Introduction

Mathematical physics repeatedly asks one question in many forms:

What remains the same while a system changes?

A planet moves, but its angular momentum may be conserved. A quantum state evolves, but probabilities are preserved. A field configuration changes under a gauge transformation, but the observable physics is unchanged. A particle may be described from different inertial frames, yet physical laws keep the same form. These are not separate miracles. They are examples of symmetry.

A symmetry is a transformation that preserves some structure. The structure may be geometric, such as distance in Euclidean space; algebraic, such as a multiplication rule; analytic, such as a differential equation; or physical, such as an action functional or Hamiltonian system. For example, rotating the Euclidean plane about the origin is a symmetry of distance, because the distance between any two points is the same before and after the rotation. Translating an isolated physical system in space is a symmetry when the laws do not depend on the absolute position of the laboratory.

Lie algebras enter when symmetries vary continuously. A continuous symmetry is not just one transformation, but a smooth family of transformations depending on real parameters. Rotations of the plane form the simplest example. For each angle \(\theta\), there is a rotation \(R_\theta\), and changing \(\theta\) slightly changes the rotation slightly. The set of all such rotations is not merely a collection; it has a multiplication law, because doing one rotation after another is again a rotation. Such objects are called Lie groups: groups that are also smooth manifolds, with smooth multiplication and inverse maps. Lie groups and their infinitesimal counterparts, Lie algebras, are a central language for continuous symmetry in geometry, differential equations, mechanics, and quantum theory (Hall, 2015; Olver, 1993).

The key idea of this book is that a continuous symmetry can often be understood by studying its infinitesimal version.

An infinitesimal change is not a tiny finite change. It is the first-order tendency of a family of changes at the identity transformation. If \(R_\theta\) is rotation by angle \(\theta\), then the derivative at \(\theta=0\),

\[ \left.\frac{d}{d\theta}R_\theta\right|_{\theta=0}, \]

records the initial direction in which the identity transformation begins to rotate. This derivative is a linear operator. In matrix form, for rotations of the plane,

\[ R_\theta = \begin{pmatrix} \cos\theta & -\sin\theta\\ \sin\theta & \cos\theta \end{pmatrix}, \]

so

\[ \left.\frac{d}{d\theta}R_\theta\right|_{\theta=0} = \begin{pmatrix} 0 & -1\\ 1 & 0 \end{pmatrix}. \]

This matrix is the infinitesimal generator of plane rotations. It does not rotate the plane by a finite angle. Instead, it tells us the instantaneous velocity field of the rotation at the identity.

This passage from finite transformations to infinitesimal generators is one of the main reasons Lie algebras are powerful. Lie groups can be curved and nonlinear. Lie algebras are vector spaces. A vector space is a mathematical setting in which we can add objects and multiply them by scalars. Therefore, once a symmetry problem is moved into a Lie algebra, linear algebra becomes available. We can choose bases, write matrices, solve eigenvalue problems, decompose spaces into invariant pieces, and compare representations.

But a Lie algebra is not only a vector space. It also has an operation called the Lie bracket.

A Lie bracket is a bilinear operation

\[ [\cdot,\cdot] : \mathfrak g \times \mathfrak g \to \mathfrak g, \]

which takes two elements \(X,Y\) of a vector space \(\mathfrak g\) and produces another element \([X,Y]\) of the same space. The bracket is not usually an ordinary multiplication. Its basic purpose is to measure the failure of two infinitesimal motions to commute.

For matrices, the model example is the commutator

\[ [X,Y] = XY - YX. \]

If \(XY=YX\), then the two linear transformations commute, and their commutator is zero. If \(XY\neq YX\), then the commutator measures the first algebraic trace of their noncommutativity. This is not just a formal trick. In a matrix Lie group, the Lie algebra is often a space of matrices, and its bracket is precisely this commutator (Hall, 2015).

Here is a concrete example. In three-dimensional space, rotate a book a little around the \(x\)-axis and then a little around the \(y\)-axis. Now reverse the order: first rotate around the \(y\)-axis, then around the \(x\)-axis. The final orientations are generally not the same. Rotations in three dimensions do not commute. Infinitesimally, the bracket of the generator of \(x\)-axis rotations with the generator of \(y\)-axis rotations gives a generator related to \(z\)-axis rotations. This is the algebraic shadow of the geometry of rotations. In symbols, for a conventional basis \(J_x,J_y,J_z\) of the rotation Lie algebra \(\mathfrak{so}(3)\), one has relations of the form

\[ [J_x,J_y]=J_z,\qquad [J_y,J_z]=J_x,\qquad [J_z,J_x]=J_y, \]

up to sign conventions depending on the chosen basis. These bracket relations encode the local structure of rotational symmetry.

The formal definition of a Lie algebra is short but deep. A Lie algebra over a field, such as \(\mathbb R\) or \(\mathbb C\), is a vector space \(\mathfrak g\) equipped with a bracket satisfying three conditions:

1. Bilinearity: the bracket is linear in each input.
2. Antisymmetry: \([X,Y]=-[Y,X]\), so in particular \([X,X]=0\) when the field has characteristic not equal to \(2\).
3. Jacobi identity:

\[ [X,[Y,Z]]+[Y,[Z,X]]+[Z,[X,Y]]=0. \]

The Jacobi identity is the least obvious part of the definition. It is the consistency law that makes brackets behave like commutators of infinitesimal transformations. For matrix commutators, the Jacobi identity follows from associativity of matrix multiplication. In abstract Lie algebras, it must be required as an axiom. Much of the structure theory of Lie algebras grows from this single identity (Humphreys, 1972).

This book begins slowly because these ideas deserve to be seen from several angles. A Lie algebra is simultaneously:

• a vector space with a special bilinear operation;
• the infinitesimal form of a Lie group;
• an algebraic device for measuring noncommuting infinitesimal motions;
• a machine for producing and organizing representations;
• a language for symmetry in classical and quantum physics.

The same object can therefore appear in different clothing. In differential geometry, its elements may be vector fields. In linear algebra, they may be matrices. In quantum mechanics, they may be operators acting on a Hilbert space. In particle physics, they may be generators of internal gauge symmetry. The goal of this book is to help you recognize the same underlying structure in all these settings.

Consider quantum angular momentum. In quantum mechanics, angular momentum components are represented by operators \(J_x,J_y,J_z\), and their commutation relations are commonly written as

\[ [J_x,J_y]=i\hbar J_z,\qquad [J_y,J_z]=i\hbar J_x,\qquad [J_z,J_x]=i\hbar J_y. \]

These relations are not arbitrary. They are a representation of the Lie algebra of rotations, with the physical constant \(\hbar\) and the factor \(i\) reflecting standard quantum-mechanical operator conventions. The algebraic relations constrain possible angular momentum states, explain ladder operators, and organize spin. Representation theory turns bracket relations into linear operators on spaces of states, which is one of the main bridges from Lie algebra to physics (Fulton and Harris, 1991; Georgi, 1999).

A representation of a Lie algebra is a way for its abstract elements to act as linear transformations on a vector space. More precisely, a representation of \(\mathfrak g\) on a vector space \(V\) is a linear map

\[ \rho:\mathfrak g \to \mathfrak{gl}(V) \]

such that

\[ \rho([X,Y]) = \rho(X)\rho(Y)-\rho(Y)\rho(X). \]

Here \(\mathfrak{gl}(V)\) denotes the Lie algebra of all linear maps from \(V\) to itself, with bracket given by the commutator. This condition says that the abstract bracket in \(\mathfrak g\) is faithfully translated into the commutator behavior of operators on \(V\). If \(\mathfrak g\) is the algebra of rotations, then a representation describes how rotations act on vectors, spinors, tensors, wavefunctions, or other spaces.

The word “representation” is especially important in mathematical physics because physical systems are rarely described only by the symmetry algebra itself. They are described by spaces on which the symmetry acts. A rigid body, a scalar field, an electron wavefunction, and a multiplet of particle states may all carry different representations of related symmetry algebras. The symmetry algebra gives the grammar; the representation tells us which sentence is being spoken.

Lie algebras also appear naturally in classical mechanics. In Hamiltonian mechanics, observables can be combined using the Poisson bracket, and infinitesimal canonical transformations are generated by Hamiltonian functions. Under appropriate hypotheses, continuous variational symmetries are connected with conserved quantities, a relationship usually called Noether’s theorem after Emmy Noether’s foundational 1918 work; modern geometric mechanics develops this relationship through momentum maps and symmetry reduction (Marsden and Ratiu, 1999; Olver, 1993). Thus, the bracket idea is not confined to matrices. It appears whenever infinitesimal transformations interact.

One of the central themes of this book is that the Lie bracket is small enough to compute with, but rich enough to remember geometry.

For example, take translations in the plane. Let \(P_x\) mean “move infinitesimally in the \(x\)-direction” and \(P_y\) mean “move infinitesimally in the \(y\)-direction.” Translating a little in \(x\), then a little in \(y\), gives the same result as doing these in the opposite order. Therefore,

\[ [P_x,P_y]=0. \]

The corresponding Lie algebra is abelian, meaning all brackets vanish. By contrast, rotations in three dimensions are nonabelian, meaning some brackets are nonzero. This difference is physically meaningful: the order of independent translations does not matter, but the order of rotations generally does. Lie algebras give us a precise way to distinguish these two situations.

The book will develop this idea in layers.

First, we build the linear algebra foundations. Lie theory uses vector spaces constantly: bases, dual spaces, linear maps, eigenvalues, bilinear maps, tensor products, exterior powers, and change of basis. These are not optional technicalities. They are the working language in which Lie algebras are computed.

Next, we define Lie algebras formally and study their first examples. The examples are the true entry point: abelian Lie algebras, matrix Lie algebras such as \(\mathfrak{gl}(n)\) and \(\mathfrak{sl}(n)\), orthogonal and symplectic Lie algebras, the Heisenberg algebra, and Lie algebras of vector fields. Each example teaches a different aspect of the definition.

Then we connect Lie algebras to Lie groups. A Lie group is nonlinear in general, but its tangent space at the identity is a vector space. With the correct bracket, that tangent space becomes the Lie algebra of the group. The exponential map then carries infinitesimal data back toward finite transformations. For matrices, this is the familiar matrix exponential

\[ e^X = I + X + \frac{X^2}{2!}+\frac{X^3}{3!}+\cdots. \]

The Baker-Campbell-Hausdorff formula explains how the product \(e^X e^Y\) can be expressed using \(X\), \(Y\), and their brackets. It is one of the clearest places where finite group multiplication and infinitesimal Lie algebra structure meet.

After that, we enter representation theory. This is where Lie algebras become especially powerful for mathematical physics. We will study modules, invariant subspaces, irreducible representations, complete reducibility, adjoint representations, derivations, automorphisms, and universal enveloping algebras. These ideas explain how a Lie algebra acts, how its actions decompose, and how noncommutative algebra can be used to analyze them.

The middle part of the book develops structure theory. Some Lie algebras are solvable or nilpotent, meaning their repeated brackets eventually become simpler in precise ways. Others are semisimple, meaning they have no nonzero solvable ideals. Semisimple Lie algebras are governed by a remarkable geometric classification using Cartan subalgebras, roots, Weyl groups, and Dynkin diagrams. This classification is one of the great achievements of twentieth-century algebra, and standard introductions show how the abstract theory leads to the classical families and exceptional Lie algebras (Humphreys, 1972; Fulton and Harris, 1991).

Finally, the book turns toward physical applications. We will examine compact Lie algebras and unitary representations, real forms and noncompact algebras, Hamiltonian and field-theoretic symmetry, quantum angular momentum, the Poincare algebra, and gauge algebras such as \(\mathfrak u(1)\), \(\mathfrak{su}(2)\), and \(\mathfrak{su}(3)\). These algebras organize important parts of modern theoretical physics, including spacetime symmetry and the internal symmetries used in the Standard Model of particle physics (Georgi, 1999).

Throughout the book, we will use a consistent learning pattern:

We begin with a concrete transformation.
We identify its infinitesimal generator.
We compute brackets.
We interpret what the brackets mean.
We then represent the algebra on a vector space relevant to geometry or physics.

This pattern keeps the theory grounded. When the notation becomes abstract, return to the model examples: translations, rotations, matrices, vector fields, and quantum operators. They are the compass points of the subject.

A graduate-level study of Lie algebras requires comfort with abstraction, but abstraction should not feel like fog. It should feel like compression. A good definition compresses many examples into one form. A good theorem tells us what must happen in every example satisfying the definition. A good representation turns an abstract algebra into concrete linear transformations. The purpose of this introduction is therefore not to prove the main theorems in advance, but to give you the central picture:

A Lie algebra is a linear home for infinitesimal symmetry.
Its bracket records noncommutativity.
Its representations turn symmetry into operators.
Its structure theory reveals hidden geometry.
Its applications reach from classical mechanics to quantum fields.

We now begin with the motivating source of the subject: symmetry, infinitesimals, and the idea that continuous motion has an algebraic shadow.

References

Fulton, William, and Joe Harris. Representation Theory: A First Course. Springer, 1991.

Georgi, Howard. Lie Algebras in Particle Physics: From Isospin to Unified Theories. 2nd ed., Westview Press, 1999.

Hall, Brian C. Lie Groups, Lie Algebras, and Representations: An Elementary Introduction. 2nd ed., Springer, 2015.

Humphreys, James E. Introduction to Lie Algebras and Representation Theory. Springer, 1972.

Marsden, Jerrold E., and Tudor S. Ratiu. Introduction to Mechanics and Symmetry: A Basic Exposition of Classical Mechanical Systems. 2nd ed., Springer, 1999.

Olver, Peter J. Applications of Lie Groups to Differential Equations. 2nd ed., Springer, 1993.