lie theory
Root Systems
You should know: hilbert spaces, group mathematics
Overview
A root system is a finite set of nonzero vectors in a Euclidean space satisfying certain symmetry conditions with respect to reflections. Root systems encode the symmetry structure of semisimple Lie algebras and provide the combinatorial data needed to classify them completely. The celebrated classification — types A, B, C, D, and exceptional types G₂, F₄, E₆, E₇, E₈ — is one of the deepest results in mathematics.
Intuition
Imagine a highly symmetric arrangement of arrows (vectors) in a plane or higher-dimensional space where reflecting the whole arrangement through any arrow's perpendicular hyperplane maps the set to itself, and the only lengths that appear are at most two distinct values. This rigid combinatorial structure — the root system — completely determines which semisimple Lie algebras can exist.
Formal Definition
Let V be a finite-dimensional real inner-product space. A root system in V is a finite set Φ ⊂ V \ {0} satisfying four axioms.
Notation
| Notation | Meaning |
|---|---|
| Root system | |
| Coroot of α | |
| Cartan integer | |
| Weyl group generated by reflections sα | |
| Positive and negative roots |
Properties
Finite Weyl group
Root length ratio constraint
Simple roots form a basis
Worked Examples
- 1
A₂ consists of 6 roots arranged at 60° intervals.
- 2
All roots have equal length. The Weyl group is the dihedral group of order 6 (symmetries of an equilateral triangle).
- 3
Simple roots are Δ = {α₁, α₂} with Cartan matrix entries ⟨α₁, α₂∨⟩ = ⟨α₂, α₁∨⟩ = -1.
✓ Answer
A₂ is a rank-2 root system with 6 roots, Weyl group S₃, and corresponds to the Lie algebra sl₃(ℂ).
Practice Problems
Show that if α, β ∈ Φ with α ≠ ±β, then ⟨α, β∨⟩⟨β, α∨⟩ ∈ {0, 1, 2, 3}.
Prove that the Weyl group of a root system Φ acts simply transitively on the set of Weyl chambers.
Common Mistakes
Assuming all roots in a root system have equal length
Only simply-laced root systems (A, D, E types) have all roots the same length. Types B, C, F₄, G₂ have two root lengths.
Confusing the root system with the root lattice
The root system Φ is a finite set of vectors; the root lattice Q = ℤ-span(Φ) is an infinite lattice. They are different objects.
Quiz
Historical Background
Root systems emerged from Wilhelm Killing's 1888–1890 classification of simple Lie algebras over ℂ, later made rigorous by Élie Cartan in his 1894 thesis. The purely combinatorial reformulation in terms of root systems and Dynkin diagrams was developed by Eugene Dynkin in the 1940s, giving an elegant and elementary approach to the classification.
- 1888
Killing classifies simple Lie algebras, discovering exceptional types
Wilhelm Killing
- 1894
Cartan rigorises the classification in his doctoral thesis
Élie Cartan
- 1947
Dynkin introduces Dynkin diagrams as a clean encoding of root systems
Eugene Dynkin
Summary
- A root system is a finite set of nonzero vectors in a Euclidean space, closed under reflections, with integer Cartan integers.
- Simple roots form a basis; every root is a non-negative or non-positive integer combination of simple roots.
- The Weyl group encodes the symmetries of the root system.
- Root systems classify semisimple Lie algebras: types An, Bn, Cn, Dn, and exceptionals G₂, F₄, E₆, E₇, E₈.
References
- BookHumphreys, J.E. — Introduction to Lie Algebras and Representation Theory (1972), Chapter III
- WebsiteWikipedia — Root system
- WebsiteMathWorld — Root System
Mathematics