Isn't 38 million wildly high?

It's big, which is why we grade it PROBABLY TRUE not FACT. But it's the central estimate of a peer-reviewed Lancet Global Health study led by a mainstream economist — a modeled 'associated with' figure with a range. Skeptics can argue it's lower; even they concede sanctions cause large excess mortality. The fight is the size, not the existence, of the graves.

Isn't it the sanctioned regimes' own fault?

Often partly — mismanagement is real and is why attribution is hard. But humanitarian exemptions fail in practice (banks overcomply), and the intent is documented: the Cuba memo and the Chile 'make it scream' order show the US set out to cause economic pain. Victim-blaming doesn't erase the written plan.

Is this just anti-American spin?

The load-bearing sources are Western: a Lancet Global Health study led by a University of Denver economist, and US government documents for the Cuba/Chile intent. You can disagree with the activists who publicized it and still reckon with the memo an American diplomat wrote in 1960.

Investigation · The Dulles Blueprint · Economic Warfare

Death by sanctions: the war that doesn’t count as war.

Q: Aren't sanctions better than war?

Sometimes, and we don't argue for bombs instead. But 'better than war' isn't 'harmless,' and a policy associated with hundreds of thousands of deaths a year isn't 'non-violent.' The point is to count their dead as honestly as a war's, not file them under diplomacy.

We reserve the word “evil” for governments that kill actively and on purpose. But a blockade that starves a country until its government falls produces the same graves — slowly, statistically, deniably. A 2025 study in The Lancet Global Health puts the toll of Western sanctions since 1970 in the tens of millions.

The headline estimate — roughly 38 million deaths associated with US and EU sanctions — is a peer-reviewed model, and we grade it as one: PROBABLY TRUE, a serious central estimate with a real range, not a hard body count. What is FACT is the part that answers the moral question: in the landmark cases, the hunger was not an accident. It was the plan — written down.

The graded claims ↓The case for sanctions

§1 · Summary Brief

What this page argues

There is a moral asymmetry at the center of how we judge states. A government that shells a city is a war criminal; a government that cuts a country off from food, medicine, finance, and spare parts until its hospitals fail and its children starve is running “foreign policy.” The dead are the same. This page is about that gap — what the political scientist Johan Galtung called structural violence: harm delivered by policy and system rather than by a fist, and therefore rarely counted as harm at all.

The occasion is new evidence. A 2025 study in The Lancet Global Health, led by the economist Francisco Rodriguez of the University of Denver, produced the first global estimate of the mortality associated with international sanctions from 1970 to 2021. Its central estimate: unilateral sanctions imposed by the United States and European Union are associated with roughly 38 million deaths — more than half of them children and the elderly — with over 800,000 deaths in 2021 alone. We grade that number PROBABLY TRUE: it is peer-reviewed and careful, but it is a modeled, counterfactual estimate with a range, and we present it as such rather than as a settled count.

What we grade FACT is the intent, because in the defining cases it is on the record. A 1960 U.S. State Department memo argued that sanctions on Cuba should “bring about hunger, desperation and overthrow of government.” In 1970, President Nixon ordered his team to “make the economy scream” in Chile — an “invisible blockade” that helped pave the way for the coup that installed Pinochet. And in a modern case, one study estimates US sanctions caused about 40,000 excess deaths in Venezuela in a single year. The deprivation, again and again, was not a side effect. It was the instrument.

We carry the other side in full: sanctions are defended as a coercive tool short of war, humanitarian exemptions exist on paper, and the mortality estimates are genuinely contested — regime mismanagement and collapsing oil prices are real confounders, and “associated with” is not “caused.” But the debate is over the size of the number, not its direction, and not over the memos. When a policy is designed to make an economy scream, the screaming is the point.

What we are NOT claiming

We are not presenting 38 million as a hard body count. It is a peer-reviewed central estimate from an excess-mortality model — “associated with,” not a death certificate — and it carries a confidence range and depends on counterfactual assumptions. We grade it PROBABLY TRUE for exactly that reason.

We are not claiming every death in a sanctioned country is the fault of sanctions; regime corruption, mismanagement, war, and commodity shocks are real and, in cases like Venezuela, substantial confounders. And we are not arguing that sanctions are always unjustified or that force would be preferable. The claim is narrower: policies that predictably and sometimes deliberately kill by deprivation deserve the same moral scrutiny as killing by force — and they rarely get it.

Death by sanctions.

We call a government evil when it kills actively and on purpose. But a blockade that starves a country until its government falls produces the same graves — slowly, statistically, deniably. A 2025 Lancet study puts the toll of Western sanctions since 1970 in the tens of millions.

1 / 7▶ Present fullscreen

§2 · Thesis

A bullet and a blockade can produce the same grave. Only one gets called murder.

The intuition this page presses on is simple. We judge harm by its method, not its result: the government that drops a bomb is condemned; the government that quietly cuts off insulin, chlorine for water treatment, and the financing a country needs to import food is doing statecraft. But a child who dies of a treatable infection because the medicine was embargoed is exactly as dead as a child killed by a missile. The moral difference we feel is mostly about proximity and visibility, not about culpability.

That is the definition of structural violence: suffering built into the arrangement, with no single trigger-puller to blame. Sanctions are its purest modern form. They are slow, so they don’t make the news as a massacre. They are statistical, so no individual death is provably “caused” by them. And they are deniable, so the imposing power can call them peaceful even while an internal memo describes the goal as hunger and collapse.

We hold the discipline line. The aggregate death toll is a serious, peer-reviewed estimate, and we grade it as an estimate. But the moral argument does not depend on the exact number. It depends on two things that are not in dispute: that sanctions kill in large numbers, and that in the landmark cases the killing was the intended mechanism.

§3 · Graded Claims

The estimate, the intent, the cases, the argument.

A 2025 Lancet Global Health study estimates US/EU sanctions are 'associated with' ~38 million deaths since 1970.

PROBABLY TRUE

The study, led by economist Francisco Rodriguez of the University of Denver and published in The Lancet Global Health in 2025, produced the first global estimate of mortality associated with international sanctions across 1970–2021. Its central estimate: unilateral sanctions by the US and EU are associated with roughly 38 million deaths, more than half of them children and the elderly; in some 1990s years the figure exceeded a million; in 2021 it topped 800,000. We grade this PROBABLY TRUE, not FACT, on purpose: it is a peer-reviewed, top-journal analysis by a careful, mainstream economist, which gives it real weight — but it is a modeled excess-mortality estimate ('associated with,' not a death count), it carries a confidence range, and aggregate figures like this depend on counterfactual assumptions that reasonable people contest. A serious, credible estimate, presented as an estimate.

In the landmark cases, the hunger was the plan — written down.

FACT

This is the part that answers the moral question, and it is documented. On Cuba: a State Department memorandum by Lester Mallory, dated April 6, 1960, argued that because Castro enjoyed broad popular support, 'every possible means should be undertaken promptly to weaken the economic life of Cuba… denying money and supplies to Cuba, to decrease monetary and real wages, to bring about hunger, desperation and overthrow of government.' On Chile: after the 1970 election of socialist Salvador Allende, President Nixon ordered his team to 'make the economy scream' — the historian Peter Kornbluh's 'invisible blockade' of finance and trade that stoked the crisis preceding the US-backed 1973 coup and the Pinochet dictatorship. In these cases, deprivation was not an unfortunate by-product; it was the stated objective.

“Every possible means should be undertaken promptly to weaken the economic life of Cuba… to bring about hunger, desperation and overthrow of government.”

U.S. State Department — Lester D. Mallory memorandum, April 6, 1960 (Foreign Relations of the United States, 1958–1960, Vol. VI, Cuba)
Peter Kornbluh / National Security Archive — 'make the economy scream' and the 'invisible blockade' of Allende's Chile (The Pinochet File)

The case-by-case record backs the direction: Venezuela, Iraq, and more.

FACT

Before the global estimate, researchers documented sanctions' toll case by case. In Venezuela, an economic study by Mark Weisbrot and Jeffrey Sachs (Center for Economic and Policy Research, 2019) estimated that US sanctions caused roughly 40,000 excess deaths between 2017 and 2018, as the country lost access to imports of food, medicine, and the inputs its oil economy depended on. In 1990s Iraq, comprehensive UN/US sanctions produced documented, widespread malnutrition and shortages of clean water, medicine, and electricity (the precise death counts from that era are contested and we do not adopt a specific figure). These are the well-known data points the new global study aggregates — the direction of the evidence is not seriously in dispute; the magnitude is what the research argues about.

The throughline: structural violence — killing by deprivation is still killing, and it escapes the reckoning.

PROBABLY TRUE

Put the estimate beside the memos and a thesis holds: a state can achieve the ends of war — a broken economy, mass death, a toppled government — without firing a shot, and largely without moral consequence, because the harm is diffuse and deniable. We grade this PROBABLY TRUE as an argued framework (Galtung's structural violence applied to sanctions), not as a proven law. It rests on facts we grade higher: the documented intent in the Cuba and Chile cases, the case-level mortality studies, and a peer-reviewed global estimate. The claim is not that sanctions are always wrong or always lethal — it is that when they kill, the deaths belong on the same moral ledger as deaths by force, and almost never make it there.

Johan Galtung's concept of 'structural violence' — harm built into political and economic arrangements rather than delivered by direct force
Black Book Audit — The Dulles Blueprint: the ladder of co-opt, buy, and overthrow that sanctions serve

§4 · The Case for Sanctions

The strongest version of the other side — and what it does and doesn’t answer.

The defense: sanctions are a coercive tool that sits below war — a way to pressure a hostile government without invading. “Targeted” or “smart” sanctions aim at regime elites, asset freezes, and specific sectors rather than whole populations, and most programs carry humanitarian exemptions that are supposed to let food and medicine through. On this view, whatever suffering follows is the fault of the sanctioned regime, which could end it by changing course.

The methodology fight: the mortality estimates are genuinely contested. Excess-death models compare a sanctioned country to a counterfactual that has to be constructed, and the answer moves a lot with the assumptions. “Associated with” is not “caused by,” and confounders are everywhere: Venezuela’s economic collapse began before the harshest 2019 sanctions, amid an oil crash and years of mismanagement. Skeptics argue figures like 38 million overstate the sanctions’ independent contribution.

What it doesn’t answer: the humanitarian exemptions routinely fail in practice, because “overcompliance” by banks and shippers — afraid of secondary sanctions — chokes off even permitted trade. The confounders temper the exact number but not the direction, and not the intent: the Mallory memo and the “make it scream” order are US government words, not modeling artifacts. You can believe the true toll is a fraction of 38 million and still be left with a policy that, by its own architects’ description, was built to cause hunger.

§5 · Why It Matters

The quiet part is a memo, not a conspiracy.

Tim’s question — what do we make of policies that have the same effect as intentional killing? — turns out to have a sharper answer than expected. In the founding cases of the modern sanctions regime, the effect was the intent. The United States did not stumble into starving Cuba or strangling Chile; it wrote down that this was the goal, and pursued it. The novelty of the 2025 study is only that it tries to count the bodies the memos always implied.

This is why sanctions belong in The Dulles Blueprint — the standard method for disciplining governments that step out of line, which runs from co-opting leaders, to buying them, to overthrowing them. Economic warfare is the rung between the bribe and the coup: cheaper than an invasion, quieter than a bombing campaign, and lethal at a scale that would be a scandal if it arrived any other way. The reason to name it plainly is not to excuse the governments on the receiving end, many of them ugly. It is that a democracy should have to account for the deaths its policies cause — especially the ones it intended.

The Dulles Blueprint hub →The Guatemalan genocide →Operation Cyclone →

§6 · FAQ

Questions worth taking seriously

Isn't 38 million a wildly high, propaganda-sounding number?

It is a big number, which is exactly why we grade it PROBABLY TRUE and not FACT. But it’s not a slogan: it’s the central estimate of a peer-reviewed study in The Lancet Global Health, led by a mainstream US-based economist not known as a firebrand. It’s a modeled, “associated with” figure with a range, and skeptics can reasonably argue the true toll is lower. What even skeptics concede is that sanctions cause large excess mortality — the fight is about the size, not the existence, of the graves.

Aren't these countries' problems really their own governments' fault?

Often partly, yes — corruption and mismanagement are real and are the main reason attribution is hard, which we say plainly. But two things cut through it. First, humanitarian exemptions fail in practice because banks and shippers overcomply out of fear of secondary sanctions. Second, the intent is documented: the Cuba memo and the Chile “make it scream” order show the US set out to cause economic pain. Blaming the victim doesn’t erase the stated plan.

Aren't sanctions still better than going to war?

Sometimes, and we don’t argue for bombs instead. But “better than war” is not “harmless,” and when a policy is associated with hundreds of thousands of deaths a year, calling it “non-violent” is a euphemism. The point isn’t that sanctions are always worse than war — it’s that we should count their dead with the same honesty we count a war’s, instead of filing them under diplomacy.

Is this just anti-American / Global South talking points?

The load-bearing sources are the West’s own: the study ran in The Lancet Global Health and was led by a University of Denver economist; the Cuba and Chile intent documents are US government records. You can disagree with the activists who publicized the study and still have to reckon with the memo an American diplomat actually wrote in 1960.

§7 · Standing Invitation

If you are named on this page

If you are named on this page, or are a party materially affected by the claims made here, and you wish to respond, correct the record, or add context, use the Contact page. Responses are published verbatim alongside the original claim, with the sender identified and the date of receipt. The channel stays open for the life of the page.

This site aggregates and grades a record that other outlets and primary sources have already put on the record. Every FACT-graded claim above is sourced to court filings, government reports, sworn whistleblower disclosures, published investigative journalism, or named-source statements. The citations are the accountability mechanism; this section is how you get on the record too.

§8 · Sources

The estimate, and the memos underneath it.

Every claim on this page grades to one of FACT · PROBABLY TRUE · SOME SMOKE · PURE SPECULATION · FALSE / MISLEADING. The ~38 million figure is graded PROBABLY TRUE as a peer-reviewed central estimate (modeled, “associated with,” with a range); the documented intent (the Mallory memo, the Chile order) and the case-level studies are graded FACT; the structural-violence framework is the argued throughline.

The Lancet Global Health (2025) — Rodriguez et al., the global estimate of mortality associated with international sanctions, 1970–2021 (primary study)
Al Jazeera — Jason Hickel, Dylan Sullivan & Omer Tayyab, “US and EU sanctions have killed 38 million people since 1970” (the accessible summary that occasioned this page)
Weisbrot & Sachs, Center for Economic and Policy Research (2019) — the ~40,000 Venezuela excess-death estimate
U.S. State Department — Lester D. Mallory memorandum, April 6, 1960 (Foreign Relations of the United States, 1958–1960, Vol. VI, Cuba): the “hunger, desperation and overthrow” objective
Peter Kornbluh / National Security Archive, The Pinochet File — Nixon’s “make the economy scream” and the “invisible blockade” of Allende’s Chile

Full method: Methodology. Home hub: The Dulles Blueprint.

Last updated September 17, 2026. The ~38 million figure is a peer-reviewed central estimate from a 2025 Lancet Global Health study (Rodriguez et al.) — modeled excess mortality “associated with” US/EU sanctions, 1970–2021, with a confidence range — and is graded PROBABLY TRUE. The documented intent (the 1960 Mallory memo on Cuba; Nixon’s 1970 “make the economy scream” order on Chile) and the case-level mortality studies (Venezuela ~40,000, Weisbrot & Sachs; the documented humanitarian toll of 1990s Iraq sanctions) are graded FACT. The structural-violence framework is the argued throughline. The methodology critique and the case for sanctions are carried in full. If a link 404s or a detail is wrong, tell us and we will fix it publicly.