> ## Content Index
> Fetch the complete content index at: https://otherwise.ghost.io/llms.txt
> Use this file to discover other available public pages before exploring further.

# Iraq’s Unpressed Backspace Key
- URL: https://otherwise.ghost.io/iraqs-unpressed-backspace-key/
- Published: 2026-07-01T19:34:17.000Z
- Updated: 2026-07-01T19:34:17.000Z
- Description: Why Article 140, Federalism, and Militia Power Must Be on the Washington–Baghdad Agenda
- Author: Khaled Salih
- Tags: Newsletter, #Migrated-1792244264359, #Import 2026-10-17 14:38

1 July 2026

\[[Download the paper here](https://www.academia.edu/169491656/Iraqs%5FUnpressed%5FBackspace%5FKey%5FWhy%5FArticle%5F140%5FFederalism%5Fand%5FMilitia%5FPower%5FMust%5FBe%5Fon%5Fthe%5FWashington%5FBaghdad%5FAgenda?ref=otherwise.ghost.io)\]

The argument of this essay is simple: Iraq’s crisis is not only militia power, oil conflict, or Kurdish grievance. It is the failure to implement the constitutional correction mechanisms that made post-2005 Iraq morally legitimate. Article 140, the Federation Council, fiscal federalism, and a properly constituted Federal Supreme Court were supposed to turn Iraq from a coercive state into a correctable federal union. Iran-backed militias and centralizing parties helped disable that architecture. If Washington wants a sovereign Iraq after the recent confrontation with Iran, it must not settle for Baghdad’s language of unity. It must ask whether Iraq is willing to restore the constitutional bargain that made unity legitimate.

**Author’s Note: Why This Argument Matters Now**

Iraq’s new prime minister is expected to visit Washington in mid-July to meet President Donald Trump at a moment when the regional balance has shifted. The confrontation with Iran has exposed again the central question that has haunted Iraq since 2005: can Baghdad become the sovereign constitutional state promised in its own Constitution, or will it remain a field in which external power, militia pressure, and centralizing habits disable federal correction?

This essay is written for that moment. It does not ask Washington to choose Kurdistan against Iraq. It asks Washington and Baghdad to remember what Iraq’s own Constitution promised: a federal, plural, voluntary Iraq in which the crimes of the old state would be corrected through law rather than preserved through delay. If the new Iraqi leadership wants international support, investment, and sovereign respect, it must show that sovereignty means more than resisting foreign interference. It must also mean honoring the constitutional bargain with Kurdistan.

The question before Washington should therefore not be only whether Baghdad can control militias. It should also be whether Baghdad is willing to restore the constitutional mechanisms that Iran-backed parties and militias, centralizing courts, and post-2005 political opportunism helped disable. Article 140, the Federation Council, fiscal federalism, oil and revenue arrangements, and the constitutional status of the Kurdistan Region are not side issues. They are the test of whether Iraq can become a state governed by correction rather than coercion.

**Abstract**

The 2005 Iraqi Constitution was not merely a legal settlement after dictatorship. It was an attempt to design a forgiving architecture for a broken state. Its deepest promise was not simply unity, democracy, or federalism as administrative technique. Its promise was that Iraq could be voluntarily reconstituted after the crimes of Baathism: Halabja, Anfal, Arabization, displacement, repression, and the long denial of Kurdish political existence. The Kurds achieved something extraordinary in the constitutional negotiations. They elevated the Kurdish question from a security problem, minority claim, or temporary autonomy into the architecture of the new Iraq itself. Federalism became the design principle by which Iraq might be rebuilt as a voluntary union rather than a compulsory state.

But the tragedy of post-2005 Iraq is that the forgiving architecture was gradually made unforgiving in practice. Once the Shia-led political order in Baghdad stabilized its control over the state institutions, many of the mechanisms that made forgiveness possible were delayed, narrowed, or disabled. Article 140 was not implemented. The Federation Council was not built. Oil and revenue arrangements remained unresolved. The Federal Supreme Court became increasingly instrumental to disputes in ways the Kurdistan Region experienced as centralizing. Budget and salary mechanisms became tools of pressure. The constitutional promise of free union became a practice of compelled unity.

This essay argues that Article 140 was not merely a territorial clause. It was also the moral center of Iraqi federalism, because it addressed the place where genocide, Arabization, Kirkuk, land, memory, and constitutional repair met most visibly. It was the Constitution’s forgiveness mechanism for the disputed territories. Its sequence — normalization, census, referendum — was designed to make choice possible after coercion. Normalization would remove the demographic distortion of Arabization; the census would establish facts after correction; the referendum would allow the people to decide once the conditions of choice had been repaired. By refusing to implement Article 140, Iraq did not merely delay a vote. It preserved the coercive reality that the vote was meant to correct.

The 2017 Kurdistan referendum should therefore be understood not as a sudden act of separatist irrationality, but as the return of the Constitution’s own unresolved premise: Iraq was to be a free union, entered into by choice. When the architecture of forgiveness was blocked, the Kurds, peacefully, appealed back to the voluntary foundation of the state. Baghdad’s answer was to treat that appeal as unconstitutional. The result was the final exposure of the post-2005 contradiction: Iraq had written voluntary federalism into its founding language, but when the Kurds tested the voluntariness of the union, the state answered with coercive subjugation, again.

⸻

**The Idea of Forgiving Architecture**

A forgiving architecture is a structure designed for fallible beings.

In engineering, a forgiving design assumes that mistakes will happen. Doors must be opened by tired hands. Roads must be used by imperfect drivers. Buildings must survive stress, weather, human error, and time. A forgiving architecture does not pretend that failure can be abolished. It designs spaces in which failure need not become catastrophe.

A constitution should do the same.

A serious constitution is not written for angels. It is written for wounded people, rival memories, competing ambitions, bad rulers, frightened minorities, exhausted citizens, dangerous passions, and inherited crimes. Its task is not to abolish conflict. Its task is to make conflict corrigible. It should allow a society to press Backspace before error becomes blood.

In this sense, the 2005 Iraqi Constitution was supposed to be a forgiving architecture.

It was written after dictatorship, genocide, war, sanctions, occupation, sectarian terror, and state collapse. It inherited a country whose political map had been shaped by imperial invention, monarchy, Arab nationalism, Baathist centralization, demographic engineering, and mass violence. No constitutional text could make such a country innocent. But a constitution could do something else. It could create a structure in which inherited errors could be corrected without requiring the destruction of Iraq.

That was the Kurdish achievement.

The Kurds did not merely secure recognition. They forced the new Iraq to admit that the old Iraq had been wrongly built. Federalism was not a decorative word. It was the architectural principle by which Iraq could be voluntarily restructured. Article 140 was not a side clause. It was the correction mechanism for territories whose reality had been falsified by state coercion. The Preamble’s language of free union was not poetic excess. It was the moral foundation of a state that could no longer claim legitimacy by force alone.

The Constitution’s promise was therefore this:

Iraq would continue only as a union of peoples who chose to share a future under a federal democratic order.

That was the forgiveness.

Not forgiveness as forgetting.

Not forgiveness as absolution.

Not forgiveness as telling Kurds to move on.

Forgiveness meant that the old state would not be allowed to continue uncorrected. Forgiveness meant that the victims of the old architecture would not be asked to live forever inside its rooms. Forgiveness meant that the Constitution would build exits from inherited coercion: federalism, regional powers, language equality, revenue fairness, disputed-territory correction, and the right to participate in reshaping the state.

The tragedy is that this forgiving architecture was gradually made unforgiving.

**The Kurdish Constitutional Achievement: Turning Survival into Architecture**

The Kurds entered post-Saddam Iraq with a unique political position.

They had survived Anfal, chemical attack, mass displacement, village destruction, Arabization, and decades of denial. After 1991, under the protection of the no-fly zone, they had also built a de facto self-governing region. By 2003, the Kurds were not merely a persecuted minority asking for protection. They were a people with institutions, parties, armed forces, a parliament, an administration, and a memory of near-destruction.

Their constitutional demand was therefore not simply autonomy.

It was restructuring.

The Kurds insisted that post-Saddam Iraq could not be rebuilt as the old centralized state with better rulers. The problem was not only Saddam. The problem was the architecture that had made Saddam’s anti-Kurdish policies possible: centralization, Arab-national state identity, unchecked sovereignty, control from Baghdad, demographic manipulation, and the reduction of Kurdish claims to security concerns.

The Kurdish achievement in 2005 was to move these issues into the Constitution.

Iraq would be federal.

Kurdistan would be recognized as a region.

Kurdish would be an official language.

Regional powers would be constitutionally protected.

Kurdistan’s existing authorities would be acknowledged.

Regional laws since 1992 would remain in force.

The disputed territories would be resolved through Article 140.

The Preamble would speak not only of unity, but of a free union chosen by Iraq’s peoples.

This was a profound success. The Kurdish question was no longer outside Iraq’s legal identity. It was placed inside the state-building architecture itself.

The new Iraq was not supposed to be the old state forgiving itself.

It was supposed to be a state rebuilt so that the old errors could no longer govern.

That is why the Constitution matters so deeply. It was not merely a compromise among elites. It was a theory of how a broken country might become legitimate.

The theory was federalism as forgiveness.

**Free Union: The Voluntary Foundation of the New Iraq**

The Preamble’s language is essential. It says that the people of Iraq, “of all components and across the spectrum,” chose freely and by choice to unite their future. It says that adherence to the Constitution preserves Iraq’s “free union of people, of land, and of sovereignty.”

This is not ordinary constitutional rhetoric.

A free union is not the same as compulsory unity.

Compulsory unity says: you belong because the state says so.

Free union says: the state is legitimate because its peoples choose to remain together under agreed terms.

This distinction is the deepest Kurdish success in the 2005 Constitution. The Kurds pushed Iraq beyond the old grammar of sovereignty. They forced into the founding text the idea that Iraq’s unity must be preserved by constitutional adherence, not by coercion alone.

That matters because the Kurds had historical reasons to fear the word “unity.” In the old Iraq, unity often meant Arabization, repression, military campaigns, denial of language, denial of self-rule, and refusal to recognize Kurdish political agency. Unity was not a neutral good. It was the vocabulary through which central power disciplined plurality.

The 2005 Constitution tried to change the meaning of unity.

Unity would now depend on federalism.

Unity would depend on choice.

Unity would depend on rights.

Unity would depend on correcting the old state.

Unity would depend on the Constitution itself.

The Preamble therefore created a profound standard: Iraq’s unity is legitimate only when it remains a free union. If the Constitution is violated, if federalism is hollowed out, if Article 140 is suspended, if regional powers are narrowed, if the budget becomes a weapon, then the state may still claim unity, but it has weakened the constitutional conditions that made unity free.

This is the clause to which the 2017 referendum implicitly returned.

The Kurds did not invent the language of voluntary union in 2017\. They found it in the constitutional founding of Iraq. Their argument was simple, even if politically explosive: if Iraq is a free union, and if the terms of that union have been persistently violated, then the people who entered the union must be able to ask whether they still consent to it.

Baghdad’s answer was: no.

That answer revealed the contradiction.

**Article 140: Forgiveness as Procedure**

Article 140 is the Constitution’s clearest act of forgiveness architecture.

It is clearest because it addresses the concrete territorial afterlife of the Baathist crime. For those close to the constitutional negotiations, its meaning was never merely administrative. It was the place where federalism, genocide, Arabization, Kirkuk, and moral repair converged.

The ethnic cleansing in and around Kirkuk was not a secondary injustice. It was one of the lasting embodiments of the regime’s genocidal project against the Kurds: a policy that did not merely kill, deport, and terrorize, but tried to rewrite the human composition of land itself. Article 140 therefore carried a moral burden larger than ordinary territorial administration. It was the constitutional procedure through which Iraq could acknowledge that Arabization was not a demographic fact to be managed, but a crime-produced reality to be corrected.

It was designed to address Kirkuk and other disputed territories, but its deeper logic was larger. It acknowledged that the old state had manufactured facts. Arabization had not merely governed territory. It had rewritten the conditions under which future political choice would take place.

That is why Article 140 required three stages:

normalization;

census;

referendum.

This sequence is a theory of rational choice after coercion.

Normalization comes first because one cannot ask people to choose inside a reality that has been engineered by force. If people have been expelled, settled, intimidated, deprived of property, manipulated in records, or administratively detached, then the existing situation is not neutral. It is the afterlife of coercion.

Normalization says: correct the inherited distortion.

The census comes next because facts must be established after correction, not before it. A census conducted inside an uncorrected reality risks counting the success of the crime.

The referendum comes last because choice becomes meaningful only after coercion has been addressed and facts have been clarified.

The referendum is the final act.

Normalization is the recovery of choice.

That is the genius of Article 140.

It was not a Kurdish trick. It was not a maximalist demand hidden inside legal language. It was a constitutional recognition that the map inherited from Baathism could not be treated as innocent. The disputed territories were disputed because the state had helped make them so.

This mattered even more because Iraq never received the full truth-telling that a proper genocide trial might have provided. Saddam Hussein was executed after the Dujail case, before the Anfal proceedings could reach the level of public recognition, judgment, and moral record that the Kurdish genocide required. The surviving leaders of the old regime were not made to stand before Iraq, Kurdistan, and the world in a completed constitutional drama of recognition and recompense for genocide. That absence left Article 140 carrying part of the burden that a full reckoning should have carried. If the courtroom did not fully name and repair the crime, then the Constitution’s correction mechanism became all the more prominent, and all the more urgent.

Therefore the state had to participate in correcting the dispute it had created.

In this sense, Article 140 was the constitutional Backspace Key.

It was Iraq’s chance to delete the coercive insertions of the old regime before asking the people of the disputed territories what future they wanted.

That is what forgiveness required.

Not revenge.

Not expulsion.

Not ethnic victory.

Correction.

**Why Non-Implementation Is Not Neutral**

The most dangerous illusion in the Iraqi case is that failure to implement Article 140 preserved stability.

It may have postponed immediate confrontation. But postponement is not neutrality when the existing reality was produced by coercion.

If a state expels people, settles others, changes administrative boundaries, alters records, distributes land, frightens communities, and then delays correction long enough, the resulting status quo begins to look normal. Later, anyone who seeks correction can be accused of disturbing peace.

This is how protected error works.

First the state manufactures a fact.

Then time passes.

Then the manufactured fact is called reality.

Then correction is called destabilization.

Article 140 was designed to prevent exactly this. Its deadline was not incidental. A deadline protects correction from being absorbed by delay. It prevents the old reality from becoming naturalized through time.

The deadline passed.

The correction did not happen.

The result was not neutrality. It was the preservation of the old map under the authority of the new Constitution.

This is why non-implementation is the central failure. Iraq did not simply fail to hold a referendum. It failed to complete normalization. It failed to establish corrected facts. It failed to create the conditions under which a referendum could become genuine choice. It failed to let the Constitution do the forgiving work it was designed to do.

A ballot is not a choice when the reality behind the ballot has been coercively manufactured.

And a constitution is not forgiving when it asks the victims of old coercion to accept the consequences of that coercion as the price of peace.

**The Shia-Led Center and the Gradual Closure of the Architecture**

The post-2005 order was shaped by the rise of Shia-led parties after decades of Baathist Sunni-Arab domination. That shift was historically understandable. Shia communities had suffered repression, exclusion, mass violence, and political dispossession. Their empowerment after 2003 was one of the central transformations of the new Iraq.

But the question is not whether Shia political empowerment was legitimate.

The question is what happened once the Shia-led Baghdad order stabilized its position inside the central state.

Step by step, the incentives changed.

In the immediate post-Saddam moment, the center was weak, the Kurds were organized, and federalism was necessary to build the new order. Kurdish demands could not be ignored. The Constitution reflected that balance.

But as Baghdad’s ruling coalitions consolidated control over ministries, courts, budgets, security institutions, oil policy, and international recognition, the center became less dependent on the forgiving architecture. What had been agreed as a federal structure of correction increasingly appeared to Baghdad as a constraint on restored state authority.

The result was gradual closure.

Article 140 remained unimplemented.

The Federation Council remained absent.

Oil and gas law remained unresolved.

Revenue became a recurring conflict.

Federal supremacy increasingly meant central interpretation.

Regional autonomy became vulnerable to fiscal and judicial narrowing.

The logic of federal partnership weakened.

The logic of majoritarian state recovery strengthened.

This is the key transformation.

The Constitution was born as a bargain among wounded components. But the practice of the state increasingly became the practice of the component that controlled Baghdad.

A federal constitution cannot survive if one community’s temporary control of the center allows it to reinterpret the founding bargain as administrative generosity.

The Kurds had agreed to a new Iraq.

Baghdad increasingly administered an old reflex in new constitutional language.

**Iran, Militias, and the Capture of the Federal Bargain**

There is another layer of the post-2005 failure that must be named carefully. The problem was not Shia political empowerment as such. Shia empowerment after Baathism was legitimate, historically necessary, and constitutionally unavoidable. A democratic Iraq could not be built by returning power to the old Sunni-Arab centralizing order.

The problem was different. Over time, parts of the Shia-led order in Baghdad became entangled with Iranian strategy, militia power, and party networks whose interests did not always coincide with the constitutional reconstruction of Iraq. Iran did not need to abolish the Iraqi Constitution in order to weaken it. It only needed to support forces that benefited from an Iraq whose federal mechanisms remained incomplete, whose courts could be used asymmetrically, whose budget could pressure Kurdistan, whose disputed territories remained unresolved, and whose armed groups could operate beside or inside the state.

This is why the militia question and the Kurdish constitutional question are connected. They are not separate files. Both concern whether Iraq is a sovereign constitutional state or a captured arena. A state that cannot monopolize legitimate force cannot fully honor federalism. A state whose armed actors answer partly to external power cannot credibly resolve Article 140\. A state whose political parties depend on militia leverage cannot build a neutral Federation Council, a fair revenue system, or a trusted constitutional court.

The Iranian-backed militia order damaged Iraq not only by threatening Americans, Israelis, Gulf states, or Iraqi dissidents. It damaged Iraq by making constitutional correction more dangerous. It turned federal disputes into security disputes. It made Kurdish autonomy appear to some central actors not as a constitutional fact, but as an obstacle to the consolidation of a Baghdad-centered order aligned with Tehran’s regional depth.

This is the opportunity created by the present moment. If Washington and Baghdad are serious about Iraqi sovereignty after the recent confrontation with Iran, then sovereignty cannot mean only removing militia rockets from the equation. It must mean restoring the constitutional bargain that militia politics helped deform. The test of a sovereign Iraq is not only whether Baghdad can say no to Tehran. It is whether Baghdad can say yes to its own Constitution.

**The Missing Federation Council: The Room That Was Never Built**

A forgiving architecture needs rooms where conflict can be processed.

Article 65 required a Federation Council. This should have been one of the most important rooms in the new Iraqi house.

In a serious federation, an upper federal chamber is not decoration. It gives regions and governorates a constitutional voice against the center. It allows territorial objections to become institutional arguments rather than crises. It forces the state to hear its components not merely as parties in the lower house, but as territorial partners in the federal order.

Iraq did not build this room.

The result was structural imbalance. Federalism existed in text, but one of its central instruments was absent. Kurdish objections to central overreach had no stable federal chamber in which to become constitutional correction. Instead, disputes moved into bargaining, budget fights, court cases, unilateral decisions, foreign mediation, threats, and crisis.

A missing institution is not empty space.

It is a wound in the architecture.

When the room where correction should occur is never built, conflict seeks other rooms. Some are legal. Some are fiscal. Some are military. Some are symbolic. Some are external.

The 2017 referendum partly emerged from this missing architecture. If Iraq had built functioning federal institutions, if Article 140 had been implemented, if revenue arrangements had been trusted, if constitutional disagreements had a credible forum, then the Kurdish appeal to independence might have taken another form.

But when the architecture of federal correction is blocked, the demand for exit becomes more rational.

The failure to build the Federation Council was therefore not merely institutional negligence. It was one of the ways Iraq’s forgiving architecture became unforgiving.

**Oil, Revenue, and the Material Meaning of Forgiveness**

Forgiveness without resources is rhetoric.

A region cannot exercise real autonomy if its salaries, budgets, oil exports, and public services depend on the discretionary goodwill of the center. Fiscal architecture is therefore constitutional architecture. It determines whether federalism is lived reality or decorative text.

The Iraqi Constitution declared that oil and gas belong to all the people of Iraq. It also required federal management with producing regions and governorates. But the word “with” was never stabilized into a trusted oil and gas law. Baghdad and Erbil developed rival interpretations. The Kurdistan Region built an oil policy around regional autonomy. Baghdad treated that policy as unconstitutional central defiance.

The unresolved oil question became a test of whether Iraq’s federalism could forgive complexity.

It failed.

Instead of creating a stable shared regime, Iraq allowed ambiguity to harden into leverage. Budget transfers were contested. Salary payments became uncertain. Oil exports became legal battlegrounds. Federal court decisions pushed in centralizing directions. Ordinary citizens became hostages of constitutional incompletion.

This matters because forgiving architecture must protect people from being punished for unresolved design flaws.

If Baghdad and Erbil disagree over oil law, a teacher in Duhok should not become collateral damage. If the Constitution is ambiguous, a civil servant in Sulaymaniyah should not lose salary because the architects failed to build a reliable revenue mechanism. If the federal bargain is incomplete, citizens should not pay the price of incompletion.

A forgiving architecture absorbs institutional error.

Iraq’s architecture transferred institutional error downward to ordinary people.

That is unforgiving.

**The Court as Architect of Closure**

Every constitutional order needs an interpreter. But in a federation, the interpreter must be trusted as guardian of the bargain, not as an instrument of one side.

There is a deeper problem here. Iraq did not properly constitute the Federal Supreme Court as the 2005 Constitution required. The Constitution envisaged a court established by law with the necessary parliamentary super-majority, but the institution that continued to operate was rooted in the temporary, centralized arrangements of the Bremer period. This means that the so-called Federal Supreme Court has often been asked to decide the most profound questions of Kurdistan’s rights, federalism, oil, revenue, elections, and constitutional authority while carrying an unresolved legitimacy problem of its own. A court whose own constitutional foundation remains incomplete is a fragile guardian of a federal bargain. This is especially grave because the same Court has later been used to adjudicate questions touching the very heart of the bargain negotiated in 2005: Kurdistan’s constitutional status, its fiscal life, its oil policy, and the limits of Baghdad’s authority.

The Federal Supreme Court of Iraq became increasingly central in Baghdad-Erbil disputes. It ruled on Article 140, the referendum, oil and gas, fiscal transfers, and Kurdistan Region electoral matters. Some rulings preserved constitutional obligations, as when the Court affirmed that Article 140 remained valid until implemented. Others were experienced by the Kurdistan Region as centralizing interventions.

The problem is not that courts should never rule against regions. A federation requires legal discipline. Erbil can be wrong. Baghdad can be wrong. Both require correction.

The problem is asymmetry.

If the Court can void the referendum but cannot make Article 140 happen; if it can invalidate Kurdish oil arrangements but cannot build the missing oil law; if it can affect salary mechanisms but cannot create durable fiscal federalism; if it can constrain Kurdistan but cannot compel Baghdad to complete the federal architecture, then the Court becomes part of the unforgiving structure.

The question becomes:

***Who corrects the corrector?***

Constitutional supremacy is legitimate when it binds the center as strictly as it binds the region. But if supremacy becomes the language through which Baghdad’s interpretation prevails while Baghdad’s obligations remain unenforced, then supremacy no longer feels like federal law. It feels like central recovery.

A forgiving architecture needs courts that repair the covenant.

Iraq’s constitutional practice too often produced courts that managed the consequences of a broken covenant while leaving the break itself unhealed.

**The 2017 Referendum: The Return of Voluntary Union**

The 2017 Kurdistan referendum is often described as a mistake, a gamble, a unilateral act, or a reckless move after the war against ISIS. There is truth in some of these criticisms. The referendum was politically risky. It faced opposition from Baghdad, Iran, Turkey, the United States, and the United Nations. It included disputed territories whose status remained unresolved. Kurdish politics itself was divided and marked by party rivalry, institutional weakness, and strategic miscalculation.

But none of this exhausts its meaning.

The referendum must also be read as a return of the Preamble.

The Constitution said Iraqis had freely and by choice united their future. It said adherence to the Constitution preserved Iraq’s free union of people, land, and sovereignty. The Kurds could reasonably ask: what happens when adherence fails? What happens when Article 140 is not implemented? What happens when the Federation Council is not built? What happens when federalism becomes central bargaining? What happens when budget and oil disputes make autonomy dependent on Baghdad’s will? What happens when the voluntary union no longer feels voluntary?

The 2017 referendum was the Kurdish answer.

It said: if Iraq is a free union, then the people of Kurdistan may ask whether they still freely choose it.

This was not how Baghdad opted to understand the matter. Baghdad treated the referendum as unconstitutional. The Federal Supreme Court declared it void. The federal government demanded control of airports, borders, and oil revenues. Iraqi forces and allied militias moved into Kirkuk and other disputed territories after the vote. The referendum did not open negotiations toward a new constitutional settlement. It triggered punishment and reversal.

This is the tragedy.

The Kurds appealed to the voluntary logic of the constitutional founding.

Baghdad answered with the compulsory logic of sovereignty.

The 2017 referendum therefore exposed the deepest contradiction of post-2005 Iraq: the Constitution spoke the language of free union, but the state practiced the language of indivisible command.

Was the Referendum Constitutionally Perfect? No. Was It Constitutionally Meaningless? Also No.

A serious argument must avoid romantic simplification.

The 2017 referendum was not legally straightforward. The Iraqi Constitution contains no clear secession mechanism. Article 1 defines Iraq as a single federal, independent, sovereign state. The Federal Supreme Court had authority to interpret the Constitution. Baghdad could argue that unilateral secession violated the constitutional order.

These points matter.

But the absence of a secession clause does not erase the Preamble’s language of free union. Nor does it erase Baghdad’s failure to implement the federal bargain. Nor does it erase the fact that the Constitution’s legitimacy, from the Kurdish perspective, depended on the fulfillment of promises that were not fulfilled.

The referendum should therefore be understood as constitutionally irregular but constitutionally intelligible.

It was irregular because the text did not provide a formal path to independence.

It was intelligible because the text did provide a moral and political premise of voluntary union, and that premise had been undermined by Baghdad’s non-implementation of the very mechanisms that made union free.

This distinction matters.

A state that refuses to create constitutional exits, refuses to implement constitutional corrections, and then condemns extra-constitutional protest as illegal has not answered the grievance. It has merely described the prison.

The Kurds may have miscalculated in 2017.

But Iraq had already misdesigned the conditions that made such a referendum appear necessary.

**Forgiveness, Exit, and the Backspace Key**

Every serious union needs a theory of correction.

Some unions survive because they are voluntary enough that exit is unnecessary. Others survive because exit is forbidden. The difference is moral and constitutional.

A forgiving architecture does not encourage constant separation. It reduces the need for separation by making correction credible. It says to wounded peoples: you do not need to leave in order to be safe, because the union itself can correct what endangers you.

That was the promise of federal Iraq.

Article 140 would correct disputed territories.

Federalism would correct centralization.

Language equality would correct Arab nationalist denial.

Regional powers would correct administrative domination.

Revenue fairness would correct fiscal dependence.

The Federation Council would correct majoritarian centralism.

The Preamble’s free union would correct the old doctrine of compulsory statehood.

When these mechanisms fail, exit becomes the Backspace Key of last resort.

This is how the 2017 referendum should be understood. It was not the first Backspace Key. It was the one invoked after the others failed.

First came federalism.

Then Article 140.

Then revenue bargaining.

Then oil disputes.

Then constitutional litigation.

Then years of postponement.

Then the referendum.

A state that disables internal correction should not be surprised when external correction appears.

The referendum was the return of the Backspace Key in its most dangerous form: not correction inside Iraq, but correction from Iraq.

**The Moral Boundary: Forgiveness Is Not Kurdish Domination**

The argument for Article 140 and voluntary restructuring must not become Kurdish romanticism.

The disputed territories are not empty moral space. They contain Kurds, Arabs, Turkmen, Assyrians, Yazidis, Shabaks, and others. Some Arabs were instruments or beneficiaries of Baathist Arabization. Others were not. Some families were settled by state policy and now know no other home. Turkmen have their own historical fears. Assyrians and Yazidis have suffered repeated vulnerability. Shabaks and other communities have often been treated as objects inside larger national projects.

A forgiving architecture must protect all of them.

Correcting Arabization must not mean humiliating Arabs.

Restoring Kurdish rights must not mean erasing Turkmen claims.

Resolving Kirkuk must not mean turning minorities into decorative witnesses.

A referendum must not reduce persons to ethnic numbers.

The person must remain larger than the map.

This is why Article 140 required careful implementation. Normalization should have been accompanied by property adjudication, compensation, minority protections, local power-sharing, security guarantees, language rights, and credible international monitoring. It should have corrected coercion without creating new coercion.

The failure to implement Article 140 did not protect minorities. It left them inside unresolved sovereignty.

The answer to complexity is not postponement.

The answer to complexity is better architecture.

**Why Baghdad’s Position Is Weaker Than It Appears**

Baghdad’s strongest argument is constitutional legality.

It can say: Iraq is one federal sovereign state. The referendum had no constitutional basis. Secession is not provided for. The Federal Supreme Court ruled against it. The federal government had the duty to preserve unity.

This argument has force.

But it is incomplete because it treats unity as if unity were independent of the constitutional conditions that made it legitimate.

The Constitution does not merely say Iraq is one state. It says Iraq is federal. It says Iraq is plural. It says Kurdish is official. It recognizes Kurdistan. It preserves regional powers. It mandates a Federation Council. It requires Article 140\. It speaks of free union. It binds unity to constitutional adherence.

Baghdad cannot select the clauses of sovereignty and ignore the clauses of forgiveness.

That is the central point.

If Baghdad invokes Article 1, it must also implement Article 140.

If it invokes constitutional supremacy, it must submit itself to constitutional obligation.

If it invokes unity, it must preserve the free character of union.

If it invokes federal authority, it must build federal institutions.

If it invokes the illegality of the referendum, it must explain the legality of indefinite non-implementation.

This is why the Kurdish argument is difficult to refute.

It does not depend on denying Iraq’s constitutional unity.

It insists that Iraq’s unity was constitutionally conditioned by federal correction.

Baghdad defended the roof while neglecting the foundations.

**The Washington Test**

This matters now because Washington often speaks of a strong, sovereign, unified Iraq. That language is understandable, but it is incomplete unless “strong” means constitutional, “sovereign” means free from militia and foreign capture, and “unified” means voluntarily federal rather than coercively centralized.

The United States should not treat the Kurdish constitutional question as a local inconvenience to be postponed in the name of Iraqi stability. That mistake has been made for too long. Stability built on suspended correction is not stability. It is deferred crisis.

The proper question for Washington is not whether it supports Iraq or Kurdistan. The proper question is whether it supports the Iraq that Iraq promised to become in 2005\. That Iraq was federal, plural, voluntary, and corrective. It recognized Kurdistan not as a concession but as a constitutional partner. It treated Article 140 not as a Kurdish demand but as an Iraqi obligation. It promised institutions through which conflict could be processed without coercion.

Any new U.S.–Iraq understanding should therefore include a constitutional test: concrete steps toward militia disarmament, implementation of Article 140 through a phased and internationally supported process, establishment of the Federation Council, a durable oil and revenue settlement, and a properly constituted Federal Supreme Court trusted as guardian of the federal bargain rather than as an instrument of central recovery.

If Washington wants an Iraq that can resist Iran, it should not strengthen the very centralizing mechanisms that Iran-backed actors have used to weaken federal Iraq. It should strengthen the constitutional architecture that allows Iraq’s peoples to belong without being swallowed.

Baghdad cannot ask Washington for help against militia capture while preserving the constitutional capture of Kurdistan’s federal rights. Sovereignty is not only the removal of foreign interference. Sovereignty is fidelity to the constitutional bargain that made Iraq legitimate after Baathism.

**The Essay’s Central Mechanism**

The mechanism of Iraq’s failure can now be stated precisely.

**First**, the Kurds succeeded in transforming the post-Saddam settlement into a federal forgiving architecture.

**Second**, that architecture depended on the idea that Iraq’s unity was free, chosen, federal, plural, and correctable.

**Third**, Article 140 served as the key correction mechanism for the most dangerous inherited territorial error.

**Fourth**, once the Shia-led Baghdad order stabilized its position, the center increasingly treated these correction mechanisms as constraints rather than foundations.

**Fifth**, the mechanisms were delayed or disabled: Article 140 unimplemented, Federation Council absent, oil law unresolved, revenue conflict recurring, federal interpretation centralizing.

**Sixth**, the Kurds appealed in 2017 to the deeper voluntary premise of the constitutional settlement.

**Seventh**, Baghdad answered with compulsory sovereignty, thereby revealing that the forgiving architecture had become unforgiving.

This is the structure.

It is not a claim of Kurdish innocence.

It is a claim of constitutional betrayal.

**The Strongest Formulation**

The strongest formulation is this:

Iraq’s 2005 Constitution offered the Kurds a forgiving architecture: a federal design through which the crimes and distortions of the old state could be corrected without destroying Iraq. The Kurds accepted the possibility of a shared future because the Constitution transformed unity into a free union, federalism into state architecture, and Article 140 into a procedure for correcting Arabization. But once Baghdad’s Shia-led ruling order stabilized, the center increasingly disabled the very mechanisms that made forgiveness possible. Article 140 was not implemented. The Federation Council was not built. Fiscal federalism became leverage. Oil became conflict. Courts became instruments of central interpretation. When the Kurds invoked the voluntary logic of the union in 2017, Baghdad treated voluntariness itself as unconstitutional. The result was the collapse of the Constitution’s deepest promise: Iraq would be one because its peoples chose correction together, not because the center could once again enforce the map.

This is the argument.

**The House That Would Not Forgive**

The 2005 Constitution tried to build a new Iraqi house.

For the Kurds, the old house had been lethal. Its rooms contained Arabization, Anfal, Halabja, village destruction, denial, and compulsory unity. The Kurds did not simply ask for a safer room inside the old house. They demanded a different architecture.

Federalism was that architecture.

Article 140 was the repair mechanism.

The Preamble’s free union was the moral foundation.

Regional powers were the walls.

Language equality was the recognition of plural ownership.

The Federation Council was supposed to be the room where regions could speak back to the center.

Revenue-sharing was supposed to make autonomy livable.

Constitutional supremacy was supposed to bind Baghdad as well as Erbil.

But the house was never completed. Some rooms were left unbuilt. Some doors were locked. Some repair mechanisms were disconnected. The center kept the master key. The victims were told to trust the architecture while the architecture ceased to forgive.

That is why the Kurdish question did not disappear after 2005.

It returned in every unimplemented article, every disputed salary, every oil conflict, every argument over Kirkuk, every court ruling experienced as centralization, every reminder that Article 140 still lived legally but not politically.

And in 2017, it returned as a referendum.

The referendum was not only a vote for independence. It was a verdict on the failure of forgiving architecture. It asked whether a union that no longer corrected could still call itself free.

Baghdad answered by declaring the question unconstitutional.

But that answer did not solve the problem. It confirmed it.

A free union cannot be preserved only by forbidding its members to ask whether it remains free.

A federal constitution cannot survive if the center disables the institutions that make federalism real.

A state cannot remember Anfal and Halabja while preserving the territorial consequences of the system that made such crimes possible.

A ballot is not a choice when the conditions of choice have been manufactured by coercion.

And forgiveness is not forgiveness when the injured party is asked to live forever inside the architecture of the injury.

Iraq’s deepest failure was not that it lacked a Backspace Key. It wrote one into the Constitution. It called it federalism. It called it Article 140\. It called it free union.

Then the state refused to let it work.

The Constitution remembered.

The architecture promised forgiveness

The center made forgiveness impossible.

And the old Iraq survived inside the new one.

**Selected Sources**

*Constitution of the Republic of Iraq*, 2005, especially the Preamble and Articles 1, 4, 65, 105, 106, 111, 112, 115, 117, 121, 126, 140, and 141.

*Transitional Administrative Law*, Article 58.

International Crisis Group, *Iraq and the Kurds: Resolving the Kirkuk Crisis*, 2007.

United Nations Assistance Mission for Iraq, reports on Iraq’s disputed internal boundaries, 2009.

Federal Supreme Court of Iraq decisions concerning Article 140, the 2017 Kurdistan referendum, oil and gas governance, fiscal transfers, and Kurdistan Region electoral matters.

Legal analyses of the 2022 Federal Supreme Court ruling on the Kurdistan Region’s oil and gas law, including summaries by Eversheds Sutherland, Norton Rose Fulbright, Al Tamimi & Co., and AMERELLER.

Reporting on the July 2026 Washington visit by Iraq’s new prime minister and the U.S.–Iraq agenda on sovereignty, militias, security, energy, and investment.

Public statements and reporting on U.S. policy toward Iranian-backed militias in Iraq and the demand that armed groups operate under state authority.