Reading the war's aims from its targets, rather than its discourse
40 days, 2,125 attacks, 130 cities, from 7 Iranian Telegram channels
Content note: This site describes deaths and destroyed civilian sites. The original posts appear only when you open an event.
Scroll to begin
What was the war for? In his first statement, on the day the war began, President Trump named its aims.
Stated aim: nuclear weapons
“Our objective is to defend the American people by eliminating imminent threats from the Iranian regime. … They will never have a nuclear weapon.”
Stated aim: missiles
“We're going to destroy their missiles and raze their missile industry to the ground.”
Stated aim: the navy
“We're going to annihilate their navy.”
Stated aim: regime change
“When we are finished, take over your government.”
Talk is cheap, but missiles are not. So instead of what they said, I looked at what they did. Each dot is one attack: 2,125 in 40 days, built with a large language model from 17,754 posts on 7 Iranian Telegram channels.
One attack can hit more than one kind of target, homes and a government building at once, for example. Split by target, the 2,125 attacks become 4,707 pieces, one for each kind of target hit. That is why the shares that follow add up to more than a hundred percent.
Stacked by kind of target, the pieces show how often each target was hit. Each bar is the share of attacks that hit it. Missiles were coded as a target type, so their bar is only an outline; the navy has no bar at all.
The gap between words and targets was large. Nuclear sites were hit in 5% of attacks, missile sites in about a quarter, and the navy does not appear in the reports at all. Of the stated aims, only regime change was hit often. The two largest kinds of target were never stated: critical infrastructure, hit in 58% of attacks, and civilian sites, in 52%.
Chart summary. A bar chart of how often each kind of target was hit over the whole war, as a share of 2,125 attacks: critical infrastructure 58%, regime change 52%, civilian sites 52%, military sites 46%, other 8%, nuclear sites 5%. Missiles were coded as a target type, so they appear only as an outline (25%); the navy does not appear in the reports at all.
So where did the bombs land? Each attack happened in a city. Here is every one of them, placed where it hit.
Now scroll to run the war day by day. Each dot lights up on the day of its attack, and the counter keeps track of the cities hit so far. Scroll back up, and the war runs backward.
On the first day of the war, 14 cities were hit. Within three weeks, 111. By the end, 130. Tehran alone was hit 548 times, and Minab 157 times. So the war was not confined to a front, or to a handful of military sites. And in 69 of these 130 cities, the kind of target hit most often was critical infrastructure.
Thin lines join crowded clusters to their true locations.
Stacked day by day, the pieces show that the attacks did not focus on one kind of target. Over the whole war, critical infrastructure was hit in 58% of attacks, regime-related targets and civilian sites alike in 52%, and military sites in 46%. For every attack on a military site, there were about 1.13 attacks on civilian sites. And the bands rose and fell together: when the campaign intensified, it intensified everywhere at once.
Map summary. Each attack is a dot at its city. Day by day the war spread: 14 cities were hit on the first day, 111 within three weeks, and 130 by the end. Tehran was hit 548 times and Minab 157 times.
Split by kind of target, each chart holds one category, and the thin outline traces the day's total, on its own scale. On days with more attacks, every kind of target was hit more, and each kept roughly the same share of the day. In other words, the campaign did not switch from one kind of target to another as it went on. It raised the intensity on all of them at once. The correlations with the daily total run from 0.82 to 0.93.
Now civilians. Civilian sites were hit on the same days as government and military sites, and in similar numbers. When government and military sites were hit more, civilian sites were hit more too. And this was not a matter of a few bad days: the pattern held across all 40 days of the war.
This suggests that civilian harm was not merely collateral. If it had been, we would expect it to appear more randomly. Instead, it moved with the rhythm of the campaign. So the civilian toll is hard to call accidental, and this pattern raises serious questions, under the principles of distinction and proportionality in international humanitarian law, about whether war crimes were committed. Together, these four kinds of target are what makes a state a state: its people, its infrastructure, its army, and its government. So the war was aimed at Iran as a whole, not at a narrow objective like destroying its nuclear program or overthrowing its government.
Chart summary. Daily charts for each kind of target over the 40 days. The kinds of target rose and fell together with the daily total: their correlations with the total run from 0.82 to 0.93. Civilian sites moved with regime targets (r = 0.84) and with military sites (r = 0.70).
The data add one more, and even darker, layer: they show not only what the war was about, but how it ended. Look at the last 5 days, from 4 April 2026 to 8 April 2026.
In those final days, the share of attacks that hit critical infrastructure rose from 54% to 75%, while the share that hit civilian sites fell from 54% to 44%. So the endgame did not match the most loudly advertised aims of the war. Its clearest logic was infrastructural. And infrastructures are not just technical systems. They are what allow hospitals to function, cities to move, homes to receive energy, and any government to govern.
The same shift shows on the map. Here are the attacks on infrastructure in the first 35 days, 27.9 a day. Keep scrolling for the final 5 days, when they rose to 49.6 a day and reached 55 cities across the country.
The targets point to a war on Iran in its totality: its people, its infrastructure, its army, and its government were hit together. And as the war approached its end, it turned to the systems that make a country livable and governable: energy, transport, health, and education. The attacks were meant to damage not only the Iran of the war, but also its ability to recover.
They targeted Iran's future.
Now the dots are yours. Filter them by kind of target, run the war at your own pace, or look up a city.
Chart summary. In the first 35 days, 54% of attacks hit critical infrastructure and 54% hit civilian sites. In the final 5 days (4 April 2026 to 8 April 2026) the infrastructure share rose to 75% and the civilian share fell to 44%. Attacks on infrastructure rose from 27.9 to 49.6 a day and reached 55 cities.