A knot is a closed curve in space that cannot be untangled without cutting, often studied in topology and physics.Knot theory classifies embeddings of a circle into three-dimensional space up to isotopy (continuous deformation without self-intersections). The key task is to find invariants that distinguish non-equivalent knots. Among them are the Alexander polynomial (1923) and the Jones polynomial (1984), which brought the author the Fields Medal. Knots naturally arise in physics: as stable field configurations (e.g., magnetic field lines or cosmic strings), and in (2+1)-dimensional systems — as world lines of anyons.
The birth of knot theory is associated with the hypothesis of William Thomson (Lord Kelvin) in 1867, who proposed that chemical elements are knots in the ether. Peter Guthrie Tait began compiling tables of knots, classifying all knots up to 7 crossings. With the discovery of the electron and the abandonment of the ether, knot theory receded into the background, but at the beginning of the 20th century, through the efforts of Henri Poincaré and other topologists, it received a rigorous mathematical foundation. In the second half of the 20th century, a renaissance occurred thanks to the discovery of the Jones polynomial (Vaughan Jones, 1984) and connections with quantum physics, the works of Edward Witten and others.
To classify knots, invariants are used — quantities that do not change under isotopies. The simplest is the number of crossings in a knot diagram. More complex ones include the Alexander polynomial, the Jones polynomial (related to braid group representations and quantum invariants), Vassiliev invariants. The Jones polynomial is computed via skein relations: it is recursively determined through the relationship of knot diagrams that differ only in a small neighborhood of a crossing. In physics, topological knots arise as stable solutions in field theories; for example, in Maxwell's equations there exist electromagnetic knots whose stability is guaranteed by the Hopf topological invariant. World lines of particles in (2+1)-dimensional systems can form knots, leading to fractional statistics.
Knots help understand how DNA and proteins are twisted, and can become a basis for super-secure quantum computing.
There exists a knot that cannot be untied, but if you copy its mirror image and join them together, they annihilate like a particle with an antiparticle.











