Weil conjectures

Weil conjectures

The Weil conjectures are four statements from number theory that André Weil put forward in 1949, establishing a deep connection between geometry and the counting of solutions to certain equations. For decades they were considered one of the hardest open problems in mathematics and were only fully proven in 1974.

The Weil conjectures are four mathematical claims put forward by the French-American mathematician André Weil in 1949. They describe how many solutions a certain type of equation has over finite fields — number systems in which only finitely many numbers exist, for example the remainders when dividing by a prime number. Weil recognized that these numbers of solutions could be described using methods that actually originate from geometry. At the time, this was a completely surprising idea. The four conjectures were proven one by one over several decades, with the last and most difficult one finally proven in 1974 by Pierre Deligne.

Why the conjectures changed mathematics

Before Weil, number theory and geometry were regarded as largely separate fields. One deals with whole numbers and equations, the other with shapes and spaces. Weil claimed that the numbers of solutions to equations behave as though they were governed by a geometric object — a so-called algebraic variety. This sounds abstract, but it had a concrete consequence: if you understand the geometry, you also understand the numbers.

This idea proved so fruitful that it sparked an entire new research program. Alexander Grothendieck, one of the most influential mathematicians of the 20th century, built an entirely new mathematical theory in the 1960s — so-called étale cohomology — specifically to make the Weil conjectures provable. This effort shows just how central the conjectures were to mathematics.

What the four statements actually claim

The first conjecture states that the number of solutions of an equation can be expressed through a compact, closed formula — a so-called zeta function. Such functions are known from analysis as a tool for elegantly summarizing infinite sums. The second conjecture describes a symmetry property of this function, reminiscent of the symmetry of mirror images.

The third conjecture connects the function to geometric characteristics of the associated object, the so-called Betti numbers, which count how many “holes” a geometric shape has. The fourth and most difficult conjecture — often referred to as the “Riemann hypothesis for finite fields” — makes a precise statement about where the zeros of the zeta function lie. Zeros are the points where a function takes the value zero; their exact location determines how regularly the solutions are distributed.

Deligne proved this fourth conjecture in 1974 and was awarded the Fields Medal for it in 1978, the highest honor in mathematics. His proof used Grothendieck’s new theory, but at one crucial point took its own, unexpected path.

Where the Weil conjectures appear today

In news about mathematics, the Weil conjectures are mostly encountered as background when major open problems are discussed. The famous Riemann hypothesis — one of the seven Millennium Problems, each carrying a million-dollar prize — is closely related to the Weil conjectures. Deligne essentially proved a finite version of it; the original hypothesis, however, remains unsolved to this day.

In modern cryptography, the science of securely encrypting data, finite fields and equations over them play a central role. Elliptic curves over finite fields — precisely the type of object the Weil conjectures concern — form the foundation of many encryption methods used today in online banking or in encrypting messages. The Weil conjectures provide the theoretical foundation for why these curves can be so well understood.

Finally, the Langlands program, a current major project in mathematics aiming to unify various subfields, stands directly in the legacy of the Weil conjectures. Many mathematicians regard the Weil conjectures as the first major example showing that number theory and geometry essentially speak the same language — just in different dialects.

Subscribe free. Unsubscribe the second it sucks.

High-signal news across AI, business, UX, and tech. Every morning.