> ## 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. # Where Does the Error Go? - URL: https://otherwise.ghost.io/where-does-the-error-go-9c4/ - Published: 2026-08-26T07:21:30.000Z - Updated: 2026-08-26T07:21:30.000Z - Description: How organizations become peoples, guardians become problems, states become natural, and Kurdistan disappears from the verb. - Author: Khaled Salih - Tags: Newsletter, #Migrated-1792244264359, #Import 2026-10-17 14:38 # *Part II Where Does the Error Go?* How organizations become peoples, guardians become problems, states become natural, and Kurdistan disappears from the verb. ## Part II Note **Where Does the Error Go?** The most useful question I know in politics is not: ***Who is wrong?*** It is: ***Where does the error go?*** An organization commits an atrocity. Does responsibility remain attached to the organization? Or does it travel into an entire people? Hamas attacks Israelis. Does the error remain Hamas’s? Or become “the Palestinians”? A Jewish militia massacres Palestinian civilians. Does responsibility remain with the perpetrators? Or become “the Jews”? The PKK commits violence. Does the error remain with the PKK? Or become “the Kurds”? This movement—from organization to people—is one of the mechanisms through which mutual radicalization becomes possible. Simon Sebag Montefiore and the TRIGGERnometry hosts identify part of this process unusually well. But the interview also exposes a second problem. Once hatred and violence escalate, the temptation returns: perhaps what the region really needs is a wise guardian. An empire. A great power. A supranational manager. Someone above the conflict who will finally get the answer right. But the guardian is fallible too. Britain made errors. The UN made errors. The Shah made errors. Khomeini made errors. Every correction mechanism can itself become protected. So Part II moves from the Nakba to Iran, from collective blame to the Shah’s closing of peaceful political correction, from Khomeini’s extraordinary political intelligence to the anti-rational system that followed, and finally back to Kurdistan. Because once again Kurds appear in the history mainly when somebody else crushes them. That is not enough. The question is not merely: ***What happened to the Kurds?*** It is: **What political explanations did Kurdish actors themselves create—and what happened to the institutions through which those explanations might have become consequential?** Part II is about error routes, guardians, Backspace Keys, institutional closure—and the moment political analysis stops treating peoples as the containers of other people’s mistakes. --- For introduction and Part one see [here](https://khaledsalih.substack.com/p/the-winners-became-countries-the). --- # Mutual Radicalization and the Error That Travels Into Peoples One of the best moments in the interview comes not from Montefiore but from the hosts. While discussing Zionist and Arab militias, one host proposes a pattern of mutual radicalization. An extremist faction attacks civilians. The attacked population says: “This is what they are all like.” Its own extremists gain credibility. They attack. The other population generalizes in return. And the loop continues. Montefiore agrees. This is a genuinely good explanation. It has mechanism. It has reach. And it reveals something central about political error. The route is: **organization acts** **↓** **victim experiences violence** **↓** **organizational responsibility expands into collective identity** **↓** **the entire people becomes suspect** **↓** **hardliners gain credibility** **↓** **retaliation** **↓** **the other people generalizes** **↓** **Protection Loop.** This is why one of my rules is: **Locate responsibility at the lowest sufficient causal level.** Hamas commits an atrocity. The error must remain attached to Hamas and responsible actors. It must not travel automatically into “Palestinians.” A Jewish militia massacres civilians. Responsibility should reach that organization, command and action. It must not become “the Jews.” PKK commits violence. Responsibility does not automatically become “the Kurds.” A Turkish government represses Kurds. Responsibility does not become “Turks” as a biological or collective category. If we cannot keep the unit of responsibility precise, we create the explanatory architecture of collective punishment. ## And Then the Guardian Appears Immediately after this excellent mutual-radicalization insight, Montefiore says something fascinating. It almost makes one wish, he suggests, for a responsible imperial power or supranational organization capable of managing the conflict. This is a perfect philosophical mistake. Because the interview has spent considerable time explaining that the guardians were themselves fallible. Britain made contradictory promises. Britain and France imposed mandate systems. Britain created political structures with unintended consequences. Britain helped elevate the Mufti. The UN designed arrangements Montefiore himself describes as deeply flawed. Britain then withdrew. And the lesson becomes: perhaps we needed a better guardian. No. The guardian is also fallible. This is one of Karl Popper’s deepest political lessons. The ancient question is: Who should rule? The wise? The king? The imperial power? The experts? The majority? The revolutionary? The nation? But every answer inherits the same problem. The ruler can be wrong. The correct question becomes: ***How do we build institutions in which bad decisions can be corrected without catastrophe?*** So my alternative to Montefiore’s benevolent guardian is not “no authority.” It is: > **no authority whose success depends upon being epistemologically sovereign.** We need: - enforceable agreements; - peaceful dispute mechanisms; - safe political loss; - institutional rights; - constitutional correction; - multiple channels of representation; - exit; - criticism; - revision. Not the perfect guardian. A political architecture that does not need one. ## The Nakba and the Danger of One Explanation Montefiore’s treatment of Palestinian displacement in 1947–48 is interesting precisely because he refuses a single mechanism. He emphasizes early flight caused by fear and war, the collapse of Arab political authority, expectations of broader invasion and instances in which Jewish leaders encouraged Arabs to stay. He then explicitly acknowledges massacres such as Deir Yassin and later expulsions, including Lydda and Ramle, which he calls an atrocity and cleansing. This is methodologically valuable. Why? Because “Why did Palestinians become refugees?” is not answered by one sentence. The explanation varies by: - place; - time; - commander; - military conditions; - fear; - massacre; - expulsion; - collapse; - anticipation; - strategy. That is what hard-to-vary explanation looks like. If we remove one mechanism, we misdescribe some locations. If we insist on only voluntary flight, we fail. If we insist on only centrally planned expulsion, we may fail elsewhere. History becomes stronger when it can accommodate causal heterogeneity without dissolving responsibility. But Montefiore occasionally overstates even here. He says, during one phase, that people “were not being driven out.” “No one” was. Absolute statements demand extraordinary evidence. The better formulation is narrower. **A substantial portion of early displacement occurred through flight under conditions of violence and fear rather than direct expulsion, while other locations and later phases involved coercive removal.** That may sound less dramatic. It is epistemically stronger. ## Explanation Is Not Excuse Montefiore’s treatment of Lydda and Ramle offers another useful lesson. He explains a strategic rationale. Israel’s territorial position was militarily precarious. Commanders wanted defensible continuity. And yet Montefiore does not conclude: therefore the expulsions were acceptable. He calls them an atrocity. This is exactly right. There is a persistent confusion in moral and political argument between explanation and exculpation. To explain why an actor committed an atrocity is not to justify the atrocity. In fact, explanation can improve accountability. If we know the actual mechanism, we know what must be prevented next time. An excuse does the opposite. It spreads responsibility until nobody can be corrected. “War is terrible.” “Everyone did bad things.” “It was complicated.” Those formulations may be true. They can also become ways of avoiding the causal mechanism. ## “Both Sides” Is Not Yet an Explanation Montefiore says at another point that there were massacres on both sides and that trying to determine who massacred whom first is somewhat pointless. Here I disagree. Chronology is not moral absolution. The first atrocity does not morally license the second. But causal sequence can matter enormously. Did event A generate fear that empowered faction B? Did massacre C produce flight from town D? Did retaliation E destroy negotiations F? Then chronology is explanatory. “Both sides” can be true while still being intellectually lazy. Symmetry should discipline moral principles. It should not erase differences in mechanism. ## The Shah, Khomeini and the Backspace Key The Iran section may be the strongest part of the entire interview. Montefiore’s central insight is devastatingly simple. The Shah modernized. Educated. Industrialized. Expanded women’s rights. Built state power. But he also concentrated decision-making, weakened opposition and progressively removed lawful routes through which dissatisfaction could alter the political order. Montefiore puts the problem beautifully: there was “nowhere to go for opinion.” That sentence contains an entire political theory. If disagreement has nowhere to go— it does not disappear. If criticism cannot move upward— it accumulates. If reform cannot occur through institutions— the price of correction rises. Eventually the system may encounter catastrophic correction. The sequence is: **fallible ruler** **↓** **opposition suppressed** **↓** **criticism blocked** **↓** **errors accumulate** **↓** **peaceful Backspace disappears** **↓** **heterogeneous opposition searches for a focal point** **↓** **revolution** **↓** **new ruler captures correction process** **↓** **new closure.** This is what I mean by Protected Error. The Shah’s deepest failure was not that he was wrong. All rulers are. His failure was that the architecture increasingly made his errors difficult to correct without removing the whole order. The revolution was therefore not evidence that problems are soluble through revolution. It showed what can happen when ordinary correction becomes unavailable. ## A One-Man State Is a Knowledge Problem Montefiore describes the Shah’s Iran as a “one-man state.” This matters beyond autocracy. If too much information, judgment and authority converge on one person, the institution becomes epistemically fragile. The person becomes tired. Ill. Isolated. Filtered. Flattered. Overburdened. Then individual cognitive limitations become state limitations. The solution is not: find a more intelligent ruler. The solution is: **move knowledge and correction into institutions.** This is the political transition I describe elsewhere as: ***Founder → Institution*** rather than merely: ***Founder → Heir.*** The finest ruler cannot know enough. No one can. The argument against excessive concentration of authority is not that rulers are necessarily evil. It is that knowledge is dispersed and fallibility universal. ## Khomeini: Intelligence Is Not Rationality Montefiore calls Khomeini a brilliant politician. I think this is an important observation. It should make nobody uncomfortable. Intelligence and rationality are not the same thing. Khomeini appears in Montefiore’s account as an extraordinary strategist. He understood ambiguity. Coalitions. Symbolic authority. Timing. Institutional sequencing. Political communication. He allowed liberal democrats, Marxists, Islamists and others to project different futures onto the same revolutionary center. Montefiore describes the extraordinary heterogeneity of the opposition that coalesced around him and the ambiguity with which he handled them. That can be highly intelligent. And anti-rational. Intelligence asks: How can I achieve the goal? Rationality must also ask: Is the goal wrong? Is the frame wrong? Can critics alter it? An institution can be brilliantly designed to prevent criticism. That does not make it rational. It makes it intelligently anti-rational. ## Strategic Ambiguity Is Better Than “Confidence Trick” Montefiore at times describes Khomeini’s achievement as a confidence trick. Again, rhetoric is replacing mechanism. The stronger explanation is: **Strategic ambiguity allowed heterogeneous groups to interpret Khomeini through their own desired future while postponing decisive disagreement about the post-Shah order.** That is much more interesting. Liberals imagined one future. Islamists another. Marxists another. Religious conservatives another. The coalition could remain broad while the governing institutional question remained underspecified. Then, after the old order collapsed, the best organized faction could transform ambiguity into authority. This mechanism travels. It applies to revolutions. Coalitions. Liberation movements. Even political parties. A shared enemy can conceal incompatible positive programs. Victory removes the shared enemy. Then the unresolved problem returns at a higher level. ## Crisis Can Become a Constructor of Closure Then Saddam Hussein invades. Montefiore shows how war enabled the Islamic Republic to consolidate. The Revolutionary Guards expand. Opponents are suppressed. Alternative revolutionary factions are destroyed. Kurds seeking autonomy are crushed. Emergency becomes institutional transformation. This is another excellent mechanism. External threat can become a **Constructor of Closure**. The sequence is: ***real danger*** ***↓*** ***emergency*** ***↓*** ***centralization*** ***↓*** ***criticism recoded as disloyalty*** ***↓*** ***alternative institutions suppressed*** ***↓*** ***temporary necessity becomes durable power.*** This is not uniquely Iranian. Every society should fear it. Security emergencies are real. So is the temptation to make their institutional consequences permanent. The correct question is: ***What Backspace Key exists for the emergency itself?*** ## And Rojhelat Disappears Again But then something revealing happens. Montefiore says the new regime crushed “the Kurds who had hoped to achieve autonomy.” And the narrative moves on. Once again, Kurdistan appears as an object of somebody else’s political consolidation. What did Kurdish parties propose? What kind of autonomy? What political institutions emerged? Why did negotiations fail? What rival Kurdish explanations existed? How did the Iranian revolution look from Mahabad, Sanandaj and the rest of Rojhelat? What did Kurds believe they were correcting? What constitutional possibility did they imagine? What was destroyed? The omission is now patterned. And a pattern deserves explanation. ## Kurdistan: The Missing Conjecture I do not think Montefiore is deliberately excluding Kurdish agency. That would be too easy an accusation. The more interesting explanation is structural. History gravitates toward successful states. They generate: - archives; - ministries; - wars; - treaties; - diplomatic correspondence; - official narratives; - recognized leaders. The stateless are harder to narrate. Their political projects fragment across archives produced by the states that defeated them. The result is a subtle asymmetry. A Turkish political project becomes: > **the creation of modern Turkey.** A Jewish political project becomes: > **the history of Zionism and Israel.** An Arab dynastic project becomes: **the Hashemite story.** A Kurdish political project becomes: > **the Kurds were promised something but did not get it.** The winners become countries. The losers become minorities. Then historians inherit those categories. That is how political victory becomes epistemological architecture. ## Restore Kurdistan to the Verb I want to reverse this. Do not begin with: Kurds were divided among Turkey, Iran, Iraq and Syria. That sentence is geographically true. It is also epistemically dangerous. Because it turns the successor states into primary reality and Kurdistan into distributed residue. Instead ask: **What political problem situation did Kurdish actors face as Ottoman and Qajar/Persian imperial orders transformed, and what competing political conjectures did they create?** Now Kurdistan returns to the verb. Kurds: Conjectured. Negotiated. Rebelled. Compromised. Disagreed. Built parties. Formed republics. Created autonomous institutions. Made terrible mistakes. Made intelligent adaptations. Changed their minds. Built administrations. Fought one another. Learned. Failed. Tried again. That does not guarantee Kurdistan a state. It restores Kurds as political agents. ## Four Arenas, Not One Kurdish Failure The Kurdish question also cannot be reduced to one institutional experience. Başûr. Bakûr. Rojhelat. Rojava. Each became a different experiment in political agency. In Başûr, Kurds eventually acquired substantial recognized self-government. The deepest question there is increasingly: What happens after agency becomes institutionalized? Can Kurdish rulers themselves be corrected? Can military command become constitutional? Can political families transfer authority into institutions? Can opposition lose safely? Can governments form? Can constitutional rules become real authority? Bakûr presents another problem. Political agency exists through: - elections; - municipalities; - parties; - civic institutions; - movements; within a Turkish constitutional system retaining decisive authority over the foundational political menu. Rojhelat presents political agency without comparable territorial institutionalization, under far tighter constraints. Rojava presents yet another architecture: extensive de facto Kurdish-led institutional agency generated under war, external alliance and an unstable constitutional environment. These are not four steps on a ladder. They are four problem situations. If a “true history” of the modern Middle East does not give explanatory standing to those Kurdish problem situations, something important remains invisible. ## Turkey Is the Comparison That Makes Kurdistan Unavoidable Ironically, Montefiore’s own explanation of Turkey supplies one of the strongest arguments for restoring Kurdistan to the analysis. Atatürk refused an international settlement because he regarded it as unacceptable to Turkish political agency. He mobilized. Won. Changed the map. Later generations call this: national liberation. Now reverse the names. Suppose Kurds say: the existing post-Ottoman settlement divides a people among states and places foundational political authority outside Kurdish correction. We may disagree with their proposed solution. But we cannot coherently reply: the map already exists. Because Montefiore has just shown us that the map was created through rejection of the map that existed before it. This is not an argument for automatic Kurdish independence. It is an argument against **Final Immunity for successful borders**. Existing political arrangements do not become good explanations merely because armies once made them durable. ## The Map Is an Explanation A political map looks physical. Colored shapes. Lines. Names. But the map contains invisible explanations. This line means: these people belong under this sovereignty. This capital may decide. This community is a minority. That one is a majority. This army is legitimate. That armed organization is rebellion. This border is international. That territorial claim is separatism. The map therefore does not merely record political reality. It embeds theories of political agency. That is why maps can be wrong without geography being wrong. The mountain remains where it is. The political explanation of who may govern whom can still fail. ## No Political Arrangement Is a Law of Nature This requires careful wording. States are real. Their armies can kill you. Their courts can imprison you. Their passports determine movement. Their borders have consequences. But their **political architecture** is not a law of nature. It is a human arrangement. Iraq is a conjecture. Federal Iraq is a conjecture. Turkey is a conjecture. Centralized Turkey is a conjecture. Syria is a conjecture. Decentralized Syria is a conjecture. Israel is a conjecture. A two-state solution is a conjecture. A Palestinian state is a conjecture. Kurdistan is a conjecture. The test is not: Which one has existed longest? The test is: **What problems does the arrangement solve?** **What problems does it create?** **Who participates in correcting it?** **What happens when it is wrong?** That is a political philosophy more useful than historical entitlement alone. ## Survival Is Not Truth One of the greatest temptations in history is to read survival backward as validation. An institution survives for 100 years. Therefore it was wise. A doctrine survives 1,000 years. Therefore it must contain truth. A state survives war. Therefore it was legitimate. A movement disappears. Therefore its idea was unrealistic. No. An idea can survive because it explains reality well. It can also survive because: - dissent was punished; - alternatives were destroyed; - children were educated into it; - exit was expensive; - military force protected it; - institutions reproduced it; - language made alternatives unimaginable. Therefore always ask: ***What selection environment produced the survival?*** This is especially important for political borders. The longevity of a state tells us that the state replicated successfully. It does not alone tell us that no better political arrangement can be created. ## The Same Rule Protects Us From Kurdish Romanticism And this principle must return home. If the existing states are conjectures, Kurdistan is a conjecture too. If Turkey receives no exemption because it exists, Kurdistan receives no exemption because it was denied. Historical injustice does not guarantee future institutional wisdom. A Kurdish state can be: corrupt. authoritarian. centralized. oppressive toward minorities. captured by parties. dominated by families. militarized. Badly governed. Of course. Kurds are human beings. That is not an argument against Kurdish political agency. It is an argument for a Kurdish political architecture capable of correcting Kurdish errors. The principle must survive the name change. Otherwise it was never a principle. ## Montefiore’s Greatest Strength: He Restores Plurality For all these criticisms, I want to emphasize what makes the interview worth engaging seriously. Montefiore resists many lazy collective categories. Palestinian politics is not one thing. Zionism is not one thing. Islam is not one thing. The Ottoman Empire is not one thing across 400 years. Iranian opposition to the Shah is not one thing. Political actors contain factions. Doctrines change. People hold multiple identities. This is extremely important. He even makes a larger cultural point at the end of the interview. Asked what the discussion has missed, Montefiore does not name another war, ruler or ideology. He talks about poetry. Arab poetry. Iranian poetry. Hebrew poetry. He says he wants his history not to become simply one bloody event after another but to capture how people actually live. This is more important than it sounds. Because political analysis itself can become anti-human. People become: Sunnis. Shia. Jews. Kurds. Arabs. Settlers. Refugees. Islamists. Nationalists. Militias. Minorities. Majorities. Then history is one armed category striking another. Poetry interrupts the abstraction. It reminds us that civilization is larger than its political errors. That is the human version of: **Protect the person. Criticize the explanation.** An Iranian is larger than the Islamic Republic. A Palestinian is larger than Hamas. A Jew is larger than the Israeli government. A Kurd is larger than Kurdish nationalism. A Turk is larger than Turkish state doctrine. That may be one of the most important corrections any history of this region can make. ## Objectivity Is the Wrong Ambition At the beginning, the hosts praise Montefiore for objectivity. I would replace that concept. Not because facts do not matter. Quite the opposite. But “objectivity” easily becomes: I escaped ideology. Nobody entirely escapes assumptions. The more useful ambition is: > **corrigibility.** A corrigible historian says: Here is the problem. Here is my explanation. Here is my evidence. Here are the rivals. Here is the part I am less certain about. Here is the counterfactual I cannot prove. Here is what would make me change the account. That is stronger than claiming to stand nowhere. I do not want history without a perspective. I want perspective without immunity. ## A Better Interviewing Standard Too The TRIGGERnometry hosts are curious, intelligent and often productive interlocutors. Their mutual-radicalization conjecture is excellent. But they repeatedly defer to expertise. At one point the host effectively says: you are the expert. Socially harmless. Epistemically dangerous. Expertise matters enormously. But expertise cannot answer: Is your explanation correct? The expert should receive better questions, not fewer. What rival explanation worries you most? What evidence would change your mind? Which claim in your book are you least certain about? Where did you have to revise your own interpretation? Which counterfactual are you tempted to overstate? What part of your political identity makes one historical interpretation especially attractive? What is the strongest Kurdish criticism of your account? That is how an interview becomes a knowledge-creating institution rather than simply excellent conversation.