Every few years, millions of people cast ballots, and we treat the result as a verdict: the people have spoken. But what exactly have they said? Is it a preference, or a judgement about what is best? And can adding up individual votes reveal something none of us knows alone?

These questions feel modern, but the Marquis de Condorcet, a French philosopher and mathematician, posed them with mathematical precision in 1785. In his Essay on the Application of Analysis to the Probability of Majority Decisions, he set out to measure how reliable collective decisions could be. What he found was both reassuring and unsettling, and it still shapes how we think about democracy today.

The Jury Theorem: When Crowds Get It Right

Condorcet began with a simple setup. Imagine a jury choosing between two options, such as guilty or not guilty, where one answer is actually correct. Suppose each juror is only slightly more likely to be right than wrong, say 60 percent of the time. Suppose also that each juror makes up their own mind independently.

Probability then does something remarkable. With three such jurors, a majority gets it right about 65 percent of the time. With a hundred and one jurors, the majority is correct roughly 98 percent of the time. Individual errors tend to cancel each other out, while the small tilt toward truth adds up. This is the mathematical heart of what we now call the wisdom of crowds.

But the theorem cuts both ways. If each voter is slightly more likely to be wrong, a larger group makes the majority almost certainly wrong. Independence also matters. If everyone follows the same leader, newspaper or rumour, a thousand voters may be no wiser than one. Condorcet's result does not promise that crowds are wise. It states precisely the conditions under which they can be.

Takeaway

Groups are not wise by default. They become wise when each member is somewhat informed and thinks for themselves, because numbers amplify whatever tendency is already there.

The Voting Paradox: When Majorities Run in Circles

In the same essay, Condorcet noticed something stranger. When there are three or more options, comparing them two at a time can produce a loop. Imagine three voters. The first ranks A over B over C. The second ranks B over C over A. The third ranks C over A over B.

Now hold head-to-head contests. A beats B, because two of the three voters prefer it. B beats C by the same margin. Yet C also beats A. Each voter has perfectly consistent preferences, but the group as a whole does not. Whichever option wins, a majority would have preferred something else. The outcome then depends on which comparisons are made, and in what order. That gives real power to whoever sets the agenda.

Condorcet proposed that the fairest winner is the option that beats every rival head-to-head, now called a Condorcet winner. But his own example shows that such a winner does not always exist. In 1951 the economist Kenneth Arrow generalised the problem. He proved that no ranked voting system can satisfy a short list of reasonable fairness conditions all at once. The phrase 'the will of the majority' assumes there is always one clear will to find. Sometimes there is not.

Takeaway

Individuals can be rational while the group they form is not. When we speak of 'what the people want', we are often describing the result of a procedure rather than a fact waiting to be discovered.

Reason Meets Politics: Condorcet and the Revolution

For Condorcet, these findings were not mere curiosities. They belonged to a larger project he called social mathematics: the belief that the methods transforming physics and astronomy could also improve law, taxation and government. Like many Enlightenment thinkers, he trusted that careful reasoning, applied openly, could replace custom and arbitrary authority. He championed public education, the abolition of slavery and political rights for women, arguing that rights belong to all rational beings.

The French Revolution offered a chance to test these ideals. Condorcet was elected to the Legislative Assembly and later to the National Convention. In 1793 he presented a detailed draft constitution with carefully designed voting procedures, but it was set aside amid bitter factional struggle. When the radical Jacobins gained power, Condorcet criticised their hastily adopted constitution and was declared an enemy. He went into hiding, was arrested in 1794 and died in prison days later.

While in hiding, he wrote his Sketch for a Historical Picture of the Progress of the Human Mind, an optimistic account of humanity's capacity for improvement. The contrast is striking. Mathematics could clarify how good decisions might be reached, but it could not make people reason calmly, independently or in good faith. The conditions his theorems required were exactly the ones that revolutionary politics destroyed.

Takeaway

A well-designed procedure can improve collective decisions, but it relies on conditions no procedure can create on its own: informed citizens, independent judgement and a willingness to reason together.

Condorcet's work reveals a quiet assumption beneath modern democracy. We treat voting not only as a way to express preferences but as a method for discovering better answers. His mathematics shows that this hope is reasonable, but only under certain conditions.

Today's debates about misinformation, polarisation and electoral reform keep returning to his questions. Are voters informed? Are they thinking independently? Does the voting method itself shape the result? Condorcet shows that democracy's reliability is not automatic. It has to be built and protected.