We thank Brian Job, Stephen Krasner, Eric Mlyn, T. Clifton Morgan, Stephen Walt, and two anonymous referees for their comments and suggestions. An earlier version of the article was presented at the 1990 American Political Science Association meetings in San Francisco.

1. We distinguish between societal-level variables, which have basically been ignored in the literature, and decision-making variables at the bureaucratic-organizational, small group, and individual levels, which have each received considerable attention. See Rosenau, James N., “Pre-theories and Theories of Foreign Policy,” in Farrell, R. B., ed., Approaches to Comparative and International Politics (Evanston, Ill.: Northwestern University Press, 1966), pp. 27–92Google Scholar. Contemporary historians have given far more attention to the causal importance of societal-level variables in the processes leading to war and the mobilization of resources for war. See Levy, Jack S., “Domestic Politics and War,” Journal of Interdisciplinary History 18 (Spring 1988), pp. 653–73CrossRefGoogle Scholar; and Barnett, Michael, “High Politics Is Low Politics: The Systemic and Domestic Sources of Israeli Security Policy, 1967–77,” World Politics 42 (07 1990), pp. 529–62CrossRefGoogle Scholar.

2. See Keohane, Robert O. and Nye, Joseph S. Jr, “Power and Interdependence Revisited,” International Organization 41 (Autumn 1987), pp. 725–53CrossRefGoogle Scholar; Nye, Joseph S. Jr, and Lynn-Jones, Sean, “International Security Studies: A Report of a Conference on the State of the Field,” International Security 12 (Spring 1988), pp. 5–27CrossRefGoogle Scholar; and Levy, Jack S., “The Causes of War: A Review of the Theories and Evidence,” in Tetlock, Phillip et al. , eds., Behavior, Society, and Nuclear War (New York: Oxford University Press, 1989), vol. 1, pp. 209–333Google Scholar. For recent exceptions, see Mastanduno, Michael, Lake, David, and Ikenberry, G. John, “Toward a Realist Theory of State Action,” International Studies Quarterly 33 (12 1989), pp. 457–74CrossRefGoogle Scholar; and Barnett, Michael, Confronting the Costs of War: Military Power, State, and Society in Egypt and Israel (Princeton, N.J.: Princeton University Press, forthcoming)CrossRefGoogle Scholar.

3. States may shun alliances in general because of domestically generated preferences for isolationist policies, and they may reject certain states as potential alliance partners because of ideological differences, religious considerations, or exclusionary trade or financial policies that are driven by domestic interests or by ethnic politics.

4. Although a formal treaty of alliance generally increases the probability that a commitment will be honored, it is neither a necessary nor a sufficient condition for mutual support on security issues. See Kann, Robert, “Alliances Versus Ententes,” World Politics 28 (07 1976), pp. 39–63CrossRefGoogle Scholar; Sabrosky, Alan Ned, “Interstate Alliances: Their Reliability and the Expansion of War,” in Singer, J. David, ed., The Correlates of War: II (New York: Free Press, 1980), pp. 161–98Google Scholar; and Kegley, Charles and Raymond, Gregory, When Trust Breaks Down: Alliance Norms and World Politics (Columbia: South Carolina Press, 1990)Google Scholar. Though common and operationally convenient, formal definitions of alliances—such as that offered by Singer, J. David and Small, Melvin in “Formal Alliances, 1815–1939,” Journal of Peace Research, vol. 3, 1966, pp. 1–32—are too restrictive for many purposesCrossRefGoogle Scholar. Note that our definition is similar to that offered by Walt, Stephen M. in The Origin of Alliances (Ithaca, N.Y.: Cornell University Press, 1987), p. 1Google Scholar.

5. Telhami's, ShibleyPower and Leadership in International Bargaining: The Path to Camp David (New York: Columbia University Press, 1990)Google Scholar is only the latest reflection of the general unwillingness to analyze systematically the role of state-society relations in shaping state security behavior. Telhami argues on pp. 3–4 that “precisely because [systemic] theory is at a high level of generality, because the assumed characteristics are minimal and exceptionally durable, because no other single theory at the same general level can explain more, and because the predicted effects … are verifiably dominant in the behavior of states, the theory provides a good starting point (or core) for a more general research program.”

6. See Thucydides, , The Peloponnesian War, trans. Warner, Rex (New York: Penguin Books, 1954)Google Scholar; Morgenthau, Hans, Politics Among Nations, 4th ed. (New York: Knopf, 1967), p. 175Google Scholar; Liska, George, Nations in Alliance (Baltimore, Md.: Johns Hopkins University Press, 1962), p. 12Google Scholar; Langer, William L., European Alliances and Alignments, 1871–1890, 2d ed. (New York: Vintage Press, 1964)Google Scholar; Holsti, Ole, Hopmann, P. T., and Sullivan, Michael, Unity and Disintegration in International Alliances: Comparative Studies (New York: Wiley, 1973), pp. 4–14Google Scholar; and Walt, The Origin of Alliances. On the “capability aggregation model,” see Morgenthau, Politics Among Nations; Gulick, Edward V., Europe's Classical Balance of Power (New York: Norton, 1955)Google Scholar; Schroeder, Paul W., “Alliances, 1815–1914: Weapons of Power and Tools of Management,” in Knorr, K., ed., Historical Dimensions of National Security Problems (Lawrence: University of Kansas Press, 1976), pp. 227–62Google Scholar; and Walt, The Origin of Alliances. Alliances may also be offensive in character and designed to overturn the status quo. See Osgood, Robert, “The Expansion of Force,” in Osgood, R. and Tucker, R., eds., Force, Order, and Justice (Baltimore, Md.: Johns Hopkins University Press, 1967)Google Scholar; and Levy, Jack S., “Alliance Formation and War Behavior: An Analysis of the Great Powers, 1495–1975,” Journal of Conflict Resolution 25 (12 1981), p. 590CrossRefGoogle Scholar.

7. For instance, the model generally ignores the possibility that a state may ally with another not primarily to aggregate capabilities against a common threat but to gain influence over the ally and perhaps restrain it from taking certain actions that might be contrary to the first state's interests. At the same time, the weaker ally may consciously accept a certain loss of autonomy in return for enhanced security. See Schroeder, , “Alliances, 1815–1914”; and Rothstein, Robert, Alliances and Small Powers (New York: Columbia University Press, 1968), pp. 49–50Google Scholar. Bismarck's alliances are often interpreted in this way, as Langer argues in European Alliances and Alignments.

8. On “foreign policy substitutability,” see Most, Benjamin A. and Starr, Harvey, “International Relations Theory, Foreign Policy Substitutability, and ‘Nice’ Laws,” World Politics 36 (04 1984), pp. 383–406CrossRefGoogle Scholar. States have other strategies for dealing with their external security threats—including appeasement, economic concessions and other forms of reassurance or co-optation, deterrence, territorial expansion, and, perhaps, preventive war—but we focus here on the arms-alliance trade-off and its domestic implications. For a discussion of alternative state strategies, see Lake, David, “The State and Grand Strategy,” University of California at Los Angeles, mimeograph, 1990Google Scholar.

9. This lack of systematic treatment of the trade-offs that exist is evident in such classic works as Gulick's Europe's Classical Balance of Power and Morgenthau's Politics Among Nations. In the last decade, however, many realpolitik theorists have followed the argument set forth by Waltz, Kenneth in Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979)Google Scholar. According to Waltz, “external balancing” through alliances is more common in multipolar systems, while “internal balancing” through arms production is more common in bipolar systems. But this hypothesis has yet to be empirically confirmed, and it ignores the central role of alliances in the bipolar systems of early sixteenth-century Europe and ancient Greece. See Levy, Jack, “The Polarity of the System and International Stability: An Empirical Analysis,” in Sabrosky, Alan Ned, ed., Polarity and War (Boulder, Colo.: Westview Press, 1985), pp. 41–66Google Scholar.

10. See Altfeld, Michael D., “The Decision to Ally: A Theory and Test,” Western Political Quarterly 37 (12 1976), pp. 523–44Google Scholar; McGinnis, Michael, “A Rational Model of Regional Rivalry,” International Studies Quarterly 34 (03 1990), pp. 111–35CrossRefGoogle Scholar; Morrow, James D., “Alliances and Asymmetry: An Alternative to the Capability Aggregation Model of Alliances,” working paper, Hoover Institution, Stanford University, 1990Google Scholar; Wagner, R. Harrison, “The Theory of Games and the Balance of Power,” World Politics 38 (07 1986), pp. 546–76CrossRefGoogle Scholar; and Niou, Emerson M. S., Ordeshook, Peter C., and Rose, Gregory F., The Balance of Power (New York: Cambridge University Press, 1989)CrossRefGoogle Scholar. One attempt to examine these relationships empirically is Most, Benjamin and Siverson's, Randolph M. “Substituting Arms and Alliances, 1870–1914: An Exploration in Comparative Foreign Policy,” in Hermann, Charles F. et al. , eds., New Directions in the Study of Foreign Policy (Boston: Allen & Unwin, 1987), pp. 131–57Google Scholar.

11. See Howard, Michael, War in European History (New York: Oxford University Press, 1986), chap. 2Google Scholar.

12. In Alliances and Small Powers, for example, Rothstein focuses primarily on European small powers that face direct threats from a great power and says little about alliances involving two or more small powers. See also Singer, Marshall, Weak States in a World of Powers (New York: Free Press, 1972)Google Scholar; Rothstein, Richard, The Weak in the World of the Strong (New York: Columbia University Press, 1977)Google Scholar; and Handel, Michael, Weak States in the International System (London: Frank Cass, 1981)Google Scholar. In “Explaining Third World Alignments,” World Politics 43 (01 1991), pp. 233–56CrossRefGoogle Scholar, Steven David develops the concept of “omnibalancing” and shows how Third World states attempt to balance external and internal threats. The great power and Eurocentric bias in the alliance literature is shared more generally by the American theoretical literature on international conflict.

13. See Jackson, Robert and Rosberg, Carl, “Why Africa's Weak States Persist: The Empirical and Juridical in Statehood,” World Politics 35 (10 1982), pp. 1–24CrossRefGoogle Scholar; and Nye, Joseph, Bound to Lead (New York: Basic Books, 1990), pp. 179–82Google Scholar. Weak states facing immediate security threats from great powers or other stronger states may have no choice but to give priority to security matters and to seek external support from another great power, but these great power threats to small states are not very common. See Rothstein, , Alliances and Small Powers, p. 247Google Scholar.

14. By political stability, we mean both the maintenance of state structures and the maintenance of state managers' own positions of political power. These are analytically distinct but can be combined for the purposes of this article.

15. See Mastanduno, Lake, and Ikenberry, “Toward a Realist Theory of State Action.” Like these authors, we assume that states have internal and external strategies as well as internal and external goals, and we also assume that internal strategies can facilitate the achievement of external goals and vice versa, though our conceptualization of state goals differs from theirs. Note, however, that while Mastanduno and his colleagues mention a hegemon's extraction of resources from the world economy to increase state wealth, they neglect territorial expansion and economic or military assistance through alliances as other means available to many states pursuing the same objective. (But see Lake, “The State and Grand Strategy.”) They also mention external validation of internal legitimacy through an alliance with a stronger state, but they neglect the potential autonomy costs of such alliances and make only a brief reference to the utility of the diversionary use of force to enhance internal legitimacy. See Levy, Jack S., “The Diversionary Theory of War: A Critique,” in Midlarsky, Manus, ed., Handbook of War Studies (Boston: Unwin Hyman, 1989), pp. 259–88Google Scholar.

16. See Most, and Siverson, , “Substituting Arms and Alliances,” p. 135Google Scholar. The termination of an alliance may, however, involve some diplomatic and domestic political costs. An alliance, once formed, may create a set of vested political interests within the state or society or may come to be seen as an end in itself, transcending the security interests or other purposes for which the alliance was initially established. See Holsti, , Hopmann, , and Sullivan, , Unity and Disintegration in International Alliances, pp. 34–35Google Scholar.

17. On the importance of internal security, see Stepan, Alfred, “The New Professionalism of Internal Warfare and Military Role Expansion,” in Stepan, A., ed., Authoritarian Brazil: Origins, Policies, and Future (New Haven, Conn.: Yale University Press, 1973), pp. 47–65Google Scholar; and Rouquie, Alain, The Military and the State in Latin America, trans. Sigmund, Paul E. (Berkeley: University of California Press, 1987)Google Scholar.

18. See Snyder, Glenn, “The Security Dilemma in Alliance Politics,” World Politics 36 (07 1984), pp. 461–95CrossRefGoogle Scholar.

19. See The Portable Machiavelli, ed. and trans. Bondanella, Peter and Musa, Mark (New York: Penguin, 1979), p. 153Google Scholar; and Morgenthau, , Politics Among Nations, p. 545Google Scholar.

20. Richard Rodgers and Oscar Hammerstein, The King and I; cited by Singer, in Weak States in a World of Powers, p. 274Google Scholar.

21. See Liska, Nations in Alliance, chap. 5; Rothstein, Alliances and Small Powers, chap. 8; and McGinnis, “A Rational Model of Regional Rivalry.”

22. See Singer, , Weak States in a World of Powers, pp. 303–6Google Scholar. On the other hand, the allocation of increased domestic resources to the military establishment would also increase its leverage and the risk of political destabilization. See Handel, , Weak States in the International System, p. 81Google Scholar.

23. Rothstein, The Weak in the World of the Strong, part 3.

24. See Knorr, Klaus, Military Power and Potential (Lexington, Mass.: D.C. Heath, 1970)Google Scholar; Gilpin, Robert, War and Change in World Politics (Cambridge: Cambridge University Press, 1981)CrossRefGoogle Scholar; and Kennedy, Paul, The Rise and Fall of the Great Powers: Economic Change and Military Conflict from 1500 to 2000 (New York: Random House, 1987)Google Scholar. Hence, it is useful to distinguish between the extraction of existing resources and the mobilization of new resources. See Mastanduno, Lake, and Ikenberry, “Toward a Realist Theory of State Action.”

25. Rothstein, , The Weak in the World of the Strong, p. 183Google Scholar.

26. See ibid., chap. 6; and Rothstein, Richard, “National Security, Domestic Resource Constraints, and Elite Choices in the Third World,” in Deger, S. and West, R., eds., Defense, Security, and Development (New York: St. Martin's Press, 1987), pp. 140–58Google Scholar. On the importance of the distribution of material benefits, see Clapham, Craig, Third World Politics (Madison: University of Wisconsin Press, 1984)Google Scholar.

27. See Organski, A. F. K. and Kugler, Jacek, The War Trap (Chicago: University of Chicago Press, 1980), chap. 2Google Scholar; and Lamborn, Alan, “Power and the Politics of Extraction,” International Studies Quarterly 27 (06 1983), pp. 125–46CrossRefGoogle Scholar.

28. See Huntington, Samuel, The Common Defense (New York: Columbia University Press, 1961), p. 1Google Scholar. This point is fully developed by Barnett in “High Politics Is Low Politics.”

29. See Eshag, Eprime, Fiscal and Monetary Problems in Developing Countries (New York: Cambridge University Press, 1983)Google Scholar; Snider, Lewis, “Identifying the Elements of State Power: Where Do We Begin?” Comparative Political Studies 30 (10 1987), pp. 314–56CrossRefGoogle Scholar; and Due, John, Indirect Taxation in Developing Countries (Baltimore, Md.: Johns Hopkins University Press, 1988)Google Scholar.

30. A direct tax is one “collected directly (or via withholding) from the persons expected to bear the burden of the tax, in the sense of a reduction of real income,” whereas an indirect tax is one “collected from persons other than those expected to bear the burden.” See Due, , Indirect Taxation in Developing Countries, p. 19Google Scholar.

31. See Neuman, Stephanie, “International Stratification and Third World Military Industries,” International Organization 38 (Winter 1984), p. 185CrossRefGoogle Scholar.

32. Although Prussia is a classic example of a militarized society in which domestic welfare is sacrificed for the sake of the military power of the state, Michael Howard notes that King Frederick William I (1713–40) organized the Prussian army “with infinite care to impose the least possible strain on the fragile economy of his lands.” See Howard, , War in European History, p. 69Google Scholar. The fear of the revolutionary potential of mass armies and the conviction that the Napoleonic style of warfare was more conducive to overturning the system of European states than to preserving it were the primary reasons that at the end of the Napoleonic wars most great powers reverted to the eighteenth-century pattern of “aristocratic officers and long-serving professional troops kept isolated from the rest of the community.” See Howard, ibid., p. 84. See also Preston, Richard and Wise, Sydney, Men in Arms, 4th ed. (New York: Holt, Rinehart, & Winston, 1979), pp. 203–9Google Scholar; and Craig, Gordon, The Politics of the Prussian Army, 1640–1945 (London: Oxford University Press, 1955), pp. 81 and 85Google Scholar. That is, the army was seen as an instrument of the dominant class for the maintenance of social order as much as a Clausewitzian instrument of national policy.

33. On conscription, mercenarism, and state legitimacy, see Giddens, Anthony, The Nation State and Violence (Berkeley: University of California Press, 1985), pp. 233–34Google Scholar; Beukema, Henry, “The Social and Political Aspects of Conscription: Europe's Experience,” in Anderson, M., ed., The Military Draft (Stanford, Calif.: Hoover Press, 1982), pp. 479–91Google Scholar; Thomson, Janice, “State Practices, International Norms, and the Decline of Mercenarism,” International Studies Quarterly 34 (03 1990), pp. 23–47CrossRefGoogle Scholar; and Tilly, Charles, Capital, Coercion, and European States (London: Basil Blackwell, 1990), pp. 80–84Google Scholar. See also Janowitz, Morris, “Military Institutions and Citizenship in Western Societies,” in Haries-Jenkins, G. and Van Dooms, J., eds., The Military and the Problem of Legitimacy (Beverly Hills, Calif.: Sage Press, 1976), pp. 127–30Google Scholar.

34. See Stepan, “The New Professionalism of Internal Warfare and Military Role Expansion.”

35. In “Explaining Third World Alignments,” Steven David refers to this as an act of “omnibalancing.” In Capital, Coercion, and European States, pp. 207–8, Tilly argues that military aid allows rulers to “bypass bargaining with their subject population” for access to the means of war making. But by turning to external support, they lose the opportunity to generate domestic consent and are likely to be viewed by their subjects as less legitimate than before.

36. We begin in 1962 because that year inaugurated a relatively stable period in Egyptian politics. It followed the dissolution of the United Arab Republic and the sweeping nationalizations of 1961, and it preceded Egypt's entry into the war in Yemen.

37. See Walt, The Origins of Alliance; and Telhami, Power and Leadership in International Bargaining. Telhami in particular goes to great lengths to distance himself from a body of Egyptian historiography that argues that Egyptian foreign policy was significantly affected by domestic political and economic factors.

38. Kerr, Malcom, The Arab Cold War (New York: Oxford University Press, 1971)Google Scholar.

39. Ibid.

40. The wave of nationalizations was a reaction to the industrial elites' lack of investment despite the tremendous enticements offered by the state. It also represented an effort to protect the Nasser regime from potential enemies after the dissolution of the United Arab Republic in 1961. See Baker, Raymond, Egypt's Uncertain Revolution Under Nasser and Sadat (Cambridge, Mass.: Harvard University Press, 1978)CrossRefGoogle Scholar.

41. See Waterbury, John, The Egypt of Nasser and Sadat: The Political Economy of Two Regimes (Princeton, N.J.: Princeton University Press, 1983), pp. 93–100CrossRefGoogle Scholar; and Sadat, Anwar, In Search of Identity (New York: Harper, 1978), p. 213Google Scholar. After reaching a high of £E 83 million in 1963–1964, net external borrowing declined to a low of £E 57 million in 1966–1967. According to Waterbury, p. 114, this corresponded to a general decline in the net deficit.

42. The Muslim brotherhood's protests in 1965 did produce some concern and a wave of domestic repression by the government.

43. This arrangement did not translate immediately into a greater level of commitment by the Soviets. After the Anglo–French attack on the Suez Canal, for example, the Soviets informed Nasser that while they admired Egypt's courageous resistance, the only thing they were willing to mobilize for the Egyptians was world opinion. See Heikal, Mohammed, The Sphinx and the Commissar (New York: Harper & Row, 1978), pp. 68–71Google Scholar.

44. See ibid., pp. 148–71 and 183; and Dawisha, Karen, Soviet Foreign Policy Towards Egypt (London: Macmillan, 1979), pp. 31 and 43CrossRefGoogle Scholar.

45. Hinnebusch, Raymond, Egyptian Politics Under Sadat: The Post-populist Development of an Authoritarian State (New York: Cambridge University Press, 1986), p. 35Google Scholar.

46. Ajami, Fouad, The Arab Predicament (New York: Cambridge University Press, 1981), p. 85Google Scholar.

47. Stein, Janice Gross, “Calculation, Miscalculation, and Conventional Deterrence: The View from Cairo,” in Jervis, R. et al. , eds., Psychology and Deterrence (Baltimore, Md.: Johns Hopkins University Press, 1985), p. 52Google Scholar.

48. See Walt, , The Origin of Alliances, p. 108Google Scholar; and Rubenstein, Alvin Z., Red Star on the Nile (Princeton, N.J.: Princeton University Press, 1977), pp. 71–73Google Scholar.

49. See Rubenstein, , Red Star on the Nile, pp. 47–48Google Scholar; and Vatikiotis, P. J., Nasser and His Generation (New York: St. Martin's Press, 1978), p. 185Google Scholar. Vatikiotis observes that “a common popular complaint in those days was ‘We accept the regime did not do much for us, but being a military one, it could at least have given account of itself in the war against Israel.’”

50. Baker, , Egypt's Uncertain Revolution Under Nasser and Sadat, p. 136Google Scholar.

51. Efrat, Moshe, “The Defence Burden in Egypt During the Deepening of Soviet Involvement in 1962–72,” Ph.D. diss., London University, 05 1981, p. 152Google Scholar.

52. Waterbury, , The Egypt of Nasser and Sadat, p. 225Google Scholar.

53. Hinnebusch, Egyptian Politics Under Sadat.

54. See al-Edel, Reda, “Impact of Taxation in Income Distribution: An Exploratory Attempt to Estimate Tax Incidence in Egypt,” in al-Khalek, G. Abd and Tignor, R., eds., The Political Economy of Income Distribution in Egypt (New York: Holmes & Meier, 1982), p. 142Google Scholar; and Cooper, Richard, The Transformation of Egypt (Baltimore, Md.: Johns Hopkins University Press), p. 40Google Scholar.

55. See Ajami, , The Arab Predicament, p. 91Google Scholar. Arms transfers and financial assistance data are generally not available on an annual basis, as Walt, notes in The Origin of Alliances, p. 219, fn. 3Google Scholar. There are a number of reasons to question the reliability of the data that are available. First, the post-1967 defense costs were placed in three different areas: the regular budget, the supplemental budget, and the emergency budget. Annual data on the emergency budget, which probably housed the majority of the post-1967 defense burden, are generally unavailable. For instance, in The Defense Burden in Egypt, p. 122, Efrat, who probably has the most reliable figures, says little more than that the burden rose from £E 60 million in 1967 to £E 399 million in 1973. Second, calculations regarding Soviet assistance would have to take into account not only the number of arms transferred but also the terms of the transfers. Egyptian Chief of Staff Saad al-Shazli argues in The Crossing of the Suez (San Francisco: American Middle East Research, 1980)Google Scholar that it is anyone's guess how much assistance actually arrived. Third, Arab financial assistance was transferred through a variety of Egyptian locales, some of which were more visible than others. And, fourth, Saudi Arabia and other oil-rich Arab states began purchasing Soviet weapons to supply the Egyptians after 1972, and this adds to the difficulties of obtaining accurate data.

56. On Nasser's new realism, see Ajami, , The Arab Predicament, p. 86Google Scholar; and Ansari, Hameid, Egypt: The Stalled Society (Albany, N.Y.: SUNY Press, 1987), p. 150Google Scholar. Ansari suggests that Nasser's change in attitude led to his acceptance of the Rogers Plan to end the war of attrition and also led to his backing of the Jordanian government's efforts against the Palestine Liberation Organization in September 1970.

57. See Rubenstein, , Red Star on the Nile, p. 38Google Scholar; and Heikal, Mohammed, The Road to Ramadan (London: Collins Press, 1975), p. 52Google Scholar. The Khartoum Agreement, however, only covered approximately one-half of Egypt's capital imports in the period immediately following the 1976 war, so the annual net loss was still from $165 million to $185 million. See Efrat, , The Defense Burden of Egypt, p. 152Google Scholar.

58. Cited by Rubenstein, in Red Star on the Nile, p. 102Google Scholar.

59. Efrat, , The Defense Burden of Egypt, p. 158Google Scholar.

60. Egypt's military industries had fallen on hard times during the 1960s and, as a result, were limited to production for the civilian economy and to the maintenance of imported weapons. See Efrat, , The Defense Burden of Egypt, p. 16Google Scholar; and Stork, Joe, “Arms Industries in the Middle East,” MERIP, 01–02 1987, p. 13Google Scholar.

61. In Red Star on the Nile, Rubenstein claims that Nasser was able to minimize the loss of autonomy. In the post-1967 period, defense outlays rose nearly ninefold in terms of current dollar amounts. See Dessouki, Ali and al-Labban, Adel, “Arms Race, Defense Expenditures, and Development: The Egyptian Case, 1952–1973,” Journal of South Asian and Middle Eastern Studies 4 (Spring 1981), pp. 69–70Google Scholar. This increase in defense spending paralleled the rise in military assistance. For instance, the value of weapons transferred to Egypt rose from about $424 million in the 1963–66 period to $2,231 billion in the 1967–73 period. See Arms Control and Disarmament Agency (ACDA), World Military Expenditures and Arms Transfers (Washington, D.C.: Government Printing Office, various years)Google Scholar.

62. Heikal, , The Sphinx and the Commissar, p. 191Google Scholar.

63. During the 1967 war, Egypt lost 80 percent of its military weapons to the Israelis. The Soviets replaced these weapons free of charge. Other weapons received a discount of 50 percent, with a 10-to 15-year payback period, low interest rates (usually 2.5 percent per annum), and repayable in Egyptian currency. This continued the terms established under the first Soviet–Egyptian weapons agreement in 1955, in which Egypt financed its weapons partly by exporting cotton and other agricultural goods and partly by long-term debt financing at low interest rates. See Efrat, , The Defense Burden of Egypt, pp. 34, 36, and 95Google Scholar. When Sadat took office in 1971, Egypt owed the Soviets $380 million in nonmilitary and $1.7 billion in military debt. See Heikal, Mohammed, Autumn of Fury (New York: Random House, 1983), pp. 86–87Google Scholar.

64. Heikal, , The Road to Ramadan, pp. 46–47Google Scholar.

65. See Rubenstein, , Red Star on the Nile, p. 22Google Scholar; Efrat, The Defense Burden of Egypt; and Karawan, Ibrahim, “Egypt's Defense Policy,” in Neuman, S., ed., Defense Planning in Less Industrialized States (Lexington, Mass.: Lexington Books, 1984), pp. 147–65Google Scholar.

66. See al-Shazli, , The Crossing of the Suez, p. 41Google Scholar; and Efrat, , The Defense Burden of Egypt, p. 148Google Scholar.

67. Rubenstein, , Red Star on the Nile, pp. 107–8Google Scholar.

68. See ibid., p. 18; and Dawisha, , Soviet Foreign Policy Towards Egypt, p. 46Google Scholar. The Soviet reinforcements did not come cheap. The Egyptians had to buy the equipment that they were to use on Egyptian territory, provide them with food and field clothing, and cover their salaries by paying Moscow the equivalent of 150 pounds sterling per soldier and 170 pounds sterling per officer. See al-Shazli, , The Crossing of the Suez, p. 138Google Scholar.

69. Dawisha, , Soviet Foreign Policy Towards Egypt, p. 47Google Scholar.

70. Telhami, Power and Leadership in International Bargaining.

71. An alternative explanation for the tightened alliance pattern might point to the idiosyncratic variables of Sadat's personal belief systems, risk orientation, and bargaining strategy. However, we side with Waterbury, Hinnebusch, Ansari, and Ajami, all of whom emphasize the continuity of Egyptian policy from Nasser to Sadat and the importance of domestic constraints over the role of personality and other idiosyncratic factors. According to Ajami, “Throughout it all Egypt's path has been navigated by two men, and there is a temptation to see the choices the two men made as idiosyncratic, personal ones. … The temptation to go after the personalities of the two “kings” in order to explain Egypt's path must be checked, for there were constants that both men had to deal with: (1) an unacceptable military defeat that both men had to try to break out of; (2) a revolutionary legacy that had generated a great deal of noise and that now had to come to terms with the world.” See Waterbury, The Egypt of Nasser and Sadat; Hinnebusch, Egyptian Politics Under Sadat; Ansari, Egypt: The Stalled Society; and Ajami, , The Arab Predicament, p. 83Google Scholar.

72. Dawisha, , Soviet Foreign Policy Towards Egypt, p. 61Google Scholar.

73. For instance, Egypt's overall deficit went from £E 194 million in 1971 to £E 315 million in 1972, and these figures do not include the bulk of the defense appropriations. See Ikram, Khalid, Egypt: Economic Management in a Period of Transition (Baltimore, Md.: Johns Hopkins University Press, 1980)Google Scholar. The observation that discontent was rising is supported by several historical narratives, including Ansari's Egypt: The Stalled Society, Rubenstein's Red Star on the Nile, and Shoukri's, GhaliEgypt: Portrait of a President (London: Zed Press, 1981)Google Scholar, and by the autobiographical accounts of Egyptian officials, including al-Shazli's The Crossing of the Suez, Heikal's The Road to Ramadan, and Heikal's The Sphinx and the Commissar. While Sadat acknowledged the domestic discontent, he attributed it to either Soviet-backed plots or journalists who were attempting to “create a sense of instability … in the country.” See Sadat, , In Search of Identity, pp. 234 and 245Google Scholar. Whatever the source of discontent, the lasting impression is that Egyptian officials felt themselves to be under greater societal constraints and pressures after 1971.

74. Shoukri, , Egypt: Portrait of a President, pp. 93–101Google Scholar.

75. See Ansari, Egypt: The Stalled Society. While Nasser attempted to control the masses, occasionally using them against the upper classes and personally dispensing and administering his view of their interests, Sadat saw the bourgeoisie as a positive force, both in the economy and the polity, and tended to view with some hostility the demands of the masses. However much Sadat attempted to portray himself as defender and protector of the common people, the institutional mechanisms used by the masses to reach Nasser rarely reached to the upper branches of decision making under Sadat. See Hinnebusch, , Egyptian Politics Under Sadat, pp. 226–27Google Scholar.

76. See Rubenstein, , Red Star on the Nile, p. 180Google Scholar; Abdalla, Ahmed, The Student Movement and National Politics in Egypt (London: Al-Saqi Books, 1985), p. 190Google Scholar; Heikal, , The Road to Ramadan, p. 20Google Scholar; and Stein, , “Calculation, Miscalculation, and Conventional Deterrence,” p. 54Google Scholar.

77. See Levi, Margaret, Of Rule and Revenue (Berkeley: University of California Press, 1988)Google Scholar. According to Levi, a ruler wishing to extract revenue must convince the population that the promised goods will be delivered, and the failure to deliver will create difficulties for future revenue extraction. This analysis shifts our attention away from the personality of Sadat and toward the more enduring problems that confronted Egyptian rulers in general.

78. See Rubenstein, Red Star on the Nile; and Ansari, , Egypt: The Stalled Society, p. 175Google Scholar. Ansari argues that Sadat's efforts to bolster his sagging support included the increased use of religious symbols and the encouragement of the formation of Islamic groups.

79. The reason behind the weapons policy shift may lie with the detente between the United States and Soviet Union, which meant that the Soviets no longer had to bargain as hard for allies. Moreover, the Kremlin's belief was that the Saudis and Libyans would pay for the weapons in hard currency, which the Soviets could then use to pay for their imports of Western technology and food. Because of these new conditions, Egypt was forced to give up its purchase of some weapons (which were later made up by Saudi Arabian and Kuwaiti assistance). See al-Shazli, , The Crossing of the Suez, p. 143Google Scholar.

80. Egypt's military establishment coordinated the Soviet departure in a manner that minimized the possible damage to its battle plans. The Soviets who operated Soviet equipment for which Egypt had no substitute were allowed to stay, provided that they remained under Egyptian command. See al-Shazli, , The Crossing of the Suez, pp. 164–65Google Scholar.

81. See Sadat, , In Search of Identity, pp. 204–32Google Scholar. In Egypt: The Stalled Society, pp. 176–77, Ansari, offers a similar interpretationGoogle Scholar. The explanation offered by Hafez Ismail (Sadat's national security adviser) in an interview with Barnett on 3 January 1991 in Cairo also points to systemic factors: Sadat was intent on sending a clear signal to the Americans that he was willing to reopen his dialogue with them.

82. See Heikal, , The Road to Ramadan, pp. 171–77Google Scholar; and Vayrynen, Raimo and Ohlson, Thomas, “Egypt: Arms Production in a Transnational Context,” in Brzoska, Michael and Ohlson, Thomas, eds., Arms Production in the Third World (Philadelphia: Taylor & Francis, 1986), p. 108Google Scholar.

83. Ansari, , Egypt: The Stalled Society, p. 176Google Scholar.

84. Rubenstein, , Red Star on the Nile, p. 196Google Scholar.

85. Ibid., pp. 241–42.

86. Safran, Nadav, Saudi Arabia: The Ceaseless Quest for Security (Cambridge, Mass.: Harvard University Press, 1985), p. 148Google Scholar.

87. See Walt, , The Origin of Alliances, p. 117Google Scholar; al-Shazli, , The Crossing of the Suez, pp. 140–43 and 157Google Scholar; and Rubenstein, , Red Star on the Nile, p. 235Google Scholar.

88. ACDA, World Military Expenditures and Arms Transfers, various years.

89. Sadat, , cited by Dawisha, in Soviet Foreign Policy Towards Egypt, p. 65Google Scholar.

90. Sadat, , In Search of Identity, pp. 246–47Google Scholar.

91. Although this might be more true for financial assistance than for weapons, the Egyptians were already experiencing a decline in weapons sales.

92. See Dawisha, , Soviet Foreign Policy Towards Egypt, p. 64Google Scholar.

93. Rubenstein and Telhami both conclude that Sadat's maneuver involved relatively low risks. Heikal argues that the Soviets perceived the situation in precisely these terms: “The military … argued repeatedly in the Politburo that there was no easy way out, and that the flow of military aid to the Arabs must be stepped up.” See Rubenstein, , Red Star on the Nile, p. 199Google Scholar; Telhami, , Power and Leadership in International Bargaining, p. 68Google Scholar; and Heikal, , The Sphinx and the Commissar, p. 253Google Scholar. Moreover, according to Hafez Ismail (interviewed by Barnett in Cairo on 3 January 1991), the military's battle plans had by this point shifted from a strategy designed to recapture the entire Sinai to one intended to achieve a limited military victory by establishing an Egyptian presence on the East Bank of the Suez. This goal was within reach without a major infusion of Soviet arms into the Egyptian arsenal.

94. We do not deny that idiosyncratic variables played a role here, but we argue that they affected policy means rather than goals. Given the domestic economic and political constraints at the time, any Egyptian leader would have been forced into greater reliance on external actors for badly needed resources. The particular strategy selected by Sadat may have been influenced by his own belief systems, risk orientation, and bargaining strategy, but it would not have been put into play at all in the absence of domestic pressures. Thus, idiosyncratic variables probably played a role, but only through their interaction with domestic variables. That is, domestic pressures and idiosyncratic variables were individually necessary and jointly sufficient factors in the Egyptian eviction of the Soviets in 1972.

95. In addition to these street protests, there was a renegade group of military officers intent on arresting the top Egyptian leadership, including Sadat. In The Crossing of the Suez, pp. 192–95, al-Shazli, attributes this planned revoltGoogle Scholar, which was aborted, to the situation of no war and no peace.

96. See al-Shazli, , The Crossing of the Suez, pp. 71, 75, and 207Google Scholar.

97. According to Hafez Ismail (interviewed by Barnett in Cairo on 3 January 1991), key members of the Egyptian high command had been reluctant to include Arab forces for two reasons. First, the Egyptians wanted to feel as if they alone brought about the coming victory. Second, those Arab leaders who might send their troops to Egypt wanted the request for forces to come immediately prior to war initiation, which obviously would have undercut the military's surprise attack strategy.

98. See al-Shazli, , The Crossing of the Suez, pp. 106 and 277–79Google Scholar.

99. Sadat, , cited by Rubenstein, in Red Star on the Nile, p. 282Google Scholar.

100. Shukrallah, Hani, “Political Crisis/Conflict in Post-1967 Egypt,” in Tripp, Charles and Owen, Roger, eds., Egypt Under Mubarak (New York: Routledge, 1989), p. 70Google Scholar.

101. See Heikal, , The Road to Ramadan, p. 20Google Scholar. Indeed, Egypt received $500 million in aid immediately after war began. See Rubenstein, , Red Star on the Nile, p. 282Google Scholar.