unitary representation of the Poincaré group



physics, mathematical physics, philosophy of physics

Surveys, textbooks and lecture notes

theory (physics), model (physics)

experiment, measurement, computable physics

Representation theory



The Poincaré group is the group of rigid spacetime symmetries of Minkowski spacetime. It is a topological group and as such has unitary representations on infinite-dimensional Hilbert spaces. For any quantum field theory in Minkowski space its space of states therefore decomposes into irreducible representations of the Poincaré group. As was first observed by Hermann Weyl, these irreducible representations encode the particle spectrum of the QFT.

We are interested in the unitary representations of a topological group GG, i.e., the continuous group homomorphisms

U:GU(H)U\colon G \to U(H)

into the unitary group of a Hilbert space HH, especially those that are irreducible in the usual sense of representation theory. The topology on U(H)U(H) here is understood to be the strong operator topology.

In this and related articles, we study such representations in the case where

G=SL 2() 4G = SL_2(\mathbb{C}) \ltimes \mathbb{R}^4

is the universal cover of the connected component of the identity of the Poincaré group, which is important in the study of quantum field theory. A more physical name for such a representation is “elementary particle”, and we will often use that term in this article. (NB: “elementary particle” will always refer to the formal mathematical

A full rounded account could become large; see the blog discussion, which despite its size was left in a still-nascent state. However, in a nutshell, the basic theorem is that (elementary) particles are classified up to isomorphism according to their mass and helicity; mass is a continuous parameter and helicity is a discrete parameter.

This theorem in commonly ascribed to Eugene Wigner and often refereed to as the Wigner classification.

Wigner classified all irreducible unitary representations of the restricted Poincare group, including the unphysical ones. The latter cannot be used to define a free quantum field theory? satisfying the Wightman axioms. Those that can are the physical ones and are characterized by a nonnegative real mass and a nonnegative half-integral spin; the zero component of the momentum has a nonnegative spectrum. Many of these are realized by particles occurring in Nature, though not as ‘elementary particles’‘ but as bound states (in a suitable approximation, e.g., QCD). From the point of view of representation theory, the center of mass of a bound state behavs just like an elementary particle. Thus elementary is meant in this generalized sense.

Relevant topics in a full account will include

Tentative notes, to be expanded on…

In the first place, physicists tend to be a little carefree with the mathematics, so this account is written from the point of view of a ‘stupid’ mathematician (for the moment Todd Trimble) who wants to get details straight and precise.

For example, physicists tend to talk about “eigenstates” as if they were elements of the Hilbert space, and other states as linear combinations of eigenstates, whereas really we are dealing with some more complicated technology like rigged Hilbert spaces or direct integrals instead of direct sums. Failure to mention such details places hurdles of communication between physicists and mathematicians. In addition, there are stylistic differences in presentation, where a physicist will happily deal with formulas replete with lots of subscripts and superscripts, whereas many mathematicians prefer dealing with more conceptual, less notation-heavy explanations.

There seem to be at least three ways of dealing with spectral theory of (unbounded) self-adjoint operators on a Hilbert space:

  • The usual Stone theory

  • Direct integrals of Hilbert spaces

  • Rigged Hilbert spaces

Rigged Hilbert spaces

A rigged Hilbert space

Induced representations

Simultaneous diagonalization.

Fact that Poincaré group is a semidirect product group.

Let p d1,1\vec p \in \mathbb{R}^{d-1,1} be a given vector in Minkowski spacetime. Write

Stab Iso( d1,1))(p)Iso( d1,1) Stab_{Iso(\mathbb{R}^{d-1,1}))}(\vec p) \hookrightarrow Iso(\mathbb{R}^{d-1,1})

for its stabilizer subgroup (often called the little group in this context, going back to Wigner). Every unitary irrep of Iso( d1,1)Iso(\mathbb{R}^{d-1,1}) of mass pp is an induced representation of a finite dimensional representation of the “little group” Stab Iso( d1,1))(p)Stab_{Iso(\mathbb{R}^{d-1,1}))}(\vec p). (recalled concisely e.g. in Dragon 16, p. 2).


The observation that the irreps of the Poincaré group correspond to fundamental particles (Wigner classification) is due to

  • Eugene Wigner, On unitary representations of the inhomogeneous Lorentz group , Ann. Math. 40, 149 (1993)

Review includes

Revised on August 24, 2016 05:18:34 by Urs Schreiber (