Atomfall

The pub is open. The pub is, in any practical sense, the only fully functioning institution within the cordon. The Home Guard chap who used to run security duties at the Windscale reactor site is in the corner, nursing a half-pint of mild that the pub's owner has, against all evident supply-chain reality, continued to keep on tap. The vicar is at the bar, asking about Tuesday's parish-council meeting. A child has wandered in from the road and is being given a packet of crisps by the pub's owner without payment being mentioned. Outside, the cordon's outer wire is visible at the edge of the village green. The light through the small leaded windows is the specific late-afternoon English light that the BBC's wartime Public Information Films managed to register as the country's emotional default. Somebody, in the back room, is playing a piano with one hand while the other holds a cigarette. The community is, against the conditions trying to close it, continuing.

This is Atomfall, the Rebellion Developments game released in March 2025, set in the Cumbrian countryside in the years immediately after the 1957 Windscale fire. The Windscale fire was an actual nuclear accident at the British Cumberland coastal site, the most serious in British history at the time. The postwar British state handled it the way the postwar British state handled most embarrassing facts: the milk from local farms was quietly destroyed, the documents were classified for decades, the village kept on.

The frame this essay wants to give the reader: every disaster narrative the reader has encountered in their cultural life has been making a specific argument about what disasters actually do to human communities. Most American disaster narratives have been arguing the same thing: the disaster reveals the war of all against all, the institutions collapse, only the strong man with the gun survives. This argument is, on the empirical record of the past century of disaster research, almost completely wrong. The actual record is the opposite. People organize. People share. People improvise mutual-aid networks of remarkable speed and effectiveness. Communities do not, on the available evidence, collapse into chaos under disaster conditions. They mostly continue, badly and locally, doing what they can. Almost no popular culture has been willing to render this. Atomfall is one of the small handful of contemporary cultural products that has.

The reader who has this frame can apply it to almost every disaster-themed piece of mass culture they encounter. The next zombie show, the next post-apocalyptic film, the next survivor-narrative novel, the next end-of-the-world game, can be evaluated on a single useful question: which of the two arguments is the product making about what disasters reveal? The product that makes the chaos-and-strong-man argument is making the argument the actual evidence does not support. The product that makes the cooperation-and-improvisation argument is making the argument the evidence does support. Both kinds of product can be entertaining. The cultural-historical work the second kind is doing is, on the available evidence, more useful, more accurate, and substantially rarer.

The empirical foundation for this is some of the most-replicated findings in the social sciences. Rebecca Solnit, in A Paradise Built in Hell (2009), assembled the documentation across several major disaster cases: the San Francisco earthquake of 1906, the London Blitz, the Halifax explosion of 1917, the Mexico City earthquakes of 1985, Hurricane Katrina, and the 9/11 attacks. Solnit's argument, drawn from the careful reading of the social-scientific record across all these cases, was that the popular media's standard catastrophe narrative is empirically wrong. The actual record is mostly cooperative. People help each other. People share resources. People organize mutual aid often before the official institutional responses arrive. The "war of all against all" narrative is, on the available evidence, mostly mythical, sustained by the convenience the narrative provides to specific institutional actors (newspapers needing dramatic stories, state authorities wanting to justify militarized responses) rather than by the empirical reality of how disasters unfold.

The disaster-sociology field, founded by E. L. Quarantelli at the Disaster Research Center at the University of Delaware in 1963 and extended by many subsequent researchers, has confirmed this picture across multiple decades of post-disaster fieldwork. The looting myths that dominate disaster media coverage are, on the available evidence, mostly mythical or substantially exaggerated. The actual disaster response in case after case has been the same pattern: mutual aid, often improvised at speed, sometimes more effective than the official institutional response. The pattern holds across countries, across disaster types, across the political and economic conditions of the affected populations. The pattern is, by this point in the research literature, not particularly controversial.

The deeper anthropological argument was made by the Russian geographer and anarchist Peter Kropotkin in 1902, in Mutual Aid: A Factor of Evolution. Kropotkin had spent years observing animal and human communities across Siberia, and he had become convinced that the prevailing late-Victorian story about evolution, which held that the survival of the fittest meant the survival of the most individually aggressive, was empirically false. The actual evolutionary record, Kropotkin argued, was substantially shaped by cooperation. The bird flocks, the wolf packs, the beaver dams, the village commons, the trade guilds, the mutual-insurance societies had been, across geological and historical time, the dominant solutions to the problem of survival. Cooperation was the species-typical response. The competition story was a particular nineteenth-century class fraction's ideological needs presented as natural law.

Kropotkin's argument has aged better than nearly any other piece of late-Victorian social science. The contemporary evolutionary biology and anthropology have, broadly, confirmed his picture. Cooperation is the default. The competition story is the ideology. The disaster does not reveal the war of all against all because the war of all against all is not the human baseline. The disaster reveals what the human baseline is, which is largely cooperative and locally organized.

This is the picture Atomfall renders. The cordoned Cumbrian village in the game is doing in fiction what Solnit and Quarantelli's disaster sociology documents communities doing in fact. The parish council still meets. The pub stays open. The vicar still preaches. The villagers trade with each other through small informal arrangements that the cordoning state apparatus has not been able to fully suppress. The Druids in the local woods and the cult of the Atom and the residual Anglican parish are all, in their peculiar ways, local attempts to organize meaning and mutual aid after the larger systems have stopped functioning. The community is messy and weird and not entirely benign (some of the local belief systems are dangerous, the way actual local belief systems sometimes are under stress). The community is, however, there. The community is what survived.

The cultural source the game has drawn from for this is not American post-apocalyptic SF. The source is British folk horror and the broader British provincial disaster tradition. The standard shorthand for the game in the early reviews was "British Fallout," which is correct as far as it goes and misses what the game is actually doing. Fallout is fundamentally an American post-apocalyptic register: wilderness re-asserting itself over the suburb, the lone scavenger with the gun, the frontier revival in a re-wilded country. Atomfall is doing the opposite. The post-disaster Cumbrian landscape is not wilderness. It is a parish that has continued. The cultural inheritance the game is drawing from is John Wyndham's Day of the Triffids (1951), Nigel Kneale's Quatermass serials of the 1950s, the BBC's Survivors (1975), and the longer tradition of British post-war disaster fiction that Adam Scovell traced under the broader label of folk horror in his 2017 study Folk Horror: Hours Dreadful and Things Strange.

The British provincial disaster tradition is structurally different from the American post-apocalyptic tradition in ways that matter for the broader argument. American post-apocalyptic fiction tends to render the post-disaster landscape as wilderness inviting the rugged individualist's frontier revival. British post-war disaster fiction tends to render the post-disaster landscape as a parish that has continued, in conditions made local and improvised by the failure of the central state. The American story is structurally invested in the lone hero. The British story is structurally invested in the small community. The two traditions are not making the same argument about what disasters do, and the difference is more substantial than the surface-genre conventions make visible.

Patrick Wright, in On Living in an Old Country (1985), made the foundational argument about why the British tradition produces this register. Wright's claim was that the British cultural relationship to its own landscape is structured by a specific combination of nostalgia and dread. The countryside is loved as a repository of national meaning and simultaneously suspected of harboring something the country has chosen not to look at. The wartime evacuation novels, the Cold War nuclear-anxiety films, and the 1970s folk-horror cycle all drew on this same combination. The same village is the source of meaning and the site of the buried thing. The country has, since approximately the early twentieth century, been working through this combination across many of its most distinctive cultural products. Atomfall is the latest entry in that tradition, transposed into the video game medium in the specific case of a post-1957 nuclear accident in the Lake District.

The mechanical design supports the cultural register. The game's economy is built on the small, the local, the lend-and-trade: the player gets through the cordon by knowing which villager will provide which thing, and by listening to the parish gossip. There is no proper currency. There are no large factions of the conventional post-apocalyptic kind. There are several small, locally-organized groups, each with its own peculiarity. The combat is sparse and ill-tempered. The investigation is mostly social. The puzzles are the kind the British detective tradition has been building for a century: who has what, who is lying about it, who is in fact in the back room of the vicarage.

What the game's design lets the player do, in the small interactive way the medium permits, is inhabit the British provincial disaster register for the length of a campaign. The player learns the village, learns who knows what, learns to negotiate the small informal arrangements that the parish has improvised after the larger systems have stopped supplying answers. The player is, in this register, doing the work the actual surviving population of a disaster-affected community would be doing. The work is mutual aid, locally negotiated, partly cooperative and partly suspicious, with the cultural-historical resources of the local tradition supplying both the trust and the limits.

The reader who finishes this essay can carry several useful frames forward. The first frame: when the next disaster narrative arrives, ask which of the two arguments it is making. The chaos-and-strong-man argument is the argument the empirical record does not support. The cooperation-and-improvisation argument is the argument the record does. Both kinds of narrative can be entertaining. The second kind is doing more accurate cultural-historical work.

The second frame: the dominant American post-apocalyptic register has been training Anglophone audiences, for several decades, in an expectation about human behavior that is, on the available evidence, mostly wrong. When this register is the audience's primary cultural exposure to disaster, the audience develops intuitions about how people behave under stress that have very little correspondence to how people actually behave. The intuitions become politically consequential in the moments when actual disasters happen and the public conversation has to decide whether to trust the cultural intuitions or the empirical record. The cultural intuitions have, on multiple recent occasions in many countries, won. The consequences have not, on the empirical record, been good.

The third frame: the contemporary cultural environment has not been producing many cultural products that render the empirically accurate picture. The few that do are, in their small way, doing useful work. Atomfall is one of them. The 2009 Cormac McCarthy Road adaptation is another, in its specific muted way. Some episodes of the 2020-2021 Station Eleven miniseries are. The films of the Iranian director Asghar Farhadi often are, in their specific contained register. The list is, on a careful look, longer than the casual impression suggests; the cultural products are there for the audience willing to look for them.

The pub stays open. The parish council meets on Tuesday. The cordoning state has not been able to suppress the small informal arrangements the village has improvised. This is, against the dominant cultural story about what disasters reveal, what disasters actually look like in the empirical record. The game has rendered it. The reader who has the frame can carry it forward into every future disaster narrative they encounter. The frame is, on the available evidence, more accurate than the alternative the dominant culture has been training. Recognizing this matters because the next time the reader encounters a real disaster in the news, the cultural training the reader has absorbed will shape how the reader reads the situation. The empirical record is the better source. The cultural products that render it are worth defending.