The Middle East has long had the dubious distinction of being one of the world’s major producers of refugees. By the beginning of 2007, the Middle East was generating 5,931,000 refugees out of a world total of 13,948,800. Over the past century, not just conflict but development projects, environmental disasters and state-mandated settlement of nomads have driven people from their homes. [1]
Several states in the region have complex histories of creating massive waves of refugees or being built by the displaced. The Greek-Turkish “population exchange” and the expulsion and genocide of Armenians mark the beginning decades of the twentieth century. Israel’s establishment in 1948 resulted in more than 750,000 Palestinian refugees. Jordan has been host to multiple influxes of the displaced, from the Circassians arriving in late nineteenth century to the Palestinians seeking refuge in 1948, 1967 and 1991, and, more recently, as many as one million Iraqis. During the Algerian war of independence, over two million were forcibly displaced by the French. In Lebanon, civil wars and invasions have produced hundreds of thousands of internally displaced over the past several decades.
Iraq is also no stranger to forced displacement. Hundreds of thousands of Kurds fled the Iran-Iraq war, the 1991 Gulf war and the murderous campaigns waged by the Iraqi state. In an attempt at demographic engineering, the Baathist regime destroyed thousands of Kurdish villages and expelled Kurds from the Iraqi north. The regime then moved Arabs into heavily Kurdish regions, where these Arabs are themselves now under pressure to leave. Thousands of Iraqis fled the country in the past two decades to escape wars, sanctions and state-perpetrated violence. But the US occupation following the 2003 invasion, precipitating as it did a cycle of sectarian and ethnic violence, has given rise to unprecedented mass displacement with clear sectarian dimensions.
The current Iraqi displacement crisis and the lack of a significant international response may presage reconceptualizations of the refugee, the spatial and administrative device of the refugee camp, and humanitarian obligations in the face of large-scale flight. The Bush administration’s attempt to redraw the region’s geopolitical map has turned Iraq into a killing field of terrifying magnitude. By 2007, more than four million displaced Iraqis—about one in six, or approximately 15 percent of the population—were either refugees or internally displaced persons (IDPs). The US occupation has led to one of the largest refugee flows in decades, a humanitarian emergency and political crisis that has been all but ignored by the US, drastically under-covered by the media and dithered over by the international community. What is novel about this refugee crisis is the discursive and spatial: the silence about the displaced, the absence of refugee camps and the minimal humanitarian assistance.
Everyone in His Enclave
By the spring of 2007, the number of Iraqi refugees was staggering, according to UN High Commissioner for Refugees (UNHCR) statistics. An estimated two million Iraqis had sought refuge across the border either in Jordan (between 750,000–1,000,000, or about 15 percent of Jordan’s population) or in Syria (1.4 million, or 7 percent of that country’s population). Thousands more were in Egypt, Lebanon, Iran, the Arab Gulf states and Turkey. Neighboring host states are increasingly closing their borders to Iraqis, in contravention of international law on the right to seek asylum. Within Iraq, over two million people are estimated to be IDPs. In the new global politics of displacement, IDPs, those who flee their homes but do not cross an international border, mushroomed from 1.2 million in 1992 to over 20 million in 2006, significantly outnumbering refugees.
Since the bombing of the Samarra’ mosque in February 2006, around 1,037,615 Iraqis became IDPs at a rate of 80,000–100,000 people per month; this figure does not include IDPs from prior to February 2006. [2] As brutal ethnic or sectarian cleansing has escalated, people have sought refuge in neighborhoods with a prevalence of their particular sect. Once cosmopolitan or “mixed” [3] neighborhoods are becoming forcibly homogenized spaces. The extreme violence—threats, torture, kidnappings, murder—it takes to effect such ostensibly homogeneous spaces is an indication of stubborn Iraqi resistance to sectarianism.
Most IDPs are from the Baghdad area and have taken refuge in central and southern Iraq or within Baghdad itself. The vast majority have sought shelter and assistance from kin or survive on fast dwindling savings. Those fleeing the war in the early months were often middle-class and elite professionals who had some capital. More recent IDPs, with fewer assets, face extreme poverty and lack of services—nutritional, educational and medical. Increasingly, southern and central governorates are overwhelmed and are asking for a halt to the influx of IDPs. In many cases, they are restricting access and tensions are running high between IDPs and locals.
In a March 2007 assessment, the International Organization for Migration determined that less than 1 percent of the IDPs are living in camps, due to their location in the harsh desert climate and the lack of services. With almost no health care, little electricity, minimal sanitation facilities and bare supplies of food and water, the desolate camps are “the last resort.” The camps are usually set up and supplied by the Iraqi Red Crescent and the Ministry of Displacement and Migration, and sometimes by local religious parties. Occasionally, it is local councils that get camps operational with assistance from the Red Crescent and the ministry. The provision of relief and protection by local religious organizations can reinforce sectarianism by creating dependency on sectarian affiliations. Sectarianism thrives when the state is unable to provide security and basic services and the gap is then filled by sect-based organizations. The Red Crescent, the main national aid organization to assist the displaced, works primarily with those in camps and collective settlements and is the “only non-sectarian group with any real structures and a countrywide presence.” [4]
Most of these camps are temporary affairs—they are often open for just a few weeks or months until residents find better accommodations. Some are spontaneous sites created by IDPs in large buildings or schools, and house very small numbers, ranging from 30 to 100 families. Where protection is available, local police provide it. IDPs increasingly report harassment and violence by locals threatened by the influx of IDPs and the resulting pressure on resources and services. UNHCR provides aid, shelter and legal advice at around seven sites in Iraq, but they have not set up camps. IDPs can register with the Ministry of Trade and receive rations, but many have faced serious obstacles transferring their registration to other locales.
The twentieth-century notion of a state for everyone and everyone in a state is being violently rewritten in Iraq, it seems, as everyone in his sectarian enclave and an enclave for everyone. US willingness to countenance such de facto fragmentation indicates a stunning and willful ignorance of the history of partitions—India-Pakistan, Palestine-Israel and Ireland, among others—with their demographic upheavals and their human costs. Dina Abou Samra makes the provocative observation that displacement as a result of US and Iraqi forces is “assessed as a short-term phenomenon, while so-called sectarian-induced displacement is viewed as a long-term trend.” [5] This imaginary Middle Eastern mosaic, which assigns ethnic and sectarian groups to particular spaces and sees them as bounded, coherent, nearly corporate groups, hearkens back to Orientalist and early anthropological elaborations of the region, and to Zionism that challenges the idea of coexistence in a plural social order. In this vision, minorities vie for control of resources, territory and power.
Spaces of Containment
Oft repeated but still worth noting is that Iraqis constitute the largest wave of displaced in the Middle East since the Palestinian refugee crisis began in 1948. Indeed, there are parallels to be found between the fragmentation of Iraq and that of Mandate Palestine. In both places, there has been a communal sorting-out; in both places, the occupying power seeks control of resources underground (oil and water) and above ground (space for military bases and/or settlements), as well as control of the skies, waterways and borders. The contemporary tactics of the US and Israel—walls, barriers, barbed wire, checkpoints, closures and curfews—bear some resemblance to each other. And both the Israeli state and the US occupation of Iraq have produced huge numbers of displaced who are marginalized—even invisible—in the narratives of these conflicts outside the region.
Interestingly, the Palestinians and the Iraqis are both outside the international community’s definition of refugee. The 1951 UN Convention Relating to the Status of Refugees defines a refugee as a person who “owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear,” is unwilling to return. UNHCR’s mandate is to coordinate international action to provide protection and relief and to safeguard refugees’ rights to asylum and non-refoulement—the principle that they may not be deported to whence they fled. People recognized as refugees are also eligible for UNHCR help in seeking “durable solutions” to their displacement: repatriation if the home country is safe, asylum in the host country or resettlement in a third country if not. Palestinians are considered outside the 1951 Convention definition and so receive aid, but no protection, from an organization created specially for them, the UN Relief and Works Agency (UNRWA). Though the 1951 definition of refugee certainly seems to fit displaced Iraqis, they have not been so classified. Whether inside or outside Iraq, they are not in refugee camps where they could perhaps capture media attention—and so, if anything, they are less visible than the Palestinians. UNHCR is providing some assistance to the few Iraqis who register, but the agency is overwhelmed by the numbers and its usual role as coordinator of relief and protection has not been activated. Most significantly, in the face of this nearly unparalleled flow of refugees, the US and the international community have largely been silent, refusing until very recently even to acknowledge the refugees or a humanitarian emergency. Jordan and Syria, non-signatories to the 1951 Convention, have yet to label the Iraqi displaced as refugees.
The absence of Iraqi refugee camps in host countries Jordan and Syria is an eye-opening indication of a major shift in international refugee policy. Reflecting their urban origins, Iraqi refugees have sought refuge, by and large, in urban areas. UN and NGO publications fairly consistently report that Iraqis will not go to camps, and UNHCR states that it is opposed to setting up camps because they are costly and can become permanent. But camps also make refugees visible. And it is important to note that while camps can confine refugees literally and figuratively, they also provide spaces for formulating new identities as well as places from which to organize politically, as transpired in camps from Afghanistan to Palestine to Mexico. Camps, in this way, are small spaces of the nation in exile. When Iraqis are scattered in urban areas such as Amman and Damascus, they may congregate in certain areas, but, unlike camps, their spaces are not set off from the larger society. While the refugees are forming “little Baghdads,” these do not yet resemble camps where social worlds can be recreated and lived daily yet radically transformed in the process. [6] The present Iraqi plight, with the international non-response and the neighboring states’ unwillingness to provide asylum, may herald a new era in warfare in which camps as features of war will disappear. Without camps, will refugees disappear as well, becoming atomized exiles rather than a collective with a voice?
While international refugee law was always geared toward protecting state sovereignty, [7] a more restrictive, state-centric global consensus to prevent refugee movements has emerged as states close their borders. New spatial devices for containing the displaced arose in the 1990s: safe havens, safe corridors and preventive zones. Safe havens crystallized in the early 1990s in Bosnia and northern Iraq to prevent refugees from crossing borders, thus protecting state sovereignty while seeming to provide relief and protection to the uprooted people. The US-led coalition’s Operation Provide Comfort established a protected zone for Kurds in northern Iraq where aid was coordinated by UNHCR. Cast as a humanitarian response, in reality the operation was intended to protect Turkey from a mass refugee influx. Paradoxically, safe havens protected the sovereignty of some states while challenging that of other states, in this case, Iraq. In Bosnia, safe havens prevented a mass movement of refugees into European neighbor states and were accompanied by safe corridors to provide humanitarian access to the besieged havens. The concept of preventive protection was replaced by the even more minimalist concept of preventive assistance. [8] Safe havens have a mixed record, at best. They can provide temporary shelter, but they can also be death traps, as in Srebrenica, where up to 8,000 Bosnian Muslim men and boys were massacred by Serb forces while the town was under ostensible UN protection (the biggest such massacre, by far, to occur in Europe since World War II). The Kurdish safe haven in northern Iraq illustrates this spatial device’s potential to form an embryonic state.
In the post-September 11 world, the displaced are thought of less in terms of their rights under international law or in humanitarian terms, and more as a matter of security, that “black hole” in which things “collapse and disappear,” that “magical term able to absorb any and all content.” [9] In much the same way that the US lumps Hizballah together with al-Qaeda, so some Beltway “experts” categorize the displaced with more threatening others. For example, Brookings Institution analysts Kenneth Pollack and Daniel Byman refer to the difficulties the US faced in stopping the “flow of dangerous people across Iraq’s border…refugees, militias, foreign invaders and terrorists.” [10] In other words, refugees are now the equivalent of terrorists. Elsewhere, Pollack and Byman refer to Iraqi refugees as “carriers of conflict.” [11] “Carrier” evokes a pathogen, bringing disease in its wake, much like Haitian asylum seekers in the US were cast as carriers of AIDS. Once objects of concern and assistance, refugees are now coded as potential criminals and terrorists who may sow instability much as Palestinian refugees in the 1950s were seen as “ripe for recruitment to communism.” [12] In coding the displaced as potential subversives, Pollack and Byman put them with the overlapping and indistinguishable categories of Islamists, terrorists and criminals. It is noteworthy that US admission rates of Iraqis and other Middle Eastern refugees have declined over the past decade, coinciding with the “war on terror,” an atmosphere in which Muslims/Arabs are always suspect. Yet if security were the issue, refugees, who are repeatedly vetted and screened before being admitted, are hardly the logical groups for terrorists to hide among.
The war on Iraq has led some to imagine new spaces of refugee containment. Pollack and Byman call for setting up buffer zones within Iraq—“catch basins”—that would prevent “spillover” of the displaced into neighboring countries where they might prove troublesome and “reduce the legal burdens” they would impose “if they crossed an international border.” [13] While water metaphors for the potential impact of mass displacement—waves, flows, floods, tides, inundations, seas of people—can be difficult to avoid, with the Iraq war they have taken a new twist. A catch basin, according to Webster’s, is “a sieve-like device at the entrance to the intersection of a sewer, for retaining solid matter likely to clog the sewer.” In this hydraulic image, the Iraqi displaced are the equivalent of sludge. Catch basins would be located in border areas close to airfields in Iraq and thus could be easily supplied by the US. In them, refugees would have neither international protection nor an international body accountable to them. The goal of a catch basin is to prevent cross-border movement and, most significantly, keep tabs on the refugees while also disarming and pacifying them. As non-refugees, akin to an ecological byproduct, the people in catch basins are not just a non-political issue, they are hardly even a humanitarian one. Their legal rights would have all but evaporated.
The specter of Palestine partly underwrites such a vision. As the paradigmatic refugees, Palestinians provide lessons for the international management of displacement. Aid workers refer to the “Palestinianization” of a refugee crisis when it is feared it will be prolonged or permanent. To express despair, Iraqi refugees talk about themselves as the “new Palestinians.” Palestinian refugees provide a valuable lesson in the human cost of remapping regions. Iraqi refugees embody a potential new “Middle East crisis,” a rallying point for political sentiment hostile to the US and its Arab allies, in much the same way that Palestinians have for decades. If militarized and politicized, it is surmised, the Iraqis could pose a threat to regional stability. In Palestinian camps and Afghan camps in Pakistan, refugees organized politically, mobilizing and recruiting for militant resistance as well. The camp could, but did not always, serve as a base for training and the launching of militant actions.
In her award-winning book Condemned to Repeat? Fiona Terry has carefully set out how refugee camps or humanitarian sanctuaries, with their connotations of being “civilian, public and neutral” can “provide advantages to guerrilla factions over purely military sanctuaries” which are “militarized, secret and political.” [14] While her suggestion is certainly not to do away with refugee camps, her observations may be used to support such an argument. Similar thinking, along with the fear of Palestinianization, may underlie the apparent interest in spatial or non-spatial alternatives to camps.
Humanitarianism Under Fire
In the absence of camps, where are the spaces of humanitarianism? How is humanitarian aid being distributed and is protection being provided? Could catch basins become the new safe havens? If so, what happens to the right to seek asylum?
Humanitarian space has all but disappeared in Iraq because of the ambient violence and the widespread sense that humanitarian organizations have lost their proclaimed neutrality, often being seen by Iraqis as complicit with the occupying forces. US forces and private contractors often present their activities as humanitarian, thus blurring the distinction between military and non-military, and putting actual humanitarian agencies and their personnel at risk. Attacks on aid organizations have compelled most to move their offices and higher-level staff to neighboring Jordan and Kuwait, where they operate in Iraq by what is now commonly referred to as “remote control.”
During the 1990s, UNHCR gradually began to provide assistance to displaced people who had not crossed borders, as states were increasingly unwilling to shoulder the burden of caring for them and pushed for more restrictions on who could claim refugee status. [15] What of Iraqis in neighboring states today? In Jordan, which initially treated Iraqis as “guests,” only 20,000 are registered with UNHCR as asylum seekers. Because the Iraqi refugees are dispersed in urban areas, the task for international relief agencies is very complex. Here the refugees are mixed with the local population, generating local or internal needs as well. UNHCR’s revised strategy for assisting refugees in Jordan and Syria is to call upon the international community to provide aid to these governments to help them cope.
“Humanitarianism” itself can be a subject of critique. In the late 1960s and throughout the 1970s, Palestinian activists insisted that Palestinian refugees were not a humanitarian issue but a political one. Humanitarian interventions, often associated with charity, were disparaged as depoliticizing what was in essence a political question. To this day, however, Palestinians insist upon remaining on UNRWA rolls, because their registration and ration cards indicate an international responsibility for them and recognition of their loss and their right to Palestine. [16] The very claim of political neutrality that has usually allowed UNHCR and others to intervene also now casts them as ignoring root causes of displacement. The prominence of humanitarian crisis in discourse about the Palestinians and the silence about the politics underwriting the crisis provide a vantage point for such a critique.
An unsettling, if not incredible quiet has attended the trauma of millions of Iraqis. President George W. Bush has yet to even mention the refugees or the IDPs in public. To acknowledge that over four million Iraqis have been displaced would be to admit to the unimaginable violence and chaos generated by the US occupation; not only has the war been lost, but it has also unleashed an enormous humanitarian disaster for which the US bears primary responsibility. US officials are aggressive in their denial of these realities. Former Ambassador to the UN John Bolton was merely blunter than his colleagues still in government when he stated that today’s Iraqi refugees have “absolutely nothing to do” with the US invasion and occupation. Furthermore, he asserted, “Our obligation was to give them new institutions and provide security. We have fulfilled that obligation. I don’t think we have an obligation to compensate for the hardships of war.” [17] While the war itself may be daily front-page news, meanwhile, Iraqi displacement is one of the least covered humanitarian crises in decades. In the West, particularly in the US, there are few visual images and almost no voices of displaced Iraqis. A startling comparison can be made to the displaced Kosovars, the Iraqi Kurds displaced in 1991 and the displaced in Darfur today. [18] Darfur is treated as a classic twentieth-century refugee crisis, perhaps because there is little risk that Darfuris will emigrate in large numbers to the West or because, in the discourse of the “war on terror,” the Sudanese regime, coded as “Arab” and “Islamic,” is responsible, making apportionment of blame and accountability logical and politically convenient. In comparison, a curtain has descended over the dismantlement of Iraq and the brutal dispersion of a significant portion of its population. Iraqi refugees are seen as a disposable byproduct.
Prior to the 2003 invasion, the US annually admitted several thousand Iraqi refugees. In the wake of defeat in Vietnam, the US was able to resettle over 130,000 Vietnamese, primarily those who had worked with US forces, in under a year. This occurred despite public opposition and a refugee resettlement program that provided many more financial benefits than it does today. Nearly 80,000 Kosovars were admitted under emergency circumstances in the late 1990s. Yet less than 500 Iraqis have been admitted since 2003 while Sweden accepted 9,000 Iraqi refugees. With criticism and pressure mounting from the international community, Congress and the UN, the Bush administration has agreed to admit 7,000 Iraqis in 2007. Most of those admitted will come from the ranks of those who assisted US forces in some capacity or those considered extremely vulnerable.
Toward Exclusion
As refugee fatigue and the recognition that masses of refugees can destabilize neighboring countries took hold over the past two decades, UNHCR began to favor repatriation over resettlement and integration as the “durable solution” for refugee crises. Yet in the Middle East, there is a double irony: The international community has never seriously considered repatriation of Palestinian refugees. At the same time, for Iraqis, resettlement is presented by UNHCR as a preferred option. Why is repatriation not on the agenda for Iraqi refugees and where are they to resettle? Without a massive infusion of aid, the absorptive capacity of Jordan and Syria may have reached its peak, not to speak of these countries’ political willingness to absorb a new population. The US is taking absolutely paltry numbers, and while European states accept relatively more, the total is not enough to make a dent.
The displaced Iraqis are emblematic of the catastrophe in the wake of the US invasion. Their uprooting seems poised to join those of the Armenians, Palestinians and Kurds as historic human tragedies that sketch the demographic and political map of the region anew. There are a host of questions that only the passage of time can answer: Will living in exile engender particular political identities? A critical question concerns the role of institutions. UNRWA has been a pivotal and transformative institution, shaping Palestinian refugee identity in manifold ways. The fact that every Palestinian refugee family received UNRWA rations, for instance, meant that they were all in the same boat. In the absence of camps and an identifiable refugee aid regime, will Iraqi refugeehood be remotely equivalent? And how will differences between Iraqis and, say, Jordanians play out politically? Especially where refugees settle among citizens, distinctions between the two can become sources of tension. Refugee influxes can drive up the cost of housing and food and put tangible pressure on services; humanitarian agencies assist refugees, but not the citizens. Will Iraqis remain in “little Baghdads,” reproducing an Iraqi identity outside the boundaries of the state while transforming urban areas in the host countries? Will sectarian affiliations intensify or will Iraqis become cosmopolitan exiles—and what are the implications for political organizing?
The exclusion of the Iraqi displaced from human rights or humanitarian discourses as used in the West poses another set of questions. Should advocates for the displaced conceptualize them as “forced migrants,” as some are doing? Or does this classification dilute the international commitment to provide assistance, protection and durable solutions? This is a period of ambiguous terms as old ones are cast aside and new ones have yet to prevail. On the one hand, the twentieth-century concept of the “refugee” arose from the displacement that followed war and exclusivist nationalisms and, on the other, it came from the subsequent emergence of administrative regimes that enumerate and govern the displaced, and in so doing construct them as a legal category and as subjects of intervention. In its very usage, “refugee” once called for international action. Will “forced migrant” eventually do the same?
It is fairly clear that the non-recognition of the Iraqi displaced as refugees portends further redefinition of the term in a way that diminishes the rights to asylum, protection and assistance. In other words, fewer and fewer people will be able to claim refugee status in the future. Closing borders, which produces IDPs rather than refugees, protects the sovereignty of potential host states and is thought to prevent regional destabilization. New spatial devices beyond the camp and the safe haven seem to be in the works. Or perhaps, there will simply be non-places for the displaced as they melt into the woodwork. Non-recognition mutes the voice of refugees and renders the nominally responsible parties oblivious to their needs. The lack of a concerted response to the Iraqi humanitarian crisis may be indicative of a gradual shift from a concern with refugee rights to increasing invisibility and exclusion on a selective basis. While some displaced (Iraqis, Palestinians and Somalis) remain unseen and hardly heard, others (Darfuris and Kosovars) are clearly visible in comparison.
Endnotes
[1] See Seteney Shami, “Mobility, Modernity and Misery: Population Displacement and Resettlement in the Middle East,” in Seteney Shami, ed. Population Displacement and Resettlement: Development and Conflict in the Middle East (New York: Center for Migration Studies, 1994).
[2] Iraqi Red Crescent statistics cited by the UN Office for the Coordination of Humanitarian Affairs and published at IRINnews.org, July 9, 2007.
[3] For a pointed and poignant examination of the term “mixed areas,” see Burhan al-Mufti, “Mixed Areas: A Dangerous Term,” Middle East Report 239 (Summer 2006).
[4] Ashraf al-Khalidi and Victor Tanner, “The Remorseless Rise of Violence and Displacement in Iraq,” Refugees (April 2007).
[5] Dina Abou Samra, “Military-Induced Displacement,” Forced Migration Review (June 2007), p. 37.
[6] See Julie Peteet, Landscape of Hope and Despair: Palestinian Refugee Camps (Philadelphia: University of Pennsylvania Press, 2005); and Laura Hammond, This Place Will Become Home: Refugee Repatriation to Ethiopia (Ithaca, NY: Cornell University Press, 2004).
[7] See Michael Barnett, “Humanitarianism with a Sovereign Face: UNHCR in the Global Undertow,” International Migration Review 35/1 (Spring 2001).
[8] Bill Frelick, “Preventing Refugee Flows: Protection or Peril?” World Refugee Survey 1993 (Washington, DC: US Committee on Refugees, 1993), pp. 5–13.
[9] Samera Esmeir, “Introduction: In the Name of Security,” Adalah’s Review 4 (Spring 2004), p. 3.
[10] Kenneth Pollack and Daniel Byman, “Iraq Runneth Over: What Next?” Washington Post, August 20, 2006.
[11] Kenneth Pollack and Daniel Byman, “Iraqi Refugees: Carriers of Conflict,” Atlantic Monthly (November 2006).
[13] Daniel Byman and Kenneth Pollack, Things Fall Apart: Containing the Spillover from an Iraqi Civil War (Washington, DC: Brookings Institution, January 2007), pp. 44–45.
[14] Fiona Terry, Condemned to Repeat? The Paradox of Humanitarian Action (Ithaca, NY: Cornell University Press, 2002), pp. 9–10.
[17] Quoted in Nir Rosen, “The Flight from Iraq,” New York Times Magazine, May 13, 2007.
[18] For a probing look at campaigns for Darfur in the US, see Mahmood Mamdani, “The Politics of Naming: Genocide, Civil War, Insurgency,” London Review of Books, March 8, 2007.
