There are now two separate debates taking place, which need to be clearly and carefully distinguished. The first is whether to replace the threat of retaliation as the basis of the US nuclear deterrence strategy with a new strategy based on defense of the United States.
The second is whether and to what extent the US should deploy active defenses with more limited objectives and capabilities.
Since the advent of nuclear weapons, presidents and other American political leaders have sought an alternative to the strategy of deterrence based on the threat of massive retaliation and to a situation where the American people are vulnerable to castastrophic destruction. President Reagan is no exception. Each president, beginning with Eisenhower, has considered the possibility of substituting defense and rejected it.
I sympathize with this desire to find an alternative. But political and military leaders also have a responsibility not to deceive themselves or to mislead, however unintentionally, the American people. No reasonable prospect for defending populations
If technology promised the certainty of a successful defense of the American population against strategic nuclear weapons, or even a high probability of overall success, it is likely that policymakers and defense analysts would agree that the US should seek that objective. But technology does not offer even a reasonable prospect of a successful population defense.
A strategy of deterrence based on defense would require that the defensive systems work to near perfection, and that we have a very high level of confidence that they will do so. That is, all but a tiny percentage of the many thousands of strategic nuclear weapons would have to be destroyed before impact. The defense would have to work not only against attack by ballistic missiles but also against attacks by air-breathing systems (bombers and cruise missiles). The various defense systems would all have to work the first time they were tried in combat. And the attacker's countermeasures would have to be negated.
Here there is a vulnerability . . . countermeasures aimed at direct attack of the defensive system would have to punch only a modest hole in the defensive systems for the ballistic missiles to be able to penetrate that particular layer of defense. Moreover, because satellites are in general more fragile than ballistic missiles, a space-based system that had any chance of attacking ballistic missiles would have a much better capability to attack another system like itself.
Besides, it is easy for the offense to multiply the number of objects and confuse the defense by including decoys, balloons, and so on. Discriminating the warheads from the decoys is a massive sensing and information-handling job, conceivable in principle but not nearly yet devised.
More is now known about discriminating warheads from decoys. But the options available to the attack have also increased, e.g., maneuvering vehicles and more sophisticated decoys.
Battle management of all the data and of the sequence of defense actions required in such a layered system will also pose an enormous challenge to information processing capability.
Activation of the system by mistake (so that it fires for example at a few test missiles launched by the other side) would not necessarily provoke a nuclear war. But it certainly could create a serious crisis.
SDI proponents claim that it is at least possible that SDI will cause a shift in the offense/defense cost-exchange ratio (the ratio between the amount that the offense has to spend to offset the effect of the defense, and the cost of that defense). But none of the proposed new layers of the SDI have been elaborated to the point where any significant calculations can be made to support such a claim. So far, the evidence is all to the contrary. The cost- exchange ratio for active defense of urban areas, even with only, say 50 percent of structures surviving in major urban centers, still seems to favor the offense by 5 or 10 to 1.
As noted, a shift in strategy to one of defense of population would also require a system of active defenses against air-breathing weapons -- bombers and cruise missiles.
Successful defense of urban-industrial targets against a nuclear attack by bombers and cruise missiles is also not feasible. The reasons are the same as or analogous to those for the outcome of defense/offense competition in ballistic missiles.
In strategic terms, these are the enormous destructive power of a single nuclear weapon, and the difficulty and cost of intercepting nearly 100 percent of the weapons aimed at the many targets to be defended.
In tactical terms, the reasons are the offense's ability to saturate defenses; to use attack corridors, electronic and other countermeasures; to employ low observable designs, decoys, and varied flight profiles; and to attack directly the vulnerable elements of the system. Could SDI enhance deterrence?
In sum, the combinations of limitations -- scientific, technological, systems engineering, cost -- and especially the potential countermeasures make the prospect of a perfect or near perfect defense negligibly low.
Might defensive systems serve two other more limited but important objectives: to enhance deterrence by ensuring the US strategic retaliatory capability; and/or to reduce the damage to the US if deterrence should fail?
The task of defending nuclear forces is much easier than defending the American population. The targets to be defended are generally fixed points, or limited areas. The degree of success required is far less than perfection, and the defense can tolerate significant levels of uncertainty; e.g., if 25 percent of the ICBM force were saved by active defense, the defensive system would be a success.
Offense countermeasures will be possible in this case also, but at least for certain kinds of offensive forces and basing schemes a defense system can have a favorable cost-exchange ratio.
An ABM defense with this more limited purpose of defending land-based missiles could be accomplished through a terminal defense and is feasible with known technology. Space-based weapons would probably not be cost effective as part of such an ABM system, though space-based sensors could be. Importance of modernizing US strategic forces
If a threat to the overall US retaliatory capability were perceived to impend, then the US should pursue active defense. But such a threat to the US retaliatory capability does not exist today and is not likely to arise during this century if the US pursues a strategic modernization program, generally along the lines of current plans, that includes all legs of the triad. So there is no need now to pursue deployment of defensive systems to ensure US retaliatory capability.
As a second limited objective, the US could pursue defense as a hedge against the possibility that deterrence might fail. If the failure resulted in the use of thousands of weapons, an imperfect defense could not significantly reduce the damage. Moreover, countermeasures to the deployment of such a defense could be designed by an attacker (including force increases) to restore fully the same level of damage planned in the absence of defenses.
If a small ballistic-missile attack were sophisticated and concentrated on a single target area, then a limited defensive system could probably be saturated and penetrated. Thus, to hedge, for all cases of small attacks, against the possibility that deterrence may fail, the US would need to deploy a defense system approaching an all-out ABM deployment.
To pursue the objective of hedging against the failure of deterrence would introduce major risks and costs.
First, in terms of the arms competition any US ABM defense would prompt Soviet countermeasures. If the USSR were to respond with her own defensive systems, the US would have to take her own countermeasures in terms of offensive forces. As a result, the arms competition -- offensive and defensive -- would be accelerated, and neither side would have much interest in arms-control limits on offensive systems.
With or without constraints on the buildup of offensive forces, defensive systems would create uncertainties as to the effectiveness of the retaliatory capabilities on each side. This could lead to pressures to preempt with nuclear weapons in a crisis. While it is most unlikely that either side could confidently believe that it had a first-strike damage-limiting capability, any such possibility could prove enormously destabilizing. And new kinds of pressures could arise: to preempt with antisatellite w eapons against the space-based defensive system, conceivably as it was being deployed, or more likely in a crisis; to preempt with nuclear weapons against the offensive forces of the other side to gain greater confidence in the performance of one's own defensive system.
The financial costs are an additional problem. The costs of ABM, air defense and civil defense to replace the deterrent strategy could approach $1 trillion.
And this problem of cost might not be symmetrical between the US and the USSR: the USSR can less afford the resource costs, but Soviet leaders are more capable of exercising the political discipline necessary to exact such sacrifices.
And if the superpowers' nuclear forces were neutralized, the allies could become concerned that they were thereby more vulnerable to conventional attack.
This consideration for the issues connected with defensive systems illustrates the importance of arms control to the US in the future.
The US should be able to avoid having to deploy defenses to ensure her retaliatory capability -- including by rebasing ICBM as necessary -- providing that she can achieve appropriate restraints on Soviet offensive forces. Such constraints, especially if they involved reductions in missile warheads and limits on new types of missiles, would also promote stability and thereby reduce the chances that deterrence would fail. Arms control, in other words, could provide the means to achieve the same objectives
sought by proponents of SDI.
As to negotiating leverage, it is not clear what SDI might provide.
The argument that actual deployment of US defenses -- at a level short of perfection -- would provide leverage to gain significant reductions in Soviet missile forces seems to fail on logic as well as on common sense. US deployments of multiple independently targetable reentry vehicles (MIRVed warheads) may have produced the ABM Treaty by convincing the Soviet Union of the futility of ABM, but it is difficult to see how American ABM will produce constraints on Soviet MIRV. The opposite effect -- an incr ease in Soviet MIRV -- seems far more likely.
The strength of the wish for a change in US strategy is understandable. But neither political leaders nor those who think about policy matters have the luxury of acting or speaking as if we were in an ideal world of perfect defenses; of an arms control regime involving reductions so deep that defense could approach perfection despite any difficulties in verification of offensive systems; or of a risk- and cost-free transition to a defense-dominated era.
Avoiding nuclear war is the vital issue of our times -- too vital to be clouded by wishful thinking.