Chapter 2 Theory I: Core Concepts and Systems PDF

CHAPTER 2 Theory I: Core Concepts and Systems core concePts International Crisis and Protracted Conict An international crisis, later identi ed as an international political earthquake, begins with a disruptive act or event, a breakpoint (trigger), that creates a foreign policy crisis for one or mor...

CHAPTER 2 Theory I: Core Concepts and Systems core concePts International Crisis and Protracted Conict An international crisis, later identi ed as an international political earthquake, begins with a disruptive act or event, a breakpoint (trigger), that creates a foreign policy crisis for one or more states; for example, the crossing of the Thag La Ridge in India’s North East Frontier Agency (NEFA) by People’s Republic of China (PRC) forces on September 8, 1962, setting in motion the China/India Border Crisis-War; and the dis- patch of Egypt’s 4th Armored Division into the Sinai Peninsula on May 17, 1967, along with its over ight of Israel’s nuclear center at Dimona in the Negev desert the same day, leading to the JuneSix-Day War. An international crisis ends with an act or event that denotes a qualita- tive reduction in con ict activity. In the cases noted above, crisis termina- tion was marked by the unilateral declaration of a cease re by China on December 1, 1962, and the end of the Six-Day War on June 11, 1967, respectively. A militarized interstate dispute [MID], the Correlates of War [COW] project counterpart of the ICB concept of international crisis, has been de ned as “a set of interactions between or among states involving threats to use military force, displays of military force, or actual uses of military force.” © The Author(s) 2018 23 M. Brecher, A Century of Crisis and Con ict in the International System, DOI 10.1007/978-3-319-57156-0_2 24 M. BRECHER The majority of post-WW I twentieth and early twenty- rst century international crises, 58%, occurred within the context of an on-going interstate protracted con ict; however, the overall frequency of crises revealed a substantial decline—from 273 international crises, with a total of 619 crisis actors during the half-century, 1929–1979, to 84 crises, with a total of 209 crisis actors during the quarter century that followed, 1990–2015. fl fl fl fl fl fi fi fi fl fi fi fi fi fl fl International crisis and protracted con ict are closely related but not synonymous. The focus of crisis is usually a single issue or a speci c epi- sode—a territorial dispute, an economic boycott, a threat to a political regime, an act of violence, etc. By contrast, protracted con ict has been de ned as “hostile interactions which extend over long periods of time with sporadic outbreaks of open warfare uctuating in frequency and intensity.... The stakes are very high.... They [protracted con icts] linger on in time.... [They] are not speci c events..., they are processes” (Azar et al. 1978). Protracted con icts are lengthy, at least 10 years, many of them several decades, centuries, or more. All uctuate in intensity. Many move from war to partial accommodation and back to violence (e.g., India/Pakistan since 1947). Other con icts have been characterized by continuous war but of varying severity (Vietnam 1964–1975). All arouse intense animos- ities with spillover effects on a broad spectrum of issues. And con ict ter- mination, where it occurs, is often complex. Even when an international crisis is very long it can be distinguished from a protracted con ict, as with the ( rst) India/Pakistan crisis-war over Kashmir in 1947–1948, one of 12 international crises, includ- ing four wars, during the India/Pakistan protracted con ict over many issues, tangible and intangible, since the end of British rule over the subcontinent in 1947. So too with the ( rst) Arab/Israel crisis-war in 1948–1949, one of 30 international crises during their largely unre- solved protracted con ict, including nine wars [to be summarized later in this book]. Using a modi ed version of the Azar et al. de nition—deleting vio- lence as a necessary condition because it did not accord with reality— ICB uncovered 33 protracted con icts since the end of World War I: for example, at the global level, the East/West con ict and, at the regional level, Ethiopia/Somalia (Africa), Ecuador/Peru (Americas), China/Japan (Asia), France/Germany (Europe), and Iraq/Iran (Middle East), among others. 2 THEORY I: CORE CONCEPTS AND SYSTEMS 25 An overall majority of international crises during the near-century, late 1918–late 2017, 58%, occurred within an interstate protracted con ict, with a notable decline over time— from 59% of 1918–1994 crises to 52% of crises from 1995 to 2015. The other international crises occurred out- side that setting; that is, they emerged in an environment without the prior condition of prolonged dispute over one or more issues and without the spillover effects of cumulative crises between the same adversaries. Operationally, for a dispute between states to qualify as a protracted con ict (con ict), there must be three or more international crises between the same pair or cluster of adversaries over one or more recur- ring issues during a period of at least 10 years (The concept, protracted con ict, is similar to that of “enduring rivalry” (ER), with three conditions: at least ve militarized interstate disputes (MIDs) between the same adversaries, each lasting at least 1 month; 25 years from the rst to the last dispute within the rivalry, and a gap of no more than 10 years between two of these disputes). This de nition of an interstate pro- tracted con ict provided the conceptual basis for the classi cation of international crises, and for the research questions that guided the analy- sis of international crises and protracted con icts. fl fl fl fl fl fi fi fl fl fi fl fi fi fi fl fi fl fl fl fl fi fi fl fl fl fi fl fl Are there differences in the con guration of crises that occur within and outside protracted con icts, and, if so, what are they? Speci cally, how does the attribute of protracted con ict affect the crisis attributes and dimensions from onset to termination? Crises that erupted within con icts were more likely than others to have been triggered by violence, to generate the perception of grave threat, and to entail the use of vio- lence in crisis management. Despite these indicators of crisis severity, the international system has often been unable to deal with these crises effectively, either through its international organizations or through the attempts at crisis resolution by major powers. The notion that international crises within protracted con icts are more likely than others to be triggered by violence derives from a con- ict’s distinctive characteristics. First, prolonged hostility between the same adversaries creates mutual mistrust and expectation of violent behavior. Second, the likely presence of several issues within an on-going interstate con ict, a characteristic of many but not all protracted con- icts, strengthens this anticipation. Third, resort to violence in the past relationship between adversarial states reinforces the belief that violence will recur. And nally, the importance of the values at stake creates a dis- position to initiate violence against an adversary. 26 M. BRECHER Conceptually and empirically, crisis is also closely linked to war. Most international crises erupt in a non-war setting. Some do not escalate to war (notable e.g., Berlin Blockade, 1948–1949, Cuban Missile Crisis, 1962). Other crises begin in a non-war setting and escalate to war later (Entry into World War II, 1939). And still others occur during a war, such as defeat in a major battle, Stalingrad, in 1942–1943, for Germany, or the dropping of atomic bombs on Hiroshima and Nagasaki, in 1945, for Japan. These intra-war crises (IWCs) profoundly affected the deci- sions of German and Japanese leaders during World War II. All types of international crisis manifest its necessary conditions, namely, more intense, or a basic change in, disruptive interactions and a perceived likely outbreak of military hostilities (or, for an intra-war crisis, a perceived adverse change in the military balance), which undermine the relationship between the adversaries and pose a challenge to system sta- bility. Moreover, the effects of the IWCs cited here were more signi cant than most non-IWCs for state behavior and the evolution of world poli- tics. In sum, a crisis can erupt, persist, and end with or without violence, let alone war. Perceptions of value threats and stress do not require war. Nor do they vanish with war. Rather, the occurrence of war at any point in the evolution of a crisis intensi es disruptive interaction, along with perceived harm and stress. Since war does not, per se, eliminate or replace crisis, IWCs were inte- grated into the overall set of international crises from late 1918 to the end of 2015 in the ICB Dataset. At the same time, IWCs have one dis- tinctive attribute, a war setting. Of the 476 international crises that then comprised the ICB Dataset, 86 cases (18%) were IWCs. fi fl fl fl fi fl fi fi fl fl fl fl The most elaborate presentation of the dataset in an ICB publica- tion, A Study of Crisis (Brecher and Wilkenfeld 1997, 2000), provided an analysis of international crises from the perspective of seven signi cant contextual attributes of the international system and its member-states: polarity and geography, as fundamental structural characteristics in which international crises unfold; ethnicity and regime type (democracy/non- democracy) as constraints and in uences on decision-making in crisis; the con ict setting (protracted con ict/non-protracted con ict), and extent of violence as criteria by which the international community judges the potential danger a crisis poses for the system as a 2 THEORY I: CORE CONCEPTS AND SYSTEMS 27 of World War I and the rst 15 years of the twenty- rst century. Each of the seven sections concluded with a summary of key ndings pertain- ing to the more than 50 hypotheses examined in A Study of Crisis, along with the signi cance of these empirical ndings for the international sys- tem as it approached the beginning of the twenty- rst century. Severity and Impact In the midst of preparation of the large-scale report on ICB empirical and analytical ndings, A Study of Crisis, a ‘ rst cut’ analysis of two cru- cial ICB concepts, by Brecher and Patrick James, was published in Crisis and Change in World Politics (1986). Its central contribution was to point the way: it was the rst published version of the concepts, Crisis Severity and Crisis Impact, which were elaborated and re ned in later Brecher publications, 1993 and 2008 (to be presented below). systeM and crisis This chapter attempts to overcome a major obstacle to a creative system orientation in international relations—a dearth of knowledge about sys- tem-level change. To accomplish this goal, two tasks are necessary. First, building upon earlier contributions, a new de nition of international sys- tem is offered and its essential properties—structure, process, equilibrium, stability—are presented and discussed. The second requirement is to cre- ate a new approach to crisis and to forge links between its unit and system levels. This, in turn, will facilitate the analysis of crises as catalysts to sys- tem change, that is, serving as international earthquakes. International System In an early critique, Zinnes (1980) argued persuasively that a satisfactory de nition of international system must address two basic questions: (1) ‘how do we know one when we see one’ and (2) ‘what distinguishes one from another’? The rst can be met by a de nition which builds upon earlier writings but restores the balance between structure and process within an integrated set of system components.1 An international system is a set of actors who are situated in a con gu- ration of power (structure), are involved in regular patterns of interac- tion (process), are separated from other units by boundaries set by a given 28 M. BRECHER fi fi fi fi fi fi fi fi fi fi fi fi issue, and are constrained in their behavior from within (context) and from outside the system (environment).2 fi fi fi whole; and third-party intervention as a potential response by the system and its actors. Each of these contextual attributes was examined with data on international political earthquakes spanning the entire twentieth century since the end Structure refers to how the actors in a system stand in relation to each other. Its basic variables are the number of actors and the distribution of power among them, from unipolar through bipolar to multi-power or polycentric. Process designates the interaction patterns among the actors of a system. The basic interaction variables are type, identi ed along a con ict/cooperation dimension, and intensity, indicated by the volume of interaction during a given period of time.3 A link between structure and process is postulated: every structure has a corresponding interaction process, and a structure creates and maintains regular interaction. International systems (and crises) do not require the physical proxim- ity of actors, though this trait is frequently present. Another distinctive property of a system, which serves to demarcate its boundaries, is issue. This concept may be de ned as a speci c shared focus of interest for two or more state actors. There are war–peace issues. K.J. Holsti (1972: 452–455) noted several issues at the base of 77 international con icts and crises from 1919 to 1965: territory; composition of a government; rights or privileges to bases; national honor; unlimited aggrandizement or imperialism; liberation, and uni cation. There are economic and devel- opmental issues. Keohane and Nye (1977, part II) analyzed shing, com- mercial navigation, offshore drilling, and military uses in the issue-area of ocean space and resources, as well as exchange rates, reserve assets, international capital movements, and adjustment, liquidity, and con dence in a regime within the international monetary issue-area. There are also political, cultural, status, and technological issues within broader catego- ries of issue-areas (Potter 1980). The inclusion of subsystems within this de nition enables us to resolve a paradox in the globally oriented concept of international system and thereby to address the other system properties, namely, boundaries, con- text, and environment. The paradox is simple yet fundamental. Every system has boundaries which demarcate members from other units. However, the global international system excludes a priori the possibil- ity of nonmember units and, therefore, of boundaries. It has the addi- tional shortcoming of negating the existence of an environment as a phenomenon distinct from the system itself. That in turn makes impossi- ble a distinction between two kinds of effects on the behavior of actors— contextual, those arising from within a system, and environmental, those from outside. As Young (1968a: 23) observed, a global system can be 2 THEORY I: CORE CONCEPTS AND SYSTEMS 29 characterized only by its context since “there is nothing outside the sys- tem which can be labeled environment.” The concept of environment, he continued, is useful when dealing with subsystems, for these “may be affected by various factors (including other organized entities) located outside its boundaries in spatial terms.” fi fi fl fi fi fi fi fl fi fi There are several usages of the concept of boundaries in international politics. They may be conceived in vertical terms, that is, boundaries in time (Rosecrance 1963, Chap. 11; Haas 1974); as horizontal, that is, in spatial terms (Singer 1971: 12–13); or diagonal, that is, time and space boundaries together (Rosenau 1972: 149). The notion of boundaries presented here is derived from the generic de nition of international sys- tem above. As These two concepts can be combined along two dimensions: extent of similarity and degree of integration. Four types of effects can be speci ed: 1. Similar-Integrative—homogeneity in religion and culture facili- tates negotiation and compromise among actors in a system; 2. Similar-Disintegrative—the presence of ethnic minorities of similar origin in contiguous states increases turmoil and the tendency to hostile behavior; 3. Dissimilar-Integrative—economic and technological heterogeneity among actors leads to increasing interdependence, specialization, and mutual cooperation; 4. Dissimilar-Disintegrative—political regimes with different ideolo- gies induce competition for leadership and spheres of in uence. The de nition of international system presented above enables us to identify a system. Other concepts are needed to distinguish among sys- tems. These are stability and equilibrium, system attributes which have been dealt with extensively in the mainstream of international rela- tions literature. In general, more emphasis has been given to stability. Moreover, its relationship to equilibrium has not been fully developed.4 The argument proposed here is the necessity of restoring equilibrium to a coequal status with stability among the attributes of an international sys- tem, as a precondition to developing the concept of system-level crisis.5 30 M. BRECHER Closely related tasks are de nitions of stability and equilibrium and a speci cation of relationships between them so as to permit us to distin- guish among international systems. fi fi fi fi fi fl fi fl fi fl The concept of change is the key to the distinction between stability and equilibrium, as well as to the organic link between them. Change may be de ned as a shift from, or an alteration of, an existing pattern of interaction between two or more actors in the direction of greater con- ict or cooperation. It is indicated by acts or events which exceed the bounds of normal uctuations or a ‘normal relations range’ (Azar 1972; Azar et al. 1977: 196–197, 207). Following Ashby (1952: 87), four types of change may be distinguished: full function—no nite interval of constancy; part function— nite intervals of change and nite inter- vals of constancy; step function— nite intervals of constancy separated by fi fi such, they make possible the spatial distinction between context and environment. Context and environment incorporate all geo- graphic, political, military, technological, societal, and cultural elements which affect the structure and process of a system, from within and from outside the system, respectively. instantaneous jumps; and null function—no change over the whole period of observation. Change may also occur in the structure of a sys- tem, namely, an increase or decrease in the number of actors and/or a shift in the distribution of power among them. Stability may be de ned as change within explicit bounds. Instability designates change beyond a normal uctuation range. These concepts may be operationalized in terms of the quantity (number) of change(s) in the structure of a system, its process or both, ranging from no changes to many changes. This continuum denotes degrees of stability. The absence of change indicates pure stability, its presence, and some degree of instability. Any system can thus be designated as stable or unstable. Instability in the international system can be illustrated by change in the volume of interaction inherent in such phenomena as wars or crises involving essential actors. The presence of one of these processes may also induce structural change and thereby accentuate system instability. fi fi fi fl fi Equilibrium may be de ned as the steady state of a system, denot- ing change below the threshold of reversibility. Disequilibrium designates change beyond the threshold of reversibility. This meaning is broader than the notion of balance of power, a widely used synonym for equilib- rium in the world politics literature. These concepts may be operational- ized in terms of the quality (signi cance) of change in structure, process or both, ranging from total reversibility to total irreversibility. This con- tinuum denotes degrees of equilibrium. Incremental change indicates a state of equilibrium which has no effect on the system as a whole. Step- level (irreversible) change indicates disequilibrium, which inevitably leads to system transformation, that is, a change in essential actors and/ or the distribution of power among them. The new system, with proper- ties which signi cantly differ from those of its predecessor, denotes a new equilibrium, that is, changes within it which are reversible. These system attributes are presented in Fig. 2.1. Every system has explicit or implicit rules of the game. Many interna- tional systems permit resort to violence as an instrument of crisis manage- ment, its legitimacy deriving from the legal sovereignty of international actors. This is evident in the inherent right of individual and collective self-defense, enshrined in the international institutions of the twentieth- century multi-power system (League of Nations) and bipolar system, and in the late twentieth and early twenty- rst century unipolycentric and renewed multipolar systems (United Nations). Violence which exceeds 32 M. BRECHER the bounds of a normal uctuation range, even when legitimized by the ‘rules of the game,’ constitutes, in our terms, instability, but not disequi- librium, unless this violence challenges the structure of the system. Acute disruptions in an existing structure or process or both may, or may not, lead to disequilibrium. This potential linkage was illuminated by Keohane (1981): “a ‘distortion’ [i.e., instability] per se—an increase in temperature in an air-conditioned room, the rise of a single powerful state in a balance of power system, or a sharp increase in price because of a sudden upsurge in demand—does not suggest that a system is in disequilibrium: rather, it tests that hypothesis by allowing us to see whether adjustments take place. Does the air-conditioning bring the tempera- ture back to the normal level, do coalitions form to counter the power of the rising state, do new sources of supply appear in response to price increases?... Disequilibrium of a system... appears only when the ‘forces tending to restore the balance’ (Arrow’s phrase in a discussion of equi- librium) fail to operate. Air-conditioning that heats a room to 100 °F.; ‘bandwagoning’ that leads to hegemony by a single power; prices that rise sharply and continuously without bringing forth new supply—these are indications of disequilibrium.” There are additional linkages. Four states of a system, along with illus- trations and systemic outcomes, are presented in Table 2.1. In sum, approaches to international systems have been assessed. A revised de nition has been proposed based upon six system compo- nents: actors, structure, process, boundaries, context, and environment. Furthermore, the two basic system attributes, stability and equilibrium, have been rede ned and the links between them speci ed. Thus, the dual task of identifying and differentiating systems has now been completed. The next section will focus on the concept of systemic crisis both within a given system and as a catalyst to system transformation. Systemic Crisis De nitions of systemic crisis, based upon concepts related to interna- tional systems, can be classi ed into two groups: process and combined interaction structure. fi fi fi fi fl fl fi fi fi Process de nitions view systemic crisis as a turning point at which there occurs an unusually intense period of con ictual interactions. According to McClelland (1968: 160– 161), “a crisis is, in some way, a ‘change of state’ in the ow of international political actions...” Elsewhere (1972: 6–7) crisis “interaction is likely to affect the stabil- ity or equilibrium of the system...” Similarly, for Azar (1972: 184), “Interaction above the... upper critical threshold... for more than a very short time implies that a crisis situation has set in.” These de nitions emphasize stages of con ictual behavior among states, different types of activity, the direction and speed of behavioral change, and shifts that indicate changes in the interaction processes. fl fl fi Well-operationalized concepts exist (Azar et al. 1972). And scales facilitate the ranking of various behavioral groups (Azar et al. 1977; Corson 1970; McClelland 1968; Tanter 1966). The shortcomings are analytical. The logic for designating the beginning and end of a crisis was not precisely indicated. Changes in process were not related to structure. There was no attempt to uncover causes and effects of systemic crisis. The result is a group of studies more valuable for their empirical ndings than for understanding the phenomenon of systemic crisis (e.g., Burgess 34 M. BRECHER and Lawton 1972; Eckhardt and Azar 1978; McClelland 1968, 1972; Peterson 1975; Tanter 1974; Wilkenfeld 1972). Combined structural-interaction de nitions view a systemic crisis as a situation characterized by basic change in processes which might affect structural variables of a system. Thus Young (1968c: 15) identi ed “a crisis in international politics [as] a process of interaction occurring at higher levels of perceived intensity than the ordinary ow of events and characterized by... signi cant implications for the stability of some sys- tem or subsystem...” Integrating structure into a process de nition serves as a good analytical starting point by specifying the essential con- ditions and effects of crisis situations. There is, however, little operation- alization of the crucial concept of structure. The result is highly abstract theoretical writings. There was another group, comprising Kaplan, Pruitt, Waltz, and oth- ers, for whom systems were characterized by normal periods of equilib- rium and stability with occasional shifts to disequilibrium and instability. Although such situations are not explicitly termed systemic crises, these transitions are clearly related to the concept of crisis. Except for Kaplan, however, emphasis was placed on the traits of a speci c system, not on changes from one system to another. A problem common to systemic crisis de nitions was the mixture of unit- and systemlevel concepts. For Young (1968c: 10, 14), “crisis con- cerns the probabilities that violence of major proportions will break out,” a point which “explicitly refers to subjective perceptions about the pros- pects of violence rather than to a more objective measure of the prob- ability of violence.” Another striking illustration was Wiener and Kahn’s (1962) 12 generic dimensions of crisis. Among them are system-level indicators such as a turning point in a sequence of events, a new con- guration of international politics as a crisis outcome, and changes in relations among actors. There were also unitlevel indicators: a perceived threat to actor goals; a sense of urgency, stress, and anxiety among deci- sion-makers; increased time pressure; and so forth. In sum, there were several shortcomings in system-level de nitions of crisis: 1. they did not integrate all the key concepts—change in interaction, type of structure, degree of disequilibrium, and instability; 2. they focused clearly on interaction processes but did little to explain their sources and diverse effects on a system; and fl fi fi fi fi fi fi fi fi fi 2 THEORY I: CORE CONCEPTS AND SYSTEMS 35 3. they mixed system concepts with unit-level components such as perception, stress, and values. Moreover, there was little attempt to link de nitions at the two levels of crisis (McCormick 1978; Tanter 1978). In an effort to overcome these weaknesses, a new de nition of inter- national systemic crisis is presented, based upon the system proper- ties discussed in the rst section of this chapter. A systemic crisis may be de ned as a situational change characterized by two necessary and suf - cient conditions: 1. an increase in the intensity of disruptive interactions among system actors and 2. incipient change within the structure of an international system, more precisely, in one or more structural attributes—power distribu- tion, actors/regimes, rules, and alliance con guration. This de nition refers to crises in the military-security issue-area only. Conditions (1) and (2) denote a higher than average increase in inten- sity of con ictual interactions and strain to the structure. By average, we mean normal uctuations as discussed earlier, that is, not beyond the bounds of the ‘steady state’ of the system. Systemic crisis encompasses change. System change need not occur by leaps and jumps, that is, crises; it may result from cumulative events. However, such change is the prod- uct of something other than a crisis. The de nition presented here speci es change in process and structure. It is also linked to stability and equilibrium, for these conditions indicate a shift in the state of a system from stability-equilibrium to instability- equilibrium or stability-disequilibrium or instabilitydisequilibrium, as illustrated in Table 2.1. In schematic terms: few distortions in process or few challenges to a structure denote low instability, whereas many changes indicate high instability; minor distortions (reversible) in process or minor challenges to a structure denote equilibrium, while major changes (irrevers- ible) indicate disequilibrium. Instability, de ned as change beyond a nor- mal uctuation range but within bounds, is present in all systemic crises; disequilibrium, that is, irreversible change, is not. 36 M. BRECHER Berlin Blockade Crisis 19481949 fi fl fl fi fl fi fi fi fi fi fi fi fi The two crisis conditions and the linkages among system properties can be illustrated by the Berlin Blockade Crisis of 1948–1949. Tension between the Western powers and the Soviet Union centered on the issue of occupied Germany. The 1945 Potsdam Agreement had divided Germany into four zones of occupation, by France, the UK, the USA, and the USSR, but had provided that they were to be treated as one economic unit under the Allied Control Council. On June 7, 1948, the three Western powers published the recommendations of the March 1948 London Conference (to which the Soviet Union had not been invited), calling for a merger of their zones in Germany. This con ict- ual-type act broke an existing, though fragile, East–West consensus on Germany and set in motion several changes in rapid succession. The Soviet Union responded on June 24 by blocking all Western transpor- tation by land into and out of Berlin. President Truman countered on June 26 with an order to step up the US airlift into Berlin, which had begun 2 months earlier, and continued with plans for the rehabilitation of Germany as part of Western Europe. Talks between the crisis actors began on August 2, 1948. An informal consensus on the future of Germany was reached by the four powers on March 21, 1949. An agreement was signed on 12 May formalizing the partition of Germany into two quasiindependent states, the Federal Republic of Germany [FRG, West Germany] and the German Democratic Republic [GDR, East Germany]. These events indicated an accommodation by the system, the May 12, 1949 event marking the end of the Berlin Blockade Crisis. fl fl In systemic crises, changes vary in quality, as well as in quantity: they are reversible in some cases, irreversible in others. Thus a sharp increase in con ictual interactions between the Western powers and the USSR clearly indicated system instability between June 7, 1948 and May 12, 1949. The Berlin crisis also affected the East–West equilibrium. Distortions were step-level in nature; that is, neither the interaction pat- tern nor the structure of the dominant system in world politics at the time was the same before and after the crisis. The agreement of May 12, 1949 illustrates this point. It left Germany divided, creating the founda- tion of two new international actors, the Federal Republic of Germany (FRG, West Germany) and the German Democratic Republic (GDR, East Germany), and tightened the polarization between the superpowers. Furthermore, the interaction pattern between the Western powers and the Soviet Union after the agreement on Berlin came into effect differed substantially from that during the occupation of Germany by the four powers. The system during the Berlin Blockade crisis was in a state of high instability leading to disequilibrium. As such, it helped to catalyze the transformation of the transitional international system of embryonic bipolarity (1945–1948) to tight bipolarity. The threshold events between phases of the Berlin Blockade Crisis, as well as the overall links between crisis conditions and the system attrib- utes of equilibrium and stability, are summarized in Table 2.2. India/Pakistan Crisis Over Kashmir 19651966 A similar analysis will now be undertaken for an international crisis at the subsystem level, the India/Pakistan struggle over Kutch and Kashmir in 1965–1966. A South Asian regional system had emerged in 1947 with the transfer of power from the United Kingdom to India and Pakistan. For almost a quarter of a century, until the sundering of Pakistan in the crisis leading to the creation of Bangladesh in 1971, India and Pakistan were the relatively equal major powers in the South Asian system, with 38 M. BRECHER several small or very small powers on the geographic periphery of the sub-continent, Ceylon (Sri Lanka) from 1948, Afghanistan from 1949, Nepal since 1950, and Bangladesh. The normal pattern of interaction between India and Pakistan was characterized by mistrust and verbal hostility, with periodic disruptions of an intensity suf cient to mark international crises, as that over the post-partition territorial issues of Junagadh, Kashmir, and Hyderabad (1947–1949) and the Punjab war scare (1951). There were also longstanding con icts over diverse issues like refugee compensation and repa- triation, and the division of river water in the Indus Valley. Among them was the princely state of Kutch. Its ruler had acceded to the Indian Union in 1947, but Pakistan claimed that the northern section of the Rann of Kutch was part of its Sind province. Incidents occurred in 1956, but Indian control over the disputed territory was quickly restored. fi fi fl The India–Pakistan systemic crisis over Kutch and Kashmir began in April 1965 and ended in January 1966. The initial breakpoint occurred on April 8, when India launched an attack on the disputed Kutch bor- der. Pakistan responded with a counter-attack the same day. Much higher-than-normal hostile interaction continued until the end of June 1965. Pakistani forces initially repelled local Indian troops. In response, on April 26, India placed its armed forces on alert, thereby escalating the crisis. A British call for a cease re and negotiations was accepted in principle on 11 May, but hostilities continued until June 30 when both parties agreed to all the terms of a UK-mediated package—mutual with- drawal of forces, direct negotiations, and arbitration if these failed to set- tle the dispute. High instability characterized the subsystem during those months, but its basic equilibrium remained unchanged. Third-party intervention led to partial accommodation of the South Asian subsystem. A second phase of this systemic crisis began in August 1965 and lasted until January 1966. The breakpoint occurred on August 5 when Pakistan-supported guerrillas in ltrated into the Indian-held part of the former princely State, Jammu and Kashmir, in an attempt to spark a large-scale uprising against India’s rule. The overall distribution of power between India and Pakistan was at stake, making the challenge to the structure of the regional system much greater than in the April–June phase over the Rann of Kutch. India responded on August 25 by sending several thousand troops across the 1949 Kashmir cease re line, capturing most areas through which the in ltrators came. The crisis escalated fur- ther on September 1, when Pakistan sent an armored column across the 2 THEORY I: CORE CONCEPTS AND SYSTEMS 39 cease re line in southern Kashmir threatening the vital road linking the Kashmir capital, Srinagar, with the plains of India. This led to a further escalation, India’s invasion of West Pakistan on September 5. The sharp increase in the volume of disruptive interaction indicated greater system instability. This was accentuated by China’s denuncia- tion of India’s ‘aggression’ against Pakistan and its ‘provocation’ on the Sikkim–Tibet border. Moreover, Peking (later, Beijing) issued an ultima- tum to Delhi to dismantle all border military forti cations and to stop all alleged intrusions into Tibet. While rejecting China’s demands on the 17th, India hinted at a willingness to make minor concessions. The next day Chinese troop movements were reported to be within 500 m of Indian border positions. However, on September 21, China withdrew its ultimatum, announcing that India had complied with Peking’s demands. This moderate decrease in con ictual interaction denoted further partial accommodation at the systemic level; change had not risen above the threshold of irreversibility. fi fi fi fi fi fi fl fi fi fi The threat of direct Chinese military involvement in a South Asian crisis generated mediation efforts by the superpowers through the Security Council. A cease re resolution in mid-September, which also provided for a UN observer group in Kashmir, was accepted by India and Pakistan. This did not, however, indicate an exit point in the system-level crisis, for both armies continued to occupy each other’s terri- tory, a situation which was soon followed by violations of their cease re agreement. Another paci c strand of third-party intervention began on September 17 when Soviet Prime Minister Kosygin offered to convene a conference in Tashkent between President Ayub Khan of Pakistan and Indian Prime Minister Shastri. The conference was held between January 4 and 10, 1966. It ended with a declaration af rming the intentions of both parties to restore diplomatic and economic relations following the withdrawal of their troops from all occupied territory, as well as the repa- triation of prisoners of war. Thus, January 10, 1966 marked the end of the crisis and a successful accommodation by the South Asian system. The challenge to its structure had been overcome, the pre-crisis equi- librium had been restored, and instability had reverted to its long-term norm of passive distrust. As with the Berlin Blockade Crisis of 1948–1949, the links between crisis conditions and the system attributes of equilibrium and stability in the 1965–1966 India–Pakistan crisis are presented schematically in Table 2.3. At the outset of this chapter, two questions were raised regarding inter- national systems: how do we know one when we see one; and what dis- tinguishes one from another? The same questions can be posed about international crises. We have already indicated how to recognize a cri- sis. It remains to explain how to distinguish one crisis from another. For this exercise, two additional concepts, severity and impact (importance), must be introduced. Severity is a composite indicator of crisis attributes from the beginning to the end of an international crisis. It refers to the volume of con ictual interactions among the crisis actors and thus denotes the extent of sys- tem instability during a crisis. Impact (Importance) is a composite indicator of crisis attributes after the conclusion of an international crisis. It refers to the quality of struc- tural change or irreversibility and, as such, indicates the effects of a crisis on the equilibrium of a system. fl 2 THEORY I: CORE CONCEPTS AND SYSTEMS 41 Severity can be operationalized by six indicators. One is the number of crisis actors: the larger the number, the more disruptive will be hos- tile interactions, the greater the likelihood of superpower or major power involvement, and the more dif cult the system’s accommodation, all pointing to greater severity. Another indicator is the geostrategic salience of the location of an international crisis in terms of its natural resources and distance from major power centers. An underlying assumption is that the broader the geostrategic salience, the more severe will be the crisis. Salience ranges from a single regional subsystem (e.g., Afghanistan– Pakistan crisis over Pathanistan, 1955) to the global system (Cuban mis- siles, 1962). A third indicator is the extent of heterogeneity among crisis adversaries, measured by the number of attribute differences in terms of military capability, political regime, economic development, and culture (maximal heterogeneity—Mayaguez, 1975, between Cambodia and the United States). Here, too, the operative assumption is that the greater the heterogeneity among adversaries, the more severe the crisis. A fourth indicator of Severity is the extent of superpower involvement in an international crisis, ranging from situations in which both the USA and the USSR are crisis actors to a crisis in which neither was involved in any form. In general, the greater the involvement by superpowers, the greater the challenge to the structure of a system and, therefore, the more severe the international crisis. A fth indicator of severity is issues. Crises may focus on one or more issues within one or more issue- areas—military-security, politicaldiplomatic, economic-development, and cultural-status. The rst issue-area creates the most severity. Moreover, the larger the number of issues, the more severe the crisis is likely to be. Finally, severity is indicated by the extent of violence in a crisis, rang- ing from full-scale war, through serious clashes short of war, to minor clashes, to no violence.6 The impact (importance) of an international crisis can be operational- ized by four indicators. One is actor change as a consequence of a crisis. This ranges from the creation or elimination of one or more actors (e.g., Bangladesh, 1971; South Vietnam, 1975), through a change in regime type (e.g., Czechoslovakia, 1948, democracy to communism), to a change in regime orientation (e.g., Guatemala, 1954, pro-Soviet to proUSA), to no change in actors or their regimes. Another indicator is the extent of alliance change owing from an international crisis, the most important being the formation or termination of an alliance (China Civil War, 1948–1949, and the PRC-USSR alliance, 1950), followed 42 M. BRECHER by the entry or exit of one or more actors into or from a formal or informal alliance (Greece–Turkey–Truman Doctrine, 1946–1947), an increase or decrease in cohesiveness in an existing alliance (Prague Spring, 1968) to no change in alliances. fi fi fi fl Power change is a third indicator of crisis importance, extending from the entry or exit of an actor into or from the ranks of the most powerful states in a system (Japan’s atomic bomb crisis, 1945), through a change in rank among the most powerful members of a Two international crises—one at the dominant system level (Berlin Blockade, 1948– 1949), the other at the subsystem level (Kashmir, 1965– 1966)—were examined in terms of several core concepts, system, stability, equilibrium, and crisis. These same cases will now be evaluated in terms of severity and importance. The Berlin Blockade crisis of 1948–1949 was the rst major direct con- frontation between the two superpowers, though both had been adversar- ies in the 1945–1946 Iran Hegemony crisis. There were four crisis actors in the rst Berlin crisis, the USA, USSR, UK, and France. Its geostrategic salience, as with all Berlin crises after 1945, was high, for it impinged on the balance of power in the dominant East–West system, as well as on the distribution of in uence in the East Europe and West Europe sub- systems. Among the adversaries, near-maximal heterogeneity is evident between France (or the UK) and the Soviet Union: while the former had a democratic political regime, the USSR had a civil authoritarian system of government; they were major military powers, it was a superpower; and cultural differences between Paris (or London) and Moscow were fundamental. As for superpower involvement, the Berlin case was at the apex of severity for, as noted, both the USA and the USSR were intensely hostile crisis actors. There were several issues at stake, including territory, hegemony, security, and status. Only with respect to the violence indicator did the Berlin case rank low: there was none. Taken together, however, its composite overall severity places the Berlin Blockade among the most severe international crises since the end of the Second World War. The impact (importance) of this crisis was no less grave. The 1948–1949 Berlin case marks the rst great divide in East–West relations. One of its 2 THEORY I: CORE CONCEPTS AND SYSTEMS 43 fi fi fi fi fl structural consequences was the crystallization of basic changes then in motion, leading to the formation of two new German states, the FRG and GDR, on the ashes of the old. Another was the change from embryonic bipo- larity in the post-World War II dominant system to tight bipolarity. As for alliance con guration, the Berlin Blockade hastened the formalization of NATO (1949) and moved the Communist states of East Europe towards the Warsaw Pact (1955). The Berlin Blockade outcome did not result in a change in the composition of the most powerful states in the dominant system or in their relative rank, but the USSR failed to achieve its objective, while the western powers did so. Berlin was more consequential, however, in changing the rules of the game: the blockade and direct confrontation indicated the end of the Potsdam phase in East–West relations; overt conictual interaction became the norm thereafter. Thus the overall impor- tance of the 1948– 1949 Berlin crisis, like its severity, was very high. fi fl system, to a change in rela- tive power, but not in power rank, among the adversaries, to no change. Finally, the importance of a crisis is indicated by the extent of change in rules of the game. There may be new rules, codi ed or tacit (Prague Spring, 1968 and the Brezhnev Doctrine), an increase or decrease in actor consensus about existing rules, or no change in rules.7 In the South Asian crisis of 1965–1966 there were, as noted, three cri- sis actors, India, Pakistan and, for a very brief period, the PRC, along with three highly involved actors, the USA, USSR, and UK; their involvement, however, was con ned to the political realm. Geostrategic salience was at the bare minimum, for the location of the crisis over Kutch and Kashmir had no relevance to any subsystem other than South Asia, let alone the dominant international system. There was limited heterogeneity between the principal adversaries, namely, in political regimes (India’s western- type democracy versus Pakistan’s military rule) and in culture (Hinduism versus Islam). There was no superpower confrontation, direct or indi- rect, only political involvement. Both military and political issues were at stake—territory and hegemony. As for violence, there was a full-scale war between India and Pakistan in September 1965. Taken together, the overall severity of the 1965–1966 India–Pakistan crisis was low. In terms of impact, this crisis ranks very low. There was no meaning- ful change in power distribution, neither in the narrow sense of the crisis outcome, which was a political compromise, nor in the rank of the two major South Asian powers, India and Pakistan. Unlike their subsequent crisis over Bangladesh (1971), there was no change in actors nor in the type or orientation of their regimes. Only the existing alliance pattern changed, with Pakistan moving from an unquali ed pro-western posture, formalized through its membership in SEATO and CENTO, to a more even-handed attitude toward the superpowers and an improvement in its relations with the USSR following the Tashkent Agreement. There was no change in the rules of the game within the South Asian subsystem: 44 M. BRECHER both in war and diplomacy, the crisis actors adhered to established rules of behavior. The impact of the 1965–1966 international crisis, that is, its overall importance, was minimal. Thus far this analysis has focused exclusively on the systems level. The next section will address the level-of-analysis problem with respect to cri- sis, that is, the crucial dimension of system change. UnitSystem Linkages In all branches of knowledge there are several levels of analysis, each with distinct concepts, research questions, and methodologies. Every level is capable of illuminating a segment of knowledge within a discipline but no more. To provide insights into a part of any whole is admirable. However, the ultimate challenge is to link the ndings at all levels into an aggregate of the whole and its parts in order to comprehend as much as possible of the total universe of knowledge in any eld.8 fi fi fi fi This perspective derives from a conviction that the competitive focus on a single level of analysis is counter-productive. To examine the two levels—unit and system—would enable us to move beyond the position of blind men attempting to grasp the elephant. In the words of Robert North (1967: 394): “As research scholars and would-be theorists in international relations we might all derive at least three useful lessons from the old fable about the blind men and the elephant. The rst is that the elephant [crisis] presumably existed; the second is that each of the groping investigators [at the unit and system levels], despite sensory and conceptual limitations, had his ngers on a part of reality; and the third is that if they had quieted the uproar and begun making comparisons, the blind men might—all of them—have moved considerably closer to the truth.” It is in this spirit that we now approach the task of linking the unit (micro) and system (macro) levels of crisis analysis. Since the early 1960s, there has been a large body of research on state behavior in international crisis, the counterpart to studies of con ictual interactions among adversary states (Hopple and Rossa 1981; Holsti 1980; Tanter 1978). They differ in de nitions, conceptual frameworks, and techniques of analysis, as they must. This chapter emphasizes points of convergence while maintaining a clear-cut distinction between the two levels and their diverse effects. A unit-level, foreign policy crisis derives from perceptions, whereas a systemic crisis is objective. Stated differently, the focus of the former is image and action by a state’s decision-maker(s), while that of the latter is reality and interaction. There is no one-to-one relationship between unit and systemic crises: the former occurs for a single state; the latter is predicated upon the existence of distortion in the pattern of interaction between– among two or more adversaries in an interstate system. A de nition of systemic crisis has been presented early in this book. From the perspective of a single state, a foreign policy crisis is a situation with three necessary and suf cient conditions, deriving from a change in its external or internal environment. All three are perceptions held by the highest-level decision-makers of: a threat to basic values, along with the awareness of nite time for response to the external value threat, and a high probability of involvement in mili- tary hostilities9 At the unit level, there are crisis actors, that is, states whose decision- makers perceive the conditions of crisis. There are parallel concepts at the system level, as presented in Table 2.4. fi fi fl fl fi fi fi fi For the threat component, the counterparts are basic values of deci- sion-makers and structure of the system. Basic values, such as existence, in uence in the global and/or regional systems, territorial integrity, eco- nomic welfare, and others are the elements which guide goals, decisions, and actions of states. Similarly, at the system level, structure provides the setting for continuity in interaction processes. Threat at the unit level indicates (subjective) perceptions by decision-makers. Challenge at the system level means an (objective) possibility of change in the structure. A challenge to the system structure may or may not materialize, just as a threat to basic values and an increase in war likelihood may or may not be realized. 46 M. BRECHER In the 1948–1949 Berlin Crisis, the threat to Soviet and USA in uence in Germany and, more generally, to the international system generated a sharp increase in con ictual interaction. This distortion, the counterpart of an increase in perceived likelihood of military hostilities, posed a chal- lenge to the existing structure of the system, namely, to the number of actors (two or more Germanys) and the tighter polarization around the superpowers as a result of the crisis. An international crisis may thus be addressed in macro-level and micro-level terms. While the former deals with a system as a whole, the latter focuses on each state crisis actor. There are situational changes in which only one state perceives a crisis for itself, that is, actions by one (or more) state(s) which trigger perceptions of threat, time pressure, and war likelihood for a single actor (e.g., the massing of Indian demonstra- tors on India’s border with Goa in 1955, creating a crisis for Portugal). In other instances, two or more states experience a crisis over the same issue, as with the Western Powers and the USSR over Berlin in 1948– 1949, 1958–1959, and 1961. The link between unit- and system-level concepts of interstate crisis may be illustrated by two different cases: when a crisis for all state actors is identical in time; and when their crises overlap but are not identical in time. Establishing this link requires the clari cation of static and dynamic concepts at both levels. The former is trigger/termination at the unit level and breakpoint/exit-point at the system level. The latter is escalation/ de-escalation and distortion/accommodation, respectively. These concepts are presented in Table 2.5. fi fl fl fi At the unit level, a trigger, a static act, is de ned as the catalyst to a for- eign policy crisis. In the 1948–1949 Berlin Blockade crisis, the trigger to the Soviet Union’s foreign policy crisis was, as noted, the publication by the Western Powers on June 7, 1948 of the recommendations of their March 1948 London Conference. The trigger for the United States, Britain, and France was the Soviet decision on June 24 to block all Western transportation, by land and sea, into and out of Berlin. In terms of a dynamic process, a trigger denotes an escalation in perceived threat, time pressure, and the like- lihood of military hostilities. The termination of a crisis at the unit level, that is, a foreign policy crisis is the point in time when decision-makers’ perceptions of threat, time pressure, and war likelihood decline to the level existing prior to the crisis trigger. In the Berlin Blockade case, the termination date for each of the four powers was May 12, 1949, when an agreement regarding West and East Germany as separate entities was signed. Thus the triggers did not coincide but the termination dates for the various actors did. In dynamic process terms, termination for crisis actors marks the nal de-esca- lation in perceived threat, time pressure, and war likelihood during a crisis. At the system level, parallel notions exist—breakpoint and exit-point as counterparts of trigger and termination. A breakpoint is a disturbance to the system created by the entry of an actor into a crisis. A systemic cri- sis erupts with an initial breakpoint event, such as the Western powers’ challenge to Moscow on June 7, 1948 regarding the integration of their zones of occupation in Germany. In dynamic terms, this change denoted distortion in the pattern of East–West interaction. Similarly, an exit-point refers to a signi cant reduction in con ictual activity, such as the formal agreement among the four powers on May 12, 1949 about the future of Germany and the lifting of the Soviet Union blockade. This change indicated accommodation, that is, a shift to a less intense level of hostile interaction than that during the systemic crisis. The duration of a system-level crisis is measured from the rst break- point to the last exit-point which, in unit-level terms, means from the trigger for the rst crisis actor to the termination by the last crisis actor. For the initial breakpoint to occur, there must be two or more adver- sarial state actors in higher-than-normal con ictual interaction. They may both or all be crisis actors simultaneously; a rare occurrence for this requires triggers the same day, as in the 1965–1966 India–Pakistan cri- sis over Kutch–Kashmir. More often, they comprise one crisis actor and one adversary who triggers the crisis; the latter may later become a crisis actor, as with Belgium and the Congo in the 1960 Congo Crisis,10 or it may not. A variant is one initial crisis actor and one adversary, with the latter joined by another in the process of becoming crisis actors, as with the USA and the USSR-cumCuba in the 1962 Missile Crisis.11 Another variation is one crisis actor at the outset with several adversaries who later 48 M. BRECHER become crisis actors simultaneously, as with the USSR and the USA– UK–France in the 1948–1949 Berlin Crisis. As for the winding down of a system-level crisis, the majority of cases reveal a simultaneous termina- tion for all crisis actors and, therefore, simultaneous accommodation by the system, as in the Berlin and India–Pakistan cases noted above. fi fi fl fi fi fl Distortion may be gradual or rapid; so too with accommodation. In general, system-level interstate crises are characterized by multiple break- points, that is, gradual distortion and, by contrast, few exit-points, that is, rapid accommodation. The reason is that the onset of a systemic cri- sis is usually a process in which crisis actors cumulatively challenge one another. The result is that breakpoints tend to differ in time and, there- fore, distortion is gradual. Accommodation, however, usually requires agreement, either formal or tacit. Thus exit-points tend to coincide in time. However, as long as any crisis actor has not terminated its foreign policy crisis, accommodation has not yet been completed: termination of the unit-level crisis for the last participant and the end of the system-level crisis are identical in time. Breakpoints and exit-points also indicate the entry and departure of actors in a systemlevel crisis. Each breakpoint denotes an increase in con ictual interaction relative to the pre-crisis phase, whereas exit-points signal accommodation at the system level. Linking unit upward to sys- tem, the effects of trigger/termination on breakpoints/exit-points are immediate and direct; that is, a trigger at the unit level always denotes a breakpoint at the system level and thus a further distortion in systemic interaction. In the Berlin Blockade case, both June 7 and June 24, 1948, which were triggers at the unit level for the Soviet Union and the three Western powers, respectively, were also immediate breakpoints in the system-level crisis. However, when systemic crisis is linked downward to actors, the effects of exit-points on de-escalation are immediate and direct for some but may be delayed and indirect for others. Stated differ- ently, not all system-level changes affect all units at once and equally in a readily identi able way. The Berlin Blockade Crisis provides an example of direct and immediate effects: the last system level exit-point, on May 12, 1949, denotes nal de-escalation for the four powers simultaneously. In general, systemic crises have more signi cant effects than unit-level crises because they pose a dual danger, namely, to the structure of the system and to its actors, whereas unit-level crises affect actors only. In sum, a system-level crisis requires behavioral change on the part of at least two adversarial actors leading to more intense con ictual interaction. 2 THEORY I: CORE CONCEPTS AND SYSTEMS 49 Although a crisis is catalyzed by behavioral actions, these actions, the trigger to a unitlevel crisis, can always be traced to their perceptual ori- gin. Here lies the organic link between the two levels of crisis. The concepts and de nitions elaborated above have several possible uses in IR, IS, and WP research. Empirical data on system-level crises can be collected, classi ed, compared, and measured. Types of systemic crises can be described and can then serve as indicators of crisis anticipation. Sources of system-level crisis can be uncovered and rank-ordered. Factors such as decision-making process, type of regime, power distribution in the dominant system or subsystem, and other state-oriented or system- derived attributes can be examined in order to explain diversity in the emergence, type, and outcome of systemlevel crises. Finally, conceptual clarity on system and crisis paves the way for the analysis of crises as inter- national earthquakes, that is, as catalysts to system change. fi fl fl fi fi fi fi notes 1. The major attempts to integrate system concepts into international rela- tions theory focused on the great powers in world politics. Moreover, they meant by international system either the global system or, more often, the dominant system, a synonym for Singer and Small’s (1972: 381) “Central Sub System,” that is, “the most powerful, industrial- ized, and diplomatically active members of the interstate system, gen- erally coinciding with the ‘European state system’.” Kaplan (1957: 4, 9) referred to a “system of action” as a set of ve interrelated variables whose relationship is characterized by behavioral regularities—essential rules, transformation rules, actor, capability and information variables— but he did not explicitly de ne an international system. For Hoffmann (1961: 207), the concept of international system is blurred by its all- inclusive nature; it incorporates the structure of the world, the nature of the forces which operate across or within the major units, capabilities, pattern of power, and political culture of the units. Rosecrance (1963: 5, 6) acknowledged the importance of international systems and treated historical systems at length but distinguished among them mainly by “signi cant changes in diplomatic style.” Aron (1966: 94, 95) appears to restrict the term, international system, to an ‘ensemble’ of political units capable of being implicated in a generalized war. E. Haas (1964: 62–63) noted the need for “de nitional clarity, verbal and operational,” among key system properties—inputs, outputs, units, environment, attributes, structures and functions—but the links were not developed. McClelland (1966: 20) distinguished between boundaries and environment but 50 M. BRECHER fi fi fi In the literature on systems—though not on international systems—pro- cess is also used to denote growth and decay, concepts which are closely linked to system transformation. The latter, though not the central focus of this chapter, will be discussed in relation to stability and equilibrium. fi 3. fi Conceptually, an international system ranges across a broad spectrum, from the global system through the dominant system to subsystems. There are two strands in the subsystems literature: geography and issue. On the rst see Binder (1958), Modelski (1961), Brecher (1963), Hoffmann (1963), Russett (1967), Zartman (1967), Bowman (1968), Kaiser (1968), Cantori and Spiegel (1970), M. Haas (1970), Dominguez (1971). Among the most careful in using a geographic criterion is M. Haas (1974: 336–356), whose empirical analysis of 21 subsystems com- bined geographic and issue criteria, providing a rare link to the second strand in the subsystems literature. On issue subsystems see Hanrieder (1965), Russett (1967), Zimmerman (1972), K.J. Holsti (1972), M. Haas (1974), Dean and Vasquez (1976), Keohane and Nye (1977), and Lampert (1980), who was the most direct in asserting the primacy of issue over geography as the basic component of subsystems. For an overview of the international subsystems literature see Thompson (1973). A later variation on the systems theme is the literature on international regimes (e.g., Krasner 1982). A regime, in the largest sense, may be termed an issue subsystem and, in narrower terms, the rules of the game within such a system. fi 2. fi fi con ned the meaning of system to interaction. Young (1968a: 6) speci- ed four essential components of a system: actors, structure, process and contextual limitations, but his distinction between structure and process is blurred. Keohane and Nye (1977: 20–21) clari ed this dis- tinction by identifying the former with “the distribution of capabilities among similar units” and the latter with “bargaining behavior within a power structure.” Waltz (1979: 40), too, asserted the need for a clear-cut demarcation of structure and interaction but, like McClelland with pro- cess, he overemphasized structure. 4. Kaplan (1957: 21, 35–36) designated his “six distinct international sys- tems” as “six states of equilibrium of an ultrastable international system”; that is, equilibrium is synonymous with system. Equilibrium is the nor- mal state of a system; and his concern was with “the expectations for sta- bility of each of the systems.” The concept of “ultrastable system” was developed by Ashby (1952: 100–122). The rst wave of analysts in the on-going debate over the relationship between systemic polarity and sys- temic stability (Waltz 1964; Deutsch and Singer 1964; Rosecrance 1966; Young 1968b) virtually omitted discussion of the concept of equilib- rium. Hoffmann (1961: 208) distinguished between two types of system, 2 THEORY I: CORE CONCEPTS AND SYSTEMS 51 ‘stable’ and ‘revolutionary,’ but he made no reference to equilibrium. Aron (1966: 100–101) barely mentioned stability and instability; and while he had an extensive discussion on equilibrium, it was treated as a policy, not a concept. For Rosecrance (1963: 220–221), “a system aim- ing at stability” comprises four elements: “a source of disturbance or dis- ruption (an input)”; a regulator; a list of environmental constraints; and outcomes. While emphasizing interactions in his analysis of nine historical systems from 1740 to 1960, he made only a passing reference to equilib- rium. Young (1968a: 42) was precise in de ning stability both statically and dynamically: “In static terms, stability refers to the continuance of the essential variables of an international system (i.e., actors, structures, pro- cesses, and context) within the bounds of recognizability over time. In dynamic terms, on the other hand, stability can be thought of as the ten- dency of a system to move in the direction of equilibrium following dis- turbances.” What is missing is the content of equilibrium. Waltz (1979: 161–162), too, was clear on stability, to which he related structure (1967: 229, fn. 18): “By ‘structure’ I mean the pattern according to which power is distributed; by ‘stability,’ the perpetuation of that structure with- out the occurrence of grossly destructive violence.” Thus a change in structure means system transformation and a new stability. Just as Kaplan equated system with equilibrium, so Waltz equated system with stability. Several international relations scholars did focus on equilibrium. In this they share the emphasis of general systems theorists and economists who identify stability and instability as “states of equilibrium” (Arrow 1968: 384, 387). Richardson’s conception of stability “referred simply to any set of conditions under which the system would return to its equilib- rium state...” (Deutsch and Singer 1964: 391). Liska (1957: 13) relied “mainly on the ideas of progressive, stable, and unstable equilibrium.” Pruitt (1969: 20, 23– 24, 36–37) addressed the relationship of these con- cepts rigorously: “Instability is de ned as the likelihood of sudden (basic) change and stability is de ned as the opposite of instability.” Moreover, “Stable relations are usually characterized by oscillations around an equi- librium point...” However, Pruitt was less clear on the meaning of change and equilibrium. On stability see also Gilpin (1981: 50–105). 5. Michael Haas’ treatise on international con ict (1974), for example, has a 23-page appendix on “De nitions of Concepts,” in which equilibrium is conspicuously absent. By contrast, Gilpin (1981: 156–185) devotes con- siderable attention to this core concept. 6. The rationale for these indicators of Severity, the scales for each, and their relative weight in the overall severity of systemic crises are elaborated in Brecher and Wilkenfeld (1988: 119– 141). 52 fi fi fi fl fi fi M. BRECHER The rationale for the indicators of the Importance-Impact of international crises, along with the crisis impact model, and the hypotheses and nd- ings on Impact are presented in Brecher (1993: 290–298, 318–334). 8. Among the pioneers of systems theory in the social sciences, Boulding (1956: 202, 201) introduced the idea of system rungs or levels. McClelland (1955: 34; 1958) was perhaps the rst to specify levels in the study of world politics. Deutsch (1974: 152–156) set out a 10level political system, including four levels in international politics. The ‘level- of-analysis problem’ was rst given explicit formulation by Singer (1961, also 1971). See also Andriole (1978). 9. A crisis de ned here refers to the war-peace issue-area. However, break- points may occur in any foreign policy issue, and the study of interna- tional political, economic, and status crises might yield no less valuable ndings. For these types, an appropriate change is necessary in the second condition speci ed above. 10. The crisis trigger for Belgium, on July 5, 1960, was a mutiny among sol- diers of the Congolese Force Publique, which rapidly turned into a gen- eral movement against Belgian and other European residents. Belgium responded on the 8th by announcing its intention to send military rein- forcements to the Congo. A crisis was triggered for the Congo two days later when Belgian troops went into action. 11. The Missile crisis for the United States was triggered on October 16, 1962 when photographic evidence of the presence of Soviet missiles in Cuba was presented to President Kennedy. The US major response, on 22 October, was a decision to blockade all offensive military equipment on route to Cuba. This, in turn, triggered crises for the Soviet Union and Cuba. fi fi fi fi 12. Brecher and Ben-Yehuda, “System and Crisis in International Politics” (1984). fi fi 7.