Algebraic Number Theory

Algebraic number theory is the branch of number theory that studies whole numbers by enlarging the world they live in.

10 frameworks, from 1870 to today · Number Theory · Mathematics

still a working view ended · each leads to its entry below

Overview

Algebraic number theory is the branch of number theory that studies whole numbers by enlarging the world they live in1. Instead of asking only about ordinary integers, it asks the same kinds of questions—divisibility, factorization, solutions of equations—inside larger systems of "algebraic numbers," such as the numbers of the form \(a + b\sqrt{-5}\) with \(a, b\) integers1. The reason to do this is that many questions about ordinary integers become tractable once they are recast in these larger systems, and many new questions appear that are interesting in their own right. The field's central drama is that the most basic property of the integers—that every number factors uniquely into primes—can fail in these larger systems, and the whole subject grew up around understanding, repairing, and exploiting that failure.

Unique factorization fails: Kummer and ideal numbers

The founding problem was Fermat's Last Theorem, the claim that \(x^n + y^n = z^n\) has no positive integer solutions for \(n > 2\)1. In the 1840s Ernst Kummer attacked it by working in the ring of numbers \(a0 + a1\zeta + \cdots + a_{p-1}\zeta^{p-1}\), where \(\zeta\) is a complex \(p\)-th root of unity234. He found that unique factorization into primes, which holds in the ordinary integers, can break down in these rings24. In the ring of numbers \(a + b\sqrt{-5}\), for instance, the number 6 factors both as \(2 \cdot 3\) and as \((1+\sqrt{-5})(1-\sqrt{-5})\), and none of these four factors can be broken down further.

Kummer's response was to invent "ideal numbers": fictitious divisors that restore unique factorization at the cost of not being actual numbers15. He used them to prove Fermat's Last Theorem for all prime exponents \(p\) that are "regular," a condition on the class number—a count of how badly factorization fails. This was the first major success of the algebraic approach, and it made the failure of unique factorization the central object of study rather than an obstacle.

Dedekind's ideals and the ring of integers

Richard Dedekind, in the 1870s, gave Kummer's ideal numbers a precise and general form125. For any algebraic number field—a finite extension of the rational numbers—he defined the ring of integers: the elements satisfying a monic polynomial equation with integer coefficients6. In \(\mathbb{Q}(\sqrt{-5})\) these are exactly the numbers \(a + b\sqrt{-5}\). Dedekind showed that although elements may not factor uniquely, the ideals of this ring do: every nonzero ideal factors uniquely as a product of prime ideals12. An ideal here is a set of ring elements closed under addition and under multiplication by any ring element; the prime ideals play the role of primes5.

This is the theorem of unique factorization of ideals, and it turned algebraic number theory into a coherent discipline. The class group measures the gap between ideals and actual elements: it is trivial exactly when the ring of integers has unique factorization12. Dedekind also isolated the abstract structure now called a Dedekind domain—a ring in which this ideal factorization theorem holds—and the associated notions of fractional ideals, the norm of an ideal, and the discriminant, which encodes which primes misbehave. These remain the vocabulary of the subject.

Geometry of numbers and the unit theorem

Two finiteness results completed the classical picture. In the 1840s, Gustav Lejeune Dirichlet proved the unit theorem: the invertible elements of a ring of integers form a finitely generated group, so that, for example, the units of a real quadratic field like \(\mathbb{Q}(\sqrt{2})\) are generated by \(-1\) and \(1+\sqrt{2}\). In 1889 Hermann Minkowski introduced the geometry of numbers, treating a lattice of points in space and proving that a sufficiently large symmetric convex region must contain a nonzero lattice point7. Applied to number fields, this yielded the finiteness of the class number and bounds on the discriminant, and it gave a geometric way to think about arithmetic—one that remains a standard tool.

Class field theory: describing abelian extensions

The next great project was to describe all abelian extensions of a number field—extensions whose Galois group is commutative8. The Galois group of an extension is the group of symmetries of the larger field that fix the smaller one; "abelian" means these symmetries commute. Class field theory, developed from the 1890s through the 1930s by David Hilbert, Teiji Takagi, Emil Artin, and Helmut Hasse, gave a complete answer: abelian extensions correspond to certain subgroups of the ideal class group, and the way primes split in an extension is governed by a reciprocity law. Artin's reciprocity law (1927) made this explicit by associating to each prime an element of the Galois group.

The theory's power is visible in a concrete case: the splitting of an ordinary prime \(p\) in the field \(\mathbb{Q}(\sqrt{-5})\) is determined by whether \(p\) is a square modulo 20. Class field theory explains why such congruence conditions always govern abelian extensions, and it unified a scattered collection of reciprocity laws into one framework.

Local methods and the local-global principle

Hasse, in the 1920s and 1930s, reorganized the subject around local fields. For each prime \(p\), one can complete the rationals to the \(p\)-adic numbers, in which closeness is measured by divisibility by powers of \(p\); this is the analogue of passing from the rationals to the reals by completing at the "prime at infinity."9 Hasse's local-global principle holds that many questions can be settled by answering them at every prime simultaneously10. The Hasse–Minkowski theorem, for instance, characterizes when a quadratic form has a rational solution by checking it over the reals and over every \(p\)-adic field11.

This reframing made class field theory cleaner and set the pattern for later work: study a global object through its localizations. The local fields themselves became objects of study, with their own ramification theory and their own class field theory.

Cohomology and the modern reformulation

In the 1950s and 1960s, Claude Chevalley, Jean-Pierre Serre, and others recast class field theory in the language of Galois cohomology—the cohomology of Galois groups acting on modules. Cohomology assigns algebraic invariants to a group action, and in this setting it turned the reciprocity laws into statements about cohomology groups, making the theory systematic and extendable. The same machinery became essential for the arithmetic of elliptic curves and for the study of Galois representations.

Iwasawa theory and p-adic L-functions

Kenkichi Iwasawa, from the late 1950s, studied towers of number fields whose Galois groups are the \(p\)-adic integers, and discovered that the sizes of class groups in such towers grow in a regular way governed by a power series. Iwasawa theory, as the resulting subject is called, connects class groups to \(p\)-adic L-functions—\(p\)-adic analogues of the analytic L-functions that encode arithmetic data12. It became one of the main bridges between algebraic and analytic number theory, and it remains a central tool in the arithmetic of elliptic curves and in the study of special values of L-functions.

Galois representations and modularity

A Galois representation is a way of letting the Galois group act on a vector space, so that arithmetic symmetries become matrices. From the 1960s onward, Serre, Pierre Deligne, and others developed the theory of such representations and their connection to modular forms—highly symmetric analytic functions whose coefficients carry arithmetic information. The high point was the modularity theorem, conjectured by Yutaka Taniyama and Goro Shimura in the 1950s and proved for semistable elliptic curves by Andrew Wiles with Richard Taylor in 1994–1995, completing the proof of Fermat's Last Theorem113. The theorem says every elliptic curve over the rationals is modular, meaning its arithmetic is encoded in a modular form113. This fused algebraic number theory with arithmetic geometry and made Galois representations a central language of the field.

The Langlands program

Robert Langlands, in a 1967 letter to André Weil, proposed a vast generalization of class field theory: a correspondence between Galois representations and automorphic representations, of which modular forms are the simplest case. The Langlands program predicts that L-functions attached to Galois representations equal L-functions attached to automorphic forms, unifying reciprocity laws, modularity, and much of the analytic theory. Parts have been proved—the local Langlands correspondence for general linear groups by Michael Harris and Richard Taylor (2001), and many instances of global functoriality—while the general conjecture remains open. It is the organizing framework of much current research.

Computational and statistical turns

From the 1960s, computers made it possible to compute rings of integers, class groups, unit groups, and Galois groups explicitly. The Buchmann–Williams and later algorithms, and systems such as PARI/GP, SageMath, and Magma, turned class group computation into a routine tool and made the subject experimental as well as theoretical. The Cohen–Lenstra heuristics (1983) predicted the statistical distribution of class groups across families of number fields, and a large body of work—including results of Davenport–Heilbronn on class numbers of quadratic fields and later work on the distribution of discriminants—established arithmetic statistics as a field in its own right, asking how often arithmetic objects of a given kind occur.

The field today

Contemporary algebraic number theory runs along several connected fronts. The Langlands program and its arithmetic applications remain the largest organizing project, with the modularity of elliptic curves over real quadratic fields proved by Freitas, Le Hung, and Siksek (2015) and ongoing work on higher-dimensional Galois representations. Iwasawa theory and p-adic methods are central to the arithmetic of elliptic curves and to the Birch–Swinnerton-Dyer conjecture. Computational methods have become indispensable, both for testing conjectures and for cryptographic applications14. Arithmetic statistics continues to grow, with heuristics now extended to Selmer groups, number field counts, and the distribution of primes. The questions that opened the subject—how badly unique factorization fails, and what that failure tells us about equations in ordinary integers—are still being answered, now with tools Kummer could not have imagined.

References

  1. Algebraic number theory (introduction, Fermat, Dedekind, Factorization into prime ideals, Failure of unique factorization, Modern theory), Wikipedia
  2. Ideal class group (History and origin of the ideal class group, introduction), Wikipedia
  3. Kummer theory, Wikipedia
  4. Cyclotomic field, Wikipedia
  5. Ideal (ring theory) (History, introduction), Wikipedia
  6. Ring of integers, Wikipedia
  7. Minkowski's theorem, Wikipedia
  8. Class field theory, Wikipedia
  9. P-adic number, Wikipedia
  10. Hasse principle, Wikipedia
  11. Hasse–Minkowski theorem, Wikipedia
  12. Main conjecture of Iwasawa theory, Wikipedia
  13. Modularity theorem (introduction, History), Wikipedia
  14. Computational number theory, Wikipedia

The frameworks of Algebraic Number Theory

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1870-Present

    Ideal Theory and Dedekind Domains

    How do ideals restore unique factorization in number rings?

    Fly to its tower in Nousfarer

  2. 1889-Present

    Geometry of Numbers

    When must a lattice contain a nonzero point?

    Fly to its tower in Nousfarer

  3. 1890-Present

    Class Field Theory

    Which abelian extensions does a number field have?

    Fly to its tower in Nousfarer

  4. 1920-Present
  5. 1950-Present

    Galois Cohomology

    What do cohomology groups say about Galois actions?

    Fly to its tower in Nousfarer

  6. 1957-Present

    Iwasawa theory

    How do class groups grow in p-adic towers?

    Fly to its tower in Nousfarer

  7. 1960-Present

    Computational Algebraic Number Theory

    How can a computer compute rings and class groups?

    Fly to its tower in Nousfarer

  8. 1960-Present
  9. 1967-Present

    Langlands Program

    Which Galois representations come from automorphic forms?

    Fly to its tower in Nousfarer

  10. 1983-Present

    Arithmetic Statistics

    How are number-theoretic invariants distributed on average?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Ideal Theory and Dedekind DomainsGeometry of NumbersClass Field TheoryLocal Methods and the Local-Global Pr…Galois CohomologyIwasawa theoryComputational Algebraic Number TheoryGalois Representations and ModularityLanglands ProgramArithmetic Statistics

Beyond the field

Where Algebraic Number Theory's ideas came from in other fields, and what they led to there: the strongest 12 of 68 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Algebraic Number Theory is a planet in Nousfarer

Its city has a tower for each of these 10 frameworks. Land, walk in, and climb one concept at a time.

Land on Algebraic Number Theory

Related fields

Fields its ideas are linked to

More in Number Theory

See where Algebraic Number Theory sits on the Map of Human Knowledge