50 years, 100+ mathematicians, 10,000+ pages — the enormous theorem
The classification of finite simple groups did not happen by accident — it was the result of a deliberate, coordinated program laid out by Daniel Gorenstein in the early 1970s. Gorenstein recognized that the scattered results of the 1950s and 1960s could be organized into a systematic attack on the problem. In a series of lectures and papers, he outlined a 16-step program that, if completed, would yield the full classification. This program gave the community a roadmap and the confidence that the problem was, in principle, solvable.
Gorenstein's program was built on a key structural insight: to classify a finite simple group G, study the centralizers of involutions — elements of order 2. An involution in G is any element g with g² = e (and g ≠ e), and its centralizer C_G(g) is the set of all elements that commute with g. The Brauer-Fowler theorem (1955) had shown that for any finite group H, there are only finitely many simple groups containing an involution whose centralizer is isomorphic to H. This meant that if you could determine all possible centralizer structures, you could enumerate all simple groups.
The program divided naturally into two halves based on a group's 2-rank (the maximum rank of its elementary abelian 2-subgroups). Groups of low 2-rank required delicate special arguments, while groups of high 2-rank were amenable to more systematic techniques. Each half required different tools and different teams of mathematicians, but Gorenstein's vision held them together into a coherent whole. By the late 1970s, dozens of mathematicians across multiple countries were working on different pieces of the program simultaneously.
The first major breakthrough was the Feit-Thompson odd order theorem(1963), which proved that every finite group of odd order is solvable — equivalently, every non-abelian finite simple group has even order. This 255-page paper in the Pacific Journal of Mathematics was, at the time, the longest proof of a single theorem ever published. Its consequence was profound: since every non-abelian simple group has even order, it contains involutions, and the involution-centralizer strategy becomes universally applicable.
Two powerful technical tools drove the program forward. Signalizer functor theory, developed by Gorenstein and John Walter, provided a systematic way to construct normal subgroups from local information (properties of p-subgroups and their normalizers). When a signalizer functor is "complete," it produces a proper normal subgroup of G — contradicting simplicity and thereby ruling out entire classes of groups. The N-group theorem of John Thompson classified all minimal simple groups (simple groups in which every proper subgroup is solvable), handling the base case of many inductive arguments.
Other essential tools included the theory of components and layers (developed by Gorenstein, Walter, and Helmut Bender), which analyzed how simple sections interact inside a group; pushing-up and amalgam methods, which extracted geometric structure from group-theoretic data; and the character-theoretic methods of Brauer, which constrained possible group orders through modular representation theory. Each technique was a sophisticated piece of mathematics in its own right, and the proof required all of them working in concert.
The interactive timeline below traces the major milestones of the classification effort, from Galois's discovery of A₅ in the 1830s through the completion of the proof in the early 2000s. Explore the key theorems, the discovery of each sporadic group, and the contributions of the major figures.
Key insight: The most intense period of the classification was 1963-1983, spanning just 20 years. The Feit-Thompson theorem (1963) opened the door, and a premature announcement of completion came in 1983. The actual completion, including the quasithin case, was not achieved until 2004 — a 21-year gap that few outside the field noticed.
In 1983, Daniel Gorenstein announced that the classification was complete. The mathematical community celebrated one of the greatest achievements in the history of mathematics. But there was a problem: the proof of the quasithin case — groups in which every 2-local subgroup has p-rank at most 2 — had never been fully written down. Geoffrey Mason had been assigned this case and had circulated a manuscript, but it was never published and contained significant gaps.
For over a decade, the classification existed in a state of ambiguity: it was widely believed to be true, and its results were freely used throughout mathematics, but a critical piece of the proof was missing. This situation was finally resolved by Michael Aschbacher and Stephen Smith, who published a complete treatment of the quasithin case in 2004 — two volumes totaling over 1,200 pages. With this work, the classification was genuinely complete for the first time, more than 20 years after the initial announcement.
The quasithin gap highlights a broader issue with the classification: its proof is so enormous — often cited as spanning over 10,000 pages across hundreds of journal articles by more than 100 authors — that no single person can verify the entire argument. This is not merely a matter of length; it is a matter of organizational complexity. The proof is distributed across papers written over three decades, using different notations and conventions, with dependencies that are difficult to trace. The mathematical community's confidence in the theorem rests on a combination of the individual verifications and the structural coherence of Gorenstein's program.
Recognizing the fragility of a proof spread across thousands of pages by hundreds of authors, Gorenstein, Richard Lyons, and Ronald Solomon began a second-generation proof in the 1980s — a unified, self-contained account intended to be published as a single coherent series. After Gorenstein's death in 1992, Lyons and Solomon continued the project. As of 2024, they have published nine of a planned twelve volumes (totaling roughly 5,000 pages so far), with the remaining volumes in preparation.
The second-generation proof aims to reduce the total length to approximately 5,000-7,000 pages — still enormous, but more manageable than the original and organized as a single logical argument rather than a patchwork of independent papers. It incorporates improved techniques discovered since the original proof, simplifies many arguments, and provides a unified notation and framework. The project is one of the most ambitious long-term undertakings in the history of mathematics.
The question of what "10,000 pages" really means deserves clarification. The original proof is not a single 10,000-page document — it is an interconnected web of several hundred journal articles, each proving a piece of the puzzle. Many of these papers serve double duty, establishing results used in the classification while also advancing general group theory. The 10,000-page figure is an estimate of the total length of papers that are specifically needed for the classification, but the boundaries are fuzzy. The second-generation proof project is, among other things, an attempt to draw a clear boundary around exactly what constitutes "the proof."