Root Systems and Dynkin Diagrams

Root systems are finite configurations of vectors in a Euclidean space that are stable under the reflections they generate; Dynkin diagrams encode them combinatorially as decorated graphs. Key result: irreducible reduced root systems are classified exactly by the connected Dynkin diagrams of types

A_{n}, B_{n}, C_{n}, D_{n}, E_{6}, E_{7}, E_{8}, F_{4}, G_{2}

. Applies to: semisimple Lie algebras and groups, Weyl groups, Coxeter groups, singularity theory, representation theory, and parts of combinatorics and physics. Originated: Wilhelm Killing (1888–1890) and Élie Cartan (1894), with the diagrammatic encoding due to Eugene Dynkin (1947).

A root system $Φ$ is a finite set of nonzero vectors in a real inner-product space $E$ such that $Φ$ spans $E$ , the only scalar multiples of a root $α$ lying in $Φ$ are $\pm α$ (in the reduced case), each reflection $s_{α}$ permutes $Φ$ , and the "Cartan integers" $⟨ β, α ⟩ = 2 (β, α) / (α, α)$ are integers. This rigid integrality forces angles between roots to take only the values $9 0^{\circ}, 12 0^{\circ}, 13 5^{\circ}, 15 0^{\circ}$ , and from this single constraint the entire classification cascades. Root systems matter because they are the combinatorial skeleton of semisimple Lie algebras: the classification of root systems is the classification of complex simple Lie algebras. The surprising fact is that an essentially geometric/algebraic problem of unbounded apparent complexity collapses to a short finite list with exactly five "exceptional" cases — $G_{2}, F_{4}, E_{6}, E_{7}, E_{8}$ — whose existence is one of the most striking accidents in mathematics.

The Core Insight

All the structure lives in the Cartan integers $⟨ β, α ⟩ \in Z$ . Because $⟨ α, β ⟩ ⟨ β, α ⟩ = 4 cos^{2} θ \in {0, 1, 2, 3}$ for non-proportional roots, the angle and length-ratio data of any pair of roots is quantized. A base (simple system) selects a minimal spanning subset such that every root is a $Z_{\geq 0}$ - or $Z_{\leq 0}$ -combination of it; the pairwise Cartan data of the base, drawn as a decorated graph, is the Dynkin diagram. The classification is then pure graph combinatorics: which connected diagrams can occur.

Formal Statement

Root System

Let $E$ be a finite-dimensional real vector space with positive-definite inner product $(\cdot, \cdot)$ . For a nonzero $α \in E$ define the reflection

s_{α} (v) = v - \frac{2 ( v , α )}{( α , α )} α = v - ⟨ v, α ⟩ α,

⟳

s_{α}

fixes the hyperplane

α^{⊥}

and sends

α \mapsto - α

; subtract twice the projection of

v

onto

α

where the Cartan pairing (linear in the first slot only) is

⟨ v, α ⟩ := \frac{2 ( v , α )}{( α , α )} .

⟳this is

2 \times

the component of

v

along the unit vector

α /∥ α ∥

divided by

∥ α ∥

, normalized so

⟨ α, α ⟩ = 2

A finite set $Φ \subset E ∖ {0}$ is a root system if:

(R1) $Φ$ spans $E$ ;
(R2) for $α \in Φ$ , $R α \cap Φ = {\pm α}$ (reduced condition);
(R3) $s_{α} (Φ) = Φ$ for all $α \in Φ$ ;
(R4) $⟨ β, α ⟩ \in Z$ for all $α, β \in Φ$ .

The elements of $Φ$ are roots; $dim E$ is the rank.

Coroots, Base, Weyl Group

For $α \in Φ$ , the coroot is $α^{\lor} = 2 α / (α, α)$ , so $⟨ v, α ⟩ = (v, α^{\lor})$ and ${α^{\lor}} = Φ^{\lor}$ forms the dual root system.

The Weyl group is the finite group

W (Φ) = ⟨ s_{α} : α \in Φ ⟩ \leq O (E) .

⟳each

s_{α}

is an orthogonal involution permuting the finite spanning set

Φ

, so

W

injects into the symmetric group on

Φ

and is finite.

A base (simple system) $Δ = {α_{1}, \dots, α_{ℓ}} \subset Φ$ is a basis of $E$ such that every $β \in Φ$ writes as $β = \sum_{i} c_{i} α_{i}$ with all $c_{i} \in Z_{\geq 0}$ (a positive root) or all $c_{i} \in Z_{\leq 0}$ (a negative root). Bases exist and are permuted simply transitively by $W$ .

Angle and Length Quantization

For non-proportional roots $α, β \in Φ$ with angle $θ$ ,

⟨ α, β ⟩ ⟨ β, α ⟩ = 4 cos^{2} θ \in {0, 1, 2, 3} .

⟳multiply the two Cartan integers; the cross-products of inner products combine to

4 (α, β)^{2} / [(α, α) (β, β)] = 4 cos^{2} θ

. Both factors are integers of the same sign, and

cos^{2} θ < 1

excludes

4

Consequently the only possible angles are $θ \in {9 0^{\circ}, 12 0^{\circ}, 13 5^{\circ}, 15 0^{\circ}}$ (plus $6 0^{\circ}, 4 5^{\circ}, 3 0^{\circ}$ for the supplement), and when $θ \neq = 9 0^{\circ}$ the squared length ratio $(β, β) / (α, α)$ is forced to $1, 2,$ or $3$ .

Cartan Matrix

Fix a base $Δ = {α_{1}, \dots, α_{ℓ}}$ . The Cartan matrix is

A_{ij} = ⟨ α_{i}, α_{j} ⟩ = \frac{2 ( α _{i} , α _{j} )}{( α _{j} , α _{j} )} .

⟳apply the Cartan pairing [2] to ordered pairs of simple roots; diagonal entries are

⟨ α_{i}, α_{i} ⟩ = 2

by definition.

It satisfies: $A_{ii} = 2$ ; $A_{ij} \in {0, - 1, - 2, - 3}$ for $i \neq = j$ ; $A_{ij} = 0 ⟺ A_{j i} = 0$ ; and $A$ is symmetrizable to a positive-definite matrix. The root system is recovered (up to isomorphism) from $A$ , hence from the Dynkin diagram.

Dynkin Diagram

The Dynkin diagram of $(Φ, Δ)$ is the labeled graph with one node per simple root $α_{i}$ , and between $α_{i}, α_{j}$ a number of edges equal to

n_{ij} = ⟨ α_{i}, α_{j} ⟩ ⟨ α_{j}, α_{i} ⟩ = A_{ij} A_{j i} \in {0, 1, 2, 3} .

⟳this is the integer

4 cos^{2} θ

from [4]; it counts the edge multiplicity (0 = no edge, 1 = single, 2 = double, 3 = triple).

When $n_{ij} > 1$ (the two roots have different lengths) an arrow points from the long root toward the short root. The diagram is connected iff $Φ$ is irreducible (not an orthogonal direct sum).

Classification — Killing–Cartan

Every connected Dynkin diagram of a reduced irreducible root system is isomorphic to exactly one of the following:

A_{ℓ} (ℓ \geq 1), B_{ℓ} (ℓ \geq 2), C_{ℓ} (ℓ \geq 3), D_{ℓ} (ℓ \geq 4), E_{6}, E_{7}, E_{8}, F_{4}, G_{2} .

⟳positive-definiteness of the (symmetrized) Cartan matrix restricts which subgraphs (cycles, high-valence nodes, long chains of multiple edges) can occur; eliminating the forbidden configurations leaves exactly this list.

Each diagram arises from a unique irreducible root system, and the simply-laced ones ( $A, D, E$ : all roots same length) have $W$ generated by reflections in equal-length roots.

The nine connected Dynkin diagrams of irreducible reduced root systems

<svg viewBox="0 0 640 520" xmlns="http://www.w3.org/2000/svg">
<rect width="640" height="520" fill="#0a0e17"/>
<g stroke="#94a3b8" stroke-width="2" fill="none">
<!-- A_n -->
<line x1="40" y1="40" x2="120" y2="40"/>
<line x1="120" y1="40" x2="180" y2="40"/>
<line x1="220" y1="40" x2="280" y2="40"/>
</g>
<g fill="#d4a853"><circle cx="40" cy="40" r="6"/><circle cx="120" cy="40" r="6"/><circle cx="180" cy="40" r="6"/><circle cx="280" cy="40" r="6"/></g>
<text x="195" y="46" fill="#e2e8f0" font-size="14">···</text>
<text x="320" y="46" fill="#e2e8f0" font-size="16">A_n</text>

<!-- B_n double edge with arrow to short -->
<g stroke="#94a3b8" stroke-width="2" fill="none">
<line x1="40" y1="100" x2="120" y2="100"/>
<line x1="160" y1="100" x2="220" y2="100"/>
<line x1="220" y1="97" x2="280" y2="97"/>
<line x1="220" y1="103" x2="280" y2="103"/>
</g>
<polyline points="255,90 270,100 255,110" fill="none" stroke="#94a3b8" stroke-width="2"/>
<g fill="#d4a853"><circle cx="40" cy="100" r="6"/><circle cx="120" cy="100" r="6"/><circle cx="220" cy="100" r="6"/><circle cx="280" cy="100" r="6"/></g>
<text x="135" y="106" fill="#e2e8f0" font-size="14">···</text>
<text x="320" y="106" fill="#e2e8f0" font-size="16">B_n (long→short)</text>

<!-- C_n arrow other way -->
<g stroke="#94a3b8" stroke-width="2" fill="none">
<line x1="40" y1="160" x2="120" y2="160"/>
<line x1="160" y1="160" x2="220" y2="160"/>
<line x1="220" y1="157" x2="280" y2="157"/>
<line x1="220" y1="163" x2="280" y2="163"/>
</g>
<polyline points="265,150 250,160 265,170" fill="none" stroke="#94a3b8" stroke-width="2"/>
<g fill="#d4a853"><circle cx="40" cy="160" r="6"/><circle cx="120" cy="160" r="6"/><circle cx="220" cy="160" r="6"/><circle cx="280" cy="160" r="6"/></g>
<text x="135" y="166" fill="#e2e8f0" font-size="14">···</text>
<text x="320" y="166" fill="#e2e8f0" font-size="16">C_n</text>

<!-- D_n fork -->
<g stroke="#94a3b8" stroke-width="2" fill="none">
<line x1="40" y1="225" x2="120" y2="225"/>
<line x1="160" y1="225" x2="220" y2="225"/>
<line x1="220" y1="225" x2="270" y2="205"/>
<line x1="220" y1="225" x2="270" y2="245"/>
</g>
<g fill="#d4a853"><circle cx="40" cy="225" r="6"/><circle cx="120" cy="225" r="6"/><circle cx="220" cy="225" r="6"/><circle cx="270" cy="205" r="6"/><circle cx="270" cy="245" r="6"/></g>
<text x="135" y="231" fill="#e2e8f0" font-size="14">···</text>
<text x="320" y="231" fill="#e2e8f0" font-size="16">D_n</text>

<!-- E_8 -->
<g stroke="#94a3b8" stroke-width="2" fill="none">
<line x1="40" y1="300" x2="280" y2="300"/>
<line x1="120" y1="300" x2="120" y2="270"/>
</g>
<g fill="#d4a853">
<circle cx="40" cy="300" r="6"/><circle cx="80" cy="300" r="6"/><circle cx="120" cy="300" r="6"/>
<circle cx="160" cy="300" r="6"/><circle cx="200" cy="300" r="6"/><circle cx="240" cy="300" r="6"/><circle cx="280" cy="300" r="6"/>
<circle cx="120" cy="270" r="6"/></g>
<text x="320" y="306" fill="#e2e8f0" font-size="16">E_8 (E_6,E_7 truncate)</text>

<!-- F_4 -->
<g stroke="#94a3b8" stroke-width="2" fill="none">
<line x1="40" y1="370" x2="120" y2="370"/>
<line x1="120" y1="367" x2="200" y2="367"/>
<line x1="120" y1="373" x2="200" y2="373"/>
<line x1="200" y1="370" x2="280" y2="370"/>
</g>
<polyline points="175,360 190,370 175,380" fill="none" stroke="#94a3b8" stroke-width="2"/>
<g fill="#d4a853"><circle cx="40" cy="370" r="6"/><circle cx="120" cy="370" r="6"/><circle cx="200" cy="370" r="6"/><circle cx="280" cy="370" r="6"/></g>
<text x="320" y="376" fill="#e2e8f0" font-size="16">F_4</text>

<!-- G_2 triple -->
<g stroke="#94a3b8" stroke-width="2" fill="none">
<line x1="40" y1="440" x2="120" y2="436"/>
<line x1="40" y1="440" x2="120" y2="444"/>
<line x1="40" y1="440" x2="120" y2="440"/>
</g>
<polyline points="95,430 110,440 95,450" fill="none" stroke="#94a3b8" stroke-width="2"/>
<g fill="#d4a853"><circle cx="40" cy="440" r="6"/><circle cx="120" cy="440" r="6"/></g>
<text x="160" y="446" fill="#e2e8f0" font-size="16">G_2 (triple edge)</text>
</svg>

Why It Holds / Derivation

Sketch of the Classification

Step 1 — Quantize angles. By [4], for non-proportional

α, β

the product

⟨ α, β ⟩ ⟨ β, α ⟩ \in {0, 1, 2, 3}

. Listing the integer solutions of

⟨ α, β ⟩ ⟨ β, α ⟩ = 4 cos^{2} θ

with

∣ ⟨ \cdot ⟩ ∣ \leq 3

yields the finite menu of (angle, length-ratio) pairs. This is the only arithmetic input. Step 2 — Reduce to a base. Choose a generic linear functional

f

; declare a root positive if

f (α) > 0

. The indecomposable positive roots (not a sum of two positive roots) form a base

Δ

. One shows

∣Δ∣ = dim E

and every root is a nonneg/nonpos integer combination. Step 3 — Simple roots make obtuse angles. For distinct

α_{i}, α_{j} \in Δ

(α_{i}, α_{j}) \leq 0

, so

θ \geq 9 0^{\circ}

and

n_{ij} = 4 cos^{2} θ \in {0, 1, 2, 3}

. Encode

(Δ, A)

as the Dynkin diagram. Step 4 — Positive-definiteness constraints. The Gram matrix of

Δ

is positive-definite. Define for each

α_{i}

the unit vector

ε_{i} = α_{i} /∥ α_{i} ∥

; then for any subset

\sum c_{i} ε_{i}

one has

∥ \sum c_{i} ε_{i} ∥^{2} > 0

. Working out this inequality forbids:

i < j \sum 2 c_{i} c_{j} (ε_{i}, ε_{j}) > - i \sum c_{i}^{2} ⟹ bounded edge multiplicities and valences.

⟳expand

∥ \sum c_{i} ε_{i} ∥^{2} = \sum c_{i}^{2} + 2 \sum_{i < j} c_{i} c_{j} (ε_{i}, ε_{j})

and demand positivity; with

(ε_{i}, ε_{j}) = - \frac{1}{2} n_{ij}

this gives concrete combinatorial bans.

Step 5 — Eliminate forbidden subgraphs. Positive-definiteness implies: (a) no cycles; (b) at most one node of valence 3 (a branch node) and no node of valence

\geq 4

; (c) at most one multiple edge, which (if present) excludes any branch node; (d) bounds on the lengths of the three arms emanating from a branch node, via the "balance" inequality. The surviving connected diagrams are precisely the list [7]. Step 6 — Realizability. Each surviving diagram is realized by an explicit root system (e.g.

A_{ℓ}

{e_{i} - e_{j}} \subset R^{ℓ + 1}

), completing the bijection.

□

Wilhelm Killing & Élie Cartan | 1847–1923 / 1869–1951

Killing, working in near-isolation in Braunsberg, sought to classify the "infinitesimal groups" (Lie algebras) and in 1888–1890 produced the first classification of simple Lie algebras over $C$ , discovering the exceptional types and inventing what we now call roots, the Cartan matrix, and the characteristic equation. His papers were riddled with errors and obscure, and he doubted his own exceptional cases. Élie Cartan, in his 1894 thesis, corrected, completed, and rigorized the entire theory, gave clean existence proofs, and built the bridge to Lie groups and differential geometry. The phrase "Cartan integer" and "Cartan matrix" honor him; Hermann Weyl later supplied the global group-theoretic and representation-theoretic superstructure (the Weyl character formula, integrality of weights). Eugene Dynkin's 1947 graphical encoding turned the classification into a teachable combinatorial picture.

Building Intuition

Think of a root system as a kaleidoscope. The reflecting hyperplanes $α^{⊥}$ chop $E$ into congruent chambers (Weyl chambers); the Weyl group shuffles them simply transitively. A base $Δ$ picks one chamber, and the simple roots are the inward-pointing normals to its walls. The whole infinite-looking symmetry is generated by bouncing off just $ℓ$ mirrors.

The integrality (R4) is the secret enforcer: it says roots "see each other at quantized angles." Two mirrors can only meet at $9 0^{\circ}, 6 0^{\circ}, 4 5^{\circ},$ or $3 0^{\circ}$ — exactly the dihedral angles that tile space. That is why the answer is a finite list: there are only so many ways to arrange a handful of mutually-quantized mirrors in a positive-definite space without the geometry "blowing up" (which positive-definiteness forbids). The exceptional groups $G_{2}, F_{4}, E_{8}$ are the rare maximal configurations where this packing is just barely possible.

Worked Examples

The Root System $A_2$ (rank 2)

Given:

E = {x \in R^{3} : \sum x_{i} = 0}

, roots

Φ = {e_{i} - e_{j} : i \neq = j}

, six in total. Take base

Δ = {α_{1} = e_{1} - e_{2}, α_{2} = e_{2} - e_{3}}

. Steps:

Inner products: $(α_{1}, α_{1}) = (α_{2}, α_{2}) = 2$ , and $(α_{1}, α_{2}) = - 1$ .
Cartan integers: $⟨ α_{1}, α_{2} ⟩ = 2 (- 1) /2 = - 1$ ; similarly $⟨ α_{2}, α_{1} ⟩ = - 1$ .
Cartan matrix $A = (2 - 1 - 1 2)$ , edge multiplicity $n_{12} = (- 1) (- 1) = 1$ → single edge, no arrow (equal lengths).
Angle: $4 cos^{2} θ = 1 \Rightarrow θ = 12 0^{\circ}$ .
Weyl group: $∣ W ∣ = 6$ , the symmetric group $S_{3}$ acting on the 6 roots; the diagram is two nodes joined by one edge.

Answer:

A_{2} : \circ - \circ, A = (2 - 1 - 1 2), ∣Φ∣ = 6, ∣ W ∣ = 6

— the Lie algebra

sl_{3} (C)

The Root System $G_2$ (rank 2, triple edge)

Given: a rank-2 system with a short root

α

(

(α, α) = 2

) and a long root

β

(

(β, β) = 6

) at angle

15 0^{\circ}

. Base

Δ = {α, β}

. Steps:

$(α, β) = ∥ α ∥∥ β ∥ cos 15 0^{\circ} = 26 (- \frac{3}{2}) = - 3$ .
$⟨ α, β ⟩ = 2 (- 3) /6 = - 1$ ; $⟨ β, α ⟩ = 2 (- 3) /2 = - 3$ .
Edge multiplicity $n = (- 1) (- 3) = 3$ → triple edge, arrow from long $β$ to short $α$ .
Generating reflections produce $∣Φ∣ = 12$ roots (6 short, 6 long) and $∣ W ∣ = 12$ (dihedral of order 12).
Cartan matrix $A = (2 - 3 - 1 2)$ , $det A = 4 - 3 = 1$ .

Answer:

G_{2} : \circ ⇛ \circ, A = (2 - 3 - 1 2), ∣Φ∣ = 12, ∣ W ∣ = 12

— the smallest exceptional Lie algebra, dimension 14.

Common Misconceptions

The Cartan pairing $⟨ v, α ⟩$ is NOT symmetric. It is linear in

v

and "anti-scales" in

α

. For roots of different lengths,

⟨ α, β ⟩ \neq = ⟨ β, α ⟩

— that asymmetry is exactly what the arrow in a multiply-laced diagram records. Confusing the two flips

B_{n} \leftrightarrow C_{n}

. A Dynkin diagram is not just a Coxeter diagram. The undirected Coxeter diagram records only angles (

n_{ij}

), so

B_{n}

and

C_{n}

have the same Coxeter diagram and Weyl group. The arrow (length data) distinguishes them, and is essential to recover the root system and Lie algebra.

B_{n}

and

C_{n}

are non-isomorphic root systems that are dual:

B_{n}^{\lor} = C_{n}

. $A_{1}$ , $B_{1} = C_{1}$ , $D_{1}$ , $D_{2} = A_{1} \times A_{1}$ , $D_{3} = A_{3}$ , $B_{2} = C_{2}$ "coincidences" are why the index ranges in [7] are restricted — they prevent double-counting the same diagram. A finite reflection-symmetric vector set need not be a root system. Take the 8 vertices of a regular octagon (vectors at

4 5^{\circ}

increments) in

R^{2}

. They are closed under their reflections, span

E

, and are reduced — but the Cartan integers fail (R4): two adjacent vectors at

4 5^{\circ}

give

4 cos^{2} 4 5^{\circ} = 2

, yet for vectors at

13 5^{\circ}

you'd need a length ratio incompatible with the octagon's equal lengths. No consistent integer Cartan data exists. This is precisely the obstruction that bans diagrams with a "cycle" of mirrors: the dihedral group

I_{2} (8)

is a perfectly good Coxeter group but not a Weyl group — it is non-crystallographic. Likewise the icosahedral group

H_{3}

and

H_{4}

, and

I_{2} (5)

(fivefold symmetry), are Coxeter but never crystallographic, by the crystallographic restriction

4 cos^{2} θ \in Z

Connections

Root systems sit at a crossroads. Via the Serre relations they reconstruct the entire Semisimple Lie Algebra from its Cartan matrix, making the classification of root systems identical to the classification of complex simple Lie algebras and (over algebraically closed fields) Linear Algebraic Groups. Forgetting the arrows yields Coxeter Groups, of which Weyl groups are the crystallographic members; this links to the geometry of Reflection Groups and hyperplane arrangements. Replacing positive-definiteness by positive-semidefinite or indefinite Cartan matrices yields affine and Kac–Moody algebras, generalizing [7] to infinite families. The simply-laced ( $A D E$ ) diagrams reappear, astonishingly, as the McKay Correspondence (finite subgroups of $S U (2)$ ), as Kleinian (du Val) surface singularities, and as quivers of finite representation type (Gabriel's theorem) — three independent classifications that all produce exactly $A_{n}, D_{n}, E_{6}, E_{7}, E_{8}$ .

Real-World Applications

Particle physics — Grand Unified Theories. The gauge group candidates $S U (5)$ ( $A_{4}$ ), $S O (10)$ ( $D_{5}$ ), and $E_{6}$ unify the Standard Model; their root systems dictate the allowed matter representations and symmetry-breaking chains.
String theory & supergravity. The heterotic string's gauge group is $E_{8} \times E_{8}$ (or $S O (32)$ ); $E_{8}$ 's 248-dimensional adjoint and its root lattice (the densest known lattice packing in dimension 8) are central.
Crystallography & quasicrystals. The crystallographic restriction (no 5-, 7-, 8-fold rotational symmetry in periodic lattices) is the same integrality that bans non-crystallographic Coxeter groups; quasicrystals exploit the forbidden $H_{3}, H_{4}$ symmetries.
Singularity theory. ADE classification of simple (du Val / Kleinian) surface singularities $C^{2} /Γ$ and their minimal resolutions, whose exceptional divisors form ADE Dynkin configurations.
Sphere packing & coding. The $E_{8}$ lattice (Viazovska, 2016) gives the optimal sphere packing in $R^{8}$ ; the Leech lattice (built from $E_{8}$ -type data) optimizes $R^{24}$ .
Representation theory of quivers / cluster algebras. Fomin–Zelevinsky cluster algebras of finite type are classified by Cartan matrices, i.e. Dynkin diagrams; mutation combinatorics drives the theory.
Integrable systems & Toda lattices. Affine Dynkin diagrams parametrize generalized (affine) Toda field theories whose conserved charges and soliton masses follow the diagram's eigenvalue data.

Historical Development

1873 — Sophus Lie founds the theory of continuous transformation groups, supplying the objects to be classified.
1888–1890 — Wilhelm Killing publishes four papers classifying simple complex Lie algebras, inventing roots, the rank, the Cartan matrix, and discovering $G_{2}, F_{4}, E_{6}, E_{7}, E_{8}$ (though with errors and tentative claims).
1894 — Élie Cartan's doctoral thesis corrects and rigorizes the classification, proves existence of the exceptional algebras, and establishes the root-system framework cleanly.
1925–1926 — Hermann Weyl develops the global theory: compact groups, the unitarian trick, Weyl group, and the Weyl character/integral formulas, anchoring representations to root data.
1934 — Coxeter classifies finite reflection groups and introduces Coxeter diagrams, separating angle data from crystallographic length data.
1947 — Eugene Dynkin introduces the diagrams now bearing his name, streamlining the classification into a few combinatorial lemmas.
1950s–1960s — Chevalley constructs the simple algebraic groups over arbitrary fields from root data ("Chevalley bases"), and the classification becomes scheme-theoretic.
1968 — Gabriel's theorem: quivers of finite representation type are exactly the ADE Dynkin diagrams, opening the modern web of ADE correspondences.
1980 — McKay's correspondence links finite subgroups of $S U (2)$ to affine ADE diagrams, deepening the unification of seemingly unrelated classifications.

Subtleties & Limits

The classification of finite root systems is complete and clean, but the surrounding landscape harbors deep open questions. (i) Conceptual explanation of $E_{8}$ : there is still no fully satisfying "reason" why the exceptional series exists; Vinberg, Deligne, and others have proposed exceptional-series interpolations (the "magic triangle," Deligne's exceptional series with uniform dimension formulas), yet a single unifying principle remains elusive. (ii) Kac–Moody and Borcherds algebras: dropping positive-definiteness to indefinite Cartan matrices yields hyperbolic and Lorentzian Kac–Moody algebras (e.g. $E_{10}, E_{11}$ conjecturally governing M-theory/supergravity); the structure of their root multiplicities is largely unknown, with no closed formula. (iii) Non-crystallographic to crystallographic: understanding precisely how $H_{3}, H_{4}, I_{2} (n)$ embed into and "fold from" crystallographic systems (Coxeter folding, quasicrystal projection) continues to generate research. (iv) Categorification: realizing root systems and their positive roots intrinsically as shadows of categorical/homological data (cluster categories, stability conditions, Bridgeland spaces) is an active program whose general structure is only partially understood.