defects.  Accordingly, each player “[has] a dominant strategy of defecting”: he thereby obtains his most-preferred outcome if the other cooperates, while he avoids his least-preferred outcome if the other defects.  Yet if both actors follow this “rational course,” then the result is the suboptimal equilibrium of mutual defection.

          A Middle Eastern version of the prisoners’ dilemma, then, might consist of Israel and an Arab actor, which both prefer a mutually-adopted peace treaty to mutual warfare.  Yet because both fear that the other will defect from any treaty, neither cooperates with the other – and war ensues.  This preference ordering may seem superficially to characterize both Arab and Israeli actors, as noted above, yet it is fundamentally inapplicable to one or both actors in each of the three conflict arenas investigated in this paper.

The Wider Conflict

          Moshe Dayan, Israel’s foreign minister during the Camp David Accords, believed in Egypt’s centrality to the Arab-Israeli conflict.  He declared that “the future is with Egypt.  If you take one wheel off a car, it won’t drive.  If Egypt is out of the conflict, there will be no more war.”  Indeed, Egypt fought in all five of Israel’s conventional wars from 1948 to 1973,[1] whereas after the 1979 peace treaty, Israel only waged war against non-state entities – as in the first Lebanon War (against the PLO)[2] and both Palestinian intifadahs.

          Thus, the fact that the Egyptian-Israeli relationship did not resemble even superficially a prisoners’ dilemma until 1973, profoundly weakens that game theoretical model’s utility in analyzing Middle Eastern geopolitics.  On the one hand, Israel did prefer mutual cooperation to mutual defection even before Egypt entered into peace negotiations: during the 1973 Yom Kippur War, Israeli Prime Minister Golda Meir asserted that “We don’t want dead on our side, we have no joy in causing the death of others.”[3]  Clearly, such rhetoric aimed at bettering Israel’s public image, but it was couched in terms reflecting a clear preference ordering: peace was preferable to war, but in the absence of mutual cooperation, Israel “has decided to live” and “to pay the price for living.”[4]

          On the other hand, Egypt’s official rhetoric up to the Yom Kippur War reflected a fundamentally different preference ordering: it valued peace with Israel less than war against Israel – in game-theoretical terms, it favored mutual defection over mutual cooperation, as do actors in what Stein terms a “deadlock”[5]:

          “The Arab national aim,” Egyptian President Nasser told other Mideast leaders, “is the elimination of Israel.”[6]  In the run-up to the Six-Day War, he proclaimed that “[the] Arab people want to fight…The mining of Sharm el Sheikh [by Egypt] is a confrontation with Israel,” and “obligates us to be ready to embark on a general war with Israel.”[7]  Such statements are particularly revealing, because they clarify two crucial determinants of Nasser’s policy toward Israel.  First, he rejected peace not because Israel might betray a treaty, but because peace with Israel was intrinsically undesirable.  Second, he identified the source of this sentiment: an “Arab national aim,” or an “Arab people.”

          Anwar al-Sadat – the Egyptian leader who made peace with Israel – articulated the pan-Arab preference ordering still more clearly, a year before the Yom Kippur War: “In our coming battle, I will not be satisfied to liberate the land [Sinai, which Israel had taken from Egypt].  Israel’s arrogance and bluster, which has been going on for 23 years [since Israel’s independence in defiance of a general Arab will, not since its capture of specifically Egyptian territory] – all this must be terminated…I am ready to pay one million men as the price for this battle.  But they too must be ready to pay a million men and more on their side.”[8]  Again, mutual defection (a war involving a million casualties on each side) outweighed mutual cooperation (a peace treaty).[9]

          Yet Sadat oversaw a profound reevaluation of Egypt’s preference ordering.  During his presidency, Egypt defined its interests in narrowly nationalistic terms, above and apart from pan-Arab issues.[10]  As it shrugged its mantle as director of the Arab fight against Israel, ideological considerations lost value, while economic troubles and territorial losses became more relevant.  In particular, Sadat’s “hope for American economic, military, and technological assistance,” and his desire to regain the Sinai, were so intense that they put him “in a weaker position than [Israeli Prime Minister Menachem] Begin” to negotiate the terms of a peace treaty[11] – so strongly did he come to prefer mutual cooperation to mutual defection.  Accordingly, Egypt’s preference ordering approximated Israel’s.

          The primary obstacle to Israeli-Egyptian cooperation thereby vanished.  The United States removed a secondary hurdle (the possibility that either actor might defect) by providing financial aid to both sides and promising to “put the weight of the United States behind Israel in the event that Egypt violated the treaty.”[12]  Though these American commitments encouraged cooperation, they arguably contorted both Egypt and Israel’s preference orderings, such that each actor most prefers mutual cooperation, even over a situation in which it defects while the other cooperates.  In other words, current-day Egypt may value American aid more highly than any benefit which might be gained by a surprise attack on Israel.  Egypt-Israel relations, therefore, approximate what Stein terms an “assurance game”[13] rather than a prisoners’ dilemma.

          The Egyptian example has mixed implications.  On the one hand, it suggests that any Arab state that can extricate itself from the pan-Arab conflict with Israel, may redefine its interests such that it acquires incentives to cooperate with Israel.  On the other hand, actors who are more deeply invested in the conflict than was Egypt (e.g., the Palestinians, and to a lesser extent, Syria), and therefore have less flexibility to reject ideological in favor of economic or territorial concerns, will find it more difficult to reverse their preference orderings and internalize a cooperative approach toward the Arab-Israeli conflict.  This hypothesis, then, may explain the breakdown in Israeli-Palestinian negotiations, to which this paper now turns:

The Localized Conflict: A First Cut

          U.S. President Bill Clinton convened the Camp David II summit in July 2000 to conclude the Israeli-Palestinian peace process which the Oslo Accords had inaugurated.  Under the American aegis, Israeli Prime Minister Ehud Barak and PA Chairman Yasser Arafat negotiated over the three issues at the heart of the conflict: (1) sovereignty over Jerusalem, which Palestinian representative Muhammad Dahlan called “the core of the conflict,” (2) the status of the West Bank and Gaza Strip, and (3) the Palestinian refugee situation.[14]  As at the first Camp David, American financial aid and security guarantees – which would have reduced the danger that either side would renege on a peace treaty, and therefore lowered the risks of cooperating – gave both actors a clearer choice between mutual compromise and mutual defection.

          Israel compromised over the first two issues – offering Arafat all of Gaza, nearly all of the West Bank, and shared sovereignty over the Temple Mount in Jerusalem.[15]  However, Israel refused to make substantial concessions on the refugee issue,[16] for fear of jeopardizing its security and independence – outside of which, peace had little value to Israel, so that Palestinian gains could come only at Israel’s expense (what Stein terms a “constant-sum” situation[17]).  Therefore, while Israel held a “prisoners’ dilemma” preference ordering toward the first two issues, its stance on the refugee issue was more conducive to deadlock.

          The PA faced no comparable existential threat.  However, its delegation treated all three issues as zero-sum games for much the same reason as had Egypt before 1973: it simply was unwilling to compromise over Jerusalem, the West Bank and Gaza, and the Palestinian refugees.  Whether because of Palestinian negotiators’ ideological convictions or their fears of being assassinated upon returning to the West Bank or Gaza, potential American aid failed to mitigate their hard-line stance.  In comparison to their need to uphold the maximal Palestinian stance on these three issues, peace was of negligible importance.[18]  Mutual defection outweighed mutual cooperation.  Official Israeli and Palestinian statements reflect this disparity in preference orderings:

          On the one hand, Israel’s then-transportation minister Amon Lipkin Shahaq said that he “[saw] no accomplishments” in Camp David II’s “[failure] to attain” any “peace agreement.”[19]  On the other hand, “PLO Political Department Head Faruq Qaddumi” just as clearly equated the summit’s failure with “success for the Palestinians.”[20]

          Yet if Israel failed where the Palestinians succeeded, then the two sides could not have pursued the same object, and if the Palestinians succeeded where Camp David II failed, then Arafat clearly chose mutual defection over mutual cooperation.  Indeed, Barak’s overriding aim was to reach a compromise, while Arafat’s was to forestall one.[21]  For the Palestinians, the negotiations over the three core issues were a game of deadlock, not a prisoners’ dilemma.

The Localized Conflict: A Second Cut

          Since the start of the second intifadah, Palestinian terrorism and its consequences have eclipsed these three basic issues.  Here again, Israel exhibits a prisoners’ dilemma mentality: it deems mutual cooperation (i.e., a return to permanent-status negotiations) to be favorable to mutual defection (i.e., Palestinian terrorism and Israeli countermeasures), though it prefers even mutual defection to Israeli cooperation and Palestinian defection (i.e., appeasing Palestinian terrorism).[22]  Further – assuming that a moderate and trustworthy Palestinian leadership exists in the person of PA Chairman Mahmoud Abbas – at least one Palestinian actor shares this preference for mutual cooperation over mutual defection (or Palestinian betrayal of a peace treaty).[23]

          Superficially, then, the prisoners’ dilemma seems to capture the gist of the situation.  Then why – assuming that American guarantees would prevent either party from defecting once both had begun cooperating – did negotiations not pick up from Oslo and Camp David II once Arafat died and Mahmoud Abbas took control of the PA?

          The answer is that, while the preceding assumptions about Israeli, Palestinian, and American preferences may reflect reality, they ignore the actors whose preference orderings are most conducive to deadlock: Palestinian terrorist groups, which, for this paper’s purposes, are treated as a single actor (Hamas).[24]  Like Arafat at Camp David II and Egypt before 1973, this actor prefers mutual defection to mutual cooperation, and, argue Kydd and Walter, adopts a rather rational strategy for achieving the former aim:

The purpose of extremist violence is not to achieve anything directly in a military sense, nor to signal that the extremist group opposes the peace treaty, which is already known….[Palestinian terrorists] understand that targeting [Israelis] with violence will increase their fear and make them increasingly less likely to implement any terms.  Thus terrorist bombs are designed to persuade a targeted group [Israel] that the seemingly moderate opposition [the PA] with whom it negotiated an agreement will not stop terrorism, and hence cannot be trusted to implement the deal.[25]

          On the one hand, this actor’s existence pushes Israeli and PA preference orderings closer to a true prisoners’ dilemma: the PA may fear that if it cooperates, Palestinian terrorists will attack Israel, Israel will therefore renege on an agreement, and the U.S. will not prevent this outcome.  Terrorism also may reinforce Israel’s concern that if it cooperates, the Palestinians will defect.  Thus both actors will keep in mind an outcome which they prefer even less than mutual defection – and may refuse to cooperate in order to preclude that outcome.

          On the other hand, the presence of a third actor rules out the possibility of a “two-actor prisoners’ dilemma,”[26] and this actor’s preference ordering does not mirror those of the other two actors.  Its very existence – manifested by the persistence of Palestinian terrorism – poses a dilemma for Israel:

          In the words of an Israeli academic: “We have a problem with Arafat if willingly he does not control violence, but also if unwillingly he does not control it.”[27]  If PA moderates cannot control Palestinian terrorists, then Israel will achieve little by cooperating with the PA.  Yet if the PA does control Palestinian terrorists, Israel will achieve little by cooperating with either Palestinian actor, since both prefer mutual defection to mutual cooperation.

          The dilemma and the conflict result from this third actor’s existence – thus if conflict is the topic of discussion, the moderate PA leadership is of secondary importance at best.  Israel’s fight is not with Palestinian moderates, but rather with terrorists.  Once again, Israeli preferences conform to a prisoners’ dilemma, while the most significant Palestinian actor inclines toward deadlock.

Conclusions

          In each of the three theaters of Arab-Israeli conflict examined above, one or more actors holds a preference ordering similar to that of actors in a prisoners’ dilemma.  Yet in none of these arenas do all actors share this same preference ordering: Egypt now prefers mutual cooperation to mutual defection, but both it and Israel most prefer mutual cooperation – and an assurance game results.  Conversely, crucial Palestinian actors in both the Camp David II negotiations and in the second intifadah favored mutual defection over mutual cooperation – and deadlock ensued.  The prisoners’ dilemma is simply an inaccurate model of Arab-Israeli strife.