterrence during the Cold War, but didn't stop the U.S. from being defeated in Vietnam (1975) or the Soviets in Afghanistan (1989).
• How risky are they? According to former Secretary of defense Robert McNamara, it was only "luck" that allowed us to avoid nuclear war during the Cuban missile crisis.
• Under what conditions could we safely dismantle nuclear weapons, and what specific advantages would that bring now?