Blaise Pascal · Mathematics
How a gambler’s puzzle - the ‘problem of points’ - drew Pascal and Fermat into the 1654 correspondence that founded probability theory, and gave the world its first rigorous grip on chance through the idea of mathematical <em>expectation</em>.
In 1654 a gambler and man of letters, the Chevalier de Méré, brought Pascal an old puzzle that had defeated mathematicians for two centuries: the problem of points. Two players stake equal sums on a game of several rounds; whoever first reaches a set number of points takes the whole pot. But the game is broken off early, before either has won. How should the stakes be divided fairly? Earlier writers - Pacioli, Tartaglia, Cardano - had guessed, splitting the pot by the rounds already won, and each knew his answer could not be justified. Pascal saw why they had failed: they were all looking at the past, at points already scored, when the honest question is about the future. From where the game now stands, what are each player’s chances of going on to win? To divide the stakes rightly, you must put a number on an uncertain future. Over the summer of 1654, in a now-famous exchange of letters with Fermat, Pascal did exactly that - and a new branch of mathematics was born.
How do you compute an expectation when the future can branch many ways? Pascal’s method was to reason backward from the end. Take the case he and Fermat worked through. Two players have each staked 32 pistoles, so 64 are on the table, and the winner is the first to three rounds. Suppose the game is stopped when the first player has 2 rounds and the second has 1. Look only at the very next round. If the first player wins it, he reaches 3 and takes all 64; if he loses it, the players are tied 2 to 2, and by symmetry each then deserves 32. So the first player is certain of 32 whatever happens, while the remaining 32 hang on a single even round. Since that round is a fair coin, the disputed 32 should be split in half - giving the first player 32 + 16 = 48 pistoles and the second 16. By dissolving each uncertain step into a guaranteed part and a fair gamble, Pascal could roll the expectation back, round by round, from the end of the game to any position at all.
What made this revolutionary was not the gambling but the method. For the first time, uncertainty itself was handled with mathematical rigour. Pascal announced the achievement in a proud memoir to the learned academy of Paris: by uniting the exactness of geometry with the unpredictability of chance, a wholly new science had appeared, one that could give demonstrations as certain as any in mathematics about events that are, in themselves, uncertain. He called it the geometry of chance. The paradox runs deep. You cannot know whether one particular coin will land heads, yet you can prove, with total certainty, the long-run value of a bet on it. Chance is lawless in the single case and law-abiding in the multitude. That double truth, glimpsed by Pascal and Fermat in a few weeks of letters, is the seed of all modern probability, statistics, insurance, physics, and risk - every field in which we reason precisely about what we cannot predict.
This is the opening of the lesson. The rest — the dialogue, the primary source, and the recall — is in the app.
You learned that the ‘problem of points’ asks how to split the pot of an unfinished game, and that Pascal and Fermat answered it in 1654 by reasoning about each player’s future chances, inventing the concept of mathematical expectation . Explain the puzzle and their solution in your own words.
Leads to Pierre de Fermat.
Begin this lesson →epoché — a humanities education that remembers you.