Documents & images

Now available from Amazon.
Now available from Amazon.
Now available from Amazon.

Partition Iraq? Imperial Iraq Strategies from Sulayman the Magnificent to Joseph R. Biden


Notes for a lecture presented at the University of Chicago, the University of Michigan and George Washington University, 11–15 January 2008. A shorter but footnoted article on this subject, illustrated with maps, was published in Survival vol. 50 no. 2, May 2008 under the title “Historical Myths of a Divided Iraq”.

Reidar Visser
Editor of www.historiae.org
Research Fellow, Norwegian Institute of International Affairs

The theme for this presentation is inspired by concern about how US policy-makers talk about Iraq. It has grown out of concern about the ways in which false arguments related to Iraq’s past are being used in discussions of US policy – often without protest from the academic community. And in particular it is driven by concern about a new round of lies about Iraq: the 2003 invasion of Iraq that was based on flawed evidence; it now looks as if the exit strategies, too, will be based on misinformation about Iraqi society. This relates not only to the so-called “soft partition” alternative, but actually to any plan that identifies the main problem in Iraq as consisting of a perennial civil war between the various “ethnic groups” in the country.

A few quotations of US policy-makers provide examples of what the problem is:

“They considered running the country more or less how the Ottomans had, with a strong central government but with the country divided, in effect, into three provinces: Kurdish, Shiite, and Sunni, each having a different governor.” (Leslie Gelb in CFR interview, 16 October 2007)

“So here is a map from 1914. This is fascinating…This is in the State of Mosul, the Kurdish north. You had the vilayet of Baghdad, the Sunni area in Iraq. You had the vilayet of Basra, the Shia State.” (Sen. Sam Brownback during a debate on the “Biden resolution” on federalism in Iraq, September 2007)

“Iraq was cobbled together by the British at the end of World War I from three different Ottoman valiyets [sic], or provinces: predominantly Kurdish Mosul in the north, mostly Sunni Baghdad in the center, and Shiite Basra in the south”. (Peter Galbraith in The End of Iraq, 2006, p. 7)

There is little doubt that prior to the invasion of a foreign country, it is prudent to look to history for advice. But with regard to Iraq, US policy-makers have not done their history lessons well. It is the aim of this paper to show that all of the statements quoted above are in fact false.

A natural place to start an overview of Iraq’s imperial masters is Sulayman the Magnificent who added Iraq to his empire in the first half of the sixteenth century. What the Ottomans found upon their conquest of Baghdad was this: a region that for more than 900 years had mostly been divided into two, not three. This is not to suggest that these administrative divisions were in any sense permanent, or that it makes sense to translate them into fine boundary lines on a map. There were certainly periods in which autonomous sub-entities emerged, or the overall system collapsed due to external pressures. Nevertheless, there was an amazing degree of continuity with regard to two centres of gravity. Firstly, the historical Iraq, extending roughly from the Gulf to Tikrit, and the home to Shiites and Sunnis alike, with the capital of Baghdad; secondly, to the west, al-Jazira or “the island” between the great rivers, continuing westwards towards Syria, and with Mosul as its key city. Despite the upheaval of medieval times, this system of two administrative non-sectarian entities had proven remarkably resilient ever since the Arab conquest of Iraq in the seventh century.

What the Ottomans did was to divide this area further – not into three, which is the number so often associated with them – but rather into four. They created Basra as a separate province in the far south, and split Jazira into Mosul and Shahrizor (later known as Kirkuk). Again this was on a non-sectarian, non-ethnic basis. Basra was never the capital for the Shiite holy cities of Najaf and Karbala, whereas the Kurdish areas were administered from Kirkuk which demographically was dominated by Turkmens.

This arrangement lasted for a couple of centuries, but already in the late seventeenth century a new trend could be witnessed: the re-emergence of Baghdad as a regional capital for the entire area between the Gulf and the Kurdish mountains. This reached its climax during the Georgian mamelukes of Baghdad in the late eighteenth century, when even Mosul at times was made a Baghdad dependency. And it was a model that was retained by the Ottomans when they made their comeback in Iraq after 1831, so that the Iraq of the 1860s was virtually undistinguishable from the modern state of Iraq. In this perspective, the idea about Iraq’s supposed “artificiality”, and the notion of Baghdad being some kind of “imposition” on the rest of Iraq, come across as lacking in historical substance.

Only in the 1884 did the much-celebrated model of a tripartite Iraq actually emerge on the ground. Three points are noteworthy in this regard: Firstly, this administrative division did not last more than thirty years. Secondly, it was never total: Baghdad retained supreme powers in a number of areas of administration, including the military, justice and customs. Thirdly, there was no correlation to sectarian identity. As before, the province of Basra was limited to the extreme south, with the majority of Shiites living in the vilayet of Baghdad. Mosul was essentially a mixed province.

These were the realities of the Ottoman administrative subdivisions of Iraq. And if we look beyond the maps, and study actual patterns of interaction with the local population, it again becomes clear that sectarian identity did not constitute any insurmountable barrier between Sunnis and Shiites. For example, in 1898, the Ottomans provided armed guards for boats with Shiite pilgrims bound for Karbala as they travelled through troublesome Shiite tribal areas. In other words: a case of Sunnis protecting Shiites against other Shiites. And even in the worst cases of sectarian bigotry, the Sunni rulers were mostly on the side of moderation: in 1801, when nomads from Arabia sacked the Shiite holy city of Karbala, the Sunni pasha of Baghdad actually punished the Sunni military commander in Karbala for not having protected the Shiite city.

The next imperial rulers of Iraq were the British, who are often quoted when there is talk about the “artificiality” of Iraq. And it is true that the British produced many colourful and imaginative maps of Iraq. But equally important is that none of these followed sectarian lines.

The big policy debate in London was never about whether Britain should support Sunnis or Shiites, but rather about whether a small coastal enclave around Basra should have a privileged position. In 1914 and 1915, Britain was primarily interested in annexing Basra, as that would complete the trucial system of the Gulf and provide security for India. When Baghdad was included in 1916 in the Sykes–Picot agreement, it was still considered secondary in importance to Basra. And again, this was not an attempt to create a Shiite state; in fact many British officials believed it would be best to leave the two holy cities of Najaf and Karbala outside the British zone. Britain’s partners in Basra were merchants, mainly Sunnis, Christians and Jews.

One map that is often given attention in discussions of British Iraq policy is the one by the legendary T.E. Lawrence in which Iraq is divided into four. Two points are of importance here: Firstly, T.E. Lawrence was utterly marginal in decision-making related to Iraq. He did not have a lot of first-hand experience from Iraq and was never taken seriously by the most important decision-makers in London or in Baghdad. Secondly, his very eye-catching map had nothing to do with sectarian identities. It was an attempt to divide Iraq between the princes of the Sunni sharifian family of Mecca, with no distinction between Shiite and Sunni areas.

Instead, what came to dominate British imperial thinking about Iraq was the vision of a unified entity from Basra to the Kurdish mountains. Its champion from 1918 onwards was Arnold T. Wilson, the acting civil commissioner of Iraq. His main argument was that Iraq constituted an organic, economically inter-dependent unit from Basra to Mosul, and, notably, that Ottoman tendencies of using Baghdad as a regional capital, proved the viability of this approach. These ideas were later inherited by Percy Cox and Gertrude Bell, and soon became the main trend in British policy-making.

To the extent that there was any challenge to this in Iraq in the 1920s, it was only the alternative of a British withdrawal to a small non-sectarian coastal enclave around Basra that received any serious attention – in addition to the fact that a few officials in the early 1920s briefly favoured the establishment of a separate Kurdish entity. Both these themes could be seen in 1920 in the plans of Hubert Young, an influential Colonial Office bureaucrat, but towards the end of the 1920s, only the idea of a separate Basra survived. It was an emergency option only, and it was finally abandoned around 1930.

On the other hand, the British ridiculed the idea of a separate Shiite state. When an opportunity for an alliance between the British and the Shiites arose in 1927, this is how the highest British official responded to the invitation of becoming a Shiite partner: “I hope that this matter may end without acute excitement of clash between Sunnis and Shiites but certain bigoted elements are as usual trying to fish in troubled waters”. These comments are not rhetoric; they are from a confidential despatch. And in another internal document, this one from London, tendencies to sectarian conflict in 1927 are described as “unpleasant”, and Shiite politicians were portrayed as “recalcitrants”.

The obvious conclusion of the historical part of this analysis is that at no point during the four hundred years of Iraqi history did the notion of an Iraq divided into three along ethno-sectarian lines receive serious consideration by the country’s imperial rulers.

A discussion of US approaches to Iraq’s territorial integrity must start with some comments on what can be described as Bush’s Biden policy. On the one hand, the Bush administration rhetoric in Iraq seems almost impeccable in its commitment to the preservation of Iraq’s territorial integrity. Official statements are sometimes almost touching when it comes to this. One example is Zalmay Khalilzad:

“Moreover, as tragic and dangerous as the ongoing violence is to our shared vision of a free and prosperous Iraq, it is not representative of the Iraqi people's sentiments toward one another. In July, a poll by the International Republican Institute, a nonpartisan organization dedicated to democracy promotion, found that 94% of Iraqis said they support a "unity" government representing all sects and ethnic communities, with only 2% opposed. Some 78% of Iraqis opposed Iraq being segregated by religion or ethnicity, with only 13% in favor. Even in Baghdad, where the worst of Iraq's sectarian violence has occurred, 76% of those surveyed opposed ethnic separation, with only 10% favoring it. The challenge of the Baghdad Security Plan and its accompanying effort at national reconciliation is to realize the overwhelming majority of Iraqis desire to live in peace with one another against the violent minority who seek to impose their vision of hatred and oppression.” (Zalmay Khalilzad, op-ed in The Wall Street Journal, 23 August 2006)

Or consider what the Bush administration’s critics, like Peter Galbraith, have to say: “The White House is dreaming of a non-ethnic Iraq” is his angry charge. He condemns “the American effort to build a unified state with a non-sectarian, non-ethnic police”.

But the practices of the Bush administration in Iraq tell a very different story. The natural place to start a discussion of this subject is with Paul Bremer. This is how Bremer addressed a group of seven Iraqi politicians in 2003: “As I have said, the process (of building a new Iraq) will be incremental and must have as its goal a truly representative group. This body is not representative. There is only one Sunni Arab among you. Everyone looked at Nasir Chadirchi. There are no Turkmen here, no Christians, no women”. What this quote shows, is that even if the Bush administration may have a commitment to Iraq’s territorial integrity, some of them are also obsessed with emphasising sectarian identities. In this particular setting, Bremer was enraged because Sunni representation was at a mere 14 percent, whereas according to his ideals, it should have been around 20 percent.

Another example of this tendency: There are about 80 MPs in the Iraqi parliament who are in favour of an Iraq divided into federal regions on the basis of sectarian and ethnic identities, roughly along the lines of the Biden proposal. They are the Kurdish parties, as well as the Islamic Supreme Council of Iraq or ISCI, formerly SCIRI. One would have thought that this minority of perhaps 30% would have been politically insignificant in comparison to the remaining 70% who favour a more unified system without any sectarian divisions in the administrative geography. But in fact these 30% happen to be the ones that have been elevated by the Bush administration as the keystones of its so-called “moderate coalition” in Iraq, and receive the lion’s share of Washington’s attention.

Finally, some in the Bush administration are so focused on the link between federalism and sectarian communities that they are virtually indistinguishable from many “soft partitionists”. A good example is Ryan Crocker and his remarks in a Senate hearing on Iraq last September: “Some of the more promising political developments at the national level are neither measured in benchmarks, nor visible to those far from Baghdad. For instance, there is a budding debate about federalism among Iraq's leaders and importantly within the Sunni community. Those living in places like Anbar and Salahuddin are beginning to realize how localities, having more of a say in daily decision-making will empower their communities”. In these comments, there seems to be scant respect for the other alternative available to the Sunnis (and indeed to all Iraqis outside Kurdistan) under the 2005 constitution: that of remaining part of the unitary state, without forming any federal region at all.

Ryan Crocker is in other words very similar to the last of the Iraq strategists that is of relevance in this discussion: Senator Joseph Biden. Is Biden an imperialist? One problem with his so-called plan for Iraq is that it exists in so many incarnations that it is sometimes difficult to pin down exactly what it is about. But a visit to Biden’s website, planforiraq.com, leaves little doubt about what he means. There are constant references to Sunnis, Shiites and Kurds, and the need for them to “have their own regions”. This is the imperialist aspect. True, the Iraqi constitution contains federal elements, but who is Biden to tell the Iraqis what particular federal entities they are to create?

It can be useful to also look at the most innocent, watered-down version of Biden’s plan: the “sense of Congress on federalism in Iraq” which was attached to the 2008 defence spending bill. The imperialist aspect of this bill is found in the portion that advocates a “conference for Iraqis” to reach a “comprehensive” settlement on federalism in Iraq. Why is this imperialism? It is contrary to the Iraqi constitution in two major ways. Firstly, the Iraqi constitution of 2005 recognises that federalism is a contested concept in Iraq: therefore federalism is left as an optional solution, not something that every part of the country needs to have. Importantly, the constitution makes it clear that the only way to form new federal entities is from below, through popular initiatives in the existing governorates. This is the logical opposite of Biden’s “congress”. The whole point of the Iraqi constitutional provisions on federalism is that federalism should never emerge as the result of a conference, where established elites would always have the upper hand. Secondly, the term “comprehensive” is unconstitutional in so far as it seems to suggest a subdivision of the entire country into federal regions. A major point in the 2005 Iraqi constitution is that those parts of the country that want to retain a unitary state structure, without federalism, are free to do so. Comparisons with the United States are of little value in this regard: Spain, Russia and arguably the UK are better places to look for parallels. This is in other words an example of how Biden’s proposals actually involve even more micro-management of Iraqi affairs than what we have seen under the Bush administration. If the objective is to get really stuck in Iraq, then this is certainly the way to do it.

In sum, two leading US politicians, Bush and Biden, are both trying to circumvent the fundamentals of Iraqi history through imposing a thoroughly artificial sectarian logic whose only sustained period of prominence in Iraqi politics dates from after 2003. Is there an alternative to all this imperialism? Yes there is. There is an alternative that would be far more in harmony with the long lines of Iraqi history: the nationalist alternative. It is however important to recognise that Iraqi politics have changed so much during the US occupation from 2003 to 2008 that the Iraqi nationalist alternative needs help to once more become politically relevant. The problem with the “threat” of an “immediate speedy withdrawal” is that Iraqi sectarian leaders will shrug their shoulders and say, “Well, you served us well for five years. It would have been nice with another five, but you have armed us generously; we now have a head start on everyone else in Iraqi society, so let’s prepare to fight a full civil war.”

To once more make the Iraqi nationalist alternative relevant in Iraqi politics, the United States should use the majority nationalist trend in the Iraqi parliament as point of departure. So far, the Bush administration has insisted on doing the logical opposite, by according preferential status to the parties with a sectarian orientation among the Shiites and the Kurds. This will always be an uphill struggle, because the parliament has shown itself again and again to coalesce around an anti-sectarian, Iraqi nationalist agenda. We saw this recently in the declaration by more than 150 MPs who used Iraqi nationalism to challenge Kurdish radicalism in the oil question and the Kirkuk issue. We are seeing it right now in the work with a law on provincial powers, where the nationalists are demanding a timetable for elections, whereas Washington’s partners are resisting this – preferring to anti-democratically hold on to the positions they acquired during the previous elections in 2005, in many cases as the result of widespread boycotts by more representative parties.

Washington should engage this nationalist trend, and above all should encourage them to translate their political position into a far more radical constitutional revision proposal than that proposed by the current revision committee (which is dominated by the sectarian parties that enjoy the support of the Bush administration). Ideas that circulate in this group include

• checks against the emergence of purely ethno-sectarian federal entities (for example size limits, or a two-thirds majority requirement for the formation of any new federal entity)
• acceptance of the idea of international arbitration over Kirkuk
• removal of the notion of “disputed territories” from the constitution
• insertion of clause that guarantees distribution of oil income on the basis of demographics only

It would be fairly easy for this group to compose a nationalist Iraqi package so attractive to Iraqi nationalist sentiment that Iraqi politicians would stultify themselves by rejecting it. Importantly, constitutional revision is unlikely to be derailed by Sunni–Shiite disagreement over legal issues (Sunni and Shiite legal codes are in fact quite similar), and could have even better prospects now that the de-Baathification issue has been channelled to other forums. This kind of deal could increase support for the Iraqi constitution and thereby offer a new start for Iraq of huge symbolic importance, among Sunnis and Shiites alike. It is often forgotten that back in 2005 the leading figure among the Shiites, the Grand Ayatollah Ali al-Sistani, tempered his support for a “Yes” vote in the constitutional referendum with an assertion that the constitution contained several weaknesses.

To further support this work, Washington should also take the opportunity to mark a very public change of policy – a turn away from sectarianism in Iraq (the term de-Bremerfication has already been heard!) – perhaps in connection with the anticipated agreement on foreign forces in Iraq, or at the time of a new incoming administration in Washington in early 2009. This could take the form of an offer of an exceptional deal: if Iraqis succeed in finalising the revision of the constitution in a more nationalist spirit, the United States and the international community would immediately support them with the following steps:

• Extraordinary aid package
• debt cancellation
• timetable for US withdrawal
• guarantee against foreign bases
• full diplomatic representation (Arab states)

It is to be hoped that in the future, the US debate about Iraq policies and alternative Iraq policies will move away from the blind alley where it currently finds itself, and instead will be re-focused on the Iraqi nationalist alternative. This is an alternative that would produce an Iraq more in harmony with itself and with its history, and therefore also with far better prospects for political stability in the long run.

E-Mail Alert Service
To receive e-mail notification of new Iraq analyses and reports from historiae.org simply enter your e-mail address here:

Copyright © 2005-2008 historiae.org & Reidar Visser

This document or quotes from it may be freely reproduced as long as www.historiae.org is credited as the original source.