FUNDAMENTAL AID: U.S. FOREIGN AID AND SECULARISM IN THE ISLAMIC MIDDLE EAST AND NORTH AFRICA

vadim glukhovsky FUNDAMENTAL AID: U.S. FOREIGN AID AND SECULARISM IN THE ISLAMIC MIDDLE EAST AND NORTH AFRICA VADIM GLUKHOVSKY There are two opposing points of view on the relationship between secularism and alignment with U.S. in the Islamic Middle East and North Africa. Professor Timothy Mitchell believes that the U.S. is more closely aligned with religious countries, while Dr. Benjamin Barber believes that U.S. interests and conservative religious interests in the region are incompatible. Using U.S. foreign aid as a measure of alignment with U.S. interests, this thesis empirically analyzes the effect that the secularism of Middle Eastern and North African governments has on U.S. foreign aid distribution in the region. The results support Mitchell s argument secularism is significantly and negatively correlated to U.S. foreign aid allocation in the region, particularly with economic aid. I. Introduction The September 11 attacks on the World Trade Center and the Pentagon brought to light a topic that had already been under intense discussion for the previous half-century among political scientists: the relationship between the United States and the Islamic Middle East and North Africa (MENA). A turning point in this discussion was a 1992 article by Samuel Huntington in Foreign Affairs called The Clash of Civilizations? This piece predicted that a conflict between civilizations (defined based on culture and especially religion) would play an integral role in the politics of the 21st century. A particular clash that Huntington highlighted was between the Islamic civilization and the Western civilization, a clash rising in intensity due to the failure of Western ideas of socialism and nationalism and hence re-islamization of the Middle East. 1 If we apply Huntington s logic to historic U.S. policy in the MENA, it may appear that the United States should be closest with those governments that both share Western ideology and depend less on Islam to justify their authority. However, an analysis by Timothy Mitchell in his article McJihad posits that 1 Samuel P. Huntington, The Clash of Civilizations, 1993, Available: http://history.club.fatih.edu.tr/103%20huntington%20clash%20of%20civilizations%20 44 Spring 2010 full%20text.htm, March 19, 2009 2009.

foreign aid and secularism in the middle east and north africa [a]s a rule, the most secular regimes in the Middle East have been those most independent of the United States. 2 Mitchell s article goes even further, stating that the U.S. in fact was accustomed to working with emirs whose power depended upon strict interpretations of Islamic law. 3 This divergence of views raises an interesting question: are governments based on fundamentalist Islam more or less likely to be closely aligned with U.S. interests? If Huntington s (and particularly Barber s, discussed in literature review) theories are correct, then the answer should be that they are less likely; while Mitchell s theory would support the idea that they are more likely. Yet there is also the question of whether secularism is a significant variable in determining U.S. relations with MENA governments. In literature on U.S. foreign aid distribution, secularism is not mentioned as having any significant bearing on policy. When dealing with abstract concepts such as U.S. relations or secularism of governments, it is necessary to compress these concepts into clearly defined categories that can be quantitatively observed. Proximity of alignment with U.S. interests, or U.S. relations, will be represented by the annual amount of U.S. foreign economic and military aid received by each country. The secularism of a country s government will be represented by an index of the percentage of females enrolled in primary and secondary education. As shown in the causal model, both of these methods of analysis have solid grounding in theory and prior use. Thus the question is two-pronged. First, do traditional theories of foreign aid distribution miss a significant variable secularism when it comes to the Middle East? Second, if secularism is indeed significant, what direction does this relationship take? Examining the relationship between U.S. foreign aid to the Islamic Middle East and North Africa and the secularism of the recipient countries should give a clearer picture of the answers. I begin with an examination of existing research on the topic of secularism and foreign aid. Then, in the causal model section, I will again analyze this piece s central questions, but this time from the perspective of existing research on the subject and a detailed justification of the tools of analysis. Next, I will present a series of hypotheses that are based on the causal model, and provide a description of the data I am using. In the results section, I will give an overview of the tests I am running, and analyze the results of the regressions. Finally, I will go over possible problems with this thesis, and conclude with an overview of the questions, results, and implications for future research raised by this analysis. The results of this thesis support Mitchell s theory of U.S. relations with the MENA. The statistical tests found lower levels of secularism in a country to be significantly correlated with higher amounts of total U.S. foreign aid received, and even more significantly correlated with economic aid received from the U.S. II. Literature Review Current literature applicable to this topic tends to fall into three distinct categories. First, there is research that directly analyzes the relationship between U.S. interests and secularism on a regional level, but only using qualitative analyses. Another large set of research focuses on the distribution of U.S. foreign aid. This literature uses both qualitative and quantitative methods, but does not account for cultural variables such as secularism, and mainly provides analysis on the systematic not regional level. A third category of literature focuses on qualitatively analyzing secularism and foreign aid in a specific country. This type of research is relatively rare for MENA countries and does not present a complete picture of the U.S. relationship on a regional level. U.S. Interests and Secularism The relationship between the U.S. and MENA is explored in Timothy Mitchell s article McJihad: Islam in the U.S. Global Order. In this article, Mitchell is primarily concerned with demonstrating that the U.S. government and American oil companies have no qualms about using fundamentalist Islamic 2 Timothy Mitchell, Mcjihad: Islam in the U.S. Global Order, Social Text 20.4 (2002). 3 Ibid. journal of politics & international affairs 45

vadim glukhovsky regimes to exploit natural resources, and that in turn these regimes have no qualms about partnering with the U.S. to collect high rents. These fundamentalist Islamic movements are thus a part of economic globalization. 4 Mitchell argues that as a rule, the most secular regimes in the Middle East have been those most independent from the United States. The more closely a government is allied with Washington, the more Islamic its politics. 5 He goes on to cite examples of secular regimes that were not American allies, and numerous examples of religious regimes that are dependent on the United States. The article classifies regimes as religious if their governments claimed an Islamic authority, whether ruled by a monarch who claimed descent from the Prophet or asserting a special role as protector of the faith. 6 Mitchell explains this phenomenon by asserting that the United States is dependent on the support of conservative political regimes that use religion to justify their power. In contrast, secular regimes in the Middle East have generally been reforming regimes, basing their legitimacy on social and economic progress instead of religion. Benjamin Barber s Jihad vs. McWorld contradicts Mitchell s hypothesis, claiming that the forces of fundamentalist religion or nationalism are fundamentally opposed to the globalizing economic interests of the United States because they are based on conflicting principles. He argues that jihad is based on a sense of community, solidarity among kinsman narrowly conceived [s]olidarity is gained through war against outsiders. 7 This definition of fundamentalist Islam implies that Mitchell s theory of a closer relationship between the United States and regimes based on fundamentalist Islam is improbable, or even impossible. Indeed, Barber goes on to state that the West had secular systems constructed to put down fundamentalism and to realize Western aspirations. 8 From Barber s conclusions, we should expect Western nations such as the United States to have close historical alignments with secular, rather than Islamic, nations. Both Mitchell s and Barber s works lack a comprehensive statistical analysis to support their conclusions. U.S. Foreign Aid Distribution Due to the analytical approach I have taken, this thesis is founded on literature on United States foreign economic and military aid distribution. As such, to demonstrate that current research on this topic is missing secularism as a significant variable, it is necessary to present an overview of the current and historical thinking in this field. Older political theory concerning the distribution of U.S. foreign aid centered on the idea that aid is distributed on the basis of power politics or national security interests. In the 1977 article A Foreign Policy Model of U.S. Bilateral Aid Allocation, McKinlay and Little claim that aid is used to establish dependency relationships between the donor and recipient nations. 9 This theory is that a donor nation wants to demonstrate its commitment to a recipient nation in order to achieve a specific political goal; for instance, to deter intervention by a hostile state [in the] recipient country s affairs or to keep a recipient state within the donor state s sphere of influence. 10 The end result of this commitment and possibly its goal is the dependence of the recipient nation on the donor country s foreign aid, and on the donor country itself. By giving foreign aid, the donor country creates a relationship of unreciprocated reliance, and is usually able to exert some form of control over the recipient country. 11 This theory of foreign aid allocation rests on the assumption that the donor has a set of interests in or in common with the prospective recipient countries, that the amount of aid increases or decreases proportionately to the amount of this interest, and that the recipient countries be of lower incomes than the donor. This theory of foreign aid directly supports the assumption that U.S. foreign aid allocation is a 4 Ibid. 5 Ibid. 6 Ibid 7 Benjamin R. Barber, Jihad Vs. Mcworld (New York: Ballantine Books, 1996). 8 Ibid. 9 R. D. McKinlay and R. Little, A Foreign Policy Model of U.S. Bilateral Aid Allocation, World Politics 30.1 (1977). 10 Ibid. 11 Ibid. 46 Spring 2010

foreign aid and secularism in the middle east and north africa good proxy for the strength of a country s alignment to U.S. interests. At the same time, this model does not account for cultural phenomena such as secularism, and thus may be missing a significant element in the allocation of U.S. foreign aid in MENA if Mitchell s and Barber s theories are accurate in that regard. Gordon Donald s 1983 U.S. Foreign Aid and the National Interest presents a similar discussion, but with more concrete examples, focusing on the United States desire to establish commitment in order to exert its political influence, specifically to counter the Soviet Union by using economic aid to decrease proliferation of communist ideas. 12 A more recent and statistically significant theory of U.S. foreign aid allocation is Bueno de Mesquita and Smith s study of the selectorate model s impact on aid. Detailed in the 2003 book The Logic of Political Survival, the selectorate model is a quantitative examination of the acquisition and maintenance of national leadership, with wide-scale applications in the field of international relations. Based on the key assumption that a leader s action will be based on what is necessary to maintain his or her power, the model consists of two key variables: the winning coalition size, and the selectorate size. More specifically, the selectorate is the set of people whose endowments include the qualities or characteristics institutionally required to choose the government s leadership and necessary for gaining access to private benefits doled out by the government s leadership. 13 The winning coalition is defined as a subset of the selectorate of sufficient size such that the subset s support endows the leadership with political power over the remainder of the selectorate as well as over the disenfranchised members of the society. 14 The model presumes that a leader placates the winning collation within the selectorate by distributing goods among them, and in this way gains their support and stays in power. Ranging from zero to one, the ratio of the winning coalition and the selectorate is an indicator of the authoritarian or democratic nature of a country s political system. The closer the ratio is to zero, the more autocratic the government; the closer it is to one, the more democratic. When applied to U.S. foreign aid distribution, the selectorate model produces some interesting results. Bueno de Mesquita and Smith, in their 2007 article Foreign Aid and Policy Concessions, claim that aid is distributed based on the difficulty of buying policy concessions from recipient nations. 15 Following the assumption that a leader will do what is necessary to appease the winning coalition to stay in power, they found that as the size of the winning coalition in the recipient state increased and the size of the selectorate decreased, the amount of aid necessary to buy policy concessions increased. Income also proved significant and states with higher income required more aid for policy concessions to be achieved. Poor countries with small winning coalitions and large selectorates were shown to require the least aid to extract policy concessions. 16 While highly useful, the selectorate model of foreign aid s sacrifice [of] detail and precision means that a variable such as the level of secularism of a country s government which may be highly significant on a regional level is overlooked in the systematic analysis. 17 Nevertheless, this model offers three important control variables to use in this analysis: selectorate size, winning coalition size, and income. III. Causal Model As explained in the literature review, Timothy Mitchell s research on U.S. relations in the Middle East proposes that the level of secularism of a country s government is a key variable in determining the extent of the country s political alliance with the United States. The most secular countries are generally the least aligned with the United States; conversely, those with more fundamentalist regimes have the closest 12 Gordon Donald, U.S. Foreign Aid and the National Interest (Washington, D.C.: National Planning Association, 1983). 13 Bruce Bueno de Mesquita, Alastair Smith, Randolph M. Siverson and James D. Morrow, The Logic of Political Survival (Cambridge, Massachusetts: The MIT Press, 2003). 14 Ibid. 15 Bruce Bueno de Mesquita and Alastair Smith, Foreign Aid and Policy Concessions, Journal of Conflict Resolution 51.2 (2007). 16 Ibid. 17 Mesquita, Smith, Siverson and Morrow, The Logic of Political Survival. journal of politics & international affairs 47

vadim glukhovsky alliances. In Mitchell s words, [a]s a rule, the most secular regimes in the Middle East have been those most independent from the United States. 18 Secularism is also a key determinant of political alignments between the United States and MENA nations for Benjamin Barber. In Barber s model, the interests of jihad are in direct opposition to McWorld. That is, fundamentalist Islam in MENA countries conflicts with the globalizing interests of capitalism, of which the United States is a critical part. 19 Thus, according to Barber s model, countries with the most secular governments should enjoy the closest political alignment with the United States. While Barber and Mitchell disagree on the actual impact of secularism on the alliances between the MENA countries and the U.S., they clearly agree that it has a significant impact on the quality of these alliances. For the purpose of this analysis, the quality of these political alignments will be measured in terms of the amount of U.S. foreign aid that is distributed to each MENA country annually. The use of this metric has a solid basis in theory, since according to Bueno de Mesquita and Smith, U.S. foreign aid is traditionally viewed as an instrument of national policy allocated very much in regard to the political and strategic considerations of the donor. 20 McKinlay and Little also claim that aid is distributed with the foreign policy interests of the donor in mind and with the ultimate intention of furthering U.S. interests. 21 There is no single, widely accepted method of quantifying the secularism of a given country s government. The method used in this study is based on a 1966 article by George Helling that attempted to measure the secularism level of the Turkish government by looking at the number of females enrolled in the education system. 22 This method is based on the assumption that governments unaffiliated with fundamentalist Islam are inclined to increase female enrollment in public schools in order to reach the nation s full economic potential, or to fulfill their promises of reform. Those governments under religious influence will attempt to curtail, or at least will not encourage, female enrollment in public schools. This assumption has much support among existing research on the topic of Islam and female education. In an article in the Middle East Report, Fatima Mernissi claims that women s increasing role in the education system of Islamic countries constitutes a revolution in the Islamic concept of the state s relation to women, while [f]undamentalists are right in saying that education for women has destroyed the traditional boundaries and definitions of space and sex roles. 23 Mernissi continues, If one looks at some of the education statistics, one can understand why newly urbanized rural youth single out university women as enemies of Islam, with its tradition of women s exclusion from knowledge and decision making. 24 Her point of view is also supported by Mitra Shavarini, who wrote in the International Review of Education that the Iranian government is worried that rising education levels of women threaten the sacrosanct family structure in Islamic society. 25 In a more extreme case, the Taliban, an extreme Islamic fundamentalist group in Afghanistan and Pakistan, have ordered the closure of all girls schools in the war-ravaged Swat district, threatening to blow the schools up. 26 The justification offered by the group s leader was that [f]emale education is against Islamic teachings and spreads vulgarity in society. 27 Thus, using the percent of females enrolled in primary and secondary education makes theoretical sense in attempting to measure the influence, or lack thereof, of conservative Islamic groups on MENA governments. Current theory of U.S. foreign aid allocation does not take secularism directly into account. Rather, as outlined in the literature review, Bueno de Mesquita and Smith argue that it is the size of a country s winning coalition and selectorate as well as the country s resources that determines how much aid that country receives from the United States. Other models of U.S. foreign aid allocation, such as 18 Mitchell, Mcjihad: Islam in the U.S. Global Order. 19 Barber, Jihad Vs. Mcworld. 20 Mesquita and Smith, Foreign Aid and Policy Concessions. 21 McKinlay and Little, A Foreign Policy Model of U.S. Bilateral Aid Allocation. 22 Bozkurt Guvenc, Secular Trends and Turkish Identity, Journal of International Affairs 2.4 (1998). 23 Fatima Mernissi, Muslim Women and Fundamentalism, Middle East Report.153 (1988). 24 Ibid. 25 Mitra K. Shavarini, The Feminisation of Iranian Higher Education, International Review of Education 51.4 (2005). 26 Zahid Hussein, Taleban Threaten to Blow up Girls Schools If They Refuse to Close, The Times 2008. 27 Ibid. 48 Spring 2010

foreign aid and secularism in the middle east and north africa McKinley and Little s or Cingranelli and Pasquarello s model, have found that aid distribution is driven by national security concerns or power politics. While it is conceivable that these theories indirectly account for secularism, only Mitchell and Barber place secularism into a primary role. Thus, if Mitchell and Barber are correct in their analysis, we can suppose that current models of foreign aid ignore, or acknowledge only indirectly, secularism as a significant factor in analyzing foreign aid distribution in the Middle East and North Africa. Thus our analysis must include not just an examination of whether the effect of secularism on the amount of U.S. foreign economic and military aid is positive or negative, but also whether it has any significant effect at all. IV. Hypotheses From the causal model, we can derive several hypotheses: Hypothesis I: Countries with lower secularism ratings will be less likely to receive U.S. foreign military, economic, or total aid than those countries with higher secularism ratings. Thus, as a country s secularism rating decreases, the amount of aid received from the United States should decrease. Null Hypothesis I: Countries with lower secularism ratings are no more likely to receive U.S. foreign military, economic, or total aid then are countries with higher secularism ratings. Thus, as a country s secularism rating decreases, there is no significant increase in the amount of aid that country receives from the U.S. Hypothesis II: Countries with higher secularism ratings are less likely to receive U.S. foreign military, economic, or total aid than those countries with lower secularism ratings. Thus, as a country s secularism rating increases, the amount of aid from the United States should decrease. Null Hypothesis II: Countries with higher secularism ratings are no less likely to receive U.S. foreign military, economic, or total aid than those countries with lower secularism ratings. Thus, as a country s secularism rating increases, there is no significant decrease in the amount of aid that country receives from the U.S. Thus, the dependent variable in this study will be annual U.S. foreign economic and military aid, from 1978-1989, measured in U.S. dollars. The independent variable in this study will be the secularism rating, measured by the percentage of female students in the country, also measured from 1978-1989. V. Data Description I will use data from 1975-1989, beginning at the first point that all of the data used is readily available and ending just before the fall of the Soviet Union. This controls for any effects that this change in the balance of power may have had on U.S. foreign aid allocation. The following countries will be considered part of the Islamic Middle East and North Africa (MENA): Algeria, Egypt, Iran, Iraq, Jordan, Saudi Arabia, Kuwait, Lebanon, Libya, Mauritania, Morocco, Oman, Sudan, Syria, Tunisia, United Arab Emirates, Arab Republic of Yemen, and the People s Democratic Republic of Yemen. Those countries that lack complete, available data have been excluded from the statistical analysis. The unit of analysis is countryyear. Secularism Data (Independent Variable) journal of politics & international affairs 49

vadim glukhovsky As stated in the causal model, secularism in this thesis will be measured using the percentage of females enrolled in primary and secondary education in each country. This data was taken from the World Bank s World Development Reports for the years 1978-1992. The data for primary school education is the percentage of all girls aged 6-11 that are enrolled in primary schooling. The representative quantity may be greater than 100 if girls from ages younger or older than those listed are enrolled. The data for secondary school education, available only beginning in 1984, is for girls aged 12-17 enrolled in secondary education. Gender breakdowns for tertiary education are not available. According to the World Bank, this data was originally taken from the United Nations Educational, Scientific, and Cultural Organization (UNESCO), and may have been gathered up to two years distant from the year indicated. Due to the fact that secondary education data is available for a limited number of years, this thesis will have three variables for secularism data, analyzed in three different regressions. The first is primary education only, available from 1975-1989. The second will be secondary education only, available from 1984-1989. The third will be an index of primary and secondary education, also available from 1984-1989. U.S. Foreign Aid Data (Dependent Variable) As stated in the causal model, alignment with the United States will be measured using the amount of foreign aid each country receives from the United States each year, with more aid indicating closer alignment or relevance to U.S. interests. Data on U.S. foreign aid distribution to each country is taken directly from The United States Agency for International Development (USAID) Green Book, which lists all economic and military aid officially given by the United States to each country every year. Three variables are based on this data: economic aid, military aid, and total U.S. aid. It is measured in 2006 US dollars. Income Data (Control Variable) The total GDP of each country is taken directly from the World Bank s Data and Statistics database for the years 1975-1989. The data is in current United States dollars, and is used in this thesis to control for the effect of income on U.S. foreign aid distribution. Selectorate and Winning Coalition Size Data (Control Variable) The data measuring winning coalition and selectorate size will be drawn from the book The Logic of Political Survival, written by Bruce Bueno de Mesquita and Alastair Smith. In this book, W (winning coalition size) and S (selectorate size) are calculated for each country, and are available for the entire time period examined in this thesis. W and S will be used to control for the elements included in the selectorate (or purchasing concessions ) theory of foreign aid distribution. Estimation Equations Equation 1 (Regressions I, II, III): U.S. (Total/Econ/Military) Foreign Aid it =a+b 1 Seculari sm1 it +B 2 GDP it +B 3 W it +B 4 S it +e 50 Spring 2010 Equation 2 (Regressions IV, V, VI): U.S. (Total/Econ/Military) Foreign Aid it =a+b 1 Secularism Index it +B 2 GDP it +B 3 W it +B 4 S it +e VI. Descriptive Statistics VII. Results See Tables I and II in Appendix. Using order of least squares regressions, the effect of secularism on the amount of U.S. foreign

foreign aid and secularism in the middle east and north africa economic and military aid received by each country is displayed in Table III (see appendix). Now, let us look at secularism alone and test the hypotheses. For the first three regressions, a significant result with a negative coefficient for the Secularism1 variable would support Hypothesis I, while a significant result with a positive coefficient would support Hypothesis II. If Secularism1 is not found to be significant, the regression lends support to the null hypotheses. Regression I analyzes the relationship between the Secularism1 variable and the total amount of U.S. foreign military and economic aid received by each country. See Table IV. Regression II analyzes the relationship between the Secularism1 variable and the total amount of U.S. foreign economic aid received by each country. See Table V. Regression III analyzes the relationship between the Secularism1 variable and the total amount of U.S. foreign military aid received by each country. See Table VI. A significant result and a negative coefficient for the Secularism Index in the next three regressions would support Hypothesis I. A significant result and positive coefficient for the Secularism Index would support Hypothesis II. If Secularism Index is not found to be significant, than the regression lends support to the null hypotheses. Regression IV analyzes the relationship between the Secularism Index variable and the total amount of U.S. foreign military and economic aid received by each country. See Table VII. Regression V analyzes the relationship between the Secularism Index variable and the total amount of U.S. foreign economic aid received by each country. See Table VIII. Regression VI analyzes the relationship between the Secularism Index variable and the total amount of U.S. foreign military aid received by each country. See Table IX. VIII. Analysis Regression II exposed a clear, significant relationship between secularism (measured by the Secularism1 variable) and U.S. foreign economic aid at the p<0.05 level. This link was further supported by Regression V, which found a highly significant relationship between secularism (using the Secularism Index variable) and U.S. foreign economic aid at the p<0.01 level. Regression IV found a highly significant relationship between total U.S. foreign aid (military and economic) and the Secularism Index variable, significant at the p<0.01 level. All of the above regressions also found a negative coefficient effect for the secularism variables, implying that a decrease in secularism leads to an increase in the amount of economic or total aid. However, Regression III and VI found no significant relationship between secularism and military aid. In addition, Regression I found no significant relationship between secularism (measured by the Secularism1 variable) and total U.S. foreign aid (economic and military). These results clearly lend support to Hypothesis I and Mitchell s theory of secularism s negative role in United States foreign aid distribution in the Middle East and North Africa. IX. Problems and Suggestions for Future Study Although the results above are based on the most reliable data available, obtained entirely from governments or intergovernmental organizations, several issues present cause for concern. First, foreign journal of politics & international affairs 51

vadim glukhovsky aid distribution data is entirely self-reported by the U.S. government. This only includes official aid, and does not include aid given unofficially, the addition of which may potentially change the results. There may also be a selection bias in the female education data. Countries for which this data was incomplete or not available were eliminated from analysis. There is the potential that the inclusion of the unreported data may change the results significantly. Unfortunately, the only widely available data on female education comes from the World Bank s World Development Reports, and thus it is would difficult to obtain more complete data. Furthermore, during the time period analyzed, data on women s enrollment in secondary education was limited. Thus a study of a more recent time period may yield more reliable results, but lose a significant portion of the valuable data available for the Cold War era. Finally, while measuring secularism using the percentage of females enrolled in primary and secondary education has a basis in prior research and is feasible logically, there exists the potential that it may not capture entirely the level of secularism of a country s government. Nonetheless, with the resources available for this study, using female education data was the most logical and theoretically-sound choice for measuring secularism. Future studies can expand on the results obtained in this thesis by adding more control variables, such as natural resources or geographic location. Adding an additional measure of secularism could also help build a more complete understanding of U.S. foreign aid distribution in MENA. For example, the inclusion of a variable based on references to Islamic law in national constitutions could provide another method of quantifying the difficult concept of secularism. Finally, the results suggest that there are highly significant local determinants of foreign aid allocation. Expanding the scope of this research beyond the MENA and secularism to include local determinants in other parts of the world has the potential to provide a valuable contribution to foreign aid distribution theory. X. Conclusion This thesis began by proposing that there is a relationship between the level of secularism of governments in the Islamic Middle East and North Africa and the relations these countries enjoy with the United States. There exist two opposing arguments about this relationship. First, Mitchell argues that the more secular a government is, the less likely it is to be closely aligned with U.S. interests. Barber s argument counters that the more religious a government is, the less likely it is to be aligned with U.S. interests. Using foreign aid distribution and female education figures, this thesis statistically demonstrated the high probability of Mitchell s theory being correct: lower levels of secularism in a country were found to be significantly correlated with higher amounts of total U.S. foreign aid received. Additionally, statistical tests found secularism to be even more significantly correlated to economic aid receipts from the United States, but not significantly to military aid receipts. 52 Spring 2010 References Barber, Benjamin R. Jihad Vs. Mcworld. New York: Ballantine Books, 1996. Custers, Peter. Islamisation and More Aid for the Generals. Economic and Political Weekly 23.26 (1988): 1317-18. Donald, Gordon. U.S. Foreign Aid and the National Interest. Washington, D.C.: National Planning Association, 1983. Guvenc, Bozkurt. Secular Trends and Turkish Identity. Journal of International Affairs 2.4 (1998). Huntington, Samuel P. The Clash of Civilizations. 1993. March 19, 2009. <http://history.club.fatih. edu.tr/103%20huntington%20clash%20of%20civilizations%20full%20text.htm>. Hussein, Zahid. Taleban Threaten to Blow up Girls Schools If They Refuse to Close. The Times (2008). McKinlay, R. D., and R. Little. A Foreign Policy Model of U.S. Bilateral Aid Allocation. World Politics 30.1 (1977): 58-86. Mernissi, Fatima. Muslim Women and Fundamentalism. Middle East Report.153 (1988): 8-11.

foreign aid and secularism in the middle east and north africa Mesquita, Bruce Bueno de, and Alastair Smith. Foreign Aid and Policy Concessions. Journal of Conflict Resolution 51.2 (2007): 251-84. Mesquita, Bruce Bueno de, et al. The Logic of Political Survival. Cambridge, Massachusetts: The MIT Press, 2003. Mitchell, Timothy. Mcjihad: Islam in the U.S. Global Order. Social Text 20.4 (2002). Shavarini, Mitra K. The Feminisation of Iranian Higher Education. International Review of Education 51.4 (2005): 329-47. Data on female enrollment in primary and secondary education comes from the World Bank s World Development Reports. These can be accessed at: http://econ.worldbank.org/wdr/ Data on GDP levels comes from the World Bank s World Development Indicators. This data can be accessed at: http://www.worldbank.org/data Data on Selectorate and Winning Coalitizion size comes from Bruce Bueno de Mesquita s The Logic of Political Survival Data Source. The Stata files can be accessed at: http://www.nyu.edu/gsas/dept/politics/ data/bdm2s2/logic.htm Appendix. TABLE I: (VARIABLES USED IN REGRESSIONS I, II, AND III) Variable Obs Mean Stand. Dev. Min Max Secularism1 (% girls in primary education) U.S. Total Foreign Aid Received (in millions 2006 USD) U.S. Economic Aid Received U.S. Military Aid Received W (Winning Coalition Size) S (Selectorate Size) 164 78.09756 25.36536 9 147 164 375.5171 1020.565 0 6149.3 164 216.8293 571.115 0 2584.9 164 158.6878 523.0074 0 3564.4 164.3323171.141502 0.75 164.6768293.4592068 0 1 GDP (2008 USD) 164 3.48e+10 4.12e+10 4.76e+08 1.99e+11 journal of politics & international affairs 53

vadim glukhovsky TABLE II:(VARIABLES USED IN REGRESSIONS IV, V, AND VI) Variable Obs Mean Stand. Dev. Min Max Secularism Index (% of girls in secondary education+% of girls in primary education) U.S. Total Foreign Aid Received (in millions 2006 USD) U.S Economic Aid Received U.S. Military Aid Received W (Winning Coalition Size) S (Selectorate Size) 56 128.6429 34.96462 51 190 56 60.05893 95.67 0 319.6 56 33.81429 57.7654 0 197.3 56 26.24464 49.87968 0 200.7 56.3482143.1404423.25.75 56.7589286.415507 0 1 GDP (2008 USD) 56 4.41e+10 4.68e+10 9.10e+08 1.99e+11 TABLE III INDEPENDENT VARIABLES AND STATISTICAL INDICATORS REG I, DEP: U.S. TOTAL FOREIGN AID REG II, DEP: U.S. ECON AID REG III, DEP: U.S. MILITARY AID REG IV, DEP: U.S. TOTAL FOREIGN AID REG V, DEP: U.S. ECON AID REG VI, DEP: U.S. MILITARY AID Secularism1-4.26 (3.17) -3.64 (1.74)* -0.61 (1.68) x x x Secularism Index x x x -.96 (.31)** -0.86 (0.17)** -0.10 (.18) GDP Winning Coalition (W) -2.12e-09 (1.82e-09) 2345.31 (640.63)** Selectorate Size (S) 131.87 (207.01) -1.38e-09 (9.99e-10) 1311.76 (351.75)** 117.18 (113.66) -7.40e-10 (9.65e-10) 1033.54 (339.93)** 14.69 (109.85) -8.04e-10 (2.26e-10)** -4.34e-10 (1.26e-10)** -3.71e-10 (1.33e-10)** 7.47 (82.59) 30.35 (45.90) -22.88 (48.63) 84.02 (28.45)** 46.78 (15.81)** # of Observations 164 164 164 56 56 56 R-squared 0.15 0.18 0.08 0.39 0.48 0.22 F (4, 159)= 6.96 (4, 159)= 8.77 (4, 159)= 3.82 (4, 51)= 8.18 (4, 51)= 11.95 *Significant at p<0.05 **Significant at p<0.01 Result format: Coefficient (Std. Error) 37.24 (16.75)* (4, 51)= 3.66 54 Spring 2010

TABLE IV foreign aid and secularism in the middle east and north africa Independent Variable: Secularism1 U.S. Total Foreign Aid Received controlling for GDP, S, and W t-value -1.34 Coefficient -4.257785 R-squared 0.1490 TABLE V TABLE VI TABLE VII Independent Variable: Secularism1 U.S. Foreign Economic Aid Received controlling for GDP, S, and W t-value -2.09 Coefficient -3.638271 R-squared 0.1807 This result is significant at p<0.05 Independent Variable: Secularism1 U.S. Foreign Military Aid Received controlling for GDP, S, and W t-value - 0.37 Coefficient -.6195143 R-squared 0.0876 Independent Variable: Secularism Index U.S. Total Foreign Aid Received controlling for GDP, S, and W t-value -3.13 Coefficient -.9572163 R-squared 0.3907 TABLE VIII This result is significant at p<0.01 TABLE IX Independent Variable: Secularism Index U.S. Foreign Economic Aid Received controlling for GDP, S, and W t-value -5.06 Coefficient -.8592338 R-squared 0.4839 Independent Variable: Secularism Index This result is significant at p<0.01 U.S. Foreign Military Aid Received controlling for GDP, S, and W t-value -0.54 Coefficient -.0979825 R-squared 0.2228 journal of politics & international affairs 55