There is no evidence that cannot be made to point both ways, by a sufficiently brazen liar, or a sufficiently deluded conspiracy theorist. When photographs of Donald Trump's inauguration in 2017 showed smaller crowds than for Barack Obama's eight years earlier, spokesman Sean Spicer insisted they were larger, to what was (at first) a perplexed press corps. Truth is a troublesome thing for an autocrat, just as Occam's Razor is a nuisance to the fantasist. Wouldn't it be great if there were a Face on Mars? But it's only an accident of lighting: and not even much of an accident, because if you photograph an entire planet of oddly-shaped hills with a blurry camera, one of them is probably going to look like a face. It's easy to misrepresent matters if there's a lot of evidence, because then you can crop the photos, use selective views only, ignore what doesn't suit, and so on. Darrell Huff's classic How to Lie with Statistics (1954) continues to be a book you should give every budding journalist for Christmas.
But matters are more complicated when, as is more usually the case, there isn't an obvious answer to the question, or even an obvious question. Consider how much film footage, photography and newsprint survives from, say, 1977. What does it tell us? The trouble is that it doesn't have a single message. Cut it one way, and you see the decline of Western civilisation in the wake of oil shocks, a collapse of post-war certainties, punk rock, etc. Cut it another, and you get Star Wars, Close Encounters of the Third Kind, The Spy Who Loved Me and other enormously fun spectaculars, with a soundtrack of Donna Summer's synthesiser hit I Feel Love. It wouldn't be hard to make two montage documentaries about 1977, one presenting it as a world hopelessly obsessed with the Second World War, and another showing an optimistic new dawn.
And when an archive of data has always been presented in broadly the same way, with the same classic moments always dredged out, it can be a valuable corrective to present a counter-narrative on the same data. Just as there should never be an unchanging canon of literature, so there is no fixed decision on what documentary footage is important, and what is not. We can genuinely benefit from the counter-archive.
We probably won't agree on what the high spots of 1977 were, but here's an early bid for 2023: Annie Ring and Lucy Bollington's new collection Citational Media: Counter-Archives and Technology in Contemporary Visual Culture, which will be published in the autumn.