The arithmetic of annihilation: Russian strikes on Ukrainian cities are not collateral damage
Odesa, 18 May 2026. Dozens of residential buildings damaged overnight in a single Russian attack. Dnipro, the same night: the roof of a twenty-four-storey apartment block catches fire. Nine people are pulled from the wreckage, among them a ten-year-old boy. These are not the statistics of a war going well for the defender. They are the product of systematic, repeated, geographically distributed strikes on populated urban areas — strikes that, when examined across eighteen months of reporting, do not read as intelligence failures or incidental overshoots. They read as doctrine.
The arithmetic is worth stating plainly. Residential buildings, grocery stores, transport hubs, medical facilities: across Dnipro, Odesa, Kharkiv, and a dozen smaller cities, the categories of infrastructure struck by Russian forces in 2026 share a common denominator. They are located in places where ordinary people live. The question of whether this reflects deliberate policy, reckless indifference, or operational incompetence is not a philosophical inquiry. It has legal consequences under the laws of armed conflict, and it has political consequences for how Western governments calibrate their continued support for Ukraine.
The doctrine, not the accident
International humanitarian law distinguishes between attacks that incidentally harm civilians — which remain subject to proportionality and precaution requirements — and attacks directed at civilian objects as such, which constitute war crimes. The repeated targeting of apartment blocks in cities like Dnipro is not consistent with an army that occasionally fails to distinguish military from civilian targets. It is consistent with an army that has decided that civilian infrastructure is a valid target category.
The legal framework is clear. Under Additional Protocol I to the Geneva Conventions, to which Russia is a signatory, attacks on civilian objects are prohibited unless and until they constitute a military objective by contribution to effective military action, and even then must satisfy requirements of proportionality and precaution. A twenty-four-storey residential building in a city of 900,000 people does not become a military objective because it is tall. The targeting logic, to the extent it can be reconstructed from open-source strike reporting, points toward a strategic function: the degradation of habitability in cities not yet contested on the ground, the dissipation of civilian morale over time, and — the most cynical calculation — the production of refugee flows that compound the logistical and political burden on Ukraine's Western partners.
The sourcing does not allow us to penetrate Russian military planning documents. What it allows is pattern recognition across dozens of reported strikes in the past eighteen months. Pattern recognition at this scale is not speculation. It is evidence.
Western responses and their gaps
Ukraine maintains an active and increasingly sophisticated air defence architecture. Systems provided by Western partners, including NASAMS and Patriot batteries, have intercepted Russian missiles at rates that would have been considered remarkable in early 2023. But protecting a city the size of Dnipro — or Odesa, or Kharkiv — against saturation strikes requires both hardware and the legal and political authorisation to use it in ways that some partners have been reluctant to grant.
The gaps in coverage are not primarily technological. They are political. Decisions about which systems to transfer, which ranges to permit, and which Russian military assets are permissible targets have varied across administrations and across the Atlantic alliance in ways that have left parts of Ukrainian cities exposed. Russia's strike planners are not unaware of these gaps. They plan accordingly.
Western public opinion has shifted in cycles — enthusiasm for military aid that spiked in early 2024, fatigue through mid-2025, renewed commitment following diplomatic setbacks in early 2026. Each cycle has corresponded with observable changes in Russian strike tempo. The inference is uncomfortable but difficult to avoid: Russian military planners treat Western political attention as a variable in their operational calculus, and they adjust civilian targeting intensity accordingly. That this is a rational strategy within their framework does not make it morally coherent. It makes it strategically coherent in the way that every war crime is coherent within the logic that produces it.
What the ten-year-old boy represents
The boy injured in Dnipro on 18 May is not a statistic in the broader contest of attrition. He is a ten-year-old boy who was in his apartment when a missile struck his building. His specific identity is not relevant to this analysis; what is relevant is the category he occupies — and the category is not military.
The framing of these attacks as incidental to a campaign of military attrition requires a willing suspension of the pattern. It requires treating the twenty-fourth strike on a residential district as somehow different from the twenty-third. It requires a parsing of Russian official statements about "military infrastructure" that no open-source analyst, no Ukrainian civil defence official, and no independent war crimes monitoring body has found credible. The burden of proof for claiming that a strike on a populated apartment block was proportionate and discriminative lies with the attacker. Russia has not discharged that burden in any systematic, publicly verifiable way.
The argument that this is the inevitable cost of a major conventional war does not hold when the target sets are examined. Military bases and logistics nodes are legitimate targets. Apartment blocks housing families are not. That the same force conducts both types of strike in the same evening does not collapse the distinction. It sharpens it.
The arithmetic continues
Ukraine's cities absorb these strikes and continue to function — not because the damage is negligible, but because the alternative is to stop. Dnipro's industrial base, its transport connections, its civilian institutions all operate under a level of threat that would prompt evacuation in most European contexts. That the city does not empty is not evidence that the strikes are tolerable. It is evidence of a level of endurance that its attackers are attempting to measure and then exceed.
The Western response, measured in hardware transfers and political statements, has not established a credible red line against the targeting of populated urban areas. That red line exists in international law. It does not exist in the operational calculations of the force launching these strikes, because the force has not been made to experience consequences for crossing it. Until that changes — through the provision of air defence systems in sufficient quantity and with sufficient authorisation, through the political will to name what these strikes are, and through the institutional mechanisms to document them for future accountability processes — the arithmetic of annihilation will continue to be calculated by those who have decided that ordinary lives in ordinary cities are an acceptable variable in their operational calculus.
The buildings were residential. The casualties were civilian. The pattern is not ambiguous, and the cost of treating it as such is measured in children.
This publication has covered Russia's full-scale invasion of Ukraine since February 2022. The reporting on civilian infrastructure strikes across Dnipro, Odesa, and other cities has been consistent in documenting patterns that, in our assessment, point toward deliberate targeting of non-military objects. Wire coverage has framed these attacks primarily in terms of air defence statistics; we have sought to foreground the character and intent of the strikes themselves.
Wire provenance
This editorial synthesis draws on the following public wire/social posts:
- https://t.me/TSN_ua/18432
- https://t.me/TSN_ua/18429
- https://t.me/TSN_ua/18427
- https://t.me/TSN_ua/18425
- https://t.me/TSN_ua/18423