The end of April 9, 2003.
The United States invaded Iraq on March 20, 2003 (with initial airstrikes beginning late on the evening of March 19 in U.S. time zones). [1, 2]
Key Details of the Invasion
Operation Name: Known as Operation Iraqi Freedom. [1]
Coalition: Led by the U.S. alongside forces primarily from the United Kingdom, Australia, and Poland.[1]
Stated Justification: The George W. Bush administration claimed Iraq possessed weapons of mass destruction (WMDs) and posed a threat following the September 11 attacks. [1]
Outcome of the Initial Phase: Major combat operations concluded quickly, with the capital of Baghdad falling on April 9, 2003, and the regime of Saddam Hussein being toppled. [1, 2]
After the Americans Leave
Iraq, Kurdistan, and the unfinished question of who commands a country
When the Cold War ended, a considerable part of the American political establishment believed that the great argument over the future had been settled. The Soviet Union was disintegrating, the United States possessed unmatched military power, and the coalition victory over Saddam Hussein’s occupation of Kuwait in 1991 seemed to demonstrate that this power could organize an international order as well as defeat an army. The question increasingly became what America should do with its predominance, rather than how long that predominance would last.
Francis Fukuyama’s 1989 essay, “The End of History?”, gave philosophical expression to the apparent exhaustion of liberal democracy’s ideological competitors. The Pentagon’s 1992 draft Defense Planning Guidance, prepared under officials including Paul Wolfowitz, expressed a different ambition: preserving American strategic supremacy by preventing a new rival from emerging. These were distinct arguments, not a single blueprint for invading Iraq. Yet they belonged to an atmosphere in which American institutions could appear to be history’s destination and American power the means of securing the journey. 1
After the attacks of September 11, 2001, that confidence acquired a more immediate military expression. Retired General Wesley Clark later recounted that an officer at the Pentagon had described a memorandum envisaging action against seven countries within five years: Iraq, Syria, Lebanon, Libya, Somalia, Sudan and Iran. Clark did not read the document, and subsequently acknowledged that it might have been a discussion paper rather than an adopted plan. His account is evidence of the ambitions he encountered in official circles, rather than proof of a settled presidential timetable for seven wars. Even with that qualification, it helps explain why Iraq was discussed as the beginning of a regional transformation. 2
That larger ambition was also expressed publicly. Testifying before a congressional committee on September 12, 2002, Benjamin Netanyahu, then a former Israeli prime minister, urged the removal of Saddam and predicted “enormous positive reverberations on the region.” He suggested that the effects could extend to Iran. This was an explicit argument for using American military power to alter the regional balance, advanced by a prominent Israeli advocate who expected the change to improve security. It belongs in the history of the case for war, although advocacy is not the same as control over the American decision. 3
The distinction between confidence and capacity became consequential in Iraq. The 1991 war expelled Iraqi forces from Kuwait and left Saddam in power. The invasion of 2003 removed his government and undertook the much larger task of reconstructing the state beneath it. Twenty-three years later, with the remaining American coalition forces scheduled to depart by September 30, 2026, Iraq is still negotiating a question that neither military victory nor a written constitution has conclusively answered: who may exercise armed power, and whose orders must they obey? 4
It is tempting to place this departure beside the continuing American presence in Germany, Japan and South Korea. Those comparisons illuminate the scale of the ambition, but they require historical care. Germany and Japan emerged from the total defeat of the Axis powers; the American presence in South Korea developed through occupation, the Korean War and a subsequent defense alliance. The Marshall Plan supported European recovery, including the western German occupation zones and subsequently West Germany. Japan’s reconstruction followed a separate occupation framework under the Supreme Commander for the Allied Powers, with Joseph Dodge’s 1949 stabilization program addressing its finances and inflation. South Korea was not a Marshall Plan recipient either. 5 Their later American bases became parts of enduring security arrangements with sovereign governments. They were not simply invasions that Washington forgot to conclude. 6
Iraq’s experience was different. The invasion was a war of choice, justified centrally by weapons claims that were not borne out, followed by occupation and an attempted political transformation amid mounting armed resistance. America spent heavily on reconstruction and security, so the distinction is not that Iraq received no rebuilding effort. It is that rebuilding proceeded through an insurgency, sectarian conflict and competing foreign interventions, while the legitimacy of the new order remained contested. American departure is not unprecedented—Vietnam and Afghanistan forbid that conclusion—but Iraq is another reminder that prolonged military presence does not necessarily produce a durable political settlement.
The human cost belongs near the beginning of this account. Iraq Body Count records more than 200,000 civilian deaths from violence since the invasion. Other methods, including household surveys estimating excess mortality, have produced substantially higher totals. A widely circulated ORB estimate exceeded one million Iraqi deaths, but its methodology was seriously challenged; it should not be presented as an established count of civilians killed. Documented violent deaths, estimated excess deaths, and deaths caused directly by American forces are different measures. Precision does not diminish the catastrophe. It prevents an argument over numbers from substituting for recognition of the people whose lives were ended or permanently altered. 7
The warning associated with Colin Powell—that breaking Iraq would make America responsible for it—concerned precisely this distance between overthrowing a government and governing the aftermath. Powell himself helped make the public case for war, but the responsibility identified in that warning extended far beyond the battlefield. Removing a ruler would leave the invading power facing the needs of the people and institutions that remained. 8
For those who survived, the collapse of Saddam’s government did not mean that the Iraqi state became a neutral institution they could immediately trust.
The army that was dissolved, and the enemy that emerged
Saddam’s dictatorship had concentrated power through the Ba’ath Party and coercive institutions whose senior networks were disproportionately Sunni Arab, without representing Sunni Arabs as a whole. After 2003, Shia parties acquired the electoral weight that Iraq’s Shia majority had long been denied, while Kurdish leaders sought to preserve the self-government they had developed in the north. Many Sunni Arabs experienced the transition through exclusion, insecurity and the loss of employment or influence.
These categories require explanation before they can explain anything else. “Sunni” and “Shia” describe religious affiliations within Islam; “Kurdish” describes an ethnic identity. Many Kurds are Sunni Muslims, but that does not align their national aspirations with those of Sunni Arab parties. Nor did membership in any of these communities determine whether an Iraqi supported the occupation, the new government or an armed organization claiming to act on the community’s behalf.
American decisions helped turn these differences into struggles over employment, security and political survival. De-Ba’athification removed party members from public institutions, while Coalition Provisional Authority Order No. 2, signed by L. Paul Bremer on May 23, 2003, formally dissolved the armed forces and other security organizations. It was an order of the American-led occupation authority under the Bush administration, rather than a presidential executive order. Much of the army had already dispersed during the invasion; this was not a single, orderly surrender of the entire institution. The subsequent legal decision foreclosed simply recalling its existing units and made clear that the old institution would not be restored. 9
The decision dismissed hundreds of thousands of military personnel, leaving men with weapons training, professional connections and increasingly uncertain futures outside the new state. After protests, the occupation announced stipends for former career soldiers and a one-time payment for former conscripts. Those payments could alleviate hardship, but they could not restore the dissolved institution or the place its members had held within it. Some subsequently joined the insurgency against the occupation and the government being assembled under its protection; others did not. 10
Meanwhile, the collapse of authority had already exposed ministries, public facilities and weapons depots to looting. The chronology matters: this began during the invasion and before Bremer’s May order. The occupation then compounded the breakdown by dismantling military institutions while failing to secure large stocks of ammunition. A later Government Accountability Office investigation connected looted munitions to the improvised explosive devices that killed and maimed people during the insurgency. The failure left armed opponents with access to both military expertise and the materials of war. 11
Among the armed organizations that developed during this period was the network led by Abu Musab al-Zarqawi, a Jordanian jihadist whose objectives extended beyond expelling American troops or restoring the old order. His organization became al-Qaeda in Iraq in 2004 and sought to provoke a sectarian war, attacking Shia civilians as well as occupation forces and Iraqi institutions. Its successors established the Islamic State of Iraq in 2006, subsequently expanded into the Syrian civil war, and adopted the name Islamic State of Iraq and al-Sham—commonly rendered in English as the Islamic State of Iraq and Syria, or ISIS—in 2013. Its ambition was to establish a state governed by its own extremist interpretation of Islam across existing national borders. 12
The connection to Saddam’s disbanded military was consequential, although it was not a straightforward institutional succession. Former officers entered the jihadist organization and contributed skills in military planning, intelligence and control of populations. One prominent example was Haji Bakr, a former Iraqi intelligence officer whose planning helped ISIS establish itself in northern Syria. The movement combined religious militancy with expertise acquired under the regime that America had overthrown; dissolving that regime’s institutions had not erased the capabilities of the people who served them. 13
ISIS was therefore neither the old Iraqi army simply renamed nor an inexplicable eruption of religious fanaticism. Its rise brought together an existing jihadist project, the destruction and dislocation accompanying the invasion, subsequent political exclusion and the opportunities created by another war across the Syrian border. Sunni Iraqis were among its recruits, but they were also among its opponents and victims. Treating them collectively as its constituency would repeat the confusion between community and armed organization that helped poison Iraqi politics. 14
By the time ISIS captured Mosul in June 2014, it had developed through more than a decade of war and insurgency. The collapse of Iraqi military formations before its advance raised an immediate question for communities facing it: if the national army could not protect them, who would?
The Popular Mobilization Forces, commonly called the PMF or Hashd al-Shaabi, expanded in response to that emergency, bringing together existing armed organizations and newly mobilized volunteers. They were predominantly Shia but included other communities. Some factions were closely aligned with Iran; others looked to Iraqi religious authorities or political leaders whose interests did not coincide with Tehran’s. Their contribution to resisting ISIS helped give them legitimacy that a description such as “foreign proxies” cannot fully explain. 15
The difficulty came in determining what that legitimacy entitled them to do afterward. Armed organizations acquired or retained political representation, access to public resources and economic interests. Some became simultaneously participants in government and centers of military authority that the government could not reliably direct. A prime minister could therefore face an organization whose political allies helped sustain his administration while its commanders resisted his orders. The problem was not merely that armed men existed outside the state; some were embedded within its institutions without surrendering their separate loyalties. 16
An American difficulty before it became an Iraqi one
Americans need not regard this as an incomprehensible feature of the Middle East. Their own revolution began with armed forces rooted in towns and colonies, and the creation of a Continental Army did not immediately erase local loyalties or establish unquestioned national command. Washington’s task included turning a collection of men fighting the same enemy into an army capable of accepting common discipline. 17
Imagine a different aftermath to that revolution. Massachusetts and Virginia help win independence, but their armed organizations retain separate commanders, develop political parties and obtain positions in the national government. Congress pays some of their expenses, yet cannot be certain that they will obey its instructions. A foreign ally maintains relationships with particular commanders, while local communities continue to regard those commanders as the men who protected them when national institutions could not.
This is an analogy, not an account of what happened in America or a claim that Iraq’s factions share the American revolutionaries’ purposes. It identifies a familiar political difficulty: the authority earned by defending a community does not automatically transfer to a national government when the emergency passes. Paying an armed organization from the treasury is not the same as commanding it.
This also explains why demanding an American withdrawal need not mean accepting Baghdad’s authority. Particular Shia parties and armed factions have pressed strongly for departure, but “the Shia south” is too broad a description of the actors involved. Opposition to foreign troops can arise from nationalism, alignment with Iran, anger over American strikes, or a desire to remove a constraint on an organization’s own power. Those motives may coincide on withdrawal and diverge immediately afterward.
September reporting describes precisely that tension: Baghdad has connected the departure of coalition forces with demands that armed factions relinquish their independent military power, while major factions have resisted. Removing the foreign army and establishing the authority of the national government are therefore two different undertakings, even when they appear on the same political timetable. 18
What Baghdad owes Erbil
Iraqi Kurdistan occupies a particular position within this unsettled arrangement. It is a constitutionally recognized federal region, not an independent country. It has its own elected institutions, laws and security forces, while national defense remains a federal responsibility. Its autonomy means substantial self-government within Iraq; it does not mean complete sovereignty over its foreign relations and defense. 19
California offers a useful first comparison. Sacramento governs many internal matters, but Washington is responsible for defending it against foreign attack. If California complained that departing British troops would leave it undefended, an American reader would reasonably ask what the United States military was doing.
That is also a reasonable question to ask about Erbil. It directs attention to Baghdad’s responsibility before passing judgment on Washington’s departure. The difference is that California operates within an established national military system, whereas Kurdish leaders remember both Saddam’s campaigns against their population and more recent confrontations with federal forces over disputed territory. Their demand for protection exists alongside apprehension about the power of the institution expected to provide it.
Cooperation is nevertheless possible. Baghdad and Erbil coordinate operations against ISIS, and federal border guards operate along Kurdistan’s international frontier. More specifically, in June 2026 federal military delegations visited the region to assess protection for energy installations. Officials discussed supplying air defenses, and Kurdish leaders welcomed the initiative. This is evidence against a sweeping assertion that Baghdad refuses even to consider defending Kurdistan. 20
It is not evidence that effective protection was delivered. Protecting selected oil facilities is a narrower undertaking than defending a city, and assessing equipment requirements is different from installing operational systems. The public sources reviewed for this essay do not establish exactly what the June initiative produced by September, or demonstrate a single reason for any delay.
That uncertainty matters because Masrour Barzani’s warning concerns a practical vulnerability. American systems have intercepted attacks around their installations, providing some incidental protection beyond their own personnel, although they have never constituted a comprehensive shield for Kurdistan. Baghdad’s constitutional responsibility does not by itself provide a replacement. Conversely, the departure of an American system does not prove that every defensive capability disappears. Reports of the drawdown have contained conflicting descriptions of what remained. 21
The Kurdish complaint is therefore neither self-evidently absurd nor sufficient proof of betrayal. Its merits depend on the protection actually available, the threats it must meet and the alternatives the parties can arrange.
An insurgency does not need a majority to survive
The ISIS question belongs within this account, but it must remain distinct from missile and drone attacks attributed to Iran or particular Iraqi armed factions. The organizations, objectives and means of violence are different. Evidence that Iran supports a Shia armed group is not evidence that it supports ISIS, and failure to establish Israeli or Chinese sponsorship does not identify Iran by elimination.
China’s expanding economic position supplies another distinction. Its companies acquired substantial roles in Iraq’s oil industry after the invasion; in 2009, CNPC and Britain’s BP signed a contract with Iraq’s state-owned South Oil Company to develop the Rumaila field. That partnership complicates the picture of China simply entering wherever Western capital withdrew. Beijing could gain commercial opportunities while Washington bore the costs of war, but economic gains do not establish responsibility for the violence. Evidence of investment, evidence of advocacy for regime change and evidence of supplying insurgents answer three different questions. 22
Describing ISIS as a criminal gang is equally insufficient. Its extortion and illicit financial activity serve an ideological and political project. A criminal organization may use violence chiefly to preserve a profitable market; an insurgency may use criminal income to sustain a struggle over who governs and under what rules. Actual organizations can combine these purposes, but the distinction helps explain why interrupting a revenue stream does not necessarily extinguish the cause for which members believe they are fighting.
Persistence also need not imply broad popular consent. A small armed network can survive through a mixture of committed supporters, intimidated residents, paid intermediaries, concealed resources and incomplete policing. It is much easier to deny such a network control of a city than to guarantee that none of its members can hide, recruit, obtain weapons or organize another attack.
The disputed territories between federal and Kurdish control have supplied opportunities of this kind. Historical reporting documents how poor coordination allowed ISIS to exploit spaces between their forces. Joint operations address that weakness, but their existence does not establish that territorial disputes or distrust have disappeared. Nor should evidence about coordination failures in 2021 be silently treated as an exact description of military conditions in 2026. 23
Weapons evidence requires the same discipline. Conflict Armament Research’s examination of more than 40,000 items recovered in 2014–2017 documented capture, diversion and commercial supply chains. A weapon’s country of manufacture did not necessarily identify the government that deliberately supplied it to ISIS. That historical investigation helps explain how an arsenal accumulated; it does not provide a complete inventory of the networks sustaining today’s cells. 24
The state after the intervention
As the September deadline approaches, Iraq cannot be understood simply as a country America is abandoning or a country finally liberated by America’s departure. Iraqi leaders have agency, responsibilities and competing interests; American policy helped create the circumstances in which those interests now contend. Neither fact cancels the other.
The United States removed a dictatorship, dissolved its military institutions and helped build a new constitutional order. Some men displaced from the old institutions subsequently helped strengthen an insurgent movement that threatened the new ones. America later returned to help defeat that movement’s territorial state, while armed organizations mobilized against it acquired their own lasting claims to political power.
Even the language of departure and return needs care. The withdrawal of the main American military force in 2011 did not mean the disappearance of every American servicemember: a small Office of Security Cooperation remained, alongside diplomatic protection arrangements. The Army’s official history records 157 military personnel in that office at the withdrawal. The intervention against ISIS beginning in 2014 established a renewed military campaign on a different scale. The prospective end of the coalition mission should therefore be described in terms of the forces and functions actually ending, rather than as either an unbroken twenty-three-year deployment or a guarantee that every American military connection will disappear. 25, 26
Those successive interventions changed who could exercise power, but they did not produce a universally trusted authority to which every armed organization would submit.
There is no satisfactory principle under which American soldiers must remain indefinitely because political settlement is unfinished. There is equally little comfort in treating the completion of a withdrawal schedule as evidence that the people remaining behind possess adequate security.
The post-Cold War confidence with which this history began assumed that certain large questions had been answered. Iraq’s experience directs attention to questions more elementary than the final form of government: whether a citizen can rely on the state for protection, whether a region can accept that protection without fearing subordination, and whether a commander will obey a government when obedience conflicts with his party, patron or private interest.
For the families who have lived through invasion, dictatorship, insurgency and displacement, the answer will not be found in the departure ceremony. It will be found in whether the authorities who promise to protect them can make that promise effective after the foreign soldiers have gone. Honoring the dead also requires an honest account of how the institutions meant to protect the living were broken, and why their replacements have remained so difficult to trust.
