That discovery was made in 1935 in Dhahran, and commercial exploitation began in 1938. By 1939, the once-impoverished kingdom was receiving an annual royalty of about £200,000 in gold.81 British, Italian, German, and Japanese oil companies soon began jostling for the favor of Ibn Saud as well. But the old desert warrior was no modern statesman and knew little about state finances and budgetary matters. By the time of his death in 1953, the country was once again on the verge of bankruptcy. Saud, his successor, was hardly more successful in his endeavors, and not until the appointment of his brother, Crown Prince Faisal, as the finance minister did the country’s finances assume a semblance of order.
The importance of oil in shaping the contemporary political history of Saudi Arabia cannot be overemphasized. Throughout the 1940s and 1950s, Aramco’s influence inside the kingdom continued to grow, to the extent that in 1948 the U.S. Department of State became concerned about the company’s assumption of extraordinary powers in dealing with the Saudi monarch and his ministers.82 Dhahran, with its ever-increasing population of American oil workers, was also beginning to look more and more like an American city—the Bakersfield of Arabia—containing two townships that each housed about five thousand Americans and had almost everything American. But the company’s Arab workers, mostly having only recently been introduced to machinery of any sort, lived in a different world, underpaid and often maltreated by their American foremen.83 By one estimate, this resulted in a turnover rate of some 75 percent between 1945 and 1960.84 Nevertheless, their purchasing power increased, as did that of Ibn Saud.
By the time of Ibn Saud’s death in 1953, Saudi Arabia had finally attained internal security and political stability. Politically, however, little development occurred, as absolute power resided first with the kingdom’s founder and then with his less able son, Saud. In fact, the once austere and frugal reign of the dynasty’s founder had degenerated into corruption and vice near the end, with broken-down Cadillacs strewn across the desert by princes who didn’t know how to or didn’t care to have them repaired.
Saud’s reign brought few changes to the ways of the royal court. His rule coincided with a time of profound disquiet in the Middle East—the Nasserite interlude of the 1950s and 1960s—and through much of that period he effectively entrusted the running of the country to his brother, Crown Prince Faisal. The royal family eventually deposed Saud in November 1964. In the reign of Faisal (r. 1964–75) a concerted effort was made to bring about political institutionalization by creating a modern bureaucracy and introducing procedural formality into the affairs of the state. As the finance minister during his brother’s reign, Faisal had also put the dynasty’s financial house in order earlier. Only then, by the mid-1960s, could Saudi rule be considered to have become consolidated.85
The Saudi Arabia of today is unrecognizable compared to the one Ibn Saud left behind, for it has made economic and industrial leaps and bounds that even the most optimistic assessments could not have imagined. But Ibn Saud’s political legacy continues to loom large in the system he left behind. That the country’s very name reflects the family’s last name represents the degree to which the Saud family permeates the life of the country not only politically but also economically, culturally, and socially. Despite the increasing differentiation of political roles within the Saudi system and the steady involvement of professional technocrats and other qualified nonroyals at the higher levels of the bureaucracy, politics in Saudi Arabia remains essentially a family venture. The state that once belonged to one man, Abdel Aziz, now belongs to a whole family, the Saudis. Absolute family rule and all that goes with it—internal conflicts, palace intrigues, patrimonial politics, potential narrowing of the political base—are still very much features of Saudi Arabian politics.
The end of the Ottoman era brought with it a fundamental redrawing of the map of the Middle East, resulting in the creation of a host of new national entities. But the termination of Istanbul’s imperial control did not necessarily mean that indigenous, national forces could now assert themselves, at least not for another twenty years or so. Even before the Ottomans had died, Britain and France had begun a contest for the spoils of the Middle East, carving it up into respective protectorates with little understanding of or regard for what the locals wanted. Thus began the era of European imperialism in the Middle East. France established protectorates in the Levant and the Maghreb, going so far as to declare Algeria an integral part of its territory and seeking to assimilate its population into its culture. British protectorates were established in Palestine and Transjordan, and Britain maintained effective suzerainty over Egypt, Iraq, and the countries of the Arabian peninsula. Under the direction and protection of Britain and France, new countries were given shape and new political systems were engineered. Eventually, again under the watchful eyes of the British, Palestine and the Kingdom of Hijaz ceased to exist. National identities and nationalist sentiments were not long to follow, as were sharp reactions to European machinations and dominance.
This chapter highlighted developments in Turkey, Iran, and Saudi Arabia in the formative decades of the 1920s, 1930s, and 1940s. In each, the dismantling of the old order brought with it a new, radically different political system. In Iran and Turkey, the new order was composed of self-declared promoters of modernity and industrialization. In each country, the modernizers sought to destroy archaic social and cultural forces that they saw as inimical to progress toward modernity. In Turkey the changes went the furthest, with the state embarking on a concerted campaign to secularize society that included the introduction of a new alphabet and a new calendar. In Iran, however, partly because of the limits of his own agendas—some might say the limits of his intellectual horizon—and the continued powers of the conservative clergy, Reza Shah’s “reforms” were not as far-reaching, though still substantial and significant.
In Saudi Arabia at the start of the 1920s, state building was at a far more embryonic stage than in Turkey or in Iran. In fact, while Abdel Aziz ibn Saud built a “state,” he did not start its construction until after he had unified Arabia in the early 1930s, and even then he built only a skeletal one—a more urban and urbane version of what he had earlier ruled as a tribal warrior. What occurred in the Arabian deserts in the 1910s and 1920s was, more accurately, a process of “nation building,” in which a series of conquests and alliances brought disparate tribes under one, increasingly national umbrella. Only after a measure of national cohesion had been accomplished could the construction of a state apparatus be started, a task that the first two Saudi monarchs undertook only haltingly.
Nevertheless, all three political systems discussed here had an important feature in common, and still do: they were created, maintained, and nurtured by one individual, someone who was not just a founding father but a personality larger than life, the embodiment of everything political. From the very beginning, each of these systems, like many others in the Middle East before and after them, was a personal creation, and, more importantly, a personal possession. Individuals came to personify systems; politics was relegated to the domain of personal relations; and institutions assumed only a secondary importance, to be bent and shaped in whatever way the nation’s father willed. Such was Middle Eastern politics from the 1920s to the 1940s, and its essence, as will be seen later, would change little for more than a half century afterward.
Just as the Great War fundamentally altered the political geography of the Middle East, so did the Second World War. Beginning in the second half of the 1940s, mandatory protectorates were given their formal independence, and a new country, the state of Israel, was born. Thus ensued one of the most traumatic phases in the life of the contemporary Middle East. The next chapter turns to that era.
3The Age of Nationalism
In the Middle East, as elsewhere, nationalism has been a powerful force shaping the destiny and character of peoples and countries. Although most conventional accounts of nationalism in the Middle East trace its genesis back to the mid-ninete
enth century, it was in the 1940s and the 1950s that nationalism became what it has been ever since, one of the most dominant forces—if not the most dominant force—in the region’s politics.
Enormous scholarly energy has been spent on defining nationalism and exploring the causes of its birth, and what follows here is of necessity brief and general.1 For the purposes of this chapter, I take nationalism to mean simply attachment on a national scale to a piece of territory, reinforced by common bonds of identity such as shared symbols, historical experiences, language, folklore, and whatever else creates a sense of commonality. At times, these common bonds include religion. This conception of nationalism has two important elements. First, there must be a definite territorial frame of reference, a piece of geography toward which a sense of attachment and loyalty is directed. This may be a result of economic ties to the land, its products having served as a source of livelihood for successive generations, or it may be more immediate and primordial, resulting from the need for shelter, personal security, and the sanctity of one’s private household. Second, this territorial attachment needs to become national in scope, a transformation often achieved only through active ideological, political, and at times even military agitation on the part of political leaders and states.
The first element of nationalism, identification with a piece of territory for economic and/or personal reasons has been a feature of human societies from the beginning of settled life, when the ability to own or at least to live and work on land became a central feature of daily living. By itself, however, this sense of attachment to land, rooted in necessity, is parochial and localized, limited in scope to units that can conceivably be as few as one or two families. What is essential is for such an attachment to become national in scope, embodying individuals not only in isolated pieces of territory but in an organically and emotionally linked territorial entity that contains various towns and cities. The organic and emotional links are reinforced by shared symbols and experiences and by other similar bonds of commonality. In other words, a nation needs to have been formed, or to at least be in the process of formation, for attachment to territory to be enhanced in scope and transformed into nationalism.
This sense of nationhood, or “becoming national,” emerges out of a variety of developments.2 Benedict Anderson traces it to the birth of “print capitalism” in Europe, first in Latin and then in more local vernaculars, and the emerging “possibility of a new imagined community, which in its basic morphology set the stage for the modern nation.”3 Similarly, Ernest Gellner maintains that the spread of industrial social organization creates a certain level of homogeneity throughout society and in cultural norms, thus resulting in the emergence of the phenomenon of nationhood and, consequently, nationalism.4 Similar material and cultural developments facilitated the social construction of nationality in the non-Western world as well, although here the deliberate role of individual personalities, whose resistance to colonial domination was often inspired in the name of a nation, was also important. Often either explicitly or in less conscious ways, these individuals mobilized people in the various cities and regions who already shared certain historical experiences and sociocultural characteristics.
In doing so, these emergent national leaders needed forums and institutions to spread their unifying messages more effectively, at times coercively. The forums often served as embryonic components of a state, through which the national project was formulated among an elite, then articulated for the masses and upheld against challenges from within and from the outside. The phenomenon of the state, therefore—or, for stateless nations, protostate organizations such as national liberation parties—is central to the development and spread of nationalist sentiments. Centrality of the state became all the more crucial in the early twentieth century, when several multinational empires—most notably the Russian, German, Austro-Hungarian, and Ottoman Empires—collapsed and gave rise to new national entities. In each of these newly independent or “successor” states, the idea of national independence had been a largely elite invention up until that point. Take, for example, the challenges facing the stewards of Turkish independence upon the death of the Ottomans. As Bernard Lewis observes, “This new idea of the territorial state of Turkey, the fatherland of a nation called the Turks, was by no means easy to inculcate in a people so long accustomed to religious and dynastic royalties. The frontiers of the new state were themselves new and unfamiliar, entirely devoid of the emotional impact made by the beloved outlines of their country on generations of schoolboys in the West; even the name of the country, Türkiye, was new in conception and alien in form, so much so that the Turkish authorities hesitated for a while between variant spellings of it.”5 Slightly to the west of Turkey, leaders of the newly independent countries of Poland, Czechoslovakia, and Hungary grappled with nearly identical dilemmas. A few decades later, so did the champions of African independence.
The popular inculcation of the idea of the new nation, and the defense of its largely artificial boundaries, is the task of the state. In fact, the state and frequently the primary actors within it, the “leaders,” emerge as the chief protectors and embodiment of national independence. It is no accident that at certain points in history, depending on the prevailing conditions within the nation and the agendas and capabilities of the state and its leaders, nationalism can boil over into jingoist militarism. At the opposite extreme, nationalism may remain dormant and untapped.
Nationalist sentiments may also be awakened by developments elsewhere. Nationalism can at times assume an antithetical nature, being formed and expressed in opposition to something. That something is often the expressed identity of another nation—another nationalism—or an external development that awakens, or reawakens, a sense of national pride and self-assertion. This is precisely what occurred in relation to both Zionist and Palestinian national identities. Zionism, it will be seen shortly, originated in the wake of and in reaction to growing anti-Semitism in Europe in the mid- to late 1800s. In the early decades of the 1900s, this new sense of national identity, by then affixed to geographic Palestine, jolted and awakened a Palestinian identity that had lain dormant for some time.6 In a sense, “Arabness and Jewishness were formulated as nationalist concepts in historically unprecedented ways.”7
In the larger Middle East, several different nationalist sentiments emerged, some sequentially, some concurrently in different countries of the region, and still others in an overlapping manner among different peoples within the same geographic territory. This latter form of nationalism developed within the Palestinian and Zionist communities in relation to the same piece of land. Earlier, from approximately the mid-1500s to the mid-1800s, Ottoman nationalism—or, more accurately, Ottomanism—held sway throughout Ottoman territories, articulated in and dictated from Istanbul. By most accounts, Ottomanism was successful in instilling a communal sense of belonging to an expansive umma (Muslim community) and in maintaining loyalty to the Ottoman state and to the sultan. But by the mid-nineteenth century, as the influence and intrigue of European powers in Ottoman territories gradually increased, especially in the Balkans, the sense of national belonging as articulated from Istanbul—of belonging to an imperial, caliphal, Ottoman nation—began to decline. In its place, more localized forms of nationalism, revolving around locally more resonant symbols and less expansive territories, emerged. At this stage Ottomanism was gradually supplanted by Turkish nationalism in Anatolia and by Arabism elsewhere in the empire. The rise and nature of Arabism, or Arab nationalism, differed from region to region in intensity, origin, and precise character.8 The earliest forms of Arab nationalism, as envisaged politically by the likes of King Hussein of Hijaz and his sons, included the Arab territories of the Ottoman Empire from the boundaries of Iran in the east to Turkey in the north, the Red Sea in the west, and Egypt in the southwest.9 Hussein’s two sons, Faisal and Abdullah, who ruled over Syria (briefly) and eventually Iraq and Transjordan, had still more narrow conceptions of national identity
and nationhood, though less out of ideological convictions than as a result of European mapmaking. Although the Ottomans allowed for considerable local autonomy, these territorially more specific versions of nationalism were more steeped in local, Arab (non-Turkic) social dynamics and cultural lore and symbolism.10
Political manifestations of Arab nationalism were eclipsed for a few years by the more powerful forces of European colonialism, which, among other things, redrew the map of the Middle East for their own administrative and political convenience. Nevertheless, during the period of European political and military domination, and largely in reaction to it, a number of Arab intellectuals began articulating nationalist ideals and sentiments through the publication of books and journals.11 Once European colonialism started retreating in the 1940s, Arab nationalism regained the opportunity to assert itself politically, this time in a much more vocal and virulent manner. The Europeans had created new Arab countries, leaving behind new states for each country, and now the stewards of these new states called on their respective nations to awaken to their full national potential. By the mid-twentieth century, there were such brands of nationalism as Egyptian, Iraqi, Syrian, Lebanese, Jordanian, and Libyan. Turkish and Iranian nationalisms had emerged a few decades earlier, articulated by the Kemalist and Pahlavi states, respectively. Hopes of resurrecting earlier, territorially more expansive conceptions of Arab nationalism—what came to be known as “Pan-Arabism”—lingered, at times motivated by more immediate political considerations. They led to territorial and political unions of Egypt and Syria (1958–61), as well as an ambitious proposed federation of Egypt, Libya, Syria, and Sudan in 1971 and another proposed union of Syria and Iraq in 1979, neither of which materialized. These supranational creations were unrealistic and at best impermanent. Their failure points to the powers of the more localized manifestations of Arab nations and the corresponding force of more locally focused nationalisms.
The Modern Middle East - A Political History Since World War I (Third Edition) Page 10