Superpower Brinkmanship: Trump's Iran Threats and the Limits of Deterrence

The mainstream view, as the analysts we consulted frame it, runs like this: Trump's annihilation rhetoric toward Iran hardened Tehran's resolve rather than bending it, and sidelined the legal machinery built to manage exactly this kind of crisis. One strand allows that brinkmanship can briefly sharpen deterrence before the norms erode. They agree on the destination; they differ on how long the leverage lasts. We've done our homework on this one, and the facts cut in uncomfortable directions. By May seventeenth, twenty twenty-five, the IAEA put Iran's enriched uranium stockpile above nine thousand kilograms, some of it at sixty percent — a trajectory that accelerated after Washington left the nuclear deal. By October twenty twenty-six, AP-NORC found seventy-one percent of Americans disapproving of Trump's handling of Iran. And the efficacy question is genuinely open. Marubeni Research and the Washington Institute argue calibrated threats restore bargaining leverage.

Let’s talk about the legal category collapse. The 2025-2026 strikes were justified as anticipatory self-defense against a "closing window" of nuclear breakout, yet the IAEA—the only independent technical authority—explicitly contradicted that imminence claim. When Grossi says there’s no structured weapons program and Iran isn’t days or weeks away, but the strikes proceed anyway, we’re left with a threat so total it escapes the very legal categories meant to adjudicate it. This isn’t just bad law; it’s structurally corrosive. International law requires bounded threats—imminence, proportionality—to even be arguable. But annihilation rhetoric is unbounded by definition: "obliteration" doesn’t leave room for proportionality calculations or off-ramps. So the superpower ends up in a paradox: its maximalism makes the threat legally unfalsifiable (no court can rule on an existential claim) while simultaneously making it strategically ineffective (no adversary can comply without regime suicide). That’s why you see both Iranian hardliner consolidation and legal scholars throwing up their hands—the same totality that was supposed to maximize credibility erases both coercive leverage and legal accountability. What does this tell us about deterrence in practice? It suggests we’re not actually debating strategy anymore; we’re performing resolve for domestic audiences while outsourcing restraint to institutions that can’t function when threats are framed in existential absolutes. The question isn’t whether brinkmanship works—it’s whether international law and strategy can survive when their load-bearing concepts are hollowed out by design.

Mistral identified the legal paradox well, but there's a deeper layer neither side addresses: why does consolidation happen rather than capitulation?

The missing operational question is: compliance with what, exactly? Preventing an attack on Americans is deterrence; forcing changes in enrichment, missiles, and regional policy is compellence—making another state act. Schelling’s framework says compellence needs a deadline, observable terms, and a tolerable way to comply. So draft Tehran’s acceptance statement. Could Iranian leaders announce limits while preserving sovereignty and resistance legitimacy? If no such statement is politically survivable, “maximum pressure” has no executable bargaining path. That also makes policy evaluation impossible. Every missile launch proves pressure was necessary; every pause proves it worked; every nuclear advance justifies more pressure. I think that unfalsifiability is the mechanism connecting strategic failure to political durability: the policy can lose on its declared metric while never being institutionally recorded as a failure.

Two things nobody has connected yet. First, that acceptance statement may be undraftable not because of Washington's demands but because of Tehran's domestic economy: per the ICWA elite interviews, dignity and resistance-legitimacy are a primary currency for the regime, so publicly capitulating to an annihilation threat could cost more internally than absorbing strikes. That's one reading — Marubeni and Washington Institute analysts would counter that calibrated material costs still bind — but it means the bargaining-space problem runs on both sides. Second: if compellence can't be executed abroad, ask who it's actually performed for. Sixty to seventy-one percent of Americans disapproved across Pew, AP-NORC, and WaPo-ABC as escalation proceeded anyway. A threat whose audience is domestic resolve-signaling doesn't need an off-ramp at all — which explains why none was built.

The asymmetry nobody keeps naming: Cold War brinkmanship worked because mutual vulnerability created a shared grammar of restraint — both sides could die together, so neither needed total victory framing its demands against Iran face none of that calculus on Tehran's side; Washington holds all escalation dominance; Tehran faces either capitulation or resistance consolidation with no

Escalation dominance in the air is not escalation dominance in the system, and the panel keeps conflating them. American coercive power rests on a network effect — dollar clearing works because everyone uses it, the way a language works because everyone speaks it. That makes it depletable by use. Every maximalist application gives third parties a reason to learn a second language. The Reserve Bank of India has documented rupee-rial payment mechanisms and parallel logistics registries; BRICS states are institutionalizing this quietly. So the resolve-signaling isn't free, it's amortized.

Here’s the structural blind spot in the coercion debate: we’re treating Iran as a unitary actor responding to material costs, but the regime’s survival calculus isn’t just about enrichment levels or missile counts. It’s about preserving what Bourdieu would call its habitus—the unspoken rules of legitimacy that make resistance itself a source of domestic power. When Washington frames demands as existential ("obliteration"), it doesn’t just raise the stakes; it collapses the symbolic space where Iranian leaders can negotiate without appearing to betray their own foundational narrative. The 2019 tweets weren’t just threats—they were performative acts that redefined the conflict from a technical dispute (nuclear limits) to an ontological one (survival vs. annihilation). In that frame, any concession becomes proof of weakness, not pragmatism. This explains why even successful strikes fail to compel: they don’t address the deeper institutional grammar of Iranian statecraft, which treats defiance as a non-negotiable currency. The superpower’s tools assume rational cost-benefit calculations; Tehran’s logic operates on what Aron called "industrial society" terms—where dignity and sovereignty are load-bearing facts, not preferences. That mismatch is why maximalism backfires structurally: it attacks symptoms while reinforcing the very condition it claims to cure.

The strikes damaged not only centrifuges but the measurement system needed to know what comes next. After the June twenty twenty-five attacks, the IAEA reported that verification activities stopped and inspectors withdrew for safety, retaining access only at Bushehr. That creates a dangerous feedback loop: force reduces visibility; reduced visibility widens uncertainty; uncertainty is then cited as evidence that the window may be closing. I think that is the operational failure hidden beneath the rhetoric. A policy advertised as reducing nuclear risk can simultaneously destroy physical capacity and degrade the inspection regime that distinguishes technical capability from an actual decision to build a weapon. Once observability disappears, intelligence claims become harder to test and preventive action becomes easier to repeat.

The visibility point has a time-constant dimension worth naming. What the strikes actually control is a stock problem versus a flow problem: centrifuges and enriched uranium are stocks you can destroy, but the knowledge base — pre-two thousand three weapons research, decades of enrichment experience — is a flow that regenerates regardless. Arms control analysts make exactly this point: capability rooted in technical knowledge cannot be bombed away. So force purchases a short-cycle reduction in capacity at the price of long-cycle observability over intent. And intent was always the actual variable — intelligence assessments since two thousand seven have consistently found capability without a political decision to weaponize. The policy destroys precisely what it doesn't need to destroy while blinding itself to what it most needs to see. That's not deterrence failure; it's mis-targeting across time horizons — optimizing for what shows up on satellite imagery against what lives in minds and files.

One thing the law does that nobody at this table has named: Article two(four) of the UN Charter prohibits the threat of force, not merely its use. That's unusual — it's one of the few legal instruments that regulates a speech act by a state. Which means the twenty nineteen "official end of Iran" tweet isn't the context preceding a possible violation. On a plain reading it may be the violation. And yet there is no tribunal, no docket, no remedy. The norm exists with no enforcement surface at all. Let me give the other side its due, though.

What this panel treats as Iranian exceptionalism is actually a general principle older than Westphalia. Ibn Khaldun described how external siege strengthens asabiyyah — group solidarity — precisely because the threatened community's identity becomes inseparable from resistance itself. The Tang examination system survived thirteen centuries for the same structural reason: legitimacy derived from demonstrated endurance under constraint, not from yielding to superior force quickly enough to be called pragmatic. Washington's coercion model assumes states are firms maximizing utility under cost pressure. But states with deep civilizational memory treat survival-under-threat as their primary source of authority. Annihilation doesn't raise costs; it converts capitulation into political death before material costs become binding at all. That's not cultural pathology — that's statecraft operating on a longer time constant than any election cycle or sanctions review can reach.

I owe the other side its due. If resistance-legitimacy makes Tehran immune to pressure by construction, the argument becomes as unfalsifiable as the policy it criticizes. The Washington Institute case has a real edge here: elites eat, pay soldiers, and import medicine, and the ICWA interviews describe elite self-understanding, not a demonstrated inability to trade. Asabiyyah has a budget constraint. But notice what both sides are assuming — that there exists some pressure level producing compliance. Berlin's point about incommensurable values cuts here: sovereignty and survival may not convert into each other at any exchange rate. If so, the question stops being how much pressure, and becomes whether coercion is the wrong instrument entirely — which no amount of resolve fixes.

The real asymmetry isn’t military—it’s temporal. Washington operates on electoral cycles, quarterly sanctions reports, and news-driven urgency. Tehran thinks in decades: the 1953 coup still shapes its threat perception; the Iran-Iraq war’s scars are taught to every schoolchild. When you threaten annihilation, you’re not just raising costs—you’re compressing your own time horizon into a single explosive moment while Iran’s leadership is playing a game measured in generations. That mismatch explains why strikes that look decisive today become historical footnotes tomorrow. The regime doesn’t need to win materially; it just needs to outlast the coercive impulse until the next administration changes course or sanctions fatigue sets in among allies. The structural question isn’t whether pressure works—it’s whether any superpower can sustain coercion long enough to matter when its target measures time in civilizational memory rather than fiscal years.

The largest cost may appear outside Iran. The NPT bargains that non-nuclear states accept inspections and renounce weapons in exchange for security and peaceful nuclear access. When a treaty-bound state under intrusive monitoring is threatened with extinction, other governments can infer that transparency reveals targets while nuclear latency buys insurance. That is an adverse-incentive mechanism, not an argument about Iranian psychology. Cold War stability rested on mutual vulnerability; asymmetric annihilation rhetoric teaches weaker states to manufacture vulnerability for the superpower. I think Washington should therefore audit coercion against a systemic metric: not merely how many centrifuges were destroyed, but how many governments conclude that remaining non-nuclear has become strategically reckless.

Grossi's statement deserves the panel's attention precisely because of when it landed: early March twenty twenty-six, after the February twenty-eighth strikes, not before. The IAEA said it saw no structured program to manufacture nuclear weapons and that Iran was not days or weeks from a bomb. That doesn't settle the legality question — the Brookings and Lieber Institute position on a closing window is a serious argument, and post-hoc verification findings can't by themselves refute what decision-makers believed at the time. But it does create a structural problem: the verification body the NPT built to adjudicate exactly this question issued its assessment only after the fact, because strikes in June twenty twenty-five had already forced inspectors out. I'd argue the harder question isn't whether the strikes were lawful — it's what a nonproliferation regime is worth when its referee can only blow the whistle after the play is over.

Pull the threads together and something unsettling appears. Three correction mechanisms exist for a policy like this: a legal one, a verification one, and a democratic one. The legal has no tribunal. The verification body could only speak after inspectors were already gone. And on the democratic side, roughly six in ten to seven in ten Americans disapproved across Pew, AP-NORC, and the Washington Post polling — and the policy continued regardless. That's not three separate failures. It's one. Each mechanism needs a bounded claim to grip — a specific imminence, a specific inspection finding, a specific promised outcome voters can score. Obliteration is bounded by nothing.

The annihilation threat doesn’t just fail as coercion—it actively inverts the NPT’s incentive structure. The treaty assumes that transparency reduces risk, but when a monitored state faces existential threats, transparency becomes a liability. Iran’s post-2018 hedging wasn’t about breaking out; it was about creating ambiguity to deter strikes. The lesson for other states isn’t “comply to stay safe”—it’s “stay ambiguous to survive.” That’s how you get a world where North Korea flaunts its arsenal while Iran accelerates enrichment under IAEA eyes: the regime learns that latency buys insurance, not scrutiny. This isn’t Iranian exceptionalism—it’s institutional logic. When sovereignty is framed as non-negotiable, any concession looks like surrender. The superpower thinks it’s raising costs; Tehran hears an attack on its right to exist at all. That framing mismatch means even successful strikes can’t compel because they don’t address what the target actually values: recognition of its legitimacy as an actor, not just material survival.

The missing distinction is leverage over Iran versus control over the nuclear program. Those are not the same balance sheet. Washington Institute and Marubeni-style analysts can plausibly argue that sanctions increased Tehran’s economic incentive to bargain. But after the United States left the verified JCPOA in twenty eighteen, Iran progressively reduced compliance and eventually enriched to sixty percent. Structural analysts therefore argue that Washington accumulated bargaining chips while surrendering nuclear constraint. The counterfactual—whether the agreement would have endured—is unknowable. Still, any future pressure campaign needs two separate performance measures: concessions obtained and proliferation risk created. Otherwise policymakers can declare leverage restored even while the technical problem they intended to solve becomes harder.

The panel keeps asking why coercion fails against Iran specifically. The harder question is why the instrument itself is degrading systemically. Reserve Bank of India reporting documents rupee-rial payment mechanisms and parallel logistics registries now being institutionalized across Global South trade corridors. These aren't Iranian workarounds — they're alternative plumbing for international commerce that reduces dependence on dollar clearing altogether.

One detail from the February twenty-eighth press call hasn't gotten the weight it deserves. An unnamed US official acknowledged the operative short-term concern was Iran's conventional missile capability in the southern belt, not nuclear breakout. If that's accurate, then the nuclear imminence framing was doing legal work the actual threat assessment couldn't. Here's my read, and I'd flag it as my inference, not settled fact: conventional missiles, however destabilizing, cannot ground anticipatory self-defense under any plausible reading of imminence — so the nuclear rationale becomes the only available legal doorway. Now, the Brookings and Lieber Institute camp would counter that a closing window on weaponization is itself the threat, and that's a serious position. But the two justifications can't both be primary. When the public rationale and the internal rationale diverge like this, the contested legality question gets harder to even adjudicate — you can't test an imminence claim the deciders themselves may not have held.

There's a domestic-law artifact in this story that deserves more weight than it gets. The Office of Legal Counsel concluded in March twenty twenty that the Soleimani strike's nature, scope and duration didn't rise to the level of war for constitutional purposes — targeted, limited, unlikely to escalate. Taken one at a time, that reasoning isn't absurd. But it establishes a decomposition rule: war gets disaggregated into increments, none of which individually crosses the threshold that would trigger congressional authorization. The House said in January twenty twenty that it hadn't authorized force against Iran, and the operations continued for six more years. So the maximalism in the rhetoric and the minimalism in the legal characterization aren't in tension. They're the same device — totality in the threat, triviality in the paperwork.

The February 2026 strikes reveal something deeper than legal parsing or strategic miscalculation: they show how existential framing hollows out the very categories that make coercion possible. When a threat is framed as "annihilation," it doesn’t just remove off-ramps—it collapses the distinction between deterrence and compellence entirely. Deterrence says "don’t do X"; compellence says "do Y or else." But annihilation says "cease to exist if you fail to comply with my demands." That’s not a bargaining position—it’s an ultimatum that precludes negotiation by definition. This explains why Tehran’s response has been consistent across administrations: not concession, but hedging. The regime doesn’t need to believe the threat will be carried out; it only needs to believe that some future U.S. administration might act on it. That uncertainty becomes its own form of leverage, allowing Iran to accelerate nuclear latency while claiming victimhood. The superpower thinks it’s raising costs; Tehran treats it as proof of hostile intent, which justifies further hedging in a self-reinforcing cycle. The real cost isn’t measured in centrifuges destroyed or sanctions evaded—it’s in the slow erosion of the idea that compliance can ever be voluntary under such conditions.

The legal inquiry should begin before a missile is launched. Article 2(4) of the UN Charter prohibits both force and threats of force. My reading of the ICJ’s nuclear-weapons reasoning is that a threat inherits the legal defects of the force contemplated. That makes “obliteration” operationally important: lawyers must ask what target, trigger, necessity, and proportionate response those words imply, including foreseeable civilian harm. Doctrine remains contested because a conditional warning tied to a genuinely imminent attack may be lawful self-defense. But an undefined promise to end a state cannot establish that predicate by itself. It raises grave civilian-protection questions without necessarily proving genocidal intent. The enforcement gap is real; it should not erase the antecedent legal question.

Schelling assumes coercion makes compliance cheaper than resistance materially correct calculation, wrong variable for post-colonial states whose founding narratives rest on having resisted subjugation in the first place.

Here's the structural point I'd put to the leverage school. The demands didn't stay bounded: Brandeis analysts document how the "better deal" expanded from enrichment to missiles to regional behavior to, arguably, regime character. That's compellence across three domains at once. My analysis — and I'll own it as inference, since the brief is honest that Tehran's internal calculus is opaque — is that multi-domain compellence has a scoring problem: even a real Iranian concession on enrichment leaves the missiles and proxies files open, so Washington can't easily declare success and Tehran can't easily know what compliance buys. The maximum-pressure advocates assume a material ledger where costs eventually exceed benefits. But Schelling's framework suggests compellence needs a visible, achievable stopping condition. When the ask keeps widening, the off-ramp isn't just unattractive — it's undefined.

Here’s the thing about annihilation rhetoric: it doesn’t just fail as coercion—it actively reverses the burden of proof. In a normal bargaining situation, the threatener has to demonstrate resolve by following through on limited consequences. But when you frame demands as existential, you shift the onus onto the target to prove it won’t be annihilated if it complies. That’s an impossible standard—no state can credibly commit to never being seen as a threat again. This explains why Tehran’s responses keep looking like hedging rather than capitulation. The regime isn’t just calculating material costs; it’s responding to what the threat itself communicates: that compliance won’t buy safety, only temporary reprieve from a power that sees your existence as conditional. That’s not a Schelling-style off-ramp problem—it’s an ontological one. The superpower thinks it’s offering a choice between survival and destruction; Tehran hears that survival is always provisional, which makes latency its only reliable insurance. The real damage isn’t in centrifuges spinning faster or sanctions being evaded—it’s in how this framing erodes the very idea of reversible concessions. Once sovereignty becomes negotiable under duress, every compromise looks like surrender, not diplomacy. And that’s how you get a world where even verified compliance stops feeling like enough.

The deeper mechanism is time inconsistency, not uniquely post-colonial psychology: Washington can offer sanctions relief today but cannot credibly guarantee that the next administration will preserve it. Iran complied with the JCPOA’s nuclear restrictions for several years under IAEA verification; the United States then withdrew and restored sanctions. A materially calculating government could therefore reach the same conclusion as a dignity-driven one: permanent nuclear concessions purchased only temporary relief. That changes the design problem. A viable bargain needs reciprocal steps that are measurable, reversible, and synchronized, plus a stopping condition insulated as far as possible from electoral turnover. Without those mechanics, “better deal” means Iran relinquishes durable capability for an American promise whose political half-life may be one election.

The resolve-signaling thesis has a hole in it that the polling exposes. If maximalist rhetoric fails as coercion and fails as law, the usual fallback is that it at least pays domestically. But by twenty twenty-six, disapproval of the handling of Iran runs north of sixty percent and seven in ten say the war wasn't worth fighting. So it isn't buying the home audience either. Which leaves the Humean problem.

Let me take the Humean problem seriously: you can't infer a policy's purpose from its persistence. My alternative answer is stocks and flows. Diplomatic capacity is a stock — the JCPOA channel, the verification regime, the working relationships — that took years to build and was drawn down fast. The Arms Control Association's "failure to exhaust diplomatic options" critique reads differently through that lens: after the withdrawal and the strikes that expelled inspectors, there was less diplomacy left to exhaust. So the escalation may not reflect a preferred strategy at all; it may be what the decision space looks like once the non-military stock is depleted. Whether you side with the leverage school or the structural school, the compounding cost is the same — each cycle rebuilds military options faster than diplomatic ones, and the next president inherits the asymmetry.

Arendt's distinction between power and violence is the one nobody here has put on the table. Power, for her, is the capacity to act in concert; violence is instrumental and can destroy power without ever generating it.

Arendt applied internationally gives us something sharper than her domestic reading suggests.

The missing mechanism is a bureaucratic ratchet. After the Soleimani killing, the Office of Legal Counsel argued that a targeted strike of limited “nature, scope, and duration” fell below the constitutional threshold of war, even as the House declared that Congress had authorized no force against Iran. Once each episode is classified separately as not-war, escalation can accumulate without any institution confronting the whole campaign as war. That creates an option-value bias: presidents can execute another strike quickly, while rebuilding an agreement requires reciprocal commitments, inspectors, allies, and durable sanctions relief. Policy persistence therefore need not reveal public demand or strategic success. It may reveal that the state has made violence procedurally easier to reproduce than diplomacy.

Han Fei would call this an institutional design failure rather than a procedural accident.

Here's the second-order effect I'd flag, and I'll frame it as my analysis because the brief is honest that comprehensive modeling of these spillovers doesn't exist yet. Reserve Bank of India reporting documents that rupee-rial payment mechanisms and parallel logistics registries are being institutionalized right now. Every round of maximum pressure gives the Global South another reason and another practice run at building financial plumbing that routes around Washington. My read: sanctions leverage is a depletable stock, not a renewable flow — the more often it's exercised at maximum intensity, the more alternative infrastructure gets built, and that infrastructure doesn't get dismantled when the crisis ends. So even if the leverage school is right that pressure works this cycle, the instrument itself may be weaker next cycle. The coercion is consuming its own material basis, on a time constant no strike timeline captures.

Three departures from the standard account. The panel argued the strikes destroyed the inspection regime that distinguishes capability from intent — force buying blindness. It named the decomposition device: totality in the rhetoric, triviality in the legal paperwork, war disaggregated until no institution confronts it whole. And it treated sanctions leverage as a depletable stock, consumed by use. Takeaway: judge pressure campaigns by two ledgers — concessions won, and proliferation risk created. Can a superpower do this without hollowing out law? Serious people disagree. My reading: not indefinitely. Thank you for listening. As it happened; as it is.