Birch and Swinnerton-Dyer Conjecture

The Birch and Swinnerton-Dyer conjecture (BSD) is one of the deepest conjectures in number theory. It predicts that a subtle question about integer solutions to cubic equations — something ancient — is controlled by a modern analytic object called an L-function.

Elliptic curves

An elliptic curve over the rational numbers is a curve defined by an equation of the form:

$y^2 = x^3 + ax + b$

with $a, b \in \mathbb{Q}$, subject to a mild non-singularity condition. The rational points on such a curve — pairs $(x, y)$ with rational coordinates that satisfy the equation — form a group under a geometric addition law. This group, called the Mordell-Weil group, has the form:

$E(\mathbb{Q}) = \mathbb{Z}^r \oplus T$

where $T$ is a finite torsion part and $r$ is a non-negative integer called the rank. The rank measures how many “independent” infinite-order rational points the curve has.

The L-function

Each elliptic curve also has an associated L-function $L(E, s)$, a complex-analytic object resembling the Riemann zeta function. It encodes, for each prime $p$, how many points the curve has modulo $p$.

The conjecture

The conjecture says: the rank of the elliptic curve equals the order of vanishing of its L-function at $s = 1$.

In particular:

• If $L(E, 1) \neq 0$, the curve has finitely many rational points (rank 0).
• If $L(E, 1) = 0$, the curve has infinitely many rational points.

There is also a sharper form of the conjecture that predicts the leading coefficient of the L-function’s Taylor expansion in terms of arithmetic invariants (the regulator, Tate-Shafarevich group, Tamagawa numbers, and more).

What is known

• The conjecture holds in many specific cases, especially for rank 0 and rank 1.
• The Gross-Zagier theorem and work of Kolyvagin gave partial results for curves with $L(E, 1) = 0$ but $L'(E, 1) \neq 0$.
• No general proof for higher ranks.

Why it matters

BSD is a prime example of the Langlands program in miniature — a conjectured deep bridge between arithmetic (integer solutions) and analysis (complex L-functions). A proof would validate a strand of research that shapes much of modern number theory.

It would also resolve, conditionally, the question of determining the rank of any given elliptic curve — something that is surprisingly hard to do directly. For cryptography (elliptic curves underpin many modern security protocols), BSD-type results shed light on the structure of the underlying mathematical objects.

BSD is a Millennium Prize Problem with a one-million-dollar reward.

Frequently asked

What is an elliptic curve?

A cubic equation of the form y² = x³ + ax + b with some smoothness conditions. Despite the name, they are not ellipses. They arose historically in computing the arc length of ellipses (hence the name) and turned out to have remarkable structure.

Why do mathematicians care so much about rational points?

Rational solutions to polynomial equations — Diophantine problems — are among the oldest and deepest questions in mathematics. Elliptic curves are the simplest case where the structure of rational solutions is non-trivial and the general theory is rich.

Continue reading

Blog

Stirling's Approximation: How Big Is n Factorial, Really?

The factorial n! grows faster than any polynomial but slower than any exponential. A single formula from 1733 pins down its size precisely — and inexplicably involves both π and e.

Read →Blog

The Collatz Conjecture: A Simple Rule, an Unsolvable Mystery

Take any whole number. If it is even, halve it; if it is odd, triple it and add one. Repeat. The conjecture says you always reach 1 — and ninety years on, nobody can prove it.

Read →Blog

Catalan Numbers: One Sequence That Counts Almost Everything

A short list of integers — 1, 1, 2, 5, 14, 42, 132 — turns out to count balanced parentheses, mountain paths, triangulated polygons, binary trees, and dozens of other things all at once.

Read →