The Centralia Files, Part I: The Fire That Never Became an Emergency

How a subsurface coal fire spent two decades not quite becoming anyone’s problem — and what that tells us about managed abandonment as policy.

The Prologue to this series established the structural condition that preceded the fire: a borough built on severed mineral rights, where the ground beneath residents’ homes belonged to extraction interests, not to the people living above it. That separation was not incidental. It determined who had legal standing to act when the ground began to burn, who bore financial liability for what happened next, and which institutions could plausibly claim the problem was simply not theirs to deal with.

Part I begins where that framework meets its first real test.

In 1962, a fire started at the edge of Centralia. It entered the mine workings beneath the town and did not go out. What followed was not an emergency response. It was something more institutional and more consequential: the normalization of a slow disaster that no single agency was structured to resolve.

What Started in 1962: A Record That Does Not Fully Settle

The most widely documented account holds that the fire began in the Centralia landfill, located in an abandoned strip mine pit at the edge of town. The landfill was operated by the borough. In May 1962, the landfill was burned — a common waste disposal practice of the era — and the fire is believed to have reached an exposed coal vein or unsealed mine opening, entering the subsurface mine workings below.

A competing account, documented by journalist David DeKok and others who have examined the historical record closely, suggests the fire may have predated the 1962 landfill burn, originating from an earlier, incompletely extinguished fire in the same area. Under this account, the 1962 burn did not start the fire so much as reactivate or spread one already present.

A third line of dispute concerns who authorized the landfill burn and whether proper precautions — specifically, the required removal of coal refuse and the sealing of mine openings in the pit — were taken beforehand. The borough council authorized the burn. Whether the site was adequately prepared is a question the record does not cleanly answer.

What the record establishes is that by summer 1962, a fire was burning in the subsurface, suppression efforts were underway, and the fire was not going out. The question of precise origin matters for liability and historical accuracy. For the institutional analysis that follows, what matters more is what happened after ignition — and why the response was so consistently inadequate. — The Centralia Files analytical framework

Early Suppression: Attempts That Did Not Hold

Initial efforts to extinguish the fire began within weeks of its detection. Workers excavated burning material and flushed water into affected areas. These followed standard mine fire suppression practice for the period and were reasonable first responses to what appeared, initially, to be a containable problem.

They did not work.

By late 1962 and into 1963, it was increasingly clear that the fire had moved beyond the immediate ignition area and into the interconnected mine passages beneath Centralia. The anthracite seams in this part of Pennsylvania had been mined extensively for decades, leaving a subsurface infrastructure of tunnels, shafts, and worked-out chambers that provided pathways for the fire to travel and oxygen to sustain it.

The Pennsylvania Department of Mines and Mineral Industries held primary state jurisdiction over mine safety and reclamation. DMMI directed and oversaw early suppression work, coordinating with borough officials and engaging contractors for various intervention approaches. Flushing operations — pumping a slurry of water, sand, and ash into affected areas — were attempted. Trenching, excavating a barrier across the fire’s projected path, was also tried.

The Bureau of Mines, the relevant federal agency at the time, provided technical assistance and funding support during the 1960s. Its involvement was advisory and financial rather than operational. The bureau did not hold direct management authority and worked through the state structure.

Each suppression attempt produced some short-term effect and failed to contain the fire over time. The mine workings were extensive enough and the fire’s behavior unpredictable enough that barrier approaches repeatedly proved insufficient. The fire moved around obstacles, found new oxygen sources, and continued spreading.

By the late 1960s, suppression had failed. The fire was still burning. And the institutional response was beginning to exhibit the pattern that would define the next decade and a half: continued engagement without resolution, and continued reframing of the problem as manageable rather than urgent.

Jurisdictional Confusion: A Structure Built for Other Problems

The governance framework surrounding Centralia was not designed for a burning mine beneath an occupied town. It was designed for mine safety inspection, mine reclamation after extraction, and the regulation of active mining operations. None of those frameworks mapped cleanly onto the problem Centralia presented.

The mineral rights had been exercised. The liability had not followed.

Centralia borough council made repeated requests for assistance throughout the 1960s and into the 1970s, documenting the fire’s progression, requesting funding, and pressing state and federal agencies to treat the situation as the emergency its residents understood it to be. Those requests produced responses. Studies were commissioned. Funding was periodically authorized for specific suppression efforts. Meetings were held.

What the requests did not produce was a lead agency with clear authority, adequate funding, and a mandate to resolve the problem rather than manage it.

Normalization: How Slow Harm Escapes Emergency Status

By the 1970s, the Centralia mine fire had been burning for over a decade. It had survived multiple suppression attempts. It had spread. And it had, in an institutional sense, become routine.

Residents of Centralia were not unaware of the fire. They smelled it. They saw steam rising from the ground in winter. Some had experienced subsidence on their properties. The Concerned Citizens Action Group — a resident organization that formed as the fire’s effects became more visible — documented conditions, organized politically, and pressed for action with increasing urgency through the 1970s. Their record of the situation did not match the administrative record being generated by the agencies nominally responsible for addressing it.

Agency records from this period tend to describe a situation under study, with suppression efforts underway and conditions being monitored. Resident accounts describe a town being slowly made uninhabitable while institutions debated jurisdiction and funding. Both sets of records are real. The gap between them is what administrative normalization looks like from the inside.

Carbon monoxide levels in some homes were elevated. Ground temperatures in affected areas were measurable and documented. The fire’s subsurface extent was being mapped, with each new study extending the estimated boundary further than the last. None of this produced a declaration of emergency. It produced more studies.

The 1981 incident that finally broke through the administrative normalization was not a policy document. It was a twelve-year-old resident who nearly fell into a collapsed sinkhole in his backyard, where ground temperatures and gas readings immediately confirmed the fire’s proximity. The incident received national media coverage. It did not create the emergency. It made the existing emergency visible to audiences that had not been attending to it.

What the First Two Decades Produced

By the early 1980s, the documented record of the Centralia mine fire showed the following: a fire burning for approximately twenty years; suppression attempts that had not succeeded; a jurisdictional structure that had not produced a lead agency or a resolution strategy; a resident population with documented health and safety exposure; and a growing body of engineering studies whose cost estimates for full remediation were escalating with each revision.

That record set the conditions for the decision that Part II of this series examines: the federal and state determination that relocating Centralia’s residents was more administratively tractable than extinguishing the fire beneath them.

That conclusion did not emerge from a single meeting or a single document. It emerged from the accumulated weight of twenty years of inconclusive response — a period during which the cost of remediation grew, the political will to fund it did not, and the population of the affected community declined to a size that made relocation financially feasible in a way it would not have been earlier.

The fire was the condition. The two decades of inadequate response were the mechanism. Together, they produced the cost-benefit logic that would, in 1983, convert an unresolved environmental crisis into a federal relocation program.