The Deal

The irony compounds across centuries. The original Domesday Book, compiled in 1086 on parchment by scribes working for William the Conqueror, remains readable after 938 years. The BBC Domesday Project of 1986, a 2.5 million pound digital survey of Britain stored on a proprietary LaserDisc format, became unreadable within sixteen years when the hardware and software needed to access it became obsolete. A heroic emulation effort rescued the data in 2002. Medieval parchment outlasted modern digital media by a factor of sixty.

Digital permanence is a lie. Digital preservation is continuous labor.

Civilization does not preserve truth by default. It preserves what power sustains.

The Forms of Loss That Shape History

Preservation failure tends to follow three patterns, and each tracks power with different precision.

Accidental loss arises from neglect, entropy, and technological change. Storage media degrade. File formats expire. Software environments vanish. Data remains physically present but functionally inaccessible. The Apollo tapes belong to this category, though “accidental” understates the institutional choices that produced the accident.

Chapter 6 documented this dynamic in the digital age: MySpace destroying fifty million songs in a server migration, Yahoo erasing thirty-eight million Geocities pages, Google killing the RSS ecosystem that sustained independent publishing. Each loss resulted from corporate decisions that treated cultural memory as a storage cost rather than a public obligation. When preservation competes with operational budgets, preservation loses. When no structural mandate requires retention, institutions optimize for the present and discard the past.

Structural loss occurs when certain lives, practices, or outcomes are never recorded in the first place. This is the deepest form of disappearance, because what was never captured cannot be mourned as missing.

Aboriginal Australians developed fire management practices over at least 60,000 years of continuous habitation, the longest sustained knowledge tradition documented anywhere on Earth. Regular, controlled burns created mosaic landscapes that prevented catastrophic wildfires while promoting biodiversity. Plant species evolved alongside these burning practices. Soil chemistry depended on them. Animal populations adjusted their territories around them. The knowledge was empirical, refined across thousands of generations, and encoded in oral traditions, seasonal ceremonies, and intergenerational apprenticeship.

British colonizers dismissed the practice as primitive. They banned traditional burning and imposed European forestry methods designed for wetter climates and different soils. The knowledge was not destroyed in the sense that a library is burned. It was excluded from the systems that colonial power recognized as legitimate: written records, scientific journals, government archives. It persisted within Aboriginal communities but was marginalized, defunded, and actively suppressed for two centuries.

In 2019 and 2020, Australia burned. Approximately 46 million acres were consumed. At least 34 people died directly. Over one billion animals perished. Firefighting costs exceeded two billion dollars. In the aftermath, Australian authorities belatedly recognized that indigenous fire practices, which had sustained the continent’s ecosystems for millennia, actually prevent the megafires that European-style fire suppression had made catastrophic. Recovery efforts now include consulting surviving elders, but much knowledge has already been lost with the practitioners who held it. What was never recorded cannot be fully recovered. And what was never recorded tracks power precisely: Aboriginal knowledge was excluded because Aboriginal people were excluded.

Intentional erasure occurs when records are suppressed, destroyed, classified, or rewritten to protect authority. This is the most deliberate form of loss, and the hardest to prove, because the evidence of destruction is typically destroyed along with the evidence itself.

Between 2002 and 2006, the CIA recorded ninety-two videotapes documenting “enhanced interrogation” sessions with detainees suspected of terrorism. The tapes showed what was done, by whom, under what conditions, and with what results. In 2005, despite ongoing court proceedings and active Congressional inquiries, the CIA destroyed all ninety-two tapes.

The official justification was protecting interrogator identities. The functional effect was eliminating primary evidence of potential war crimes. The Senate Intelligence Committee eventually documented the interrogation program through secondary sources: internal cables, emails, memoranda, and testimony. Their 6,700-page report, completed in 2014, remains largely classified, with only a 525-page executive summary released to the public. But the tapes themselves, which would have shown exactly what was done in unambiguous visual form, are gone.

No one was prosecuted for destroying the evidence. The destruction violated no statute, because the rules governing evidence preservation had been written to exclude precisely this scenario. The archive protected power by forgetting what power wished forgotten. In a courtroom, destroying evidence is obstruction of justice. In the archive of history, it is standard procedure.

These three forms of loss, accidental, structural, and intentional, follow a pattern that transcends technology. Whether parchment or magnetic tape, oral tradition or digital file, what survives is what serves power or escapes its notice. The Apollo tapes were erased because space history competed with active missions for budget. Indigenous knowledge was excluded because indigenous people lacked political standing. CIA videotapes were destroyed because they documented acts their creators wanted deniable.

Preservation is not a neutral technical operation. It is a political one. The question is never simply “can we save this?” The question is “does someone with resources want this saved?” When the answer is yes, archives find funding, formats get migrated, emulation happens. When the answer is no, the data degrades, the knowledge dies, the evidence disappears.

The archive serves power by default. Making it serve accountability requires deliberate structural design.