PAUL K. DAVIS,
SANTA MONICA, CA,
October 21, 1991.

Hon. Joseph Biden
Chairman, Senate Foreign Relations Committee,
Washington, DC.

DEAR SENATOR BIDEN: I was asked by your staff to prepare background material on the CFE treaty from the perspective of someone who, over the years, has directed analytic studies about the military balance in Europe and the potential benefits and dangers of one or another arms control agreement. Since the Warsaw Pact has disintegrated and the Soviet Union has itself become a thing of the past, my comments were to be primarily for the historical record. I was asked to draw on my work to provide for that record some examples of how analysis contributed and to comment on some of the strategic principles and concerns that remain valid. The attachment to this letter is my response, but let me observe here that analysis contributed to the CFE process in many different ways, helping to: establish objectives; conceive and analyze broad arms-control measures; define units of measure and verification constraints; and work out fine points, some of which were quite important. Most of the analysis was performed within the national governments or in NATO. My own examples, however, draw on work done over a several-year period by RAND's federally funded research and development centers (FFRDCS), particularly the National Defense Research Institute, which is sponsored by the Office of the Secretary of Defense and the Joint Staff.

Sincerely,

PAUL K. DAVIS,
Corporate Research Manager,
Defense Planning and Analysis, RAND.

THE CONTEXT FOR CFE

To understand the context of CFE and the role of analysis, it is useful to start with a brief discussion of the classic Central Region military balance. Western analysts focused on that balance for decades, with innumerable studies on the subject, the vast majority of which were high pessimistic for a wide variety of reasons -- some specious and some valid. In the 1950s, the assessment was so pessimistic that NATO thought in terms of defense at the Rhine. However, as the result of the German army building up its strength, a reassessment of the balance that used Armored Division Equivalents (ADEs) to reflect differences in size and capability between Pact and NATO divisions, and an assessment of what Soviet forces would actually be available in the Central Region, it was concluded by defense analysts in the Kennedy administration that conventional defense should be feasible with reasonable defense budgets. Although our European allies were never enthusiastic about developing a robust conventional defense, believing that deterrence depended fundamentally on the threat of nuclear escalation, NATO nonetheless developed a significant conventional defense in the 1960s. NATO later adopted a strategy calling for a forward defense on or close to the borders of Germany with Poland and Czechoslovakia.

Competitive modernization and gradual changes in deployment capabilities and operational strategy continued throughout the 1970s and most of the 1980s. Interestingly, there was no consensus even on whether the trends over this time were favorable or unfavorable, because different studies emphasized different aspects of the balance and made different assumptions. Nonetheless, as of the mid-1980s, most assessments credited the Pact with better than a 2:1 advantage in ADEs and there was widespread pessimism about NATO's ability to conduct a conventional defense.¹ This was the context in which the CFE negotiations began.

It should be noted that some analyses were substantially more upbeat. In particular, books and articles by John Mearsheimer, Barry Posen, William Kaufmann, and Joshua Epstein all argued that the conventional balance was nowhere near as adverse as commonly depicted, and that NATO should be capable of defending itself conventionally.² These studies were based on methods that had been used throughout the 1970s.³ They made different assumptions, however, and also used different data.

My own views on the balance and the potential value of conventional arms control solidified as the result of a major classified study I conducted with colleague Robert Howe in 1986-1987. This drew on analytic war gaming methods developed at RAND under the sponsorship of OSD's Director or Net Assessment, Andrew Marshall, and reflected my long-standing concern about the treatment in analysis of uncertainty and strategy. What follows are some extracts from a 1988 paper in response to a Congressional request, a paper based on the earlier study.⁴ The material is relevant, because it describes a very different way of looking both at the balance and at the potential impact of arms control.

A basic question in any discussion of the military balance is which balance one is addressing: the balance of "inputs" such as defense expenditures and manpower under arms, the balance of combat equipment such as tanks, the balance or force readiness and mobilization potential, or * * * the balance as measured by likely war outcomes if deterrence failed. This paper is concerned primarily with the last of these, and with the challenge of addressing that warfighting balance in the face of massive uncertainty rendering it meaningless to talk about allegedly "best-estimate" scenarios.

The beginning of wisdom about this balance is recognizing that war outcomes are sensitive to scores of factors, rather than the handful regularly discussed. Assessment should consider a vast range of plausible scenarios, where scenario is construed broadly to mean a set of assumptions about, for example, political-military context, warning times, mobilization times, alliances, operational strategies, force effectiveness, sheer quality of leaders and their troops, and even the "laws" of combat that determine rates of advance and attrition. Moreover, analysis should be based on a gaming approach, at least in structure, because the confrontation of opposing strategies and tactics is fundamental to warfare, and wars seldom look like those in standard planning scenarios.

The results of such multi-scenario analytic wargaming defy reductionist analysis: simulated war outcomes often change drastically with what might naively be considered to be small changes of assumption, and even the relative value of alternative improvement measures varies substantially from scenario to scenario. Measures or capabilities critical in some circumstances are almost irrelevant in others.

These wild fluctuations are not analytic artifacts, but rather a manifestation of something that professional military officers and historians have known since time immemorial, that war is an incredibly complex phenomenon characterized by uncertainty-except, for example, in instances where one side has overwhelming force (a situation that does not obtain in Europe). Moreover, tactics, strategy, and other human factors matter greatly. Caution should be exercised in using common analytic methodologies that obscure these basic aspects of warfare in the Search for well-behaved and simply explained results. What can be said from initial experience with multi-scenario analytic war gaming applied to Europe's Central Region? Some personal conclusions so far are as follows:

• There exist both good cases and bad-i.e., cases in which NATO could today hold its own and defeat a Pact invasion, and cases in which the Pact would sweep to a quick and decisive victory.
• Claims that NATO is hopelessly outnumbered, or that conventional defense would surely fail, are wrong.
• Claims that "NATO could probably thwart a Pact invasion today" are potentially quite misleading; disaster-ridden scenarios are "probably" about as likely as, or more likely than, the good scenarios. Moreover, the balance is demonstrably quite fragile. Many "good cases" turn into "bad cases" with only modest changes of assumption.
• Those most optimistic about the balance usually assume away the operational-level and political-military problems about which many strategists, including myself, are most concerned. Indeed, the optimists often confuse their readers about the difference between what "could be" or "should be" and what "is."
• NATO should be particularly concerned about scenarios in which the Soviets raise general readiness for war over a long period of time and then, perhaps after a period of tension but no acute crisis go to war with only a short mobilization and enough ambiguity of political signals to disrupt the coherence and decisiveness or NATO's reaction to initial warning. In retrospect, that would constitute the strategic surprise emphasized in Soviet military thinking.
• The military challenge for NATO is to make such Pact-favorable scenarios less plausible-by eliminating potential Achilles' heels (which include a probable lack or flexibility and decisiveness in top-level command-control because of the political constraints placed on NATO's military commanders and planners during peacetime), by increasing the robustness and sustainability of its defenses, by creating additional challenges to preferred Soviet operations, by planning feasible responses to strategic warning, and by continuing to develop innovative weapon systems and operational concepts. Affordable opportunities for improvement abound.
• Another challenge for NATO is new: tile era of potential large-scale conventional arms control. Instead of attempting to "limit damage" or avoid it, the United States and NATO should approach arms control ambitiously-but with hard-headed and operationally rich military analysis to guide negotiations. Arms control focused on readiness and other operational issues has the potential virtually to eliminate the threat of a short-mobilization attack (including variants in which the Soviets conduct extensive premobilization training). Such scenarios should probably be the most worrisome currently for NATO. Going further, force reductions could be stabilizing if sufficiently large and strongly asymmetric, but dangerous otherwise. The long-discussed move toward "defensive defenses" by both sides is also more worthy of study than it has been in the past. * * *

The last bulleted item is noteworthy, because, at the time the research it describes was accomplished (1985-1987), there was widespread cynicism about conventional arms control. This was a legacy of the MBFR era, an era of endless negotiations that were not seriously intended to succeed (and which succeeded in not succeeding). As the result of analysis, however, it was clear that there were major opportunities for improving the warfighting balance through arms control, if merely the Soviets were interested. One contribution of this study, then, was to help increase interest in arms control.

In this same period, RAND urged policymakers to approach the CFE negotiations with clear objectives focusing on improving military stability, as judged from analytic war gaming and related methods. Increasing military stability was defined as improving prospects for the defender being able to successfully defend his territory, even in "bad case" scenarios. A widely briefed study led by James Thomson concluded that with this objective for NATO (as contrasted with objectives, say, for confidence building or reductions per se), any reduction formula should be highly asymmetric and that NATO should insist on equal ceilings for final forces.⁵ That study was unwelcome to many proponents of arms control, who feared that an overly tough NATO position was unnecessary and would doom prospects. For a variety of reasons, however, NATO's Ministers decided to propose and insist upon an equal ceilings approach. It was generally agreed that the analysis had played a significant role in this decision.

OPERATIONAL ARMS CONTROL

Work proceeded apace by the governments on negotiating "structural" reductions, by which was usually meant the destruction of equipment. The principal issues were the extent of the reductions, the units of account, definitions, and verification. While these negotiations continued, some observed that there was also an important complementary role for what is sometimes called "operational arms control," by which was meant arms control affecting the deployments, readiness, and practices of forces, rather than their structures.⁶ There had been a long history of related discussions under the rubric of confidence-building measures and, later, confidence and security building measures (CSBMs), but these had taken place in separate negotiations often dominated not by military analysis but by political reasoning. What follows is an excerpt from a study in 1988.

In the past, operational arms control has been largely associated with confidence building measures with limited ambitions. This study argues, however, that operational arms control has the potential to substantially improve NATO's military security. This potential has gone largely unappreciated in the past, because its effects are not seen in standard NATO-conservative scenarios, but are rather most evident in scenarios developed to represent the possible viewpoint or a Pact military commander charged with developing an invasion plan. Operational arms control could greatly increase the risks seen by such a commander.

While operational arms control has considerable potential by itself, at least in the abstract, this study concludes that it should no longer be treated as a subject apart, but rather should be integrated with structural measures. One concern here is that the Soviet Union might gain most of its principal arms-control objectives, such as an improved East-West atmosphere leading to increased Western investments underwriting perestroika, with operational arms control agreements that would not even address the quantitative imbalance that is at the heart of NATO's security problem. Another problem is that certain operational arms control measures that seem intuitively virtuous to their proponents could in fact cause considerable mischief, and could even reduce NATO's military security. It is with this backdrop that the present study provides a top-down discussion of what NATO's objectives for operational arms control might be, what strategies it might follow in seeking those objectives, and what principles might guide Its assessment and tuning of particular proposals, The next paragraphs summarize conclusions of that discussion.

Objectives
The objectives for operational arms control should be the same as those for conventional arms control more generally, which in turn should be a subset of high-level coalitional objectives:

• To deter aggression and coercion
• If war begins, to defeat the aggressor and restore the prewar boundaries while minimizing the devastation of war and avoiding use of nuclear weapons; or, failing that, to re-establish deterrence and convince the aggressor to withdraw
• To increase crisis stability
• To increase arms-race and political stability, and to reduce accordingly the peacetime cost of armaments
• To avoid misunderstandings and miscalculations that might lead to unintended crises or conflicts
• To maintain alliance unity

As discussed in the text, in examining arms control initiatives NATO should place more relative emphasis on improving its military security, and somewhat less relative emphasis on the less critical though worthy objectives of avoiding accidental wars and building confidence. The preferred approach can then be captured by a single paramount objective.* * * :

• To improve the actual and perceived prospects of the defender should war occur.

This formulation is consistent with the growing emphasis on stability, but makes explicit the connotation intended.

To achieve the above objective, arms control should focus on the alliance's actual military problems. The study draws on previous military analysis that characterized the warfighting balance in terms that identify possible failure modes for NATO's conventional defense. Some of the failure modes can be addressed only by unilateral improvement measures (e.g., procuring adequate munitions), but many can be addressed by a combination of arms control and defense programs. The possible failure mode in which NATO succeeds in enforcing its operational strategy of forward defense, but loses a war of attrition by virtue of the Pact's quantitative advantages or what proves to be Pact superiority of doctrine or strategy, must be addressed primarily by defense programs increasing forces and their technical effectiveness, and by structural arms control in the form of highly asymmetric reductions involving destruction rather than the mere withdrawal of Soviet and other Pact forces. There is little role for operational arms control in this domain.

By contrast, operational arms control could significantly improve the defender's prospects in other ways summarized by Fig. S.1, which constitutes a broad strategy motivated by observations from analytic war gaming. Notable elements of the Strategy are as follows:

• Avoiding strategic and operational surprise. Some constraints would not only make surprise less likely by assuring warning signs (which would probably exist anyway), but more importantly would decrease the ambiguity of such signs and thereby greatly increase the likelihood of an early cohesive response.
• Raising risks for the attacker. Some of the constraints would increase the objective vulnerability of a would-be attacker to preemption (e.g., tactical-air strikes) during his final preparations, and would thereby tend to deter destabilizing preparations in crisis (even if the defender, like NATO, had it highly defensive doctrine and mindset, and no capability to initiate a serious offensive).
• Improving the defender's tactical odds. Some of the constraints would increase the defender's tactical odds on D-Day by increasing the complexity and reducing the likely efficiency of the attacker's immediate pre-attack operations. This could be critical for NATO by reducing the likelihood of immediate penetrations of the zone of prepared defenses on which it so strongly depends.

Fig. S.1 indicates a broad strategy for using operational arms control, but to develop concrete proposals requires taking into account numerous considerations and asymmetries. Of these, the most important appear to be the following:

• The Pact has an excess of ground forces overall and an excess of forward-deployed forces in particular. This is the most fundamental threat to NATO's security.* * *
• Because of its quantitative disadvantages and shortfalls, NATO depends more heavily than the Pact on high operational readiness. This stems also from the strategy of forward defense and terrain factors.* * *
• Similarly, many formulas for geographical disengagement would on balance hurt NATO, because NATO's defense depends upon having prepared defenses in depth everywhere along the border, total familiarity of defending forces with the local terrain, and holding forward. In a mobilization race re-introducing forces to the forward regions, the attacker would probably be asymmetrically benefited.* * *

• Operational arms control should henceforth be integrated with structural arts control, because the net effects of each will depend on the other.
• The sides should strive for an outcome in which both sides have (or at least have the right to have) enough high-readiness in-place or early-arriving reserves to provide a basic level of defense (coverage of the front, plus reserves).* * * The principle * * * is symmetric: having high-readiness forces adequate for defense is "good," despite the intuition of many people that high readiness is "bad." * * *
• The excess-force concept implies that readiness limitations should be conceived and formulated in terms of ceilings on the number of forces in each of several readiness categories. These ceilings should permit initial defense.
• Although hypothesized changes in force composition and doctrine might make it possible in the future to improve stability by preferentially reducing "offensive" systems and imposing disengagement zones, the most straightforward and compelling way to improve the defender's prospects (and stability) Is by eliminating or prohibiting excess forces generally, and high-readiness excess forces as a first priority. No changes in composition are required.

Some of the ideas expressed in this 1988 study were reflected in discussions about verification, zone limits, and follow-on negotiations. Some have proven useful in other arms-control settings (e.g., Korea), and in thinking about how to deal with the emerging European situation in which the former Soviet Union (and Russia in particular) retains a great deal of military equipment, but with much of it no longer associated with serious military units.

Fig. S-1 -- Potential role of operational arms control in improving defender prospects

THE OPERATIONAL MINIMUM AND DEEP CUTS

Another analytic contribution to the CFE debate involved the issue of the so-called "operational minimum." For many years, it was conventional wisdom that successful defense "required" a minimum of defensive forces on the order of 40 divisions. When discussions began about possible deep cuts (in a regime of nominal parity), that conventional wisdom generated the view that NATO could not cut its forces more than perhaps 5 percent or so, even if the Pact accepted the principle of parity. If this had been valid, it would have greatly limited the degree to which the sides could hope to achieve a "peace dividend."

In analysis conducted early in 1989, my colleagues and I explained the discrepant views and defined the circumstances under which deep cuts would and would not be risky from NATO's perspective. Although the work involved mathematical models, war games, and other paraphernalia of the analytic business, the resulting principles were simple. The following material is extracted from one of several papers deriving from that study, the results of which were initially surprising and controversial, but which were subsequently confirmed by government war games and other studies.⁷

The CFE-mediated process of moving toward parity will greatly improve force-posture stability in the Central Region. However, many wars have been fought at parity, and the defender has often lost. Thus, parity is no panacea. To make things worse, there is a major difference between nominal parity as achieved in negotiations for large aggregations such as the ATTU, and dynamic parity (force levels vs. time) in the Central Region in time of war.

At low force levels such as 50 percent of NATO's current levels (counting all forces NATO would expect to be able to employ in the first month of a Central Region war), force-posture stability might be high or low, depending on details of force structures, deployments, readiness, and so on, as well as on details of strategy. There is no theoretical reason to believe that a low-force-level regime should be stable or unstable, although the importance of operational-level defensive maneuver would become increasingly important at force levels below what has been called the operational minimum. That minimum should be regarded as a fuzzy threshold rather than a sharp one, but to the extent that it is meaningful analytically, its value is probably between 22 and 27 equivalent divisions or 27-34 average NATO divisions (roughly 6,600-8,100 main battle tanks), which is significantly lower than the figure often quoted and which would suggest that CFE II could call for reductions to roughly 30 percent below the levels that would obtain under NATO's CFE I proposal. However, a low-force-level regime could be highly unstable if NATO did not make fundamental changes in force posture, command-control, and operational strategy. A scaled-down version of today's layer-cake posture and related operational concepts would be highly inappropriate. Thus, force-posture stability could worsen as reductions proceed much below NATO's current levels. On the other hand, at sufficiently low levels-below what might be called the offensive minimum, an attacker would probably be unable to win a permanent strategic victory, even if he won the initial battles. This would be a strong deterrent.

Returning to the larger issue of what constitutes stability, it is worth observing that deep cuts accompanied by appropriate unilateral restructuring and a host of CFE stabilizing measures making operational surprise difficult if not impossible would probably improve international relations, crisis stability, and arms-race stability-thereby lowering the likelihood of crisis or attempted coercion. Even if, despite the restructuring and stabilizing measures, the effect of deep cuts on force-posture (military) stability were ambiguous, because war outcomes at low force levels depend more on operational competence in maneuver and command-control, and because of the problem of "out-of-area forces," the net effect on security-related stability might be unambiguously favorably-quite aside from the economic benefits that many are hoping to enjoy. And, when to nations this is added the trend for increasing independence of Eastern European nations that augurs poorly for the military cohesiveness of the WTO in time of crisis or conflict, the prospects for deep cuts are considerably more favorable than seemed possible even a year or two ago.

Since deep cuts are unlikely in the immediate future (if for no other reason that the Soviet Union's likely inability to absorb quickly more cuts than those necessary under CFE I), there is time for NATO to make necessary adjustments in its force posture, and for the sides to negotiate ceilings and other measures that would enhance the stability of the low-force-level regime. These measures (often called operational arms control) might include the substantial or complete withdrawal of Soviet forces and attack infrastructure from Eastern Europe (e.g., excess artillery ammunition, above and beyond that appropriate for defense, constraints on the number of high-readiness reserve units, ceilings and constraints on out-of-area ground forces, and constraints on certain types of large-scale exercises and movements). In addition, the sides may evolve toward force postures with a smaller ratio of attack-suitable to defense-suitable weapons and units, while retaining high mobility and flexibility. This might be encouraged by negotiating numerical limits on mobile air defenses and avoiding severe constraints on transport helicopters, attack helicopters, and thinly armored personnel carriers, since these systems improve the defender's mobility but are extremely vulnerable in the opponent's rear areas. At a unilateral level, NATO's nations should all procure the equipment needed to assure the ability quickly to create tactical obstacles in depth-not just at the IGB, but at whatever point the defending force finds it useful to employ obstacles tactically.

Finally, it should be noted that the developments in Eastern Europe may be making the current CFE process obsolete. Those contemplating approaches for whatever "CFE II" turns out to be should avoid concepts that would have the effect of impeding the natural normalization processes. In particular, concepts based on an image in which the intra-German border (IGB) is the natural middle and the purpose of measures is to assure force-posture stability with respect to conflicts at the IGB between monolithic WTO and NATO alliances are severely flawed from a strategic perspective. Stability in the broader sense of this paper would hardly be enhanced by agreements that encouraged the Soviet Union to leave large forces in Eastern Europe.

These final arguments became even clearer once the Soviets agreed to leave Eastern Europe. Strategically, this creates a buffer zone protecting Western Europe from anything remotely like operational surprise-the same kind of buffer zone that the Soviets had sought and maintained because of fear of being attacked again by the Germans or NATO. Consider: even if the Soviets (or Russians) have many divisions distributed throughout their interior, more divisions than NATO could readily counter in a "fair fight," the Soviets (or Russians) would now have to prepare those divisions for war and then deploy them over long and vulnerable LOCs through potentially (almost certainly) hostile areas of ex allies. And, unless NATO had been stupid during this period, the aggressor forces could be attacked by air forces. As we saw in the war with Iraq, modern air forces can be devastating.

LONG-TERM EAST-WEST STABILITY

What follows is the summary of a 1990 report on long-term East-West security in Central Europe.8 The report was apparently useful for a variety of reasons that included its observation that future arms control might depend less on formal agreements than on orchestrated unilateral decisions, and its suggestions regarding the continuing value of operational arms control measures. It (or the briefings that preceded it by some months) was probably the earliest place that the buffer concept was discussed in some detail, along with its implications for military stability. Although much has happened in the last year, many aspects of this material remains valid (as of October 1991), so long as one thinks in terms of the former Soviet Union rather than the Soviet Union.

The long-term security of Europe depends on continued progress on two fronts, East-West security issues and "other" security issues, the latter including economic developments to reduce discrepancies among nations, the harmonious emergence of a united Germany, and the moderation of ethnic disputes within Eastern Europe. This report is concerned only with long-term East-West security and the related concept of "stability." Our approach, however, is strongly tuned to the existence of the other class of problems.

The recent changes in Europe have transformed the strategic landscape and altered what can be accomplished with respect to security. In this report we propose a framework or new NATO objectives and a strategy for accomplishing them. Our approach starts with the desire to achieve long-term stability, by which we mean a state characterized by robust security, predictability, the absence of crises and dangerous international tensions, a "reasonable" defense burden that is either constant or shrinking, and public satisfaction with the situation.

To achieve this objective we recommend thinking in terms of five subordinate objectives, as shown in Fig. S.I. First, NATO must continue to deter, but without provocation, a Soviet invasion of Western Europe-the traditional mission that now seems easy to accomplish. Second, NATO must recognize that the security of both the Soviet Union and Western Europe is being greatly enhanced by the emergence of a strategic buffer zone composed of Poland, Czechoslovakia, and Hungary. Thus, it is now important, even before Soviet forces leave Germany and Eastern Europe, to plan on long-term deterrence of Soviet reentry into Eastern Europe-again, without provocation or souring the trends toward greatly improved East-West relations and the integration of the Soviet Union into Europe. Fortunately, this is not a zero-sum game: to the contrary, it is strongly in Soviet interest also that the strategic-buffer should emerge and continues.

The third element of strategy is to maintain strategic equivalence in the East-West competition, maintaining a good balance of forces and maintaining a vigorous program of research, development, and doctrinal evolution. This strategic equivalence is essential if the military balance is to be de-emphasized in national politics and discredited as an instrument of foreign policy.

The fourth element is to deter rearmament, a challenge that NATO has never needed to concern itself with in the past, but which may be especially important in the years ahead. Nuclear forces have an especially important role here that has no historical precedent. And, finally, we have the objective of reducing the sources of conflict and tension.

As summarized in Fig. S.1, there are many measures that can be pursued for each of the subordinate objectives. These involve a mix of unilateral military improvements, arms control beyond CFE 1, and political and economic activities. Some of the measures are strongly interlinked and controversial. For example, we recommend reducing force levels to roughly 60 percent of CFE I levels, but this should be coupled with reductions and restrictions on forces outside the ATTU, which is likely to mean the need to consider some measure or naval arms control. We also recommend joint Western and Soviet security guarantees for Eastern Europe and a state of a defensively armed neutrality for the buffer states.

In contemplating the strategy as a whole, which is represented by the totality of Fig. S.1, it is useful to see it as having three aspects. There is a core strategy, an environment-shaping strategy, and a hedging strategy embedded in Fig. S.1. The core strategy is essentially to purse the highly favorable current trends that are restructuring Europe and the character of NATO. The environment shaping includes building the Soviet Union into Europe rather than continuing to treat it strictly as "threat," using arms control to reduce the feasibility of large-scale military campaigns of aggression, removing some of the sources of tension, and developing military capabilities with great flexibility. The hedging strategy includes continued R&D, rapidly deployed deployable U.S. air forces with advanced munitions, and realistic plans to permit mobilization if necessary. As the text makes evident, despite the very favorable trends, there is ample need and opportunity for environment shaping and hedging that go well beyond CFE I: the end of history is not yet quite upon us, even in Europe.

ASSESSMENT OF THE CFE TREATY

The CFE Treaty has now been signed, although not yet ratified. How does it look? There are really two issues: is the treaty in our interest and could the treaty have been better? My own view is that the CPE Treaty is essentially a pure win for the West (and still valuable), but it is also beneficial to the former Soviet Union: this was not a zero-sum game, when viewed appropriately. That the Soviet government recognized and acted upon this, and subsequently went even further in permitting the unification of Germany and agreeing to withdraw Soviet forces from Eastern Europe, is an extraordinary example of the power of rational thinking.

The biggest single problem with the treaty was NATO's failure to prevent the Soviets from moving equipment from the Central Region to the other side of the Urals, thereby evading requirements for destruction. In other periods, this would have been considered a monumental political blunder on the part of the NATO governments, whether or not the Soviets completed their movement of equipment before tile cut-off date. NATO policymakers had believed, until late in the game, that the Soviet Union would in fact be destroying that equipment rather than moving it outside the Atlantic to the Urals (ATTU) region.

Even now, to assure irreversibility of the favorable trends, it is highly desirable that NATO reach agreement with the former Soviet Union on constraints that will prevent whatever government emerges from maintaining and inappropriately exploiting that equipment. Despite the disintegration of the Soviet Union, Russia or the residual central government will retain a very large army quite capable of reentering the Balkans and Eastern Europe, and perhaps threatening Western Europe in the future, if the NATO nations reduce unilaterally as much as many observers expect. It therefore seems highly desirable to pursue, "operational arms control" constraints constraining the size, readiness, and positioning of the residual army of the former Soviet Union. Interestingly, Some western states of the former Soviet Union may very much favor such constraints, because their own independence will be at issue as well.

Beyond this, the biggest military challenge is for NATO to maintain its military competence, while developing a building block approach to planning that leaves it ready for a wide range of contingency operations (e.g., intervention to support Poland against a threatened invasion by the Russians), rather than an implausible and stereotyped defense at the German border. In the absence of a credible near-term threat, the tendency may be to unreasonably reduce NATO's force levels and avoid attending to the many fundamental issues of strategy and command and control doctrine that would determine NATO's effectiveness in time of crisis or war -- problems that cannot, because of organizational realities, be resolved quickly upon receipt of strategic warning. Recent developments involving a possible French-German force, and other indicators of interest in a continental approach to security, do not bode well for the continuing vitality and competence of NATO.

LESSONS LEARNED

Having taken this quick tour through selected examples of how analysis contributed to the CFE arms control process, it is now appropriate to ask whether there are some lessons learned, lessons that might assist us or other nations as they enter other arms control negotiations (e.g., in Korea, the Mideast, or South Asia). My recommended principles would be these:

• Derive proposals from political-military objectives and analysis, not political intuition alone. Pay particular attention to military stability, not merely one or another notion of what might build confidence. Good confidence building measures will usually increase military stability. Bad ones may worsen it.
• Establish and stick with equitable principles. Do not accept claims that the other side is irrational, immature, paranoid, or otherwise incapable of understanding objectively reasonable principles (although recognizing that its perspective on facts may well be different).
• In developing objectives and principles, avoid having arms control codify unnatural borders or other injustices.
• Insist on "adequate" verification, without permitting verification "requirements" to become the tail that wags the dog for reasons unrelated to security.
• When evaluating the value of potential arms control measures, conduct realistic operational military analysis that recognize the historical follies of defenders, the friction of war, and the potential value of measures that may have no value in "standard" scenarios.
• With this in mind, consider both structural and operational arms control. Recognize the value of operational arms control measures in making surprise attacks and other specialized strategies exceedingly difficult for the aggressor to accomplish. Aggressors will violate arms control measures when the time comes, but their doing so can be a powerful alarm greatly increasing the likelihood of timely response to warning.
• Insist that intelligence assessments use comparable criteria in assessing the two sides' capabilities, including readiness. Be cautious against exaggerating the threat, in part because doing so may obscure the potential value of the arms control measures being considered (e.g., if one unreasonably assumes a current capability for roll-out-of-bed-and-go-to-war surprise attacks, then measures to prohibit actions that would make surprise attacks easier seem unuseful).
• When conducting such analyses, consider a wide range of scenarios and assumptions, not just the current best-estimate case. Multi-scenario analysis, in all its dimensions, should be fundamental, not a nice-to-have feature of assessment, and measures with modest value in some cases would have substantial value in others.

Hon. Joseph Biden,
US. Senate, Washington, DC.

DEAR SENATOR BIDEN: An important and promising conventional arms treaty has emerged in the thaw of the Cold War. The Conventional Forces in Europe (CFE) Treaty, which sets ceilings on conventional weapons systems, was signed on November 16, 1990, and was submitted to the U.S. Senate for ratification approval on July 9. We urge you to take advantage of the historic changes that have been taking place in the Soviet Union and Eastern Europe by voting to consent to ratification of the CFE Treaty.

In addition to modest cuts in NATO armed forces, the CFE Treaty will prompt substantial reductions in the forces of the Soviet Union and Eastern Europe. 30,000 conventional weapons are slated to be destroyed under the treaty which has already been signed by all 22 NATO and former Warsaw Pact countries.

The CFE Treaty, made possible by the end of the Cold War and improved relations with the Soviets, helps put to rest any possibility of a massive surprise attack against Western Europe. In addition to its overall ceilings on major conventional firms, the Treaty establishes geographic sub-ceilings designed to prevent the U.S.S.R. from deploying the bulk of its forces in its westernmost territory. By scaling back the number of conventional forces available for large scale offensive action, the CFE Treaty stabilizes a military and geographical region that has been a fulcrum point of tension in the past, and initiates an important shift from escalating to reducing conventional arms levels.

The Congressional Budget Office estimates that the Treaty will have a one-time $130-$385 million cost and much smaller ongoing costs per year (for verification and other purposes). These costs are overshadowed by the potential savings from troop cuts in the area, savings which could be as high as $14 billion if the 300,000 troops stationed in Europe are reduced to 100,000, action the House recommended in May 1991. In addition to its contributions to political and military stability, and its important place in an overall movement to reduce military armaments, the CFE Treaty provides some necessary economic relief by revising outdated military assumptions.

But while the CFE Treaty promises to play an important part in casing a post-Cold War transition and promoting international arms stability, its work is far from complete. The recent arms control proposals by President Bush are welcome and address some of the areas that the CFE Treaty leaves unresolved. But the President's proposals do not go far enough, and there will be a need for continued discussion and negotiation on such issues as conventional force modernization, naval arms control, and acceptable European troop levels.

The CFE Treaty is a significant arms control agreement. With its provisions for 25 percent reductions in the arms holdings of former Warsaw Pact nations, the destruction of almost 20,000 weapons by the Soviet Union, and its implicit international pledge to work against arms escalation, the Treaty supplies hope for continued arms control. The accomplishments of the CFE Treaty must be codified and expanded upon in reducing arms levels even further. The warming of relations between NATO and former Warsaw Pact nations and the continued efforts to reduce arms Proliferation should serve as mutually reinforcing and complementary movements. Within this context the CFE Treaty is a crucial step toward international arms stability.

Hon. Joseph Biden,
US. Senate, Washington, DC.

DEAR SENATOR BIDEN: I am pleased to have this opportunity to support the Treaty on Conventional Forces in Europe.

As a State Department official until February 1990 I chaired the Interagency Working Group that developed Washington's CFE position and represented the United States at many of the NATO meetings on Allied proposals; since then I have followed the negotiations as a Senior Associate at the Carnegie Endowment working on European security and United States-European relations.

The two most striking things about the CFE Treaty are, first, that the United States got everything it wanted and in some cases more-and at a lower price than it was prepared to pay; and second, that proposals designed before the collapse of the Soviet Union's East European empire and before any of us working on CFE understood the scope of changes in Soviet foreign policies are so well suited to Europe's new security situation. Some provisions in the Treaty may be even more useful now than when NATO proposed them. Additional limits on Soviet military capabilities and potential that some private experts have advocated were rejected by the U.S. Government because our military did not want to accept reciprocal constraints on its own forces or military activities.

The Treaty's major accomplishments, especially what it will do to reduce any possible threat to NATO Europe from Soviet conventional forces, are well known to the Committee. Perhaps I can best contribute to its deliberations by providing some background on why NATO's proposals emerged as they did-including why some things were not included-and how all major Western goals were realized, albeit sometimes in changed forms to take account of a rapidly changing European security situation. In the process, I would like to draw attention to one accomplishment that has received less attention, than it deserves: how the Treaty serves the security interests of Eastern Europe's new democracies as well as that of NATO Allies.

There never was any illusion that one agreement could cure all Europe's security problems. The "meat and potatoes" American aim in this Treaty was to eliminate Warsaw Pact superiority in Europe in the weapons most suited to seizing and holding territory, and to reduce Soviet weapons in Eastern Europe. Anything else was seen as "gravy".

Well before negotiations began the Soviet Union and its then-allies proposed an outcome of NATO-Warsaw Pact parity at 10-15 percent below the existing level of armaments of the side that had less. Equality of outcome had been the agreed East-West goal in the MBFR negotiations, but disputes over starting levels-and therefore over what cuts would be necessary to reach equality-made agreement impossible. But in early 1989 the Warsaw Pact published its version of the conventional weapons holdings of members of both alliances, a version which was very close to NATO's count. Differences seemed to be over what to include (most importantly, a Soviet desire to exclude Warsaw Pact aircraft comparable to those they would include for NATO) rather than from an attempt to misrepresent the figures. This Warsaw Pact publication convinced Western officials that the Soviets were serious, and that the MBFR agony of negotiating a "prior data" agreement could be avoided. The differences which did need to be resolved-some of them important ones-could be handled in negotiations over definitions and "counting rules", i.e., over what systems to include, rather than over who had how many of them.

After decades of worrying about Soviet numerical superiority in conventional arms in Europe, all Allies were eager to pin do n an Eastern offer of parity at only slightly below existing NATO levels. All agreed we should not risk forfeiting that prize by overloading negotiations that were seen as the first phase of a process. The "KISS" principle of arms control ("Keep It Simple, Stupid") was often invoked. Tanks were easily identified as the single most important ground-gaining weapon; armored combat vehicles as the weapon which could most easily substitute for tanks; and artillery as the third principle component of ground combat capability (especially as Moscow then seemed to be shifting emphasis somewhat from armored to motorized rifle units).

The Allies wanted to exclude aircraft from this first phase, in part because Soviet ground weapons were what most concerned the West, but also because of concern that Soviet desires to exclude their own air defense planes which could also be used in a ground attack role could tie up the negotiations. (Western negotiators privately cautioned their Soviet counterparts that Moscow was making a mistake to push for inclusion of aircraft, because any serious discussion would disprove Soviet assertions that the West had a numerical advantage in this area that would offset the Pact's numerical superiority in ground weapons. As Committee members know, negotiations after NATO agreed to include aircraft and helicopters vindicated that Western argument.)

The chief problems for NATO in devising an initial proposal were how to reduce Soviet weapons in Eastern Europe without unacceptably constraining United States and other Allied forces in Germany, and how to divide the Atlantic-to-the-Urals area into subzones that would meet the political as well as military concerns of all Allies. Washington doubted the latter could be done, and would have settled for equal alliance-to-alliance ceilings for the entire Atlantic to the Urals area with a sublimit only on weapons on foreign soil in Europe (thus limiting Soviet forces in the subzone we cared most about).

While NATO Allies easily agreed on military goals and the broad outlines of an eventual Treaty, most also had national political concerns: the Americans to stress NATO collectivity, the Germans not to be singled out for special constraints or to allow any "arms control dividing line" between themselves and France, Norwegians and Turks to avoid any impression that NATO cared less about their security than about that of the Central region, and the French to somewhat obscure CFE's bloc-to-bloc nature.

Paris never questioned that there would be collective Alliance ceilings, or that NATO-Warsaw Pact parity was the desired outcome. But no French politician wants to be accused of abandoning De Gaulle's legacy by a back-door resubmersion of French forces into NATO's integrated military structure. The French therefore insisted that the Treaty talk about "Groups of States Parties", with members of each group listed by name, rather than about members of alliances; that a "Sufficiency Rule", limit the portion any one state may have of the weapons permitted to all CFE participants; and that there be a system of subzones in which the most immediate collective ceiling covering French forces could be portrayed as being based on the territory of members of the Western European Union rather than of NATO (even though the "WEU collective ceiling" is then merged into a NATO collective ceiling).

One interesting, and I think little appreciated, part of the CFE story is how useful some aspects of NATO's proposal which initially were included to meet these French concerns proved to be as Europe's security situation changed. The fact that the Treaty being negotiated in Vienna put obligations on "Groups of States Parties" rather than on Alliances made it unnecessary to renegotiate everything when the Warsaw Pact came apart-a renegotiation than might not have been possible as hard-liners in the Soviet Union grew increasingly unhappy about the Gorbachev-Shevarnadze approach to CFE.

And the French-proposed "Sufficiency Rule" is responsible for the fact that there are specific limits on Soviet, as opposed to Warsaw Pact, forces. Some American officials initially had been wary of the Sufficiency Rule, lest it distract from the principle of collective NATO obligations. But in the end Washington endorsed it, in part to dramatize the fact that when negotiations began the Soviet Union had more tanks, armored vehicles, and artillery than all NATO states and all non-Soviet Warsaw Pact states combined. When the Treaty is fully implemented the "Sufficiency Rule" will limit the Soviet Union to approximately one-third of the total; a limit that will require reduction of some 70 percent of the tanks it had west of the Ural mountains when negotiations began.

In fact, while the declared Western aim was NATO-Warsaw Pact parity in key ground-gaining weapons, the combination of overall limits and the Sufficiency Rule will make a Soviet Union shorn of allies interior to NATO in Europe. No future Soviet leadership could hope to launch a successful attack on NATO Europe without reconstituting forces in ways that would violate the CFE Treaty in ample time for NATO to react-and that would free the West from any Treaty-constraints on doing so.

As already noted, the subzone limits that constrain concentration of Soviet weapons in any one part of the ATTU area were important politically as well as militarily to several Allies. (And not just the French; Norwegian security officials know that if there is a Soviet military threat to their country it won't come from tanks. But they needed to be able to show their citizens that Norway's NATO allies cared about their neighborhood as well as about the Central front.)

But these subzone limits also have the added advantage not only of limiting any possible return of Soviet forces to Eastern Europe, but also of limiting weapons within the Soviet Union itself in active units close to Eastern Europe's new democracies. Let me explain how this works.

One of the hardest puzzles those designing NATO's CFE proposals had to solve was how to force cuts in Soviet weapons stationed in Eastern Europe without accepting a reciprocal obligation that would have cut into US POMCUS stocks. This was especially challenging after Gorbachev announced in his December 7, 1988, U.N. speech that the Soviet Union would unilaterally reduce its tanks in Eastern Europe to below the level of U.S. tanks in Western Europe. This was not an issue that divided the Allies. The French were especially inventive in looking for a formula that would brand Soviet forces in Eastern Europe as "occupiers", While making clear that Americans in Western Europe were welcome protectors. But every imaginable alternative -- including the one Washington favored and NATO finally adopted -- was flawed.

After it became clear that the Soviet military would be leaving Eastern Europe, the "Stationing Rule" was dropped from NATO's proposal. It would have been unwise-and deeply offensive to the new East European democracies-to agree to any limit above the "zero" the latter were then negotiating with Moscow, and a reciprocal zero obviously would have been unacceptable to the West.

But the prospect of Soviet withdrawals under bilateral agreements with East European governments did not make limits on "stationed" forces unnecessary. Any future Soviet leadership should have to violate a Treaty with the United States, Canada, and as many European states as possible ever again to use military force to threaten or intimidate Moscow's former satellites. As Secretary Baker and others have pointed out, the CFE Treaty's prohibition of forces on foreign soil in Europe except at the express invitation of the host government should make violation of the bilateral withdrawal agreements a violation of the CFE Treaty as well. That is important. But the record of trumped-up "invitations" to Soviet forces, from the Baltics in 1940 to Afghanistan in 1979, does not make it alone entirely satisfactory.

Thus I think it worth drawing attention to the fact that the CFE Treaty's regional subceilings will have virtually the same effect on Moscow as NATO's original "Stationing Rule," but without putting an specific limits on American or other Allied "stationed" weapons. Each "Group of States Parties" will be limited to 7,500 tanks, for instance, in active units in Europe's critical Central region (Poland, Czechoslovakia, Hungary, Germany and the Benelux states). And by internal agreement among the Eastern Group of States Parties, Poland, Czechoslovakia, and Hungary will have 4,000 of that Group's 7,500, limiting Moscow to 3,500. That of course is far too many. But it is very close to the 3,200 limit NATO originally proposed for "stationed" tanks on each side, and considerably less than the roughly 9,000 tanks the Soviet Union had in the Central region when negotiations began.

The Treaty also will limit the weapons Moscow can have on its own territory adjoining Eastern Europe's new democracies. Each "Group of States Parties" can have only 10,300 tanks (to continue with them as our example) in active units in the next largest subzone-on the Eastern side, the Central region plus the U.S.S.R.'s Baltic, Byelorussian, Carpathian, and Kiev Military Districts. Since Poland, Hungary, and Czechoslovakia will be allotted 4,000 of that total, the Soviet Union will be limited to 6,300 in the parts of its own territory closest to its former satellites. That again is more than one would like; certainly enough for some future Soviet leadership to threaten its nearest neighbors. But it is less than half the 13,400 tanks Moscow had when negotiations began.

There is of course no plausible formula that could keep the Soviet Union, with its 22 million square kilometers of land area to defend, from having forces larger than those of Hungary, Czechoslovakia, and Poland, with a combined total of just over half a million. But the CFE Treaty will sharply reduce the number of weapons Moscow keeps on its own territory nearest its former satellites. And it establishes that what forces Moscow keeps within the U.S.S.R. near the borders of others is of legitimate concern of all CFE participants.

The Treaty also gives Eastern Europe's new democracies the right to inspect Soviet forces in the Soviet Union (and, it they want, to cooperate with each other in doing so by forming joint inspection teams) as well as the right to inspect each other. This can make a modest but useful contribution to easing suspicion and tensions among former members of the Warsaw Pact.

The changing European security situation also made it possible to drastically cut Soviet weapons in Europe with so few reciprocal cuts in NATO's arms. The Allies initially proposed equal limits some 10-15 percent below NATO's existing levels in ground weapons, at 15 percent below NATO levels in combat aircraft and helicopters. But when Moscow realized that it would be facing life without allies, it decided it wanted to keep more tanks, etc., than it initially had envisioned. Rather than retreat from its own much-trumpeted proposal for parity in outcome, it suggested higher ceilings for each alliance in most weapons categories than NATO had possessed when negotiations began. As a result, if one discounts weapons added to Bonn's total by incorporation of eastern Germany into the Federal Republic, NATO will need to cut only about 2,600 tanks, and could actually increase its holdings of all other Treaty Limited Equipment.

Such increases of course will not happen unless the European security situation takes a dramatic turn for the worse; most NATO allies will instead be cutting forces and relying on the longer warning of hostile Soviet intent we now would have in order to reconstitute them if necessary. But the point is that the CFE Treaty would pose no impediment to that reconstitution, even before Moscow freed us from constraints by violating it. NATO can make its own decisions about its force needs, for all practical purposes free of CFE Treaty constraints.

There is much the Treaty does not do. But in every important case the reason is that the United States did not want to accept reciprocal limits on its own forces. Let me touch briefly on the most frequently alleged sins of omission.

1. There is no limit on Soviet forces east of the Ural mountains.

• True. But getting one would have required limiting our own forces in the United States, including those designed for contingencies such as the Persian Gulf. Even if global limits on United States and Soviet forces are deemed feasible and desirable, they should not be set in a negotiation about Europe.

1. By Moscow's own count, some 67,300 Soviet weapons in Treaty Limited categories were moved east, out of the ATTU area, before signature.

• True and unfortunate. But one consequence of Soviet efforts, after Treaty signature, to exclude weapons held by Coastal Defense and Naval Infantry units from CFE limits will be greater accountability for these weapons. As part of the agreement resolving that dispute, the Soviets have given Washington a detailed breakdown of how many weapons in each category removed from the ATTU area will be destroyed (under "procedures that will provide sufficient visual evidence"), used to replace older weapons now used by military units in the Eastern U.S.S.R. (which will in turn be destroyed), or held to replace weapons of those Eastern units as the latter wear out. Moreover, Moscow has undertaken to supply information about the locations and numbers of each type of weapon in question, and not to store them in sets for military formations that would facilitate their rapid return back across the Urals. This is not a perfect outcome. These 67,300 weapons were eliminated from Europe because of the CFE Treaty, but special constraints on their disposition and use in the East is less desirable than their destruction or conversion under the Treaty would have been. But American negotiators are to be commended for getting a pretty good handle on weapons that were legally removed from the Treaty area. Moscow's undertakings regarding those weapons should not only impede their return to Europe, but also ensure that they are not transferred to potential third-world trouble makers, which was one reason NATO proposed destruction of excess weapons in the first place.

1. Failure to require reductions by large units will permit Moscow to retain in the ATTU area a military infrastructure that would facilitate rapid "breakout" from the Treaty if weapons were moved back across the Urals.

• True, but Washington adamantly opposed any requirement to take its own reductions by units of any size, or to put any limits on NATO's residual military units. We insisted on complete freedom to take our cuts by thinning weapons out of units rather than by removal of whole units, and to subsequently restructure our forces, inter alia by redividing them into as many units as we in the future might choose. Now that most NATO countries will be reducing active duty forces, relying instead on the ability rapidly to reconstitute them, I suspect that limits on our own and our allies' ability to leave skeleton units and logistics assets in place would be even less acceptable to us.

1. There is no limit on production of Treaty Limited Equipment; newly produced weapons could legally be stored in the zone to facilitate "breakout" from Treaty limits.

• True. NATO allies were unwilling to accept limits on production of weapons, some of which are intended for export-and especially not if American arms industries, located outside the ATM area, were not similarly constrained. Post-Desert Storm concern about weapons proliferation may now make it possible to get some constraints on weapons exports, but if so it should be done on a global basis, not in an agreement limited to Europe.

1. Reducing the Soviet military threat to Central Europe will somehow increase it to NATO allies on the flanks.

• This certainly is not true quantitatively. Soviet tanks in the Leningrad Military District which borders Norway, and in the Odessa, North Caucasus, and Transcaucasus Military Districts near Turkey, will be limited to 5,700 in contrast to 7,236 that were in those Districts when negotiations began. And 1,000 of those tanks must be kept in storage in the Odessa MD and the southern part of the Leningrad MD, not directly adjacent to any NATO ally. The 6,700 Soviet tanks that can remain adjacent to flank Allies very likely will lie modernized, as newer equipment from Moscow's Central region is substituted for older equipment now held on the flanks. But it was NATO that invented the idea of "canading" weapons (i.e., to move modern weapons which are over subzone ceilings in the Center to the flanks and destroy older weapons located there), and Turkey, in particular, stands to benefit from it.

1. Moscow was allowed to renegotiate the Treaty, to exclude weapons in Coastal Defense and Naval Infantry units.

• Not true. This is a phony issue. Washington never intended to specify which weapons from which kinds of units had to be destroyed to meet Treaty ceilings, and indeed would have rejected any such limit on its own freedom. Moscow may think it has saved a bit of face by saying the weapons in Coastal Defense and Naval Infantry units can remain intact while an equivalent number from other types of units are destroyed. But that is perfectly consistent with NATO's original proposal and with the letter and intent of the Treaty.

Finally, I would like to comment briefly on how those of us designing the NATO proposal thought about verification. No one ever imagined that every tank, much less each artillery piece, in the European U.S.S.R. could be counted. Even if that were possible it would not be worth the drain on the American military's shrinking manpower-or the U.S. taxpayer's pocket-to do so. We would be able to detect militarily significant cheating in ample time to respond, even with a much less rigorous monitoring regime than this Treaty includes. How much cheating is politically significant is to some degree a subjective judgment, and one that would have to be made in the context of the time. Even a little deliberate cheating certainly would make one question why it was happening. But the web of inspections under this Treaty should reveal any significant pattern of small-scale cheating, even below the level of "militarily significant". I hope it also will contribute-to the broader goal of exposing the Soviet military to their Western counterparts, on every level and in every way possible. That has been an unexpected side benefit both of the Stockholm CDE observation and inspection regime and of the INF Treaty.

Could we have done better? Could the West, by seizing the initiative and proposing to move immediately to NATO-Warsaw Pact parity at half NATO's level (as Gorbachev was suggesting for a subsequent phase of negotiations) have won deeper cuts in the Soviet military, at no cost to ourselves beyond the unilateral cuts that NATO members, as it turns out, will be making anyway? I thought so while I was in government, and argued unsuccessfully for just such a proposal. But I now think that in view of growing unhappiness among the Soviet military about cuts being forced on them, and indeed the simple physical problem of absorbing redundant Soldiers into Soviet civilian society, a proposal for still deeper cuts not only could have tied up the negotiations; it also might have played into the wrong hands in intern debates about the Soviet Union's future.

I also would like to have seen a total ban on Soviet forces from Eastern Europe, and have argued elsewhere why I think Moscow might agree to that in return for banning only U.S. ground combat units from Europe. But neither our NATO allies nor, I suspect, any member of this Committee, would be willing to pay that price.

Could we do better now, if the Senate for some reason found this Treaty inadequate and asked the Administration to try again? Definitely not. The point should not have to be belabored that 1989, when both NATO and the Warsaw Pact introduced strikingly similar CFE proposals and 1990, when negotiations were concluded, were the time of maximum Soviet will and ability to agree to sweeping cuts in their own forces in return for no meaningful NATO cuts.

I do not mean to suggest that this is just the best Treaty now possible. By any standard, it is an extraordinarily good one. To again pose a military threat to NATO, any Soviet leadership would have to violate it in ways that would provide the "smoking gun" that should persuade even the most timid Western leader to take action. Its verification provisions are more than adequate to detect cheating that might pose a military threat, and also should substantially increase our understanding of Soviet military forces. The reverse will also be true, and is very much in our interest. Hopefully, the inspection regime will add to the growing web of military-to-military contacts and increased mutual understanding that can have a confidence-building spinoff. The Treaty also makes modest but useful contributions to enhancing the security of the new democracies in Eastern Europe. And it does all this while requiring no meaningful cuts in NATO forces or constraints on its flexibility. I am proud to have had a part in shaping its terms and congratulate those who brought the negotiations to so splendid a conclusion. I urge its prompt ratification.

Sincerely,

JENONNE WALKER,
Senior Associate, Carnegie Endowment for International Peace.

Next Section

Return to Top
Return to Table of Contents