Shlomo Mizrahi, Abraham Mehrez, Arye Naor




        This paper suggests a two-level game analysis of Israel’s strategy toward peace during the 1990s. The paper shows how various paradoxes in Israeli society create domestic obstacles and internal opposition that weaken Israel’s bargaining position toward neighboring countries. Treating domestic parameters in these countries as a given, we argue that Israeli leaders can hardly use this weakness to manipulate information in the bargaining process, because neighboring countries can observe Israel’s internal processes. Therefore, attempts by Israeli leaders to create the impression that they are willing to adopt a conflictual approach towards neighboring polities, especially the Palestinians, without actually creating the necessary internal conditions for such a policy, may finally lead to a sub-optimal equilibrium for Israel – in terms of territory and deterrent ability – since it will have to compromise under difficult conditions. Several practical implications as to the preferred bargaining process under these conditions follow. 


During his service as Secretary of State in the 1970s, Henry Kissinger once stated that Israel did not have a foreign policy, it had only domestic policy. By that he meant that Israel’s foreign policy is primarily the result of internal conditions and constraints. Yet since this claim was made, students of Israel’s foreign and strategic policy, especially within the discipline of international relations, have not dealt with this issue systematically. Most studies focus on the balance of power between Israel and the neighboring countries as measured by armed forces, territory and security budget (Karsh and Mahler, 1994). Kissinger’s observation seems more correct than ever in the 1990s, however, in light of events since the signing of the Oslo Agreement.

        This observation was reaffirmed recently, in May 2000, when Israel unilaterally withdrew from Lebanon. The withdrawal followed strong internal pressures that were interpreted by many as a significant decline in the willingness of the Israeli public to pay the price of conflict. Indeed, Sheik Nassralla, the leader of the Southern Lebanese guerrilla organization Hizbullah, called on the Palestinians to observe how weak Israel had become despite its military strength.

        This paper analyzes the impact of internal socio-political and economic processes in Israel on its strategic position in the Middle East, and especially its peace strategy during the 1990s. We assume that Israel is facing a bloc of hostile countries composed of players such as Iran, Iraq, Syria, Libya, and parts of Palestinian society. The interests of these players may differ in certain aspects, and they do not necessarily coordinate their strategy towards Israel. We assume, however, that their basic calculations concerning Israel are similar, and therefore concentrate on domestic variables that influence Israel’s strategic choices while treating domestic parameters in the hostile bloc as a given.

        The paper uses the general concept of “nested games,” where players’ interests and actions in one game are influenced by their involvement in others (Tsebelis, 1990; Colomer, 1995). Specifically, the paper applies the idea of a two-level game, as developed for analyzing international relations and foreign policy (Schelling, 1960; Putnam, 1988; Iida, 1993; Evans, Jacobson and Putnam, 1993; Schneider and Cederman, 1994; Mo, 1995; Schultz, 1998). Two-level game literature has introduced solid micro foundations to the theory of international bargaining. Most importantly, this research tradition has shown that the amount of uncertainty in the international system is not a given but can be manipulated both for the better and for the worse. This ambiguous potential is the essence of a two-level dilemma in world politics in which domestic politics affects international behavior both positively and negatively, and vice versa – clearly, international conditions also affect domestic politics, which again affects foreign policy. To avoid a cyclical argument, we concentrate on explaining foreign policy based on internal conditions. The opposite direction of this mutual dependency between foreign and domestic policy will be discussed only when necessary.

        A major debate in the literature is whether domestic obstacles weaken or strengthen the state’s bargaining position in international negotiations. Putnam (1988) has shown how negotiators might claim successfully that domestic opposition prevents them from concurring in an international agreement. Iida (1993) questions this argument and relies on sequential bargaining to analyze how domestic constraints impact the negotiations between two states, given various assumptions regarding information. One basic result is that a country’s bargaining leverage does not necessarily increase when its domestic constraints become more severe. When there is complete information about domestic constraints, the constrained negotiator has a bargaining advantage only if the constraints are severe. When there is asymmetric international information, the constrained negotiator will benefit only if the foreign negotiator strongly believes that the home negotiator is severely constrained. Finally, Iida (1993) shows that when there is incomplete domestic information (on the side of the home negotiator), the constrained negotiator has a bargaining advantage if the probability of successful ratification increases with the share that this side receives. This opens up the possibility for misinterpretation, which may lead to efficiency loss. On the other hand, Schultz (1998) shows that as there is more domestic competition in a state (e.g., in democratic regimes), the ex ante probability of war decreases, since a strategic opposition party helps reveal information about the state’s preferences. In this paper, we argue that the Schultz model is more accurate than the Iida model for analyzing the interaction between Israel and the neighboring countries. Given Israel’s democratic regime, neighboring countries can easily obtain information about Israel’s internal processes.

        Empirical studies also question Putnam’s argument. According to Evans et al (1993: 409), leaders “did try to strategically misinterpret their own politics, but not as often as expected, and with much less success.” In Moravscik’s view (1993: 159), bluffing is rare because governments might be able to predict the actions of an eventual cheater: “…among modern information-rich democracies, it is extremely difficult for negotiators to mask their true domestic win-set, even in a sensitive area of national security like weapons procurement.” In other words, as the clarity of a state’s domestic obstacles increases, the ability of the state’s leaders to manipulate this information in the bargaining process declines. Such clarity is more likely in democracies, but in many cases it also exists in non-democratic systems.

        With regard to Israel’s strategic choices and foreign policy in the Middle East during the 1990s, this paper shows how domestic obstacles and internal opposition weaken Israel’s bargaining position towards the hostile bloc it faces. We also argue that using this weakness to manipulate information is unlikely to be advantageous, because of the characteristics of two-level interaction, as mentioned above.

        During the 1990s Israel faced several challenges – especially the Gulf War in 1991 and the armament of states like Syria and Iran – which threatened to upset the balance of power between it and several neighboring countries that comprise a hostile bloc. Those events, as well as the Palestinian uprising (“Intifada”) since 1987, led Israeli leaders to devise a peace policy that found expression in the Oslo Agreement signed in September 1993 (Peres and Naor, 1993). However, due to internal opposition within both Israeli and Palestinian societies, the peace process, which brought great hope to the region, gradually slowed down. Moslem and Jewish fundamentalists committed terrorist attacks on civilian populations. The Israeli Prime Minister who made peace, Yitzhak Rabin, was murdered at a political rally by a Jewish extremist, and Palestinian suicide bombers took the lives of many Israelis. Thus, the public mood was more open to the right-wing campaign that brought Benjamin Netanyahu to power in 1996. According to the International Institute for Strategic Studies (IISS), “The air of optimism generated by the famous Rabin-Arafat handshake on the lawn of the White House in September 1993, dissipated long ago… official Israeli statements refer to the process as going through a critical stage. The Secretary-General of the Palestinian Cabinet, Ahmed Abdel Rahman, has gone much further. In March 1998, he announced the death of the peace process.” (IISS, 1998: 144).

        The shift in Israel’s approach to the peace process cannot be attributed solely to the fact that a right-wing government governed Israel from May 1996 to May 1999. We argue, rather, that it is not only ideological considerations that explain the slow-down of the peace process but also several paradoxes within Israeli society. Since the Palestinians, as well as other Arab countries, observe these processes, Israeli leaders are hardly able to use domestic obstacles to manipulate information in the bargaining process, as might be theorized according to Schelling (1960: 22) and Putnam (1988). Therefore, a precondition for an Israeli leader to halt the peace process for any reason and take a conflictual approach toward the hostile bloc is to shape the preferences of Israeli society. This includes convincing the different population segments that Israel is playing a non-cooperative game with its neighbors – especially the Palestinians – and creating a willingness to tolerate the high cost of violent conflict. We show, however, that the deepening polarization of Israeli society in various dimensions makes it very hard to create such beliefs. Therefore, attempts by Israeli leaders to create the impression that they are willing to adopt a conflictual approach towards the hostile bloc, without actually creating the necessary internal conditions for such policy, may finally lead to a sub-optimal equilibrium for Israel – in terms of territory and deterrence ability – since it will have to compromise under difficult conditions. Once Israeli leaders understand these limitations, it is highly probable that they will form a peace strategy that takes into account the different interests and sensitivities of the neighboring states. Further, the analysis suggests that a consensus in Israeli society may be achieved if there is a sequential bargaining process on crucial issues, with decisions on each issue being taken one at a time, followed by implementation and evaluation of outcomes.

        The paper is organized as follows. The following section uses games in presenting the development of power relations and equilibria in the Middle East until the 1990s. The next section discusses domestic conditions that influence Israel’s strategic choices, as well as its peace policy. Subsequently, we model and explain the possible impact of those domestic conditions on Israel’s strategy toward the peace process with the Palestinians.  

The Development of Power Relations in the Middle East

         The main characteristic of the relations between Israel and neighboring countries has been that these countries did not recognize Israel’s right to exist as an independent state. Until 1977 this was the position of all Arab countries; since then several states – Egypt, Jordan, the Palestinians, and some Persian Gulf states – have joined the peace process to some extent. Hostile countries, however, such as Syria, Libya, Iran, and Iraq, still do not recognize Israel’s right to exist. As for the Israeli side, it has recognized all Arab states, other than the Palestinians’ right to an independent state. Moreover, until 1967 the conflict was not about “the occupied territories” but concerned, rather, most of the area of Israel as one large occupied territory. The minimal demand was for Israel to give up territories it occupied in the 1948 Independence War and to allow the return of Palestinian refugees. The maximal demand, of course, was the abolition of Israel as an independent state. As a result, the Israeli national security conception has been defensive at the strategic level and offensive at the operative level (Horowitz, 1975). The defensive approach at the strategic level has relied on conventional and non-conventional deterrence, motivating a reciprocal arms race between Israel and its neighbors (Aronson, 1984).

        Given the existential nature of the conflict until the early 1970s, the power relations between the Arab countries and Israel during that period can be best described as a zero-sum game. First, we will illustrate this argument by specifying the preferences of Israel and its neighbors, excluding the Palestinians, until the mid-1970s. Then we will explain how, since the mid-1970s, the game between Israel and some neighboring countries has been transformed into a symmetrical prisoners’ dilemma (see also: Brams, 1994: 85-7, 101-2). Finally, we suggest a game for analyzing the power relations between Israel and the Palestinians.

        In the first stage, the players – Israel (I) and a given neighboring country (N) – are modeled as unitary players, i.e., they are assumed to be homogeneous societies. In this game each side has two strategies. Israel can cooperate (C) with the neighboring country by signing a peace treaty which satisfies the neighboring country's territorial (or other) demands to some extent, or not cooperate (D), meaning that it does not give up any territory but has to invest in building deterrent ability. The neighboring country can cooperate (C) with Israel by recognizing its legitimacy to exist and signing a peace treaty, or not cooperate (D), meaning that it does not recognize Israel and has to invest in the arms race. The combination of these strategies creates four possible outcomes:

         C-C: Both sides cooperate.

        D-C: Israel does not cooperate while the neighboring country cooperates.

        C-D: Israel cooperates while the neighboring country does not cooperate.

        D-D: Neither side cooperates.

         The players’ preferences for these outcomes are as follows: Israel mostly prefers D-C, because then it benefits from the neighboring country’s cooperation without giving up any territory (a). Israel’s least preferred outcome is C-D, because then it gives up territory without benefiting (d). It pays a price both in terms of territory and by further investment in building deterrent ability. Israel prefers mutual cooperation (b), C-C, to mutual defection (g), D-D, because then it can attain peace for its citizens and reduce the cost of an arms buildup. The underlying assumption is that Israel recognizes the benefits of peace even at the cost of territorial compromise. As will be shown later, if this is not the case for both sides, the conflict is even deeper than that described by a zero-sum game or by the prisoners’ dilemma. Israel’s order of preferences is:

                        a = (D,C) > b= (C,C) > g= (D,D) > d = (C,D) 

        The neighboring country has calculations similar to those of Israel for D-C and C-D. Yet, as explained, until the mid-1970s the neighboring countries did not recognize Israel’s legitimacy to exist and therefore preferred mutual defection to mutual cooperation. That is, the neighboring country in this game did not believe it could attain any benefits from mutual cooperation while it could benefit from escalating the conflict. It follows that the neighboring country prefers D-D to C-C; thus, its order of preference is:

a = (C,D) > b= (D,D) > g = (C,C) > d= (D,C) 

Figure 1: A Zero-Sum Game between Israel and a Hostile Neighboring Country

                                       Hostile Bloc

                                                                  C                 D

                                          Israel     C        b, g             d, a     

                                                               D        a, d            g, b     

        The players’ order of preference is presented in a game matrix in Figure 1 where a>b>g>d. This is a zero-sum game where one player’s win is the other’s loss. Technically, if the order is presented in terms of numbers, e.g., 1,2,3,4, the sum in each cell is the same. In this game, both players have a dominant strategy of non-cooperation (D) leading to the unique Nash equilibrium (g, b). This means that in a situation of conflict the neighboring country is better off than Israel since it does not recognize any benefits from cooperation.

        It follows that any change in the equilibrium outcome, which existed until the mid-1970s, required a preference change by the neighboring country. The change came about due to certain changes in attitude after the Yom Kippur war of October 1973 (Stein, 1985). This and other processes we will not discuss here changed the attitudes of some neighboring countries about cooperation (Mansur, 1985). They began recognizing the advantages of mutual cooperation, meaning that the zero-sum game presented in Figure 1 was transformed into the symmetrical prisoners’ dilemma presented in Figure 2.  

Figure 2: A Symmetric Prisoners’ Dilemma between Israel and a Neighboring Country 

                           Neighboring Country

                                                                  C                 D

                                          Israel     C        b, b            d, a     

                                                               D        a, d            g, g     

In this game, neither Israel nor the neighboring country is motivated to begin cooperating on its own. Therefore, both sides stay with their dominant strategy of non-cooperation and the equilibrium remains very stable. It is commonly argued that the players can reach a Pareto-optimal outcome (b, b) when a third party intervenes and creates incentives for (or forces) cooperation.

        Third party intervention is usually discussed in the literature with respect to intrastate, often ethnic, conflicts (Licklider, 1993; Gurr, 1993; Gottlieb, 1993). Walter (1997), for example, studied 41 civil wars between 1940 and 1990, and showed the importance of third-party intervention in finding successful negotiated solutions. She argues that negotiated settlements do not fail because bargains cannot be struck but, rather, because it is almost impossible for the combatants themselves to arrange credible guarantees on the terms of the settlement. Regan (1996) also studied all intrastate conflicts since 1944, showing that it is the characteristics of the intervention strategy rather than the characteristics of the conflict that largely determine the success of the intervention. Thus, third-party intervention is required both to create incentives for cooperation and to guarantee the terms of compromise.

        Indeed, under the new circumstances created by the 1973 war, the intervention of a third party became possible and American mediation led to the first peace treaty in the Middle East – between Israel and Egypt. In that peace process both incentives and guarantees were needed.

        To complete the analysis of the historical conditions, we now suggest a game to describe the power relations between Israel and the Palestinians until the early 1990s. As explained, in that period neither Israel nor the Palestinians recognized the other’s right to form an independent state. Therefore, the essence of the conflict was existential, meaning that neither side recognized the advantages of mutual cooperation. The order of preferences that corresponds to this situation is that of the neighboring country in Figure 1. If neither side recognizes the advantages of mutual cooperation, the D-D outcome is preferred to the C-C one.  

Figure 3: The Power Relations between Israel and the Palestinians when Both Sides Do Not Recognize the Advantages of Mutual Cooperation


                                                                    C              D

                                          Israel     C        g, g           d, a     

                                                               D        a, d          b, b              

   Figure 3 shows that, under the above-mentioned condition, not only is there a dominant strategy of non-cooperation for both sides but also the Nash equilibrium is Pareto-optimal (b, b). In other words, both sides believe they win something in a D-D situation while they would only lose in a C-C situation. This means that even if there is a third party who wants to force an agreement, it will have to invest a lot of resources, because the cost of mutual cooperation is higher than the cost of mutual non-cooperation. In comparison, in the prisoners’ dilemma the choice of D-D is due to Nash equilibrium calculations leading to a situation of mutually hurting stalemate situation (Zartman, 1991). In such a case, a third party’s intervention can evidently help the sides achieve cooperation.

        This equilibrium analysis does not refer to domestic variables because until the mid-1970s Israeli society and political culture were uni-dimensional on the security issue  (Arian, 1985; Barzilai, 1996; Sened, 1996). Internal conflicts were covered by the belief that it was necessary to defend the country, as long as the Labor party governed Israel, from 1948 to 1977. This homogeneous political culture began to change in 1977 when the Likud party formed a coalition for the first time. This electoral change expressed and triggered the polarization of Israeli society in several aspects. As a result, Israel’s strategic choices and calculations in its relations with neighboring countries have been transformed.  

Internal Processes Influencing Israel’s Strategy toward Peace in the Late 1990s 

The socio-political and economic processes during 1977-1998 highlight five dimensions that influence Israel’s power and strategic choices in the international scene: The socio-economic dimension, the ethnic-religious dimension, the geographical dimension in terms of center-periphery relations, the security dimension in light of the Arab-Israel conflict, and the dimension of arms buildup, both conventional and non-conventional. In this section, we describe the polarization in each dimension and the mutual dependence between them. According to opinion polls, 30% of the Jewish-Israeli population regard the increased internal tension among various segments of the people as the most important problem on Israel’s public agenda, and 31% so regard the slowdown in the economy, while only 19% regard the stalemate in the peace process as the most important problem facing Israel (Ya’ar and Hermann, 1998). Since the polarization in the various dimensions merge with each other, any Israeli government potentially faces significant domestic difficulties in building a consensus around a foreign policy.

        The Socio-Economic Dimension: Traditionally the Israeli economy has been characterized as highly centralized due to the socialist political culture (Horowitz and Lissak, 1989). The 1990s, however, have been characterized by privatization processes, with various social and economic consequences. These processes include market liberalization, deregulation, transfer of control and management to stockholders, and attempts by international companies to enter the Israeli market (Office for Economic Planning’s Report, 1994). Another aspect of these processes is the creation of flexibility and mobilization in the labor force, thus intensifying socio-economic inequality. Further, due to security problems during the Palestinian uprising (“Intifada”), the Israeli economy became dependent on cheap imported labor to replace cheap Palestinian workers. The large number of imported workers from Africa, South-East Asia and Eastern Europe created significant social, demographic and moral problems due to their very low wages, inequality and lack of basic social and labor rights (Kondor, 1997). This labor policy, which was encouraged by the government, also created significant difficulties for the Palestinian economy, which was highly dependent on the Israeli economy (Roy, 1995).

        The growing socio-economic gaps together with rising unemployment became one of the main issues dividing Israeli society, creating a potential for conflict. Further, when the polarization of this dimension merges with polarization in other dimensions, the potential for conflict intensifies. This leads us to the second dimension listed above.

        The Ethnic-Religious Dimension:  Israeli society is made up of Jewish immigrants from many countries. From the 1920s to the 1940s, these immigrants came mainly from Europe, thus creating a predominantly Western-oriented culture (Horowitz and Lissak, 1989). During the 1950s, after the establishment of the State of Israel, there was large-scale immigration from Arab and Muslim states, thus changing the proportion between the Western-oriented population segment, usually termed “Ashkenazi” and the Eastern-oriented population segment, usually termed “Sephardi.” The arrival of  Sephardim in a predominantly Western-oriented culture created many difficulties for them in becoming established (Horowitz and Lissak, 1989:117). Over the long term, this ethnic division merged with the socio-economic polarization:  The lower classes were mostly composed of Sephardim, and this intensified their feelings of discrimination and deprivation.

        Furthermore, many Sephardim were religiously observant, and thus the orthodox-secular polarization in Israeli society also merged with the previous two (Liebman, 1997; Horowitz and Lissak, 1989). On the other hand, national-religious Ashkenazim also share the popular feeling of traditional Sephardim against the individualism that characterizes the willingness to negotiate with the Palestinians and to recognize the PLO. In the previous decade, Israel (together with Portugal) had been excluded from Western individualism, on the grounds of its collectivist culture (Hofstede, 1983; Huntington, 1996: 71). As Liebman (1997: 102-103) notes, in the current era of growing individualist ethos, religious Zionists are the sector most committed to the values of Israel’s civil religion. The religious import of their political and cultural approach gives it a sense of holiness that separates religiously orthodox people from the rest of society. The merging of polarization in the two dimensions discussed so far (i.e., the socio-economic and the ethnic-religious dimensions) intensifies the conflicts between these population segments. It also presents great difficulty in terms of mobilizing the entire society for a given cause, because the bonds that maintained a certain national consensus until 1977 no longer exist. The polarization in other dimensions further intensifies the problem.

        The Geographical Dimension of Center-Periphery Relations: Interestingly enough, polarization in the geographical dimension also fits the other aspects discussed thus far. Many new immigrants to Israel during the 1950s were sent to the periphery, especially to development towns in the south and north of Israel, while the political, economic and geographical centers (i.e., Tel-Aviv and Jerusalem) were dominated by the upper and middle classes (Arian, 1985; Horowitz and Lissak, 1989; Lipshitz, 1996; Waterman, 1996). As a result, the periphery is dominated by traditional and religious, lower-class Sephardim. It follows that the geographical division of Israeli society also fits the other aspects of polarization and inferiority, as well as their political consequences.

        The Security Dimension relative to the Arab-Israeli Conflict: This dimension has been dominant in Israeli society for most of the century (Arian, 1985; Arian and Shamir, 1990; Barzilai, 1996; Sened, 1996). The question of territorial compromise in exchange for peace has been at the center of political debate since the beginning of Zionism. Other questions, such as Israeli-Jewish identity, usually merged into this dimension. In this respect, until 1977 there was a national consensus on the policy adopted by the Labor-led government (Arian and Shamir, 1990). However, the socio-political and economic processes discussed so far also influenced this dimension. Since the early 1980s – especially since the 1982 Israeli-Palestinian war in Lebanon – the polarization between right and left in Israeli society intensified (Arian and Shamir, 1990; 1994; Horowitz and Lissak, 1989).

        To a large extent this polarization fits the divisions in the other dimensions – lower-class voters, the Sephardim, voters in the periphery, and religious voters traditionally support right-wing parties (Yuchtman-Yaar and Hermann, 1998; Shamir and Arian, 1999). Shamir and Arian (1999) present a logistic regression based on a longitudinal analysis of electoral cleavages from 1969 to 1996, and an analysis of the 1996 election. They show that religious, Sephardim, less educated, and lower status workers voted for the right-wing Likud and religious parties, whereas the left (Labor and Meretz) has had a disproportionate share of secular, upper class Ashkenazi voters. Since voting patterns significantly correlate with the preferences concerning the peace process, this cross-sectional characterization fits the polarization in the security dimension. This argument is also supported by an ongoing monthly opinion poll done by Yaar and Hermann (1993-2000), beginning in August 1993. These polls, also called the peace index, basically examine the public’s attitudes toward the peace process given ongoing events and the divisions in Israeli society. As explained earlier, the range of alternatives in Israel’s policy toward neighboring countries is ultimately reduced to a dichotomous choice: A person is either for or against giving up territory in exchange for peace. This distinction is the basis for the poll questions. The participants sampled are representative of the Jewish population of Israel. According to this continuing opinion poll, the religious population in general, and the Ultra-Orthodox in particular, has assumed the role of the radical right-wing symbol for everything touching on the peace process (Ya’ar and Hermann, 1997a). Among the Ultra-Orthodox only 20.5% support or greatly support the process; among those defining themselves as Religious 43% support the process. On the other hand, 82% of traditionalist and 78% of secular groups declared their support for the process (Ya’ar and Hermann, 1997b).

        Thus, low-class voters, Sephardim, peripheral voters and religious voters traditionally support right-wing parties. Although some of these voters do not completely accept the right-wing attitude to the Arab-Israeli conflict, they vote for right-wing parties based on their calculations and preferences in the other dimensions (Arian and Shamir, 1990; 1994). As a result, Israeli society faces the paradox that a small majority of the population favors the peace process, but this is not clearly expressed in the political division of power.

Table 1 shows there is long-standing support for the peace process, but Netanyahu still won the May 1996 election even though he challenged the peace policy of his predecessors, Rabin and Peres. It was only in the May 1999 election that the left-wing candidate, Barak, took over from Netanyahu, and that supporters of the peace process received nearly 50% of the seats in the parliament. Thus, since May 1999 the near-50% support for the peace process has been expressed in the political division of power.

Table 1 also shows certain changes in the support for the peace process over the months. Although these changes are not statistically significant, they are usually attributed to events and developments in the peace process. For example, violent attacks by the Palestinians in the territories or by Hizbullah in Lebanon are clearly followed by declining support for the peace process. Thus, the two-level dynamic also works in the opposite direction. Not only do internal processes influence foreign policy but international events and developments also influence internal beliefs and processes. It is often argued, for example, that the slowdown in the peace process during 1994-95 can be attributed to the murderous car bombs exploded by extreme Palestinians, which killed many Israelis.

        It follows that the deep polarization in the socio-economic, ethnic-religious and geographical dimensions project strongly onto the security dimension, thus creating great domestic difficulty for any peace policy. Furthermore, the deep polarization between different social segments creates domestic difficulties for any foreign policy, because Israel’s leaders cannot create a consensus for a militarist policy either. As expressed through their behavior during the Gulf War and the long-standing conflict in Lebanon and the occupied territories, many of Israel’s citizens are no longer willing to pay the high price of a non-consensual war (Barzilai, 1996; Ya’ar and Hermann, 1998). This approach also affects the fifth dimension mentioned above – arms buildup, both conventional and non-conventional.


Table 1: The Level of Support in the Peace Process with the Palestinians among   the  Jewish-Israeli Population – August 1993-April 2000



In favor (percent)


 In favor      (percent)

August 1993


June 1997


June 1994


September 1997


December 1994


December 1997


March 1995


March 1998


September 1995


September 1998


December 1995


December 1998


March 1996


April 1999


June 1996


August 1999


September 1996


November 1999


December 1996


April 2000


March 1997


August 2000


        Source: Yaar and Hermann (1993-2000) 

        The Dimension of Non-conventional Deterrence: For many years Israel maintained a policy of obscuring its nuclear capability, by stating it would not be the first country to introduce nuclear weapons into the Middle East (Aronson, 1992). Nevertheless, it became common belief that Israel had an impressive nuclear endowment. Based on his impressions from Egyptian leaders, Peres attributed Egypt’s decision to make peace with Israel to that capability, in part (Peres and Naor, 1993: 4-5). It did not, however, prevent Egypt from launching a limited war against Israel four years earlier. According to some accounts, in 1973 Israel already had nuclear arms, which could be deployed from aircraft and missiles (Hersh, 1991: 215-6; Paul, 1995). Israeli deterrent calculations were based on the possession of superior conventional and nuclear capability, and the Israeli leadership implicitly declared this capability to be its ultimate deterrent against an Arab attack:  Before the October 1973 war, leaders such as Defense Minister Moshe Dayan hinted at the Israeli nuclear deterrent and made ambiguous nuclear threats (Evron, 1990; Bar-Joseph, 1982; Feldman, 1982; Freedman, 1975). Dayan reportedly believed that the Arab states would not initiate a war before the early 1980s, and until then Israel’s nuclear capability likely would act as a deterrent against conventional attack. From 1967 to 1973, the Arab leaders and the media talked unceasingly of Israeli nuclear capability and the implications of it (Van Creveld, 1993: 108-110; Evron, 1973: 19-31). The Egyptians were also presumed to have received intelligence information on Israel’s nuclear weapons and strategy from Soviet spies who had penetrated the state’s defense and intelligence establishments (Hersh, 1991: 219).

        However, non-conventional deterrence capability did not deter Egypt and Syria from starting a conventional war in October 1973. Stein (1985) suggests that Egyptian internal politics were much more important than Israel’s conventional or non-conventional strength. Paul (1995), on the other hand, attributes this and other similar cases to the concept of “nuclear taboo,” the notion that nuclear weapons are characterized by their non-use. As Schelling (1994: 110) argues, the main reason for the uniqueness of nuclear weapons is the perception that they are unique and that once introduced into combat, they cannot be “contained, restrained, confined, or limited.” Given this “nuclear taboo,” Israel’s nuclear ability cannot guarantee that a limited war will not break out. Therefore, the influence of Israeli society’s deep polarizations on its willingness to enter into a conventional war is a central parameter in analyzing the Arab-Israeli conflict.  

        These polarizations, which cannot be hidden from neighboring countries, is a source of weakness for Israel’s position in the region. Israel’s leaders can indeed claim, as they often have, that they cannot proceed with the peace process due to internal opposition. But neighboring countries can also observe the strong opposition to a non-consensual war. Domestic obstacles, therefore, can hardly be used to manipulate information in the bargaining process. We now discuss the effect of these obstacles on the conflict between Israel and the Palestinians. 

Israel’s Strategic Choices Relative to the Peace Process with the Palestinians in the Late 1990s 

        In this section, we analyze Israel’s strategic choices with regard to the peace process with the Palestinians in the late 1990s. Given the two-level game approach, we first explain the preferences of the two sides as they have been shaped until the late 1990s. Based on these preferences, we outline the basic strategic choices open to Israeli leaders. We then expand the analysis to explain the bargaining mechanism preferred for Israel.

        Applying the two-level game to the Palestinians, their position was influenced by international events such as the 1991 collapse of the Soviet Union and the subsequent loss of superpower support by Syria and other Arab states, as well as the Palestinians’ own loss of support from the Gulf States following the 1991 war with Iraq. These events created the basic conditions for the Palestinians to move toward a cooperative strategy as expressed by the signing of the Oslo Agreement (Zartman, 1997).

        Yet, since the signing of the Oslo Agreement, dissatisfaction with the Israeli-Palestinian peace process has grown significantly among the Palestinian population (IISS, 1998: 146-148). As expressed in the Palestinians’ own observations, the economic situation in their autonomous regions deteriorated (Roy, 1995) and the Israeli government’s new policy since May 1996 has left the impression that their demands will not be met soon enough. As one military officer said, “When people are hungry, policy disintegrates” (Limor, 1998). As a result, they hardly trust Israel’s promises and commitments. This change in attitude, which mainly took place during 1996-1998, means that the Palestinians’ order of preference in the prisoners’ dilemma is as presented in Figure 2. They recognize the advantages of cooperation but believe they are playing a non-cooperative game with Israel, meaning that they prefer conflict (D-D) to the option of being the sole compromiser (D-C).   

        Because of the internal polarization discussed earlier, however, Israel has a different order of preference in its game with the Palestinians. This polarization means that Israeli society is divided in respect to the peace process. A significant right-wing segment of society actually views the Israeli-Palestinian conflict in terms of the prisoners’ dilemma described in Figure 2. Among them is a small “farther right” group of Israelis who, in line with the Figure 3 game, genuinely prefer mutual conflict with the Arabs to mutual cooperation. On the other hand, supporters of the peace process are more inclined toward cooperation with the Palestinians and clearly are not willing to pay the price of what they regard “a non-consensual war”. This means that their order of preference is similar to that in the Chicken game, where conflict is the worst outcome (Taylor, 1987). In a symmetrical Chicken game, there are two equilibria in pure strategies:  A player will cooperate if he/she believes the other will not cooperate, but will not cooperate if he/she believes the other will.

        It follows that, given domestic conditions, Israeli leaders can adopt neither a conflictual approach toward the Palestinians nor a coherent peace strategy. All they can know with certainty is that a majority of the population supports some version of compromise, and that to those in favor of the peace process a violent conflict with the Palestinians constitutes the highest cost and therefore may also deepen the polarization, towards a total disintegration of Israeli society. On the other hand, right-wing parties and their supporters are split. Since most of them understand that the Oslo Agreement is irreversible (Sprinzak, 1998), they are willing to make some compromises, meaning that mutual cooperation is preferred to mutual defection. Thus, for them as well a violent conflict is ordered low in their preferences, while a reasonable compromise in their view can be accepted. This approach is expressed, for example, in the relatively low mass mobilization against compromises made by Netanyahu and Barak. In addition, in the May 1999 elections right-wing parties which strongly opposed any compromise in the peace process lost many seats in the parliament, and they now constitute only 7-10% of the seats. Overall, the hard core of strong opposition to the peace process is composed of religious settlers numbering about 50 thousand people. Although they constitute a strong interest group, they have gradually understood their power is limited. Furthermore, following the assassination of Itzhak Rabin in 1995 their modes of protest have modified significantly, and the intensity of their protest activities has declined  (Yuchtman-Yaar and Hermann, 1998a). This opposition may also use party tactical/electoral/coalition calculations to bring about the government’s disintegration following a given move in the peace process. But these are stages in the adaptation of public attitudes to peace and of the division of political power in the parliament. Based on these indications, we argue that a clear majority of Israeli public and politicians recognize the need for compromises in negotiating with the Palestinians. This means that the long-term potential for societal disintegration as a result of concessions is lower than the disintegration potential as a result of conflict.

        Thus, it is almost impossible to create a consensus for a conflictual strategy toward the Palestinians, while a reasonable consensus can be achieved over certain concessions. Therefore, the option of violent conflict is the worst possible outcome for Israeli leaders in their relations with the Palestinians. As explained in the previous section, the polarization in all dimensions converges; as a result politicians and observers have difficulty isolating the attitudes and motivations in each dimension separately. Yet, both Israeli and Palestinian leaders clearly observe the unwillingness of the Israeli public to pay the high price of violent conflict. They can both interpret the preference ordering of Israeli leaders as being similar to that of the Chicken game. Hence, due to domestic obstacles, Israeli leaders’ order of preference in their game with the Palestinians can be represented as follows:

a = (D,C) > b = (C,C) > g = (C,D) > d = (D,D)

The combination of the players’ order of preference leads to an asymmetric game as presented in Figure 4.  

Figure 4: Asymmetric Power Relations between Israel and the Palestinians


                                                                  C                 D

                              Israeli Leaders    C        b, b            g, a     

                                                               D        a, d            d, g  

       In this game, the Palestinians have a dominant strategy of non-cooperation. Israeli leaders, on the other hand, have an order of preference in the Chicken game, meaning they will cooperate if the Palestinians do not cooperate but will not cooperate if the Palestinians do. If Israeli leaders recognize the Palestinians’ order of preference, they can expect them to choose non-cooperation. Then the best possible strategy for Israeli leaders is cooperation, leading to a unique Nash equilibrium with pure strategies (g,a).

        This equilibrium outcome expresses the asymmetry between Israeli and Palestinian societies in terms of their willingness to tolerate the costs of a violent conflict. Since Israeli society is less willing than Palestinian society to bear such costs, it can be expected to achieve sub-optimal results as long as the non-cooperative game continues and the Palestinians are willing to enter into a violent conflict. Furthermore, this analysis implies that possible attempts by Israeli leaders to express an order of preference in the prisoners’ dilemma rather than recognizing the weaknesses of Israeli society may indeed lead to violent conflict, in which Israel will have no choice but to compromise under difficult conditions. Given the asymmetric game presented in Figure 4, the straightforward option for Israeli leaders is to influence the Palestinians’ order of preference in such a way that violent conflict will become the worst outcome for them as well. This means increasing the potential losses from a violent conflict by increasing its costs, as well as increasing the benefits from cooperation – i.e., accelerating the peace process rather than slowing it down. Alternatively, Israeli leaders may try to create a broad consensus in Israeli society with respect to the preferred strategy, whether a peaceful or conflictual one, toward the Israeli-Palestinian conflict.

        Moreover, since the Palestinian and Israeli communities are highly integrated, Israeli leaders can hardly use domestic obstacles to manipulate information in the bargaining process. They can claim, as they often have, that due to these obstacles they cannot proceed with the peace process, but the Palestinians can also see their difficulties in creating a consensus for a militarist policy. Therefore any attempt to manipulate information and create the impression that Israel has an order of preference in the prisoners’ dilemma is not likely to succeed. Rather, it may trap Israeli leaders in their own manipulation, thus creating a cognitive dissonance. On the other hand, Palestinian negotiators successfully use internal opposition to argue that they cannot make significant concessions. Yet if, as a result, Palestinian-Israeli relations deteriorate to a violent conflict, Israeli leaders will have to compromise under difficult conditions as expected by the asymmetric game presented in Figure 4.

        A more complex analysis of the game in Figure 4, however, enables us to draw practical conclusions about the preferred bargaining mechanism for Israel, as well as possible ways to influence the attitudes of the Israeli public through this mechanism.

        The game in Figure 4 describes the core of the strategic dilemma that Israel is facing. Although the bargaining in the peace process is very complex, with many decision points on specific issues and many issues on the table, ultimately it can be reduced to several final decisions that will have to be made on key issues:  The size and location of territories to be under Palestinian rule, the Jerusalem problem, the refugees problem, division of water resources, border controls, and military limitations on the Palestinian state. We argue that given the conditions existing in the late 1990s, at each of these crucial decision points Israel is likely to face the strategic dilemma described in Figure 4. In other words, no matter how long the bargaining continues and whatever tactical moves the sides make, at the final point of decision the Palestinians are likely to adopt a conflictual approach in order to force Israel to make concessions. As long as this game continues, Israel is likely to make these concessions.

        Furthermore, given the Israel public’s unwillingness to pay the price of conflict, dividing the peace process into many points of decision on small matters, such as the release of three or a dozen Palestinian prisoners, creates a situation in which the public sees a violent conflict over such points as unnecessary and therefore non-consensual conflict. Israeli leaders therefore make the concessions. Thus, by creating many decision points over small points, Israeli leaders actually create a situation where the asymmetric game presented in Figure 4 is played again and again but its outcome does not lead the Israeli public to change preferences. The concessions at each point seem too minor to justify a conflict and thus even when the game is repeated many times, the outcomes at one stage do not change the conditions for the next round. This cumulative effect led Israeli Prime Minister Barak, for example, to move from a willingness to give the Palestinians 40% of the West Bank to a willingness to give 80-90%. This means that breaking the peace process issues into many decision points, both in terms of sub-issues and in terms of time, creates a situation in which there is no difference between the meta-game and the one-stage game. Thus, Figure 4 can describe them both.

        It directly follows that Israel has strong interest in immediately reaching the final decision points over the crucial issues in a sequential bargaining process. That is, the best strategy for Israeli leaders is to push for a time-constrained bargaining process seeking agreement on each crucial point individually rather than looking for a package deal covering all the issues. Then, if concessions on an isolated crucial point are followed by a conflictual approach by the Palestinians, the Israeli public can be expected to see a possible conflict over the next crucial issue as a consensual one. The model thus concludes that the bargaining mechanism should be composed of time constrained discussions on a key issue, implementation of the agreement on it, evaluation of the outcomes by both sides, followed then by another time constrained discussion on a key issue, implementation, evaluation, and so on. In this way, both the Palestinians and Israelis will have indications about the other’s intentions on the basis of specific actions, rather than subjective interpretations and beliefs.  


In this paper we have shown the impact of domestic processes on Israel’s strategy toward peace during the 1990s. Such processes intensify the polarization between different segments in Israeli society and limit the possibility that its leaders can create a consensus for any policy towards the Arab-Israeli conflict – especially a consensus for violent conflict with the Palestinians. Since the Palestinians, as well as other Arab countries, observe these processes, Israeli leaders can hardly use domestic obstacles to manipulate information in the bargaining process.

        Although the empirical setting analyzed in this paper is very complex, we believe that a theoretical game approach can help make the players’ choices very clear. Furthermore, by using games we are bounded by certain assumptions and terminology that make the analysis clear and well founded. For example, changes in bargaining position can be attributed to many factors. By specifying the players, their choices and the mutual dependence between them using simple, precise language, we can point out explanatory variables. In this respect, the two-level game analysis clearly helps explain the complex world of international relations. We believe it is very hard to generalize through formal models any hypothesis regarding the impact of a two-level interaction. Rather, we demonstrated how internal polarization may create an order of preferences in the Chicken game. Further research should proceed through a comparison of detailed case studies.     


Arian, Asher. 1985. Politics in Israel. Chatham: Chatham House.

Arian, Asher and Michal Shamir, eds. 1994. The Election in Israel – 1992. Albany: SUNY.

Arian, Asher and Michal Shamir, eds. 1990. The Election in Israel – 1988. Albany: SUNY.

Aronson, Shlomo. 1992. The Politics and Strategy of Nuclear Weapons in The Middle East. Albany: State University of New York Press.

Aronson, Shlomo. 1984. “The Nuclear Dimension of The Arab-Israeli Conflict: The Case of The Yom Kippur War.” Jerusalem Journal of International Relations, Vol. 7, 107-42.

Bar-Joseph, Uri. 1982. “The Hidden Debate: The Formation of Nuclear Doctrines in The Middle East.” Journal of Strategic Studies, Vol. 5, 205-27.

Barzilai, Gad. 1996. Wars, Internal Conflicts and Political Order. Albany: State University of New York Press.

Brams, Steven J. 1994. Theory of Moves. Cambridge: Cambridge University Press.

Colomer, Josep. 1995. Game Theory and The Transition to Democracy: The Spanish Model. Aldershot, UK: Edward Elgar.

Evans, Peter. B., Harold K. Jacobson and Robert D. Putnam, eds. 1993. Double-Edged Diplomacy. Berkeley: University of California Press.

Evron, Yair. 1990. “Opaque Proliferation: The Israeli Case.” Journal of Strategic Studies, Vol. 13, 44-63.

Evron, Yair. 1973. “The Arab Position in The Nuclear Field: A Study of Policies Up to 1967.” Cooperation and Conflict, Vol. 18, 19-31.

Feldman, Shay. 1982. Israeli Nuclear Deterrence. New York: Columbia University Press.

Freedman, Lawrence. 1975. “Israel’s Nuclear Policy.” Survival, Vol. 17, 114-120.

Gottlieb, Gideon. 1993. Nation Against State. New York: Council on Foreign Relations.

Gurr, Ted R. 1993. Minorities at Risk: A Global View of Ethno Political Conflicts. Washington, DC: United States Institute for Peace.

Hersh, Seymour M. 1991. The Samson Option: Israel’s Nuclear Arsenal and American Foreign Policy. New York: Random House.

Hofstede, Gary. 1983. “National Cultures in Four Dimensions: A Research-Based Theory of Cultural Differences among Nations.” International Studies of Management and Organization, Vol. 13, 52.

Horowitz, Dan. 1975. Israel’s Concept of Defensible Borders. Jerusalem: Hebrew University of Jerusalem.

Horowitz, Dan and Moshe Lissak. 1989. Trouble in Utopia. Albany: State University of New York Press.

Huntington, Samuel P. 1996. The Clash of Civilizations and The Remaking of World Order. New York: Simon and Schuster.

Iida, Keisuke. 1993. “When and How Domestic Constraints Matter? Two-level Game Uncertainty.” Journal of Conflict Resolution, Vol. 37, 403-426.

International Institute for Strategic Studies (IISS). 1998. Strategic Survey 1997/98. London: Oxford University Press.

Karsh, Efraim and Gregory Mahler, eds. 1994. Israel at The Crossroads: The Challenge of Peace. London: British Academic Press.

Kondor, Yoav. 1997. Alien Labor in Israel. Jerusalem: National Security Institute. (Hebrew).

Licklider, Roy. 1993. “How Civil Wars End: Questions and Methods.” In Roy Licklider, ed., Stopping The Killing: How Civil Wars End. New York: New York University Press.

Liebman, Charles S. 1997. Religion, Democracy and Israeli Society. Amsterdam: Harwood Academic Publishers.

Limor, Yoav. 1998. “Project ‘May 99’.” Maariv (a Hebrew daily) September 15, 1998, p. B23.

Lipshitz, Gabriel. 1996. “Core Versus Periphery in Israel Over Time: Inequality, Internal Migration, and Immigration.” In Yehuda Gradus and Gabriel Lipshitz, eds., The Mosaic of Israeli Geography. Beer-Sheva: Ben-Gurion University of the Negev Press.

Mansur, Salim. 1985. “The Egypt-Israel Relationship: An Anatomy of Wasted Decades.” In Paul Marantz and Janice Gross Stein, eds., Peace Making in The Middle East. Totowa, New Jersey: Barnes and Noble.

Mo, Jongryn. 1995. “Domestic Institutions and International Bargaining: The Role of Agent Veto in Two-Level Games.” American Political Science Review, Vol. 89, 914-924.

Moravcsik, Andrew. 1993. “Introduction: Integration International and Domestic Theories of International Bargaining.” In Peter B. Evans, Harold K. Jacobson and Robert D. Putnam, eds., Double-Edged Diplomacy. Berkeley: University of California Press.

Office for Economic Planning’s Report. 1994. Jerusalem.

Paul, T. V. 1995. “Nuclear Taboo and War Initiation in Regional Conflicts.” Journal of Conflict Resolution, Vol. 39, 696-717.

Peres, Shimon with Arye Naor. 1993. The New Middle East. New York: Henry Holt and Company.

Putnam, Robert D. 1988. “Diplomacy and Domestic Politics: The Logic of Two-Level Games.” International Organization, Vol. 42, 427-60.

Regan, Patrick M. 1996. “Conditions of Successful Third-Party Intervention in Intrastate Conflicts.” Journal of Conflict Resolution, Vol. 40, 336-359.

Roy, Sara. 1995. The Gaza Strip: The Political Economy of De-Development. Washington, DC: Institute for Palestinian Studies.

Schelling, Thomas. 1994. “The Role of Nuclear Weapons.” In L.B. Ederington and M.J. Mazarr, eds., Turning Point: The Gulf War and U.S. Military Strategy. Boulder, CO: Westview.

Schelling, Thomas. 1960. The Strategy of Conflict. Cambridge, Mass.: Harvard University Press.

Schneider, Gerald and Lars-Eric Cederman. 1994. “The Change of Tide in Political Cooperation: A Limited Information Model of European Integration.” International Organization, Vol. 48, 633-662.

Schultz, Kenneth A. 1998. “Domestic Opposition and Signaling in International Crises.” American Political Science Review, Vol. 92, 829-844.

Sened, Itai 1996. “A Model of Coalition Formation: Theory and Evidence.” The Journal of Politics, Vol. 58, 350-372.

Shamir, Michal and Asher Arian. 1999. “Collective Identity and Electoral Competition in Israel.” American Political Science Review, Vol. 93, 265-277.

Sprinzak, Ehud. 1998. “The Politics of Paralysis: Netanyahu’s Safety Belt.” Foreign Affairs, Vol. 77, 18-28.

Stein, Janice Gross. 1985. “The Fundamentals of Peacemaking: A Retrospective Analysis. In Paul Marantz and Janice Gross Stein, eds., Peace-Making in the Middle East. Totowa, New Jersey: Barnes and Noble.

Taylor, Michael. 1987. The Possibility of Cooperation. Cambridge: Cambridge University Press.

Tsebelis, George. 1990. Nested Games. Los Angeles: University of California Press.

Van Creveld, Michael. 1993. Nuclear Proliferation and The Future of Conflict. New York: Free Press.

Walter, Barbara F. 1997. “The Critical Barrier to Civil War Settlement.” International Organization, Vol. 51, 335-364.

Waterman, Stanley. 1996. “Geography and The Israeli Electoral Reform. In Yehuda Gradus and Gabriel Lipshitz, eds., The Mosaic of Israeli Geography. Beer-Sheva: Ben-Gurion University of the Negev Press.

Ya’ar, Efraim and Tamar Hermann. 1993-2000. The Peace Index. Tel-Aviv: Tami Shteinmatz Center, Web site: http://www.tau.ac.il/peace/.

Ya’ar, Efraim and Tamar Hermann. 1998. The Peace Index – July 1998. Tel-Aviv: Tami Shteinmatz Center.

Ya’ar, Efraim and Tamar Hermann. 1997a. The Peace Index – January 1997. Tel-Aviv: Tami Shteinmatz Center, Web site: http://www.tau.ac.il/peace/.

Ya’ar, Efraim and Tamar Hermann. 1997b. The Peace Index – August 1997. Tel-Aviv: Tami Shteinmatz Center.

Yuchtman-Yaar, Efraim and Tamar Hermann. 1998. “The Latitude of Acceptance.” Journal of Conflict Resolution, Vol. 42, 721-743.

Zartman, William I. 1991. “Conflict and Resolution: Contest, Cost and Change.” Annals of The American Academy of Political and Social Science, Vol. 518, 11-22.

Zartman, William I. 1997. “Explaining Oslo [Israel-PLO Negotiation].” International Negotiation, Vol. 2, 195-215.