'Lines Drawn on an Empty Map': Iraq’s Borders and the Legend of the Artificial State (Part 2)

[This article is the second in a two-part series on the making of Iraq`s borders. Click here to read Part 1.]

In attending to how local actors shaped Iraq’s formation as a nation-state after World War I, the point is not to deny the power of British imperial forces, or the violence they unleashed on Iraqis during the occupation (1914-1920) and Mandate (1920-1932) periods. On the contrary, I would contend that one effect of the artificial state narrative is precisely to efface British imperial violence while simultaneously denying the impact of non-British, and anti-British, actions. One way this works is by imagining that Iraq’s borders were created on an “empty map” in a European drawing room and not—as all nation-state borders everywhere have been created—through the resolution of competing claims to territory and sovereignty by deployments of power, including acts of insurgency and counterinsurgency.

Three moments in the early formation of Iraq’s borders—specifically those with Syria, Najd (present-day Saudi Arabia), and Turkey—may help illustrate some of the ways in which the process worked. The British played significant roles, and so too did residents of Iraq, Syria, Najd, and Turkey.

Iraq and Syria

The Iraq-Syria border was rather mobile from the end of the war in 1918 to Iraq’s formal independence in 1932, but the concept of Iraq and Syria as separate states was widely accepted. It is often forgotten that the San Remo conference, which was held in late April 1920, was in part a hastily convened response by the colonial powers to the Arab conference in Damascus in early March, which had proclaimed the independence of Syria and of Iraq as constitutional monarchies under two different sons of Sharif Husayn, Faysal and Abdallah, respectively. The Iraq declaration was issued by the Iraqi branch of al-Ahd, often referred to as the “Arab nationalist” party. Formed in late 1918 when the original group split into two, al-Ahd al-Iraqi was led by Iraqi ex-Ottoman military officers based in Syria; by 1919 it also had an active branch in Mosul and a less active one in Baghdad. Its official platform called for “the complete independence of Iraq” within “its natural borders,” which it defined as extending from the Persian Gulf to the bank of the Euphrates north of Dayr al-Zur in present-day Syria and to the Tigris near Diyarbakir in present-day Turkey—that is, rather more territory than included in the Ottoman provinces of Basra, Baghdad, and Mosul[i]. The group also pledged to work within a loosely defined “framework of Arab unity”; this part of its platform is better understood as Arabist than Arab nationalist, as it did not involve any specific territorial or state-oriented imaginary.

By 1919, then, the two branches of al-Ahd were calling for two independent territorial states—Syria, with its capital in Damascus, and Iraq, with its capital in Baghdad. Throughout the 1920 Iraqi Revolt against the British Mandate—which started in May and June, partly in response to San Remo, and involved large areas of northwestern, central, and southern Iraq—this was also the official platform of the other major Iraqi nationalist party, Haras al-Istiqlal (the Guardians of Independence), based in Baghdad and with significant support in the southern Shi`i shrine cities.[ii]. What the two parties diverged on was not the demand for an independent Iraqi state stretching from the Persian Gulf to somewhere north of Mosul, distinct from Syria, and with its capital in Baghdad—all of that they agreed on—but rather the question of what kind of foreign assistance the future Iraqi state would rely on. Al-Ahd al-Iraqi’s platform specified that it would rely solely on British assistance, while the platform of Haras stated that independent Iraq could request the assistance of any foreign power it pleased.

This understanding of Iraq and its borders had also been the single point of agreement among notables in the three Ottoman provinces who were involved in the plebiscite organized by Britain in late 1918 and early 1919. While the plebiscite was presented as an effort to ascertain the opinion of the locals on what kind of government they wanted, officers of the British occupation army were directed to produce the desired results by pretty much any means necessary. Thus, in most places, a dozen or even fewer notables believed to be loyal to Britain were convened and instructed to answer favorably to a set of questions about Iraq’s future. What was remarkable about the plebiscite was not the duplicity of British colonial officials, or the fact that they managed in many places to elicit the outcome they wanted. Rather, it was that in four of Iraq’s most important cities—Baghdad, Najaf, Karbala, and Kadhimiya—they failed to locate even a small group of compliant notables who would deliver a unified opinion accepting British governance over Iraq. In all of these cities, the text of the plebiscite was altered in some way—against explicit British instructions—in order to reject the British occupation and assert Iraqis’ desires for complete independence. The only part of the text that was not rejected or altered in any of the responses was its definition of Iraq as a state stretching “from the northern border of Mosul province to the Persian Gulf."[iii]

There was no map drawn at San Remo in 1920; the agreement explicitly postponed the determination of the borders. But one thing the European powers did at San Remo was ratify the concept of Iraq and Syria as two states, while Sykes-Picot had divided present-day Iraq and Syria into either three or four states. The 1920 agreement was thus more in line with local nationalist demands, and in particular with the recently declared independence of Iraq and Syria, as well as with historical and linguistic understandings in the Arabic-speaking world of Syria and Iraq as geographical areas, and sometimes as states, loosely centered on Damascus and Baghdad, respectively. San Remo was also an attempt to contain those nationalist demands, by promising the two states only conditional independence, moving toward eventual full independence, under the tutelage of the Mandate governments, France and Britain. The main conflict between local nationalists and the European colonial powers by 1920, at least as far as the Iraq-Syria border was concerned, was thus not over the division of Arab lands into separate states but rather over the degree and timing of those states’ sovereignty. Of course, things were different on Syria’s western and southern borders, given the conflicts over the separation of Palestine, Lebanon, and Jordan from Syria—the first by the time of San Remo, the other two later—and these conflicts may have shaped our understanding of Iraq’s formation.

In terms of establishing the actual whereabouts of the Iraq-Syria border, the main question (leaving aside Mosul for the moment, since that was primarily an Iraq-Turkey, not an Iraq-Syria, dispute) was over the Ottoman province of Dayr al-Zur. In Sykes-Picot, Dayr al-Zur had been placed on the French side of the line between the A and B territories, but the fact that it ended up in Syria was almost a historical fluke.[iv] In November 1918, conflicts between residents of Dayr al-Zur and the officers of the Arab army in Syria led local notables to appeal directly to Britain to annex the region to the occupied territory of Iraq. British troops duly arrived and did so. But soon the residents became resentful of the British occupation as well, and in 1919, petitioned Damascus for re-incorporation into Syria.

Ironically, it was the Iraqi nationalist officers of al-Ahd al-Iraqi who were ultimately responsible for the inclusion of Dayr al-Zur within Syria. They hoped to use the region as a base for launching attacks from Syria on British occupation forces in Iraq—and that is what they did, thereby helping to spark the 1920 revolt.[v] In 1923, Baghdad-based Iraqi nationalist Muhammad Mahdi al-Basir explained the Dayr al-Zur decision: “Iraqis [in Syria] were working for the liberation of Iraq, even if that required annexing much of its land for the Syrian government."[vi] Leading British officials, including Acting Civil Commissioner in Iraq at the time, A.T. Wilson, later asserted that Britain’s acquiescence at Dayr al-Zur—i.e., the evacuation of its troops and relinquishing of the province to the Arab army in Syria—helped precipitate the entire 1920 revolt, not only by providing the Iraqi nationalist officers in Syria a base for cross-border military operations but also by giving other opponents of the British Mandate within Iraq a sense of Britain’s vulnerability.[vii]

Iraq and Najd

Iraq’s southern border with Najd (present-day Saudi Arabia) was left completely undefined in most of the international agreements during and after the war, though it does appear in one early map of the Mandate territories (see below). This map—which is titled “Mandates in Arabia” and has been displayed in many popular and scholarly accounts of Iraq as an artificial state—is a good example of some of the pitfalls involved in tracing the history of post-Ottoman borders. At some point, the map started being dated to 1920 and attributed to either the Treaty of Versailles or the San Remo Agreement, thus suggesting that the borders it shows had been established by then. But this date is impossible for numerous reasons, including the map’s depiction of a Transjordan Mandate, which did not exist even as a concept until 1921. In fact, the map was drawn by US geographer Lawrence Martin in 1924, as his personal interpretation of all of the border agreements up to and including the 1923 Treaty of Lausanne. In the case of the Iraq-Najd border, however, it actually depicts the situation in early 1922, or at least one plausible interpretation of it. Martin was apparently unaware of that border’s redrawing in the Uqair Protocol signed between Iraq and Najd in December of that year.[viii]

[1924 map of Mandate territories. All of the borders with dashed lines are defined as “Undetermined” in the key.
Image from The Treaties of Peace 1919-1923 (New York: Carnegie Endowment for International Peace, 1924), 2:966.]

In any case, both the Iraq-Najd and the Transjordan-Najd borders on this map appear considerably south of their current locations. In Sykes-Picot, we may recall, the territory envisioned as British-ruled Iraq extended even further south on the eastern side, encompassing the coastline and a good inland chunk of the Arabian Peninsula down to Qatar. The main factor that undid both of these plans, fixing the border in its more northward location, was the military expansion of Abdulaziz ibn Saud of Najd and his Ikhwan forces during these years. In a series of treaties with Abdulaziz from 1920-1927, British officials acknowledged his ongoing territorial conquests by accepting progressive contractions of their envisioned territories. Britain eventually fought to prevent further northward expansion of the Saudi state, launching a massive air offensive into recognized Najdi territory to that end in 1927-1928. Britain’s line in the sand, as it were, was its determination not to lose what remained of the corridor linking its Iraq and Transjordan mandates, which in the British view had already been dangerously narrowed by Abdulaziz’s recent conquests.[ix] The image below shows a rough outline of the Iraq-Najd border as of 1927, when the British air force finally halted the Saudi advance.

[Iraq in 1927. Image from “The Boundaries of the Nejd: A Note on Special Conditions,”
Geographical Review 17, no. 1 (January 1927): 130.]

Rather than a line on an empty map, the border was defined in the treaties as a series of lines connecting known waterholes or wells in the desert. The placement of the border on one side or another of each well determined the nationality of the nomadic people living in the borderlands. If a well was placed on the Iraq side, the members of the tribe to which that well was locally understood to belong became Iraqi subjects; otherwise, they became subjects of Najd. The placements themselves were not arbitrary either, since British and Iraqi officials and Abdulaziz all had very strong ideas about which tribes they wanted and did not want as subjects, and negotiated over those that were claimed by both. The tribes also had some say in the matter; a few successfully resisted their new nationality and were transferred to the other side.[x]

Iraq and Turkey

The dispute over the Ottoman province of Mosul was by far the most significant border question in the formation of modern Iraq. When Britain occupied Mosul on 3 November 1918, and then declined to leave, it was breaking two agreements: its Sykes-Picot Agreement with France, according to which Mosul was to be under French influence, and the Armistice of Mudros, signed on 30 October 1918, which officially ended World War I in the Middle East by ending hostilities between the Ottoman Empire and the Allied powers. Since Britain occupied Mosul four days after the signing of the armistice—i.e., when it was not at war with the Ottoman Empire—first the Ottoman and then the Turkish state naturally refused to recognize the legal validity of the occupation and thus Mosul’s incorporation into Iraq.

[Treaty of Sèvres, 1920. Image from The Treaties of Peace 1919-1923
(New York: Carnegie Endowment for International Peace, 1924), 2:789.]

While Britain ultimately succeeded in including Mosul within Iraq, an important factor in establishing the Iraq-Turkey border, and one that is often forgotten, was the Turkish War of Independence. The history of Turkey’s southern border from 1918-1923 is far too complicated to recount here, but suffice it to say that the Treaty of Sèvres, which was forced on the Ottoman state in August 1920, left Turkey as a small rump state in central Anatolia. The British- and French-influenced territories were extended northward well into present-day Turkey through a semi-independent, possibly Kurdish, state or zone; the map above shows Lawrence Martin’s interpretation of Sèvres. This treaty helped fuel the Turkish War of Independence, in which Kemalist forces defeated the Allies and reclaimed much of the land that had been lost since the 1918 armistice, pushing the borders south once again. In this case, the borders were thus imposed on the European powers rather than by them. If the Kemalists had been defeated, we might now be hearing that Europeans drew the borders of the modern Middle East at Sèvres rather than in San Remo or Sykes-Picot.

The 1923 Treaty of Lausanne, which ended the Turkish War of Independence, established most of Turkey’s present-day borders, but not the one with Iraq, since neither Turkey nor Britain would budge on Mosul. The agreement specified that the two parties would attempt a peaceful resolution of the dispute, and if that failed within nine months, the question would be referred to the League of Nations, which it was. After a commission appointed by the league toured Mosul in 1925 to ascertain local opinion, it recommended that the province be given to Iraq. Turkey appealed the decision, but it was upheld and in 1926 Turkey signed the agreement establishing its border with Iraq. King Faysal of Iraq proclaimed that the treaty “fixes our political existence, internally and externally."[xi] He was right, of course, since mutual recognition—usually of a resolution to competing claims to territory—is what a border is, or in any case what makes one possible. It takes at least two.

Both British and Iraqi officials had worked hard for eight years to break Mosul away from Turkey. The British air force had engaged in near-continuous air bombing campaigns, often described in British primary sources, and repeated in many secondary sources, as defensive actions against Turkish “incursions” into Iraq. Since Mosul did not yet belong to Iraq, however, this may be a questionable description. The bombings targeted both pro-Turkey borderland communities and Kurdish separatists. As one British report explained:

[The] Arab and Kurd . . . now know what real bombing means, in casualties and damage; they now know that within 45 minutes a full sized village . . . can be practically wiped out and a third of its inhabitants killed or injured by four or five machines which offer them no real target, no opportunity for glory as warriors, no effective means of escape.[xii]

More diplomatic measures were also used. The Mandate government-issued statements from Baghdad promising that minority languages and rights in the north would be respected, and delegations of Iraqi officials toured Mosul and Sulaymaniyah to persuade residents they would be better off in Iraq than Turkey, especially as the date approached for the arrival of the League of Nations commission.

In fact, from 1918 to 1926 Mosul acquired profound significance in the formation of an Iraqi national identity, a process that has received inadequate attention and is thus poorly understood. Already in the 1918-1919 plebiscite, as both Iraqi and British observers noticed at the time, the only unanimous point of agreement was that “Mosul is part of Iraq,"[xiii] and in the coming years nationalist poets in Baghdad and Basra waxed lyrical about Mosul as the “Jewel of Iraq.” Iraqi officials mobilized these sentiments to counter nationalist critiques of the Mandate relationship; Prime Minister Abdul Muhsin al-Saadun asserted that Mosul was the “head” of the “body” of the Iraqi nation, and King Faysal declared that the Mosul question was “a life or death matter for our beloved country."[xiv] British officials were also cognizant of them, repeatedly using Mosul as both carrot and stick for bringing nationalists in Baghdad and the southern regions into line. Most dramatically, they threatened not to support Iraq’s case at the League of Nations on the Mosul dispute if the Iraqi government did not ratify the treaty with Britain ensuring the latter’s authority over Iraq’s affairs. There was much debate among Iraqi nationalists at the time over whether Britain was bluffing—after all, it had major oil interests in the province—but many did not think it was a risk worth taking, especially after Turkey offered Britain the same rights to oil in Mosul that it would have with Iraq. Faysal asserted that it would be “a terrifying gamble to take with the most glorious part of our nation."[xv] When the Iraqi Constituent Assembly finally ratified the Anglo-Iraqi Treaty in 1924, under intense British pressure, it made its approval conditional on “the British government’s protection of Iraq’s rights over the province of Mosul in its entirety." [xvi]

As for Mosulis, they were deeply divided between pro-Turkey and pro-Iraq sentiments throughout these years, with some moving back and forth across the divide according to their evaluation of rapidly changing conditions. For some, especially Armenians and other Christians, communal affiliations often did drive political allegiance; Turkey was a hard sell after the genocide and the Turkish-Armenian War. Others—and perhaps especially the Kurds—were split, a division complicated in the Kurdish case by the existence of a third faction proposing an independent Kurdish state. This faction was more or less vocal at different moments and in different regions, but never unified enough to stand a chance against the British air force. For Arab Mosulis, communal/linguistic affiliations and even Arabist commitments did not necessarily determine political alignments. For example, a dispute broke out between the Mosul and Syria-based branches of the Arabist-oriented al-Ahd al-Iraqi when the Mosul group announced that many Mosulis were looking to Turkey and suggested that Iraqis, in general, would be better off allying with the Kemalists (and the Bolsheviks) than with Hijazi Arabs such as Sharif Husayn and his sons.[xvii] For all Mosulis, the dispute was constantly complicated by the fact that there were two conflicting imaginaries of “Iraq”—the British one and the independent one.

That “independent Iraq” was often imagined—both by Iraqi nationalists and by the British—as an “Arab state” was very significant for Iraq’s non-Arab populations. Arabism and Iraqi nationalism were not opposing ideologies in the 1920s, as some versions of the artificial state narrative assume; this is to write a later political divide back onto this period. Rather, Arabism shaped the formation of an Iraqi national identity and thus the formation of Iraq as a nation-state. Many pro-Iraq Kurds thus leaned toward the continuation of some kind of British administration rather than the creation of an independent “Arab” Iraq. British officials, of course, encouraged these leanings, though it should be noted that they also regularly referred to Iraq, including Mosul, as an “Arab state,” and from the beginning had asserted the importance of establishing an “Arab administration” in the country as a means of indirect British rule. 

It might be noted that the possibility of Mosul’s incorporation into Syria was not a major topic of discussion from 1919-1926, since that was not one of the options on the table, Sykes-Picot notwithstanding.[xviii]. Decades-long scholarly speculation about whether Mosul might have been more “naturally” joined with Syria is thus beside the point. The actual historical problem was that Britain was militarily occupying the province, and wanted it to be part of Iraq—as did Iraqi government officials and anti-British Iraqi nationalists—while Turkey had a very strong legal claim to it given that the British occupation had been in violation of international law. Nobody seriously proposed that a secret wartime agreement between Europeans had given Syria or France any internationally or locally recognizable claim to Mosul, despite some short-lived grumbling by French officials about the wartime actions of their ally (smoothed over in any case by 1920, after Britain promised France a future share in Mosul’s oil). The notion that there is some Archimedean vantage point from which international borders can be, or ever have been, created is obviously nonsense. But it is a notion that the “fabricated in Europe” narrative has helped to sustain, perhaps in spite of itself, by perpetuating the idea that the act of drawing a line—any line at all—on a map in Europe has the quasi-magical power to create a border in Asia. That historians have contributed to this story may have something to do with what Neil Smith has argued was the increasingly “anemic” conception of geography within the academic discipline of history, at least in the United States, over the course of the twentieth century.[xix]

The Emergence of the Artificial State Narrative

While the narrative of Iraq as an artificial state resonates with similar discourses on other postcolonial states—especially in the Middle East and Africa—it has its own particular history. It emerged in the earliest years of the state’s formation, and was connected to claims of the ungovernability of Iraqi subjects: the rural ones because they were tribal and uncivilized, the urban ones because they were ethnically and religiously heterogeneous. It was originally a colonial narrative, used to justify Britain’s continuing occupation of Iraq. As early as 1922, the London Times expressed the emerging suspicion that there was something scandalous about Iraq: “No common purpose yet animates these heterogeneous communities . . . Mesopotamia, with its vague frontiers and mixed population, was treated as a nation, as an embryo State, to be ranked with the modern democracies included under the League of Nations."[xx] This argument, it bears remembering, was an explicit response to Iraqi nationalist demands for sovereignty. Acting Civil Commissioner A.T. Wilson had been in charge of putting down the 1920 nation-wide revolt that demanded the evacuation of British troops and the “complete independence” of Iraq within its “natural borders,” at the cost of over eight thousand Iraqi lives and five hundred British and Indian ones, according to British estimates. Yet in his 1931 memoirs, he claimed that in the period 1917-1920, nationalism “was not an important element in Iraq . . . the idea of Iraq as an independent nation had scarcely taken shape, for the country lacked homogeneity, whether geographical, economic, or racial."[xxi] He even cited older British sources to support these claims, quoting the Marquess of Salisbury in 1878 that “Asiatic Turkey contains populations of many races and creeds, possessing no capacity for self-government and no aspirations for independence." [xxii]”

Iraqi officials invoked similar narratives for various purposes throughout the 1920s and 1930s. In 1932, the year Iraq became formally independent and joined the League of Nations, King Faysal proclaimed:

In my opinion, and I say this with a heart full of sorrow, an Iraqi people does not yet exist in Iraq. There are only throngs of human beings lacking any national consciousness, immersed in religious traditions and falsehoods, disunited, susceptible to evil, inclined toward anarchy, and always prepared to rise up any government whatsoever.[xxiii]

This quote has been widely invoked over the years as further evidence that Iraq itself is some kind of scandal. Faysal’s next lines are often not included: “This being the situation, we want to fashion these multitudes into a people that we will cultivate, guide, and educate . . . This is the people I have taken it upon myself to bring into being." [xxiv] Like most nation-builders of his time, Faysal understood very well that nation-states are constructed—which, for nation-builders, means that they have to be constructed.

But it might also be productive to focus less on what such quotes say about Iraqis than on what they were invoked to do in the context in which they were uttered. Of course, Iraqis in the first decade of Iraq’s existence were not rising up against “any government whatsoever,” as Faysal claimed. They were rising up against Faysal’s government, including the British Mandate system that propped it up. The discourse of Iraqis as inherently ungovernable submerged all conflicts in Iraq into a single, ready-made explanatory narrative, as the discourse of the artificial state continues to do today.

In the 1920s and 1930s, what this discourse may have been most concerned with submerging and forgetting was the 1920 revolt. The contemporary sources on this revolt simply do not support the notion that Iraq was a colonial imposition. The consistent demand of the insurgents, from Mosul to Baghdad to the southern Shi`i shrine cities, was Iraq’s complete independence—al-istiqlal al-tamm—within what they called its “natural borders,” which they defined as extending from north of Mosul to the Persian Gulf.[xxv] This does not mean that all Iraqis participated in the revolt, but rather that most of those who did were protesting the British military occupation and the British Mandate, not the drawing of Iraq’s borders and its formation as a nation-state. At the time, the notion of Iraq’s inauthenticity was almost exclusively a colonial discourse.

Fortunately for Iraqis, the small cohort of British officials who argued after the war that the new state borders should be based on ethnosectarian ones lost the battle; Iraq therefore witnessed no major projects of ethnic cleansing in the 1920s. As I have suggested, many factors contributed to this outcome, including Iraqi nationalist demands, British imperial interests, and the actions of Iraq’s neighbors. Critics of the argument also frequently pointed out the extreme difficulty of implementing an ethnosectarian vision in this region. Indeed, it was this vision that drew its logic from the fantasy of an empty map, or a map filled with empty homogeneous space: empty of history; of claims to territory and other resources; of neighbors speaking different languages; of multiethnic villages and virtually any conceivable city; of existing provincial and international borders; of previously concluded treaties and agreements; of local and international laws; even of mountains, rivers, deserts, and oil deposits. Nothing but empty space and fixed ethnosectarian identities.

But that point was made a hundred years ago. What I have also been suggesting is that the narrative of Iraq as an artificial state emerged out of the very historical conflicts and processes it was then retrospectively deployed to explain, as well as to explain away. Rather than historicize the narrative, by exploring its emergence in the years after World War I, scholars and countless other commentators have used and re-used it to empty Iraq of history.

[This article is the second in a two-part series. Click here to read Part 1.] 

[For extremely helpful comments on earlier versions of this article, I thank Beth Baron, Hala Fattah, Samira Haj, Dina Rizk Khoury, and Omnia El Shakry.]

[xxv] For an English-language discussion of these sources, see Abbas Kadhim, Reclaiming Iraq: The 1920 Revolution and the Founding of the Modern State (Austin: University of Texas Press, 2012), especially chap. 4, “The Journalism of the Revolution.”