The Western consensus on deterrence, stability, and arms control forged by Robert McNamara lasted a decade and a half - a long time indeed in the nuclear era. But by the 1980s it was fraying under new technological and political strains.
The technological strains arise from the inexorable progress both in old and new weapons. Two decades ago technology was the friend of arms control, producing spy satellites to keep reliable tallies of the adversary's arsenals, preserving the stable balance of invulnerable weapons. Today technology is the opponent of arms control, concealing or obfuscating the adversary's arms inventories and again making major weapons vulnerable.
In any crisis situation, the new technology undermines predictability. It increases anxiety and puts a premium on fast rather than considered action. It threatens, in short, the breakdown of deterrence.
All this would not seem so dangerous if concurrent political developments had somehow acted as a counterweight. But as the technical capacity to control arms has been deteriorating, so has the spirit of political control of arms. World politics has maximized rather than minimized the growing technological risks.
In the peculiar logic of the nuclear age, the greatest damage wrought by technology has been the unanticipated increase in missile accuracy in the 1980s.
This wasn't brought about by any great discovery or scientific breakthrough. It resulted instead from what has been called ''technology creep.'' A host of small, tinkered improvements suddenly added up to a leap in precision that caught everyone - and especially the theorists of deterrence - by surprise.
Inertial navigation made possible the gyroscope logging of a vehicle's position by measuring acceleration from point of origin in three dimensions independent of outside readings, weather, or direction change of the vehicle. Geodesy, the mathematical determination of the curvature, shape, and dimensions of the earth and landmasses, became accurate enough to provide highly detailed mapping of routes to targets.
To these developments were added incremental improvements in computer and engine microminiaturization; order-of-magnitude acceleration of data processing; efficiency of solid propellant rocket booster motors; American NAVSTAR satellite position fixing of submarines to within 10 meters (11 feet) in three dimensions regardless of weather; real-time satellite reconnaissance of ground information; new materials; preprogrammed terminal homing computers in warheads; computer terrain-matching; infrared, radar, and other sensors that search out targets; and rapid retargeting of both intercontinental ballistic missiles (ICBMs) and submarine-launched ballistic missiles (SLBMs).
This swift evolution was no monopoly of the United States. The US pioneered the changes, but the Soviet Union has generally been only half a generation behind, testing weapons with performance characteristics equivalent to weapons the US is already deploying. US ICBMs currently have accuracies of 600 to 1,000 feet; Soviet ICBMs, of 1,000 to 1,500 feet, with the Soviet Union expected to match the present American achievement within a few years. These accuracies verge on the absolute precision at which not even ''superhardened'' concrete could protect missile silos.
Ordinarily, greater accuracy would be a welcome development. Ordinarily it would mean, for example, that military targets could be hit with less collateral loss of civilian life. But in the topsy-turvy world of nuclear deterrence this is exactly what is most dangerous. The overall nuclear balance - and therefore deterrence - depends on keeping both superpowers convinced that they could not wipe out the other's missiles in a first strike - and that their own military establishment and civilian populations would pay the gruesome price if they were so foolish as to try.
The new accuracy, on the contrary, holds out the tantalizing hope that just around the corner there may be a ''disabling'' first strike - a capacity to demolish the opponent's best weapons in a surprise attack. Under certain circumstances that prospect could look tempting. Deterrence of war could be weakened.
This prospect has been enhanced by the multiplier effect on the new accuracy of an earlier technological innovation: multiple independently targetable warheads (MIRVs) on single missiles. With the original single-warhead missiles, an attacking nation would always have expended more of its own missiles than the enemy missiles it would have knocked out, given the low accuracies and therefore low kill ratios of the time. The attacker would have always left himself with a risky inferior residual balance for any further combat.
With MIRVs, however, an attacker could use, say, 10 warheads from a single missile (at the standard two warheads per target) to wipe out five of the opponent's ICBMs in a first strike - a very favorable ratio of 1 to 5. The surviving balance after the initial strike would then strongly favor the attacker in any subsequent exchanges.
It was this post-first strike score card that gave rise in the late '70s to concern about an American ''window of vulnerability'' and a Soviet ''window of opportunity.''
As persuasively argued by Paul Nitze, then founder of the Committee on the Present Danger, and now chief US negotiator at the Geneva Euromissile arms control talks, this window would arise in the early 1980s. Given the disparity between the nuclear forces of the US (more warheads) and the Soviet Union (more missiles, including far more land-based missiles, plus much greater megatonnage of strategic nuclear explosives), the Kremlin could shortly launch a surpise attack on the US and wipe out 90 percent of America's land-based ICBMs, Mr. Nitze contended. (The US, with its far fewer ICBMs, could not wreak the same damage on the Soviet Union in any first strike until the late 1980s, since the accuracy required currently resides only in land-based ICBMs.)
After such a Soviet first strike, Mr. Nitze continued, the US would be crippled, because its few remaining ICBMs and its inferior residual nuclear megatonnage would be outclassed by the remaining Soviet ICBMs. Washington would of course not have enough ICBMs left to have the choice of retaliating against Soviet missiles. More significantly, it might not even dare to retaliate against Soviet cities - since it could do so only in the knowledge that American cities would then receive a much more devastating counterretaliation.
To Mr. Nitze's military concern about the early 1980s was added the political concern of Richard Pipes, a Harvard historian. Mr. Pipes, who is now a National Security Council adviser, suggested that precisely because Moscow now feels pressed by severe economic problems, potential crisis with restive nationalities in the Soviet empire, and an imminent succession struggle, the Kremlin might be tempted to exploit its window of opportunity and whip up anti-Western chauvinism to divert Soviet citizens' attention away from domestic troubles.
To classical arms controllers this whole scenario seemed far-fetched. Politically, they argued that the Soviet Union has never displayed the kind of wild adventurism the scenario presupposes - and especially not in a cautious succession period.
Militarily, they thought too that Moscow would never launch a bolt-out-of-the-blue attack, given the imponderables. The towering uncertainties include firing missiles over the tricky North Pole in trajectories that have obviously never been tested; coordinating a massive simultaneous launching of ICBMs, which has equally obviously never been tested; and banking on the US to accept millions of deaths and still decide rationally that retaliation is unprofitable.
Further, the classical arms controllers pointed out that only a quarter of US warheads are on vulnerable immobile ICBMs. Even if the Soviet Union did knock out 90 percent of these, there would still be more than 6,000 US warheads left on invulnerable mobile submarines and bombers. The latter do not yet have the accuracy of the ICBM. But they would flatten Moscow, Murmansk, and Vladivostok, along with all but the most hardened silos and command posts. And the Kremlin displays no more eagerness than does the White House to gamble away its entire civilization for the sake of political one-upmanship. Therefore deterrence holds.
Nitze's point was and is taken very seriously by arms controllers, however, in its more general application. That is, the adversary's theoretical first-strike capability against one's own most accurate, flexible, and reliable missiles makes everyone much more jittery and unpredictable.
The Russians wield this theoretical threat over US ICBMs in the early 1980s. The US will wield this theoretical threat over Soviet ICBMs in the late 1980s, when its MX and Trident II submarines will have attained accuracies close to the present-day precision of the ICBM.
Theoretically, again, this should prevent a worse threat to the Soviet Union in the late 1980s than it does to the US today, since the Soviet ratio of vulnerable land-based missiles and invulnerable bombers and submarine missiles is just the reverse of America's. Three-quarters of Soviet strategic warheads are on land and will be subject to first-strike destruction. In addition, Soviet bombers do not have the range of US bombers and - most important - Soviet submarines are noisier and much more easily tracked and destroyed than their American counterparts.
To counterbalance this theoretical weakness, however, the Soviet Union will always have Western Europe as a very effective hostage against any US first strike (as it did during the most dramatic period of American missile superiority in the 1960s). While Soviet retaliation against any American first strike could conceivably be limited to ''only'' several tens of millions, Soviet retaliation against Western Europe could not be minimized, even if all Soviet ICBMs were demolished.
It thus seems that the mutual balance of terror still holds, asymmetrical in detail but symmetrical in horror. It also seems, however, that both superpowers have new fears that the balance may not be holding. And this can lead to rash judgments, hair-trigger alerts, and missile ''launch on warning'' rules that could leave decisions about war and peace up to computers - and computer error.
This phenomenon is worrisome because of the increased possibility of war by accident. It is even more worrisome because of the increased possibility of war by miscalculation. In an interaction of technology and politics it produces what has been called ''crisis instability.''
That is, in a tense confrontation (like the 1962 Cuban missile crisis, the 1973 Mideast war, or potentially the 1982 Mideast wars) there might be an impassioned escalation of encroachments and commitments to allies and tests of credibility. Clashes of wills and interests could reach a point at which one or both sides might fear that the other was about to launch a nuclear attack. (In the abstract this sounds outlandish, of course, but then so did a British-Argentine war a few months ago.)
If this suspicion took firm hold, the ''use 'em or lose 'em'' pressures for a preemptive strike could build up. If a nuclear war is inevitable, then the whole momentum of bureaucratic decisionmaking (and of military-party complexes even more than military-industrial complexes) could well be to fire first and ensure that the ''unavoidable'' holocaust burns the other side more than oneself.
This is a grim prospect. And the over-the-horizon technology of the 1990s promises no relief. Antisatellite weapons are on the way which could threaten the reliability of second more than first strikes. The elements of nuclear-age ''C3I'' - command, control, communications, and intelligence - are all so fragile that they compound anxiety in any period of tension and would surely deepen the ''fog of war'' in any outbreak of nuclear hostilities.
Oceans will probably become ''transparent'' through real-time computer processing of wave, sonar, and heat analysis. This would make SLBMs almost as vulnerable to a first strike tomorrow as the obsolescent ICBMs are today. That would remove the cushion of SLBM assured second strike that provides what guarantees of deterrence we have today. And that in turn would enormously increase the pressures for preemption in any crisis.
Nor do experts hold out much hope that technology will soon - if ever - favor strategic defense over offense and restore mankind to a less uncomfortable balance based once again on defending populations rather than holding them hostage.
Interviews suggest that most Western specialists do not think that lasers or the Soviet research pet of particle beam weapons can ever produce a comprehensive defense against missiles.
If they did, in fact, they might turn out to be ''destabilizing'' in the short run before they ever got to a ''stabilizing'' long run. If one side got these weapons appreciably before the other, this could well create a new perceived window of vulnerability/opportunity during the transition period when one side's population became invulnerable while the other's remained vulnerable.
Less speculatively, technological progress can soon give us mobile land-based intercontinental missiles and other, highly ambiguous, new weapons which will make distinctions and verification difficult in the extreme.
These include weapons that blur the ''firebreaks'' between conventional and nuclear weapons (like small 500-pound fuel-air bombs that can produce a nuclear-matching blast of 900,000 per square inch pressure). They include weapons that blur the firebreaks between strategic and nonstrategic nuclear delivery systems (such as the Soviet Backfire bomber and America's forthcoming Pershing II). They include carriers that can be loaded interchangeably with nuclear or conventional warheads (such as America's forthcoming, relatively inexpensive, cruise missile).
Unless there are some predeployment arms control argeements, then - on limits and built-in features to make these weapons countable - each side is going to be left in great uncertainty about the other side's military capability. General staffs will make the worst-case assumptions about opponents out of prudence. A fresh arms race will take off without even the modest restraints of SALT I and the unratified but still observed SALT II.
In addition, what might be called technology seep will inexorably spread the bomb to other owners outside the two superpowers and four (or possible six) medium powers that now possess it. Newcomers might possible turn out to have as many inhibitions about actually using nuclear weapons as old-timers have diplsayed in the past 37 years. But the fact that those nations that most want the bomb tend to be engaged in regional squabbles (or in putting down domestic unrest) bodes ill for proliferation. So does the risk that some ambiguous third-nation firing of a nuclear explosive could be mistaken for superpower use and trigger retaliation reflexes.
The questions that have been hovering in the shadows ever since the beginning of the nuclear age are therefore emerging again to haunt mankind with renewed urgency:
What if the increasing pressures for breakdown of nuclear deterrence grow too great?
What if people forget with time and the habit of peace (in the northern parts of the earth, at least) just how terrible a nuclear holocaust would be?
What if deterrence fails?
Next: Nuclear weapons and the individual Suggested further reading
''US Defense Policy in the 1980s.'' Daedalus. Vol. 109, No. 4, fall 1980 and Vol. 110, No. 1, winter 1981.
Nuclear War, What's in It for You?, by Ground Zero, New York: Pocket Books, 1982.
Strategic Survey and The Military Balance, International Institute for Strategic Studies, London, current and back issues. Also occasional papers.
International Security, Harvard University Center for Science and International Affairs, Cambridge, Mass., current and back issues.
Cold Dawn: The Story of SALT, by John Newhouse. New York: Holt, Rinehart & Winston. 1973.
''Deterring Nuclear War,'' by Elizabeth Pond. The Christian Science Monitor, Aug. 26-29, 1980.
''Restructuring the SALT Dialogue,'' by Jane M. O. Sharp, International Security. Vol. 6, No. 3, winter 1981-82.
World Armaments and Disarmament Yearbook. Stockholm International Peace Research Institute, current and back issues.
Endgame: The Inside Story of SALT II, by T. Strobe Talbot. New York: Harper-Row, 1979.
Annual Reports of Secretary of Defense, Undersecretary of Defense for Research and Engineering, and Chairman of the Joint Chiefs of Staff. United States Defense Department, current and past issues.
The SALT Experience, by Thomas W. Wolfe. Cambridge: Ballinger. 1979.