Novels, Lookup Tables, and Rice Pudding
I grew up surrounded by stories. When I was six years old, my mother read to me in the bath: first the Hobbit, then the entire Lord of The Rings trilogy. I used to annoy her by banging on the side of the bath to simulate the Uruk-Hai war drums. I was lucky enough to inherit the books that my parents inherited from their parents, old stories and new: Rosemary Sutcliffe, Rudyard Kipling, Ladybird books, box after box, shelf after shelf.
One of my precious Christening presents was a single-volume illustrated hardback copy of The Chronicles of Narnia; not just The Lion, The Witch, and the Wardrobe, but the baffling Magician’s Nephew (what on earth was Colney Hatch?), the exhilarating but unsettling Dawn Treader, and the frankly terrifying Silver Chair. And so I feel a bit of an affinity to C.S. Lewis.
I’ve written about Lewis before, but this is a great quote:
"Those of us who have been true readers all our life seldom fully realize the enormous extension of our being which we owe to authors. We realize it best when we talk with an unliterary friend. He may be full of goodness and good sense but he inhabits a tiny world. In it, we should be suffocated."
Someone else said that “you must learn from the mistakes of others—you’ll never live long enough to make them all yourself.“ - and I think that’s basically the point Lewis is making here. The range of possible human experience is so big, that relying only on first-party data is severely limiting. Books make our world bigger. We have an immense literary inheritance, a real-life Bene Gesserit Other Memory.
That idea provides context for this tweet:
Given enough data and compute, you could, perhaps derive everything yourself; Deep Thought managed to do it without data!
And to this end they built themselves a stupendous super computer which was so amazingly intelligent that even before its data banks had been connected up it had started from I think therefore I am and got as far as deducing the existence of rice pudding and income tax before anyone managed to turn it off.
But those deductions are jolly expensive, and we all operate under constraints. Few-shot learning is much easier than zero-shot. And when a good computer scientist comes across expensive computations, she caches them and performs lookups when she needs them. Our literary inheritance is a lookup table. War and Peace contains data that you will never get to experience first-hand, and the Tolstoy’s real genius, the stored results of his computations on that data.
When you read a great novel, or a work of history, you’re absorbing the stories about human behaviour; how people work, how they react in different situations, how they feel. At this point, I want to start naming some stories, just to illustrate the point - and OK I will! King Alfred and his cakes; Lady Macbeth and her ambition; Sonya and her sacrifice; Grushenka and her heartbreak; Oedipus and his prophecy; Heracles and his crime; these stories are our treasures, carefully-generated lookup tables that seventy generations have passed down to us. All those mad, lonely authors, all those brave men and women, they’ve given us this gift of culture, and now we don’t have to make their mistakes again. It’s all there, if you care to look, fiery with its beauty, the incredible detail of human experience.
It’s reminiscent of Amia Srinivasan’s interaction with Derek Parfit: having been elected to the All Souls Prize Fellowship, a seven-year, fully-salaried, no-strings-attached opportunity to research anything you want in the greatest research institution in the world, she asked one of the most eminent philosophers in the world how she ought to spend her time.
Parfit – who had been a fellow there since 1967 – was appointed as my college adviser... he wanted to talk about what I intended to do with the seven years of my fellowship. He suggested I spend the first year reading novels, ‘sowing seeds’.
Novels are to an All Souls philosopher as zone 2 cardio is to the Olympic rower.
There’s an assumption being made: that we can learn something about ourselves and the people around us by reading nineteenth-century Russian novels. Is human nature the same in all times and all places? I probably lean towards ‘yes’; for an articulation of that view, take this passage (which I’ve quoted here before):
Anyone wishing to see what is to be must consider what has been: all the things of this world in every era have their counterparts in ancient times. This occurs since these actions are carried out by men who have and have always had the same passions, which, of necessity, must give rise to the same results.
Machiavelli, Discorsi III xliii
Humans are really hard to model, and ideas are really hard to develop; in practice, you’ll never figure it all out from first principles. The lesson to take from Ramanujan is that trying to derive mathematics from a single textbook in an isolated village sucks - move to Cambridge, and work with Hardy! Throw yourself into the lookup tables, and perform computations on their results. In perhaps my favourite history book of all time, J.N. Figgis wrote:
The sonorous phrases of the Declaration of Independence or the Rights of Man are not an original discovery, they are the heirs of all the ages, the depositary of the emotions and the thoughts of seventy generations of culture. 'Forty centuries look down upon you' was a truer picture of the mind of the Revolution than the military rhetorician knew or cared to know.
J.N. Figgis, Studies of Political Thought: From Gerson to Grotius, 1414-1625