I WAS INTRODUCED to nuclear war in the mid-1950s, hiding under my school desk during civil defense drills, hoping the Russian bombers would never come. It never crossed my mind then that I would someday be working on nuclear strategy at the White House, hoping still that nuclear war would never come -- and realizing how easily it might.

I came to Washington in the 1960s to work for a defense think tank, and within a few months I found myself at an Albuquerque conference, sitting in a Holiday Inn bar listening to war stories. Nuclear war stories. That's part of the nuclear trade -- making fun of yourselves, trying to find ways not to take yourself too seriously, even having a good time now and then.

My favorite tale that night was about Rarotonga, a dreamy island in the South Pacific. For several months in the early 1960s it was an outpost for a two-man crew manning a radar that observed atomospheric weapons tests, clearly a hardship assignment. The men's only contact with civilization was a weekly supply plane, which kept breaking down on the island.

In fact, it broke down so often that the regional military commander sent a special mission force to find out what was up. The special mission discovered that Rarotonga was a Polynesian paradise with lush tropical fruits and affectionate maidens straight out of a Gauguin painting. Rarotonga was taken off the hardship duty list.

Within a year or so, my think-tank studies of weapons effects gave way to studies of the weapons themselves and to communications systems and missile warning systems. Then came nuclear "exchange" calculations: our missiles against their missiles, their missiles against our bombers, their subs against our bombers -- endless combinations.

There were no people involved in these "exchanges," only calculations. It was a curious fiction, never discussing the humans at the military installations or the industries or the cities. I guess that made it easier on the targeteers in Omaha, the people there in charge of launching the missiles or the bombers, and the analysts like me.

I recall one Saturday a colleague came into the think tank office with his wife to find me sticking different- colored pins -- representing different-sized weapons -- into a map of the Soviet Union. Add a pink pin for Minsk -- another 200,000 dead. My colleague's wife was horrified. But when the pin went into Minsk or Moscow, to me it was just a pin. I didn't see people working or children playing. I assumed that someone above me in the system thought about those things. I just stuck in the pins.

In 1969, the Strategic Arms Limitation Talks (SALT) began, and I found myself trying to find the combinations of weapons limitations and verification provisions that would be acceptable to us and our allies as well as to the places where the pins were being stuck. At first blush the problem looked easy to me. But an older colleague told me I had a lot to learn. He was, of course, correct.

Within a month I had met the first of a small but not uninfluential community of people who violently opposed SALT for a simple reason: It might keep America from developing a first-strike capability against the Soviet Union. I'll never forget being lectured by an Air Force colonel about how we should have "nuked" the Soviets in the late 1940s before they got The Bomb. I was told that if SALT would go away, we'd soon have the capability to nuke them again -- and this time we'd use it.

As the SALT negotiations began in earnest, I dug into many studies at the think tank for the Pentagon -- and almost immediately came face to face with the ultimate questions of the nuclear war trade: How much is enough? What is the "threshhold of pain" for the Soviet decision-makers? What level of destruction will deter Soviet attack? Is it measured in industrial capacity? In war machines? In Soviet citizenry? In some arcane combination of these and other factors which a careful reading of Russian history and of recent articles in Red Star would divine?

My rite of passage was complete. The scientist -- whose main interest in graduate school was trying to obtain commercially useful energy from controlled fusion -- had become the policy analyst playing nuclear war. The policy analyst went to the White House.

I was at the White House's National Security Council only a few months when it was time for a SALT negotiating session to begin in Geneva. One of Secretary of State Kissinger's division heads asked me to draft a set of instructions for the American delegation. I asked what to put in the instructions -- and was told just to do a draft on my own, with one cover memo to Kissinger and another from Kissinger to the president.

Three days later I got the package and the instructions back. The person who had asked for the draft had not changed a word. Nor had Kissinger. Nor had the president. The instructions were on their way to Geneva. I swallowed hard.

Those people above me who were supposed to be thinking about the Big Questions were relying on me to think about those things. I was to make decisions in the nuclear war trade, not just stick in pins. So I began to think about many things.

I thought about the fact that nobody else around the White House seemed to understand nuclear war issues better than I did; knowing my limitations, that did not reassure me. I thought about the organizational chaos at the White House, the haphazard way decisions often were reached. I thought about the minimum amount of time the president had to spend on nuclear war issues, his ultimate responsibility. I thought about the comment of a former presidential science adviser, similarly struck by the way major decisions are made, who asked, "Where are the grown-ups?"

His comment is apt. There is a good deal of childish behavior in the White House, including temper tantrums. As I worked my way up the nuclear war trade, I kept looking for the place where people "had it together." The last place I expected to find adults losing control of themselves was in White House rooms with nuclear war planners. But there the tantrums were -- directed at officials of other countries, at briefing books, at staff, at other high U.S. officials, at almost anything you can think of. I had hoped that the White House's nuclear war business was in the hands of people who were rational and calm under pressure. I was learning.

In time I learned to live with all of this. But to friends -- who regularly asked questions like "Not going to get blown up soon, are we?" or "Things in the nuclear war business okay?" -- I confided that it was the ultimate example of "in the land of the blind, the one-eyed man is king." Many of them thought I was joking. I wasn't.

As the shock of these experiences gradually wore off, I joined with some of the most selfless and dedicated people I have ever met in trying to help the president perform the hardest job in the world. I watched three presidents who were deeply concerned about the problem of preventing nuclear war leave the White House with a sense of frustration. Each sought to leave the American people with a legacy of security with respect to nuclear war, a confidence that nuclear war would not happen. Each failed.

I felt that same sense of frustration and failure, especially in early 1980, when the struggle to save SALT II and the work of three administrations ended with the Soviet invasion of Afghanistan. I had expected to spend the first few months of 1980 carrying the case for the treaty -- "modest, but useful," in the words of the Joint Chiefs of Staff -- to the Senate floor. I knew it would be a real challenge: I had discovered, to my dismay, that most of the senators on the Armed Services and Foreign Relations committees -- those making critical decisions, to say nothing of endless speeches -- lacked even a rudimentary understanding of the nuclear war business.

When SALT II was defeated, I had some time to ponder how we had gotten ourselves into the awful mess we were in.

The factor that stood out in my mind was the seeming lack of understanding of just how great the chance of nuclear war really was.

I knew the weapons systems, the communications systems, the warning systems. I understood the exchange calculations and the nuclear policy questions. I had seen how the White House and the so-called chain of command operated -- and assumed the Russians were probably worse. I knew how poorly we understood the Russians -- and how poorly they understood us. I could see the rising problem of nuclear proliferation vastly increasing the risk of superpower confrontation.

Adding it all up was unsettling. There was altogether too much opportunity for machine error, for inadvertent human error, for errors in judgment. Nuclear war could occur far more easily than people in the White House, in the Congress and in the country at large seemed to realize.

In Thomas Pynchon's prize-winning novel "Gavity's Rainbow," two of the major figures, a statistician and a Pavlovian psychologist, debate the driving force behind human events. The statistician claims it's mostly random and unpredictable -- a lot of balls bouncing off each other governed primarily by the laws of probability. The Pavlovian argues for a much more deterministic world -- a world dominated by cause and effect, stimulus and response. I vote with the statistician.

If nuclear war comes and any historians survive, they will marvel at the role of chance in its genesis, it escalation, and its grim conclusion.

Some chance events -- which have taken us closer to the brink than is realized -- have of course already occurred. There was the mid-1960s incident in which U.S. radar mistook the rising of the moon for a missile attack. There was the 1979 mishap in which a computer with a practice Soviet missile attack tape on it was accidentally introduced into an operating missile warning system. There was the 1980 accident in which a microchip failed in a computer at Strategic Air Command headquarters in Omaha and the B-52s almost took off.

These, remember, are some of the unintended happenings that are known -- and only ones on our side. They can be multiplied by presumed mishaps on the Soviet side and by additional mistakes in other nations which have acquired, or are in the process of acquiring, nuclear weapons. It is by no means inconceivable that next time, rather than on a calm day when we and the Soviets are merely at our normal levels of enmity, a false alarm will occur in an atmosphere of crisis, with somebody suddenly heading for the Hot Line and trying to explain that it was just a mistake.

Ah, yes, the Hot Line. How many people know that it's a slow teletype machine, and that its use suffers from the usual problem of getting a good translation? I had witnessed two incidents in the SALT negotiations in which the United States and the Soviet Union had profoundly misunderstood each other in this fashion.

The first was at Vladivostok in 1974, when President Ford and Secretary Kissinger had come home in triumph with an agreement that was found to be no agreement at all when the sides tried to write it down in agreed language. A similar incident took place in the early months of the Carter administration, when an agreement on limiting new types of ICBMs evaportated into thin air over a language disagreement.

What if one of these "misunderstandings" took place in a crisis as the sides tried to control further escalation, rather than in the midst of a seven-year negotiation? In the nuclear war business, we cannot afford to lose anything in the translation.

It was also chance that these thoughts coincided with the birth of my second child in early February of 1980.

There's something in the birth of a child -- or the death of a loved one -- that is a reminder of both the miracle and the fragility of life. I had been educated as a physical scientist, but evolution and the majesty of life hav

e always left me in awe. The ability of the computers we call brains to reason and conceive, and of the extraordinarily "well-designed" machines we call bodies to build machines which build skyscrapers, missiles and spaceships, is humbling. Now there she was, a new person, a new being, demanding the right to live, to find out "why she came." And here I was, thinking of the risks of nuclear war.

I held forth on all this to a friend late one night when most sensible people have gone home or to bed. I railed away at the absurdity of the situation we Americans found ourselves in -- living in an imperfect world with imperfect machines and imperfect people making decisions on subjects they only partially understood. Something had to be done to change all this.

Clearly, at the root of the problem is that the public has scarcely any reliable information with which to develop thoughtful opinions about American nuclear policy. Policymakers, therefore, have little serious sense of public opinion to guide them (to say nothing of the fact that many officials frequently lack a genuine grasp of the issues themselves).

Sure, there are polls on the nuclear question. Yes, the American people are concerned about nuclear war. Yes, they want to pursue arms control negotiations. No, they don't trust the Russians. But that's where the information stops.

Nowhere do these polls tell us about the difficult decisions and trade-offs that are involved. We know from polls, for example, that two-thirds of the nation wants to pursue arms control with the Soviet Union -- and that at the same time two-thirds doesn't trust the Russians to adhere to such accords. Do Americans, therefore, want arms control negotiations or not? The polls, in their simplicity, have been part of the problem, not the solution.

A larger part of the problem, however, is that no effort has been made by the government to maintain public concern and understanding about the fundamental problems of nuclear war. Perhaps this is understandable. What president is going to send a message to the nation that he and his colleagues in Washington are losing their grip on the nuclear war issue? Public interest groups have made some effort, but they are small, uncoordinated, often suspected of being "softheaded lefties," and expend most of their energy in Washington.

It was clear that something was wrong, that the link between policymakers in Washington and the people we served was far too weak. We didn't understand their fears and frustrations; they didn't understand the complicated bases of our decisions. Only by providing careful and thorough information to public and to officials can we avoid the hysteria we often find on both the extreme left and right.

My interest in doing something about all this waned, though, as daily life took over again -- until chance intervened once more in the form of the abortive attempt to rescue the American hostages held in Iran in April of 1980.

The day after the raid, as we all waited to see how the Iranians would react and what fate held in store for the hostages, I encountered a friend, a general, in the dimly lit halls of the Old Executive Office Building. We both knew all too well that the favorite Pentagon war game scenario for the start of World War III was a crisis in Iran. Now we had one. What if the Iranians killed the hostages? What would the Russians do if we retaliated?

We talked about the uncertainties, and as the conversation drew to a close, he said, "You know, I called my kids last night." He hesitated and then continued. "I never call my kids." His kids were grown up, and I knew what he was saying: Was this it?

The final chance event that confirmed my determination to help correct our flaws involved another military officer. It happened at a meeting in the Pentagon when a Navy captain mused on the state of the world and offered the view that people in this country and Europe were getting much too excited about the consequences of nuclear war. He argued that people were "talking as if nuclear war would be the end of the world when, in fact, only 500 million people would be killed."

Only 500 million people. I remember repeating it to myself: Only 500 million people.

Then he went on to argue that within a generation, genetic engineering would make people immune to radiation. I reached for my hat, suddenly knowing how Woody Allen felt in "Annie Hall" when he excused himself from a conversation with her brother with the plea that he had "an appointment back on planet Earth."