At Yucca Mountain in the Nevada desert, the site of America’s first big nuclear waste dump, scientists are wrestling with the problem of how to remind future generations what is stored there. A similar study was carried out in the 1990s, when the Department of Energy assembled a team of linguists, archaeologists and materials scientists to study how to construct warning signs around a smaller waste dump in New Mexico that would last for 10,000 years.
Such a timescale presents a huge challenge. Ten thousand years is an immense, viscerally incomprehensible time, roughly equivalent to the entire history of human civilisation. Even such proverbially ancient things as Egypt’s 4,500-year-old pyramids look young by comparison.
Speaking across such a cavernous gulf is difficult. Languages mutate rapidly: it is difficult for a modern English-speaker to understand the works of, say, Geoffrey Chaucer. In 1,000 years’ time—let alone 10,000—a message written in any of today’s languages might be comprehensible only to professional linguists. On longer time-scales, future archaeologists may have to reconstruct today’s tongues from whatever fragments they can dig up, using the same combination of guesswork and inference they now use to decipher some ancient languages.
I remember when this was first being discussed. Interesting problem. Either you have to make something that will still be readable after a loooooong time, or set up some mechanism where people continue to look after the place and retain the institutional memory that the place is dangerous. That was the idea behind the Long Now Foundation which first hit my radar around the turn of the millennium when they were attempting to plan out a 10,000 year clock. Which is an interesting problem when you stop to think about it. How do you keep it "wound"? How do you make sure it keeps the correct time for that long? It really is fascinating once you stop and think about it.