- Atlas
- Mathematics
- Knot Theory
Knot Theory
A knot, to a mathematician, is a closed loop of string in three-dimensional space, with its two ends joined so it cannot come undone.
10 frameworks, from 1870 to today · Topology · Mathematics
- Knot tabulation 1870-Present
- Knot group 1910-Present
- Braid groups and closed braid representations 1923-Present
- Skein theory and polynomial invariants 1928-Present
- Seifert surface 1934-Present
- Physical and biological applications of knot theory 1985-Present
- Quantum invariants and quantum groups 1989-Present
- Finite type invariant 1990-Present
- Floer homology and gauge-theoretic invariants 1990-Present
- Categorification and Khovanov homology 1999-Present
Overview
A knot, to a mathematician, is a closed loop of string in three-dimensional space, with its two ends joined so it cannot come undone12. Two knots are considered the same if one can be deformed into the other without cutting the string or letting it pass through itself12. The central question is deceptively simple: given two tangled loops, how can you tell whether they are the same knot or genuinely different ones? A shoelace and a simple circle look different, but proving that no amount of clever manipulation will turn one into the other requires real mathematics. Knot theory is the study of these questions—how to distinguish knots, how to classify them, and what their structure reveals about the three-dimensional spaces they inhabit. It matters because knots turn out to connect to deep questions in topology, to the physics of quantum field theory, to the chemistry of DNA, and to the structure of matter itself.
Origins in Tabulation and the First Tables
Knot theory began not as a branch of mathematics but as a practical problem in the physical sciences. In the 1860s, the physicist William Thomson—later Lord Kelvin—proposed that atoms might be knotted vortices in the ether1. If different chemical elements corresponded to different knot types, then classifying knots would amount to classifying matter. This hypothesis, though wrong, motivated the first systematic study of knots. Kelvin's collaborator Peter Guthrie Tait undertook the task of tabulating knots by drawing them and trying to determine which were distinct. Working largely in the 1870s and 1880s, Tait produced tables of knots up to a certain number of crossings—the points where the curve passes over or under itself in a planar drawing. He was joined in this effort by the Reverend Thomas Kirkman and later by Charles Little, who extended the tables considerably.
The method was combinatorial: draw all possible knot diagrams with a given number of crossings, then eliminate those that could be simplified by a sequence of local moves—twisting, untwisting, or sliding a strand over another. These moves, later formalized by Kurt Reidemeister in 1927, are the three Reidemeister moves, and they remain the foundation of combinatorial knot theory. Tait's tables were remarkably accurate, though some errors were later found. The tabulation project established a key idea: knots could be studied through their diagrams, and the problem of classification was at least approachable by systematic enumeration.
The Fundamental Group and the Knot Group
The first genuinely powerful invariant came from algebraic topology. In the early twentieth century, topologists had developed the fundamental group—an algebraic object that captures the essential loops in a space3. In 1910, Max Dehn applied this to knots by considering the complement of a knot: the three-dimensional space that remains when you remove the knotted loop. The fundamental group of this complement, called the knot group, turned out to be a sensitive invariant14. If two knots have different knot groups, they are different knots.
Dehn's insight was profound: the knot itself is just a curve, but the space around it carries rich structure. The knot group encodes how loops in the complement can wind around the knot. For the unknot—the simple circle—the knot group is infinite cyclic, generated by a single loop that goes once around. For more complicated knots, the group becomes non-abelian and much harder to analyze. This approach dominated knot theory for decades. It was powerful but difficult: computing knot groups and distinguishing them required sophisticated group theory, and the groups themselves were often unwieldy.
Braids and the Algebra of Tangles
A parallel development came from Emil Artin's work on braids in the 1920s and 1930s. A braid is a collection of strands running from top to bottom, each crossing over or under others, with the ends fixed. Artin showed that braids form a group—the braid group—with generators and relations that can be written down explicitly. This gave knot theorists a new tool: any knot can be represented as a closed braid, obtained by connecting the top ends of a braid to the bottom ends. Alexander's theorem, proved by James Waddell Alexander in 1923, states that every knot or link can be represented this way5.
Braid theory provided an algebraic handle on knots. Instead of working directly with tangled diagrams, one could work with braid words—sequences of generators—and use the group structure to simplify and compare. This approach was particularly fruitful in the 1980s when Vaughan Jones discovered new invariants through braid representations, but its roots go back to Artin's foundational work.
Seifert Surfaces and Knot Genus
Another strand of development concerned surfaces. In 1934, Herbert Seifert proved that every knot bounds a surface—a two-dimensional sheet whose boundary is the knot itself. These Seifert surfaces provided a geometric way to study knots. The minimal genus of such a surface—the number of holes in the simplest surface the knot can bound—became an invariant called the knot genus6. The unknot bounds a disk (genus zero), while more complicated knots require surfaces with handles6.
Seifert's algorithm for constructing these surfaces from a knot diagram is still taught today. The genus invariant, though coarse, connects knot theory to the topology of surfaces and to questions about how knots sit inside three-dimensional space. It also foreshadowed later developments in which knots were studied through the surfaces they bound and the four-dimensional spaces those surfaces inhabit.
The Polynomial Revolution
The middle of the twentieth century saw knot theory become somewhat marginalized within topology, but a revolution came in 1984 when Vaughan Jones, working on operator algebras, discovered a new polynomial invariant. The Jones polynomial assigns to each knot a Laurent polynomial in a variable, and it distinguishes many knots that earlier invariants could not. Its discovery was unexpected and its consequences enormous.
The Jones polynomial was soon generalized. The HOMFLY polynomial (discovered independently by several mathematicians in 1985) and the Kauffman polynomial (1985) subsumed earlier invariants and provided new ones. These polynomials are defined through skein relations—local rules that relate the polynomial of a knot diagram to the polynomials of diagrams obtained by changing a crossing or smoothing it out17. This skein-theoretic approach made the invariants computable and revealed deep connections to statistical mechanics and quantum field theory1.
The classical polynomial invariants—the Alexander polynomial (1928), the Jones polynomial, and their generalizations—remain central tools. The Alexander polynomial, discovered by Alexander in 1928, was the first polynomial invariant and is still used today, though it does not distinguish all knots.
Quantum Invariants and the Physics Connection
The Jones polynomial turned out to have a physical interpretation. In the late 1980s, Edward Witten showed that the Jones polynomial arises naturally in Chern-Simons gauge theory, a quantum field theory in three dimensions. This connection between knot invariants and quantum physics was a revelation. It suggested that knots were not just topological curiosities but were deeply tied to the structure of quantum field theories.
This led to the development of quantum invariants—a family of invariants derived from quantum groups and representation theory. The Reshetikhin-Turaev invariants, developed in the early 1990s, generalized the Jones polynomial and connected knot theory to the representation theory of quantum groups. These invariants are often defined through skein theory but have deep algebraic and physical meaning.
Gauge Theory and Floer Homology
The gauge-theoretic perspective also led to new invariants through Floer homology. In the 1980s and 1990s, Andreas Floer developed a homology theory for three-manifolds based on gauge theory. This was adapted to knots by several mathematicians, leading to knot Floer homology—a categorification of the Alexander polynomial. Knot Floer homology, developed by Peter Ozsváth and Zoltán Szabó in the early 2000s, is a powerful invariant that detects the genus of a knot and has applications to questions about knot concordance and four-dimensional topology.
These gauge-theoretic and Floer-homological invariants are computationally intensive but theoretically deep. They connect knot theory to the topology of three- and four-dimensional manifolds and to questions about smooth structures.
Vassiliev Invariants and Finite-Type Theory
In the 1990s, Victor Vassiliev introduced a new approach: finite-type invariants, also called Vassiliev invariants. The idea is to study how knot invariants behave when a crossing is changed or when a knot is allowed to pass through itself (creating a singularity). By considering the space of all knots and studying invariants that vanish on knots with many singularities, Vassiliev defined a hierarchy of invariants. The Jones polynomial and its generalizations turn out to be of finite type, and this framework provided a unified way to understand many classical invariants.
Khovanov Homology and Categorification
The most recent major development is categorification. In 1999, Mikhail Khovanov discovered that the Jones polynomial could be "categorified"—replaced by a homology theory whose Euler characteristic is the Jones polynomial. Khovanov homology assigns to each knot a bigraded homology group, and the Jones polynomial is recovered as a graded Euler characteristic. This was a breakthrough: it showed that polynomial invariants were shadows of richer algebraic structures.
Khovanov homology has since been generalized and connected to other invariants. It has applications to knot concordance, to the study of smooth four-manifolds, and to questions in representation theory. The categorification program continues today, with researchers seeking to categorify other invariants and to understand the deeper structures they reveal.
Applications Beyond Pure Mathematics
Knot theory has found applications in the physical and biological sciences. In biology, DNA is a long molecule that can be knotted and linked; understanding DNA topology is essential for understanding how enzymes called topoisomerases untangle DNA during replication and transcription. In chemistry, knotted and linked molecules have been synthesized, and knot theory helps predict their properties. In physics, knot invariants appear in the study of quantum field theories, statistical mechanics, and even in proposals for topological quantum computation1.
The Present State
Today, knot theory is a vibrant field with multiple active approaches. Computational methods allow researchers to tabulate knots to high crossing numbers and to compute invariants efficiently. The classical polynomial invariants remain fundamental, but they are now understood through the lens of quantum groups, skein theory, and categorification. Floer homology and gauge theory provide deep connections to low-dimensional topology. Vassiliev invariants offer a unifying framework. And the applications to biology, chemistry, and physics continue to grow.
Open questions abound. The classification problem—determining when two knots are the same—remains unsolved in general, though invariants provide powerful tools. The relationship between different invariants is still not fully understood. The categorification program is ongoing. And the connections between knot theory and quantum field theory continue to yield new insights. Knot theory, which began as a Victorian curiosity about atoms, has become a central meeting point for topology, algebra, physics, and biology1.
References
- Knot theory (introduction, Knot equivalence, History, Knot invariants, Knot polynomials, Applications), Wikipedia
- Knot (mathematics), Wikipedia
- Fundamental group (introduction, History), Wikipedia
- Knot group, Wikipedia
- Alexander's theorem, Wikipedia
- Seifert surface (Genus of a knot), Wikipedia
- Skein relation, Wikipedia
The frameworks of Knot Theory
Each framework is a way of working the field has used: when it held, and the question it answers.
- 1870-Present
- 1910-Present
- 1923-Present
Braid groups and closed braid representations
How can braids represent and distinguish knots?
- 1928-Present
Skein theory and polynomial invariants
What polynomial rules distinguish knots?
- 1934-Present
- 1985-Present
Physical and biological applications of knot theory
How do knots arise in physics and biology?
- 1989-Present
Quantum invariants and quantum groups
How do quantum groups produce knot invariants?
- 1990-Present
- 1990-Present
Floer homology and gauge-theoretic invariants
What homologies from gauge theory distinguish knots?
- 1999-Present
Categorification and Khovanov homology
How can polynomial invariants be categorified?
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.
Quantum invariants and quantum groups grew out of Skein theory and polynomial invariants
Witten's 1989 Chern-Simons interpretation of the Jones polynomial and the subsequent Reshetikhin-Turaev construction generalized skein-theoretic invariants using quantum groups.
Carried by Edward Witten; Nikolai Reshetikhin; Vladimir Turaev.
Categorification and Khovanov homology grew out of Skein theory and polynomial invariants
Khovanov homology was constructed as a categorification of the Jones polynomial, replacing the polynomial with a bigraded homology whose Euler characteristic recovers it.
Carried by Mikhail Khovanov.
Knot group influenced by Knot tabulation
Dehn's 1910 construction of the knot group was motivated by the need to distinguish knots in Tait's tables, and the group was tested against tabulated examples.
Carried by Max Dehn.
Skein theory and polynomial invariants influenced by Knot group
Alexander derived his polynomial from the knot group by considering the infinite cyclic cover, so the first polynomial invariant grew directly out of the group-theoretic approach.
Carried by James Waddell Alexander.
Skein theory and polynomial invariants influenced by Braid groups and closed braid representations
Jones discovered his polynomial through representations of braid groups, using the trace on the Temperley-Lieb algebra that arises from braid representations.
Carried by Vaughan Jones.
Finite type invariant influenced by Skein theory and polynomial invariants
Vassiliev's finite-type invariants were constructed by extending polynomial invariants such as the Jones polynomial to singular knots, using skein relations.
Categorification and Khovanov homology influenced by Quantum invariants and quantum groups
Khovanov's construction was motivated by the quantum-group and physical interpretation of the Jones polynomial, and it categorifies the Reshetikhin-Turaev invariants.
Carried by Mikhail Khovanov.
Braid groups and closed braid representations influenced by Knot tabulation
Alexander's theorem that every knot is a closed braid was proved as a tool for the classification problem that tabulation had posed.
Carried by James Waddell Alexander.
Seifert surface influenced by Knot tabulation
Seifert's 1934 construction of spanning surfaces gave a geometric invariant, genus, that could be computed from the diagrams of tabulated knots.
Carried by Herbert Seifert.
Seifert surface influenced by Knot group
Seifert surfaces were used to study the knot group, since the complement of a Seifert surface gives a presentation of the group.
Carried by Herbert Seifert.
Floer homology and gauge-theoretic invariants influenced by Skein theory and polynomial invariants
Knot Floer homology was designed as a categorification of the Alexander polynomial, so the polynomial invariant provided the model and the target.
Carried by Peter Ozsváth; Zoltán Szabó.
Beyond the field
Where Knot Theory's ideas came from in other fields, and what they led to there: the strongest 12 of 29 links.
Knot group influenced by Fundamental Group and Covering Space Theory in Algebraic Topology
The knot group is an instance of the fundamental group of a complement, and its definition and computation depended on the fundamental group and covering space machinery developed by Poincaré and others.
Carried by Henri Poincaré; Max Dehn.
Knot group grew out of Knot theory in Topology
The knot group was defined by applying the fundamental group to the complement of a knot, so it grew directly out of the topological study of knots and their diagrams.
Carried by Max Dehn.
Braid groups and closed braid representations grew out of Knot theory in Topology
Braid groups and closed braid representations were introduced by Alexander as a way of representing knots and links, growing directly out of the classical study of knot diagrams.
Carried by James Waddell Alexander.
Skein theory and polynomial invariants influenced Quantum topology in Topology
Skein theory and polynomial invariants, from the Alexander polynomial onward, were the direct ancestors of the Jones polynomial and the quantum invariants that define quantum topology.
Carried by James Waddell Alexander.
Skein theory and polynomial invariants grew out of Knot theory in Topology
Skein theory and polynomial invariants grew directly out of knot theory: Alexander's polynomial was defined from skein relations on knot diagrams, and later polynomials were built the same way.
Carried by James Waddell Alexander.
Seifert surface grew out of Knot theory in Topology
Seifert surfaces were introduced by Seifert as a tool within knot theory, growing directly out of the problem of studying knots through their spanning surfaces.
Carried by Herbert Seifert.
Physical and biological applications of knot theory grew out of Knot theory in Topology
Physical and biological applications of knot theory grew directly out of knot theory itself, applying its invariants and classification to knotted polymers and DNA.
Quantum invariants and quantum groups influenced by Quantum group in Lie Theory
Drinfeld and Jimbo's quantum groups gave the R-matrices and ribbon Hopf algebra structure from which Reshetikhin and Turaev built their knot invariants in 1989, making the quantum-group formalism the engine of the whole framework.
Carried by Vladimir Drinfeld; Michio Jimbo; Nikolai Reshetikhin.
Floer homology and gauge-theoretic invariants grew out of Gauge-theoretic topology in Topology
Floer homology arose directly out of the gauge-theoretic topology of the early 1980s: Floer's 1988 construction used the Yang-Mills gradient flow and the Chern-Simons functional, building on Donaldson's instanton moduli techniques.
Carried by Andreas Floer; Simon Donaldson.
Floer homology and gauge-theoretic invariants grew out of Gauge Theory of Smooth Four-Manifolds in Geometric Topology
Floer homology was created to extend Donaldson's instanton gauge theory of smooth four-manifolds to three-manifolds, using the anti-self-dual connections and moduli spaces that Donaldson had developed.
Carried by Andreas Floer; Simon Donaldson.
Categorification and Khovanov homology grew out of Quantum topology in Topology
Khovanov's construction categorifies the Jones polynomial and the Reshetikhin-Turaev quantum invariants, so it grew directly out of the quantum topology programme.
Carried by Mikhail Khovanov.
Categorification and Khovanov homology grew out of Quantum topology in Low Dimensional Topology
Khovanov homology was constructed as a categorification of the Jones polynomial, the central invariant of the quantum topology programme, replacing the polynomial by a bigraded chain complex whose Euler characteristic recovers it.
Carried by Mikhail Khovanov.

Knot 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 Knot TheoryRelated fields
Fields its ideas are linked to
- Topology 9 links
- Low Dimensional Topology 4 links
- Abstract Algebra 3 links
- Algebraic Topology 3 links
- Geometric Topology 3 links
- Group Theory 2 links
- Lie Theory 1 link
- Molecular Biology 1 link
- Algebraic Combinatorics 1 link
- Geometric Group Theory 1 link
- Algebra 1 link