|Anticipatory Defense: An Attempt at Structuring the Problem|
Anticipatory Defense: An Attempt at Structuring the Problem
Invited contribution to the Hanns-Seidel Stiftung Workshop "Anticipatory Defense: Basic Principles, Regional Priorities; Military Implications", in Wildbad Kreuth, Bavaria, May 27, 2003. Address by Professor Dr. Reiner K. Huber (à“), from the Institut fà¼r Angewandte Systemforschung und Operations Research, Fakultà¤t fà¼r Informatik, Università¤t der Bundeswehr Mà¼nchen. Photo European-Security.
Professor Dr Huber with Professor Dr Klaus Lange
The notion of preemption is one of the core elements of the new National Security Strategy (NNS) of the United States. There, the traditional definition of the anticipatory use of force in the face of an imminent attack is extended to encompass preemption by using force without evidence of a clear and present danger to national security in order to eliminate a potential threat to the United States before it materializes.
In her Wriston Lecture, delivered to the Manhattan Institute on 1 October, 2002, Condoleezza Rice pointed out that the NNS considers preemptive military action only after all other means, including diplomacy, have been exhausted: "Preemptive action does not come at the beginning of a long chain of effort. The threat must be very grave. And the risks of waiting must far outweigh the risks of action". [Ric 02]
Nevertheless, the preventive actions taken by the United States in Afghanistan and, in particular, Iraq have led to considerable irritations in transatlantic and, one should add, inner-European relations as well. Whatever may have motivated some of Europe's leaders to denounce U.S. actions against Iraq, public response suggests that most Europeans did regard the threat not to be very grave and, therefore, preventive action simply a "war of aggression" rather than an act of anticipatory defense. In the public debate before and during the war, attempts to explain the rationale underlying the concept of preemption were mostly met by rather emotional assertions about presumed American motives and inappropriate historical analogies.
Therefore, it is considered highly meritorious that the Hanns-Seidel-Stiftung provides a forum to discuss the emerging concept of anticipatory defense and its military, political, and international legal implications. The author of this paper hopes to contribute to the debate by attempting to structure the problem by analyzing a simple analytical model of threat perception on order to investigate the fundamental military-strategic logic and some of the problems underlying the concept of anticipatory defense in general.
The threat perception model analyzed in this paper has been first proposed by an international study team commissioned by NATO to investigate hypotheses related to conventional military stability in Central Europe. The core question was how military organizations in a region must be sized, structured, equipped, deployed, and operated so that no party to the regional international system had a reason to perceive its security to be threatened by the other parties. Particular attention was to be paid to strengthen crisis stability between NATO and the Warsaw Pact by uncovering conditions that might provide either side with military incentives for preemptive use of force in a crisis [RSG 18 and Hub 96].
The model assumes that the threat assessment of a party à‚Â– a state or an alliance à‚Â– vis-à -vis another party basically implies a three-staged process that may be captured by a simple binary scheme as depicted in Figure 1. Accordingly, the military situation between the two parties is considered as stable if unilateral assessments by each of both parties arrived at the conclusion that the other party denoted by X either
- had no intentions to attack or
- were not inclined to take high military risks or
- could be repelled if it were to attack.
Fig 1 Threat Perception vis-à -vis Party X
Of course, it goes without saying that in most cases threat assessments will not arrive at definite answers à‚Â– yes or no à‚Â– as suggested by Fig. 1. Rather, they must be expected to be characterized by some degree of uncertainty that may be expressed in terms of probabilities (defined in terms of values between 0 and 1) as shown in Figure 2.
If it is assumed that the security requirements of the party assessing the threat posed by party X are satisfied if the probability for the unstable state does not exceed a certain threshold value K, then the following condition holds:
with the notations:
- K = threshold probability satisfying security requirements
- PI = probability that party X intends to attack
- PA = probability of attack by party X succeeding
- P* = minimal attack success probability required by party X (risk threshold)
- W = probability of successful defense against party X.
Fig 2: Probability Tree of Threat Perception vis-à -vis Party X
Condition (1) is satisfied if either PI = 0 (there are no aggressive intentions perceived on part of X) or G(PA )= 0 (military risk associated with attack by X is perceived to be to too high since PA < P*).
If, however, X must be regarded as not being risk-averse (G(PA) = 1), i.e., there is uncertainty about the value of the risk threshold P*demanded by X, or the value of P* is low enough to satisfy the condition PA à‚Â³ P*, the minimal probability of repelling an attack by X that meets the security requirements vis-à -vis X results by solving equation (1) for W as
Because the threshold value W* is largely insensitive to variations of PI if security requirements are high (e.g. K à‚Â³ 0.95), there is no need for the notoriously difficult assessment of an opponent's intentions by assigning a value to the probability PI. Moreover, if a party were certain that it would be attacked by X if an opportunity arose (i.e. PI = 1), condition (3) simplifies to
As envisaged at the time, conventional stability required that the threat assessments of both parties à‚Â– NATO and WP à‚Â– arrived at the conclusion that the mutual conventional force structures and operational/tactical doctrines meet condition (1). Therefore, it is safe to state that the military situation in the Central European region was not very stable during most of the Cold War, in particular in crisis situations.  The apparent Cold War stability was lastly due to the nuclear arsenals of the United States and the Soviet Union which involved the risk that military conflict between the two camps might escalate ending in mutual destruction of both sides.
- Conclusions from the Model
Even though the model has originally been designed to develop easily reproducible criteria for assessing conventional stability in a well defined bipolar context, it does nevertheless capture the essential ingredients of assessing, in numerical terms, the degree of potential threat posed by any type of antagonist, provided there is reasonable historical evidence to estimate P*, and the dependent variables PA and W in equation (1) on the basis of expert judgments, historical evidence, and computational or simulation experiments . Short of numerical results obtained from such sources in the context of a defined threat, however, the model does provide a structure for deriving the principal options that parties have for meeting their security requirements given the nature of the international security environment.
In a cooperative international environment, the majority of parties share the common objective of acting in a manner that is conducive to international stability. Cooperative security environments are characterized by parties which agree on confidence building measures and negotiate, and conclude, formal agreements on arms limitations and force deployment constraints so that the security requirements of all parties captured by the conditions W à‚Â³ W* (security criterion) or PA < P* (sufficiency criterion) are met. 
In an uncooperative security environment, however, parties are forced to satisfy their security requirements unilaterally. And if the environment is outright hostile, parties are inclined to perceive many an action on part of presumed opponents as attempts to increase their offensive potential vis-à -vis themselves to the degree that PA à‚Â³ P* and W Â® 0.
Such a perception of parties who consider themselves as potential victims of aggression would be justified if there is strong evidence that the respective opponents
- are building up their arms inventories;
- develop means and procedures for surprise attack, and
- pursue policies to reduce the victim's defense potential by subversion and/or through policies of alienating allies from the victim.
Excluding the option of appeasement, the potential victim's response options are aimed at strengthening its own defensive potential so that W É¨19; W* and, at the same time and reducing the offensive potential of opponents to the degree that PA < P*.
There are essentially five principal categories of options for responding to perceived threats:
-- limiting the offensive potential of opponents through agreements, with the governments of like-minded countries, on export limitations of crucial technologies and arms (arms export controls and counter-proliferation policies) including arms embargos,
-- increasing the active defense potential for defeating threats in a reactive manner,
-- reducing the vulnerability of the active defense potential and of high value targets as well as populations,
-- deterrence by threatening punishment in case of aggression, and
-- preemptive/preventive counter-force attacks to neutralize and /or destroy threat potentials.
The first three categories include the traditional options of a defense-oriented (status quo) party for responding to emerging and immediate threats. They are legitimized by international law and the charter of the United Nations and therefore not controversial, at least in principle. However, their implementation to a degree that satisfies the security criteria may involve economic costs and expenditures that the parties cannot or may not be willing to shoulder.  Moreover, they may be of limited effectiveness, especially in case of threats posed by non-state actors and governments of rogue states. This is also true for deterrence because essential prerequisites for it to work seem to be missing in those cases. 
That leaves preemption and prevention in order to mitigate or eliminate the threat before it becomes effective. The difference between preemptive and preventive response is that the former is directed against an immediate threat, i.e. a clearly identified threat in the process of being activated for attack, and the latter against an emerging threat that may become an immediate threat sometime in the future unless dealt with now. Both make good strategic sense and are in many cases superior in cost-benefit terms to not acting or waiting until an attack has begun. However, while preemption is supported by international law and the just war tradition, prevention has no basis in current international law. 
Thus, since it implies both preemptive and preventive use of force, the concept of anticipatory defense is bound to cause controversy. Therefore, an important question to be answered is under which circumstances anticipatory defense must be considered the only option that a potential victim has left to meet its security requirements and bring about some degree of stability.
Figure 2 suggests that this would be the case if the values of both the probability of successful defense W and the attacker's risk threshold P* were to become small which implies that the value of the victim's security threshold K must be small as well (K É¨14; 0) in order to satisfy condition (1). In other words, the threat perception model tells us that there is no security for the victim when W É¨14; 0 and P* É¨14; 0. He is prepared either to live with the unstable situation and surrender once the threat materializes, or use force in a preventive manner in order reduce the threat potential to a degree that his security requirements are met.
If we interpret an attack not as large scale aggression Â– as was implied when the model was originally formulated Â– but as a massive attack against soft targets such as the one on September 11, 2001, it is the existence of weapons of mass destruction (WMD) and stealthy means and mechanisms for their delivery Â– such as terrorists or ballistic missiles (BM) in the absence of an effective anti-ballistic missiles (ABM) system Â– that have the potential of bringing about conditions in the real world which correspond and in our model. Therefore, a closer look shall be taken at both threats.
It is not individual terrorists or groups of terrorists that would be the object of preventive defense by military means but their training infrastructure and shelter available in weak and failed states as well as states and factions that support them.
The trends of global demographic, environmental, economic, societal, and technological developments Â– as identified, for example, by the United Kingdom's "Project Insight" [Ham 99] or Germany's project "Armed Forces, Capabilities and Technologies in the 21st Century"  Â– suggest that the terrorist problem will most likely worsen as the number of weak and failing and states must be expected to grow. This is because, in addition to Islamic fundamentalism, a fundamental change in warfare has taken and still is taking place. Wars between states as in the first part of the last century are increasingly replaced by intra-state (civil) wars between ethnic and religious factions and groups interested in ongoing conflict for economic reasons, and by international terrorism and organized crime. The observation that only 15-20 percent of all wars and warlike conflicts since 1945 were conducted between states has led political scientist Herfried Muenkler to propose the hypothesis of "privatization of war" [Mue 01]. As a consequence, the world will increasingly witness situations characterized by what von Johannes Kunisch (1973) refers to as "small wars" similar in nature to those fought in the European middle ages with feudal levies and short-term contract mercenaries (condottieri) before standing armies emerged [Kun 73]. 
Thus, the state as the only legitimate party for conducting war is about to be replaced by non-state actors such as warlords, guerrilla groups, and criminal organizations which live on war and, therefore, have no interest in ending war and violence. Muenkler considers the terrorist attacks of September 11 on the World trade Center and the Pentagon as dramatic manifestations of that trend the consequences of which may be dire indeed if weapons of mass destruction (WMD) come into play. Unless appropriate means of prevention are developed, failing and weak states like the Taliban's Afghanistan will be used and even hijacked by such actors to provide staging grounds for their operations. And rogue states may not resist the temptation supply such groupings with know how and WMD, or even employ them, unwittingly or covertly, as mercenaries in the hope of avoiding discovery and evading retaliation [Hub 03].
In order to cope with these trends, military intervention, including the provision of security for rebuilding the economic and social infrastructure of such states (nation building), to deny terrorists the use of weak and failing states as platforms for their purposes must be considered an essential mission of anticipatory defense.  Operations of this kind should provide little opportunity for international legal controversy because their objectives are to a large degree humanitarian as weak and failing states are incapable of protecting elementary human rights of their citizens.
Another important element of a preventive defense strategy in the face of WMD is proposed by Paul Davis in form of a credible declaratory policy threatening anyone who even tolerates WMD-related terrorism. "Not only active supporters (should be punished), but even those states and factions that merely tolerate the terrorists or indirectly facilitate their acquisition of WMD. The purpose would be so to alarm heads of state and of sub-state organizations that they would work actively to get rid of elements that might bring destruction upon them." [Dav 02, pp. 40]. Davis also concludes that terrorists themselves, while not to be individually deterred in the traditional sense, may be "influenced" to eventually give up by forcefully and relentlessly holding at risk and attacking, not necessarily by military means, what is dear to them. To this end he recommends to develop an orchestrated "Broad-Front Strategy" that cuts across all of the normal boundaries of war Â– military, diplomatic, economic, and law enforcement.
Other than the five declared nuclear powers, nearly 25 states have acquired, or are about to acquire, ballistic missiles (BM) and/or the know-how and technology required for their domestic production ([Rum 98], [Wil 00], [Sch 01]). Most notable among them are India, Pakistan, and the rogue states North Korea, Iran, Syria and Libya. Even though today's delivery accuracy of ballistic missiles produced in emerging missile states may be rather low, in conjunction with nuclear, bacteriological, or chemical warheads they represent a formidable threat to urban targets and military facilities alike. Most of the current arsenals of rogue states consist of SRBM (Russian Scud and Scud improvements) with ranges up to about 600 km. However, there are several indigenous IRBM development programs such as the north Korean Taepo-dong and the Iranian Sahab. If launched from North Africa, the 2,000 km range (presumably operational) Taepo-dong-1 and the Sahab-4 would be sufficient to cover most of Southern Europe, the Balkans, and Turkey. The IRBM Taepo-dong-2 and Sahab-5 expected to become operational about 2005 will have a range of 5,000-6,000 km capable of reaching, from the Middle East, targets anywhere in Europe, Western Siberia, Central and South Asia and North and Central Africa. North Korea and Iran are suspected of developing ICBMs with ranges of up to 12,000 km that could become operational within 10-15 years.
From the viewpoint of international law, BM present a particular challenging problem for anticipatory defense. This is because, in order to preserve its security, the country which perceives itself as a potential victim of missile attacks has only the option of a preventive attack against the BM and/or its WMD warheads and their production and storage facilities unless, of course, it has deployed an effective ABM system.  It must be remembered, that an operational BM system represents a threat not only for a particular country, but for the entire region within its range. Which country is being targeted will only be known sometime after the missiles have been launched so that their trajectories can be determined. And even if launch preparations were discovered early, and countries were willing to preempt on the suspicion of being targeted, it probably would be too late.
Nevertheless, despite the fact that all of them will soon be within range of BM from the Middle East, Europeans as yet show little enthusiasm about both, deploying ABM systems and considering preventive response as an option. For one thing, besides being rather expensive, development and procurement of ABM systems is still a controversial issue even though the debate has abated as the publics begin to realize that none of the dire predictions about the consequences of the U.S. withdrawal from the ABM treaty have materialized so far. For the other, an effective preventive response may very likely require new and even more controversial weapons (such as low-yield nukes) as production and storage facilities of BM systems, in particular of their WMD-warheads, will be relocated into deep underground bunkers.
From the foregoing we conclude that anticipatory defense must be regarded an important, and very likely the only, concept to provide for security in face of the observed global trends related to international security, the problems that international law poses today notwithstanding.
The standards of current international law Â– which protects states and governments, even those that deprive citizens of elementary human rights Â– will be up-dated eventually, if not on the basis of the precedent of the Kosovo war, at the latest when painful evidence abounds that international terrorism is a global phenomenon that threatens civilian populations on a massive scale not only in the United States, but everywhere.
In order to minimize the fallout for international order and civilian populations, the legitimacy of the preventive use of force should be made contingent to at least three conditions being met:
- (1) clarity of protective purpose,
- (2) capability to keep collateral damage at a minimum, and
- obligation to restore whatever damage the operation may have caused.
Professor Dr. Reiner K. Huber
 NATO analysts concluded that the vast numerical superiority and structure of WP forces assured them a probability of success when attacking NATO's fairly static "Forward Defense" that exceeded the high operational/strategic risk threshold (P* = 0.9) of the traditionally risk-averse Soviet forces. On the other hand, Soviet analysts interpreted the numerical inferiority of the heavy-armor ground forces of NATO in conjunction with the small operational depth on NATO territory as an indication that NATO was preparing for a preemptive attack in a serious crisis. In their assessment, surprise and moving forward to gain the operational depth, for an efficient employment of armor against a numerically superior enemy, east of the demarcation line were the essential elements of "Forward Defense". Thus, a high degree of crisis instability has apparently persisted during the Cold War because the Soviets had come to the conclusion that they must be able to preempt NATO's presumed preemption in a serious crisis (statement of a Soviet participant at a 1986 Pugwash meeting in Starnberg).
 The reader is referred to Huber et al. [HFL 99] for example of estimating Â– in response to an assessment of the security risk associated with NATO enlargement by Professor Vitaly Tsygichko of the Russian Academy of Science [Tsy 97] Â– the relative degree to which Russia's national security may be impaired by regional threat potentials along the periphery of its territory.
 The security criterion is appropriate in situations when there is no evidence to assume that the military risk aversion of the potential opponent in question is high. The sufficiency criterion is adequate if there is overwhelming evidence that the opponent in question is highly reluctant to take military risks. Huber and Schindler have shown that it is impossible to satisfy every party's security requirements in a multi-state international system on the basis of the security criterion if the situation is characterized by distrust among parties. However, military stability of a multi-state international system is entirely feasible on the basis of the sufficiency criterion especially when non-military relationships among the risk-averse states are cooperative [HuS 93].
 For example, arms control suffers from the difficulties of achieving agreement among potential supplier states on technology limitations. These difficulties are related to: economic competition and differences in political goals and interests; problem of enforcing compliance with agreed-upon limitations if rewards of non-compliance exceed possible sanctions by some significant margin; and the declining efficacy of controls as the technologies spread beyond the group of signatories to the arms control agreement [Pay 97].
 Deterrence is considered to be inherently unreliable vis-à-vis proliferating states in the post-Cold War era. It is difficult to establish reliable rules for deterrence with governments of proliferating states whose cultural and social environment are not well understood in most cases, and whose behavior is evidence of their unfamiliarity with the essential prerequisites of effective deterrence which emerged during the Cold War: rational behavior of antagonists; mutual understanding of motives; effective communication between antagonists; credibility of threats [Pay 97].
 The NNS does not distinguish between preemptive and preventive use of force. Rather, it extends the meaning of preemption to encompass military action "even if uncertainty remains as to time and place of attack", i.e. preventive use of force. Because of its possible implications for international order, critics consider this aspect of the NNS as worrisome [LAR 02].
 The "Zentrum fuer Analysen and Studien der Bundeswehr" (ZAS - Center for Studies and Analyses of the Bundeswehr) was commissioned by the German Ministry of Defense in1999 to analyze what type of military would required in the 21st Century. It was supported by experts of the relevant scientific fields as well as industry and the military. A publication of the findings will be available from the ZAS.
 See also the treatise by Carl von Clausewitz on the aims a belligerent party adopts, and the resources it employs [Cla 84, pp.585-594].
 This is very likely what German Defense Minister Struck had in mind when he explained Germany's new defense policy guidelines by stating that Germany is also defended in the Hindu Kush.
 Unless armed with WMD-warheads, BM are not very effective weapons. Therefore, production and storage facilities for WMD-warheads are high priority targets for preventive attack. The first and only successful operation of this kind in history was carried out in 1981 by Israel on the Osiraq reactor nearing completion at the Tuwaitta nuclear research center near Baghdad. Carried out with devastating precision by 14 attack aircraft, the raid had been carefully planned and rehearsed for some time [Arm, 1993]. The international reaction at the time was quite negative. However, today one wonders what course history would have taken had this raid failed and Iraq would have been in possession of a few nuclear weapons when it occupied Kuwait.
[Arm 93] Armitage, Michael: History of Airpower. In: Dupuy (Ed.): International Military and Defense Encyclopedia. Washington-Ne York 1993: Brassey's, pp.82-93
[Cla 84] Carl von Clausewitz: On War, Book Eight, Chapter Three. Indexed Edition edited and translated by Michael Howard and Peter Paret. Princeton 1984: Princeton University Press
[Dav 03] Davis, Paul K. and Brian Michael Jenkins: Deterrence and Influence in Counterterrorism Â– A Component in the War on al Qaeda. Santa Monica 2003: RAND
[HFL 99] Huber, Reiner K., Friedrich, Gernot and Jaroslav Leszczelowski: A New Paradigm for Estimating Russian Force Requirements? On Tsygichko's Model of Defense Sufficiency. European Security, Vol.8, No.3 (Autumn 1999), pp. 101-123
[Hub 96] Huber, Reiner K.: Military Stability of Multipolar International Systems: Conclusions from an Analytical Model. In: Models for Security Policy in the Post-Cold War Era (R.K. Huber and R. Avenhaus, Eds.). Baden-Baden 1996: Nomos Verlagsgesellschaft, pp. 71 - 81
[Hub 02] Huber, Reiner K.: Ballistic Missile Defense: A Risk for Stability and Incentive for New Arms Races? In: International Seminar on Nuclear War and Planetary Emergencies 26th Session (A. Zichichi and R. Ragaini, Eds.), Singapore 2002: World Scientific Publishung Co., pp. 71-84
[Hub 03] Huber, Reiner .K.: The Transatlantic Gap: Obstacles and Opportunities for Closing it. In: Transforming NATO Forces: European Perspectives (C.R. Nelson and J.S. Purcell, Eds.), Washington 2003: Atlantic Council of the United States, pp. 59-78
[HuS 93] Huber, Reiner K. and Otto Schindler: Military Stability of Multipolar Power Systems: An analytical concept for its assessment, exemplified for the case of Poland, Byelarus, the Ukraine and Russia. In: International Stability in a Multipolar World: Issues and Models for Analysis (R.K. Huber and R. Avenhaus, Eds.). Baden-Baden 1993: Nomos Verlagsgesellschaft, pp. 155-179.
[Kun 73] J. Kunisch, J.: Der Kleine Krieg: Studien zum Heerwesen des Absolutismus. Frankfurter Historische Abhandlungen. Wiesbaden 1973: Steiner
[LAR 02] Lieber, Keir A. and Robert J Lieber: The Bush National Security Strategy. Foreign Policy Agenda. Washingto 2002: U.S. Department of State, pp.32-35
[Mue 02] Muenkler, H.: Die brutale Logik des Terrors: Wenn Doerfer und Hochhaeuser zu Schauplaetzen von Massakern werden Â– Die Privatisierung des Krieges in der Moderne. SZ am Wochenende, Nr. 225, 29./30 September 2002, p. 1
[Pay 97] Payne, Kieth: Diplomatic and Dissuasive Options (Counter-Proliferation, Treaty based Constraints, Deterrence and Coercion). In: Ranger (Ed.): Extended Air Defence and the Long-Range Missile Threat. Bailrigg Memorandum 30, 1997: Lanchester University, Centre for Defence and International Security Studies, pp. 38-43
[Ric 02] Rice Condoleezza, : A Balance of Power that Favors Freedom. Foreign Policy Agenda. Washingto 2002: U.S. Department of State, pp. 5-9
[RSG 18] Research Study Group 18 on Stable Defence: Stable Defence Â– Final Report. Panel 7 on The Defence Applications of Operational Research. Brussels 1995: North Atlantic Treaty Organization Â– Defence Resarch Group
[Rum 98] Rumsfeld, Donald H. et al: Report of the Commission to Assess the Ballistic Missile Threat to the United States. Washington 1998
[Sch 01] Schilling, Walter: Die Proliferation von ballistischen Raketen und Massenvernichtungswaffen. Europäische Sicherheit Nr. 5, Mai 2001, pp. 49-51
[Tsy 97] Tsygicko, Vitali N.: A Model of Defense Sufficiency for Estimate of Stability in a Multipolar World. In: Proceedings of the International Seminar on Nuclear War and Global Emergencies, 22nd Session (Goebel, Ed.). Singapore 1997: World Scientific, pp. 168-171
[Will 00] Wilkening, Dean A.: Ballistic-Missiles Defence and Strategic Stability. Adelphi Paper 334, London 2000: The International Institute for Strategic Studies.