Simon Singh on Science

Went to quite a good talk this lunchtime by Imperial alumnus Simon Singh, author of Fermat's Last Theorem and Big Bang. He spoke about how filmmakers communicate science to a mass audience, using his BAFTA award-winning Horizon programme on Fermat as an example. Its introduction - a voiceover by Andrew Wiles - mentions math just once, and is really a parable of a man's working life. And there's drama in those first two minutes: having had the film crew sit with him every day for a week, Wiles was sufficiently relaxed in their presence to give a rather emotional statement about the moment he realised he'd never solve a problem as great again!

A big issue is the precision of the narrative versus its accuracy. Part of the film featured an interviewee speaking about the futility of using computers to check a formula when there's an infinity of numbers to check. Andrew Wiles' proof of Fermat is actually about prime numbers, of which there are infinitely many, so the interviewee mentioned "prime numbers" repeatedly. Too precise, according to Singh; you lose no accuracy in the principle of checking an infinitude by simply mentioning "numbers". And, with editing, that's what the film did.

One of Simon's most interesting points was the need for a cast of characters throughout a film, whom viewers can easily differentiate. An interviewee who can comment on all the other interviewees is helpful too; the viewer sees the story through their eyes, which are more personal than the narrator's.

Afterwards, Simon told me that he'd only decided to become a filmmaker in the last six months of his PhD, when he was applying for post-doc positions. His biography reads as if the academics at Cambridge and CERN make you feel that you're useless to your subject if you're not as good as them. If they indeed have this attitude, then I think they're flat out wrong. Luckily, Simon - having been teaching and writing for some years - found his feet at the BBC and has never looked back.