Group Averaging Procedure in de Sitter QFT

• Group averaging procedure is a mathematical method that projects Fock space states onto a de Sitter-invariant subspace via integration over symmetry groups.
• It leverages refined algebraic quantization and harmonic analysis techniques to rigorously define invariant inner products and ensure convergence of the group integrals.
• The procedure is crucial for satisfying linearization-stability constraints in de Sitter quantum field theory, thereby enabling consistent quantization of fields in curved spacetimes.

A group averaging procedure is a mathematical process that constructs, by explicit averaging over a symmetry group, a subspace of group-invariant physical states together with an invariant inner product. In the context of quantum field theory in de Sitter space, group averaging is essential due to the linearization-stability constraints: only quantum states invariant under the de Sitter group are physically admissible, but standard constructions rarely provide such states beyond the vacuum. Group averaging creates a physical Hilbert space comprised of de Sitter-invariant states by integrating (or "summing") over the symmetry group applied to Fock space states. The procedure’s convergence and efficacy depend critically on the analytic properties of the unitary group representations involved.

1. Mathematical Definition and Rigging Map

The group averaging procedure is rooted in the refined algebraic quantization (RAQ) framework, wherein one starts from an auxiliary (or kinematic) Hilbert space $\mathcal{H}_\mathrm{aux}$ supporting a unitary representation $U(g)$ of the symmetry group $G$—here, the de Sitter group $\mathrm{SO}_0(D,1)$ for $D$-dimensional de Sitter space. The group averaging “rigging map” $\eta$ is defined as

$\eta(\psi) = \int_G dg \; U(g) |\psi\rangle,$

where $dg$ is the Haar measure on $G$. The group-averaged inner product between (say) auxiliary states $\psi_1, \psi_2 \in \mathcal{H}_\mathrm{aux}$ is then

$$\langle \eta(\psi_1) | \eta(\psi_2) \rangle = \int_G dg \; \langle \psi_1 | U(g) | \psi_2 \rangle.$$

Physically, this projection extracts the invariant content of the state by integrating over all group actions. If $|\psi\rangle$ is already invariant ($U(g)|\psi\rangle = |\psi\rangle$ for all $g$), it is unchanged by $\eta$. In de Sitter QFT, this is required by the gravitational constraints.

2. Representations and the Need for Group Averaging

For quantum fields on de Sitter space, the underlying group is $\mathrm{SO}_0(D,1)$. States are built as Fock space vectors from single-particle representations of this group, which are classified into principal, complementary, and discrete series. For scalar fields of positive mass, the principal series representations are relevant.

A fundamental obstruction highlighted by Higuchi is that the standard Fock space contains no de Sitter-invariant states except for the vacuum. Generic $N$-particle Fock states transform nontrivially under the group. As a result, the physical sector—i.e., de Sitter-invariant states—cannot be obtained simply by taking the original Fock space and restricting to invariant vectors. Group averaging is necessary to define and construct a physically admissible Hilbert space.

3. Explicit Construction for Free Fields

For a free scalar field, consider the $N$-particle sector of Fock space. States are wavefunctions $\Psi(k_1, \dots, k_N)$, where each $k_j$ labels a mode compatible with the de Sitter group’s spherical harmonics decomposition (or de Sitter harmonics), subject to smoothness or spectral constraints depending on the result being established.

Group averaging of such a state proceeds by acting with $U(g)$ on each momentum argument and integrating over the group: $$\Psi_\mathrm{av}(\vec{k}_1, \dots, \vec{k}_N) = \int_G dg \; \Psi(g \cdot \vec{k}_1, \dots, g \cdot \vec{k}_N).$$ The inner product on the averaged space is correspondingly given by a double group integral over the product of two such functions, yielding a positive-definite invariant inner product provided convergence is ensured.

The central technical task is to characterize the convergence of the group integral, which reduces to the asymptotic properties of the group representation matrix elements and, ultimately, to detailed harmonic analysis: generalized Legendre functions, hypergeometric functions, and multiplet methods (e.g., Gelfand–Tsetlin bases) are required to analyze and ensure the group averaging map is well-defined for smooth $N$-particle states or those built from finite linear combinations of de Sitter harmonics.

4. Convergence and Physical Hilbert Space

A major result established is that group averaging converges for $N$-particle states provided $N$ is “sufficiently large.” For general smooth wavefunctions, the required $N$ depends on the representation (mass parameter) and spacetime dimension. For wavefunctions with finite spectral support (e.g., finite sums of de Sitter harmonics), group averaging converges for all $N$. This convergence is analytically controlled by the decay properties of the boost matrix elements and the angular quantum numbers; detailed formulas for the group elements and their decompositions (e.g., via Bargmann, Dixmier, Takahashi, and the functional analytic methods of Wolf and Wong) are employed to rigorously justify these results.

Once convergence holds, the group-averaged states define a Hilbert space of de Sitter-invariant states with a calculable, positive-definite inner product. This space is “new” in the sense that it does not coincide with the $G$-invariant sector of the usual Fock space, except for the vacuum.

5. Connection to Unitary Representation Theory

The group averaging procedure is underpinned by the classification and explicit construction of the unitary irreducible representations of the de Sitter (or more generally, Lorentz) group $\mathrm{SO}_0(D,1)$. These are organized into:

• Principal series: Associated with massive fields (“sufficiently heavy”).
• Complementary series: Related to lighter fields and “intermediate” masses; certain positivity constraints apply.
• Discrete series: Special cases, present in restricted dimensions or for specific quantum numbers.

Techniques from harmonic analysis—such as analytic continuation of Legendre functions, explicit basis expansions (Gelfand–Tsetlin, Wolf’s recursion, Wong’s hypergeometric parametrizations)—allow one to write down the group action explicitly and to reduce the group averaging integrals to tractable expressions. Understanding the decay and asymptotics of the matrix elements is essential for establishing the existence of invariant states beyond the vacuum via group averaging.

6. Implications for Quantum Gravity and Linearization Stability

In perturbative quantum gravity about a de Sitter background, linearization-stability constraints enforce that only states with vanishing de Sitter charges—that is, fully invariant states—couple consistently to gravity. As the Fock representations fail to contain invariant excitations beyond the vacuum, group averaging offers the only viable method for constructing physical state spaces compatible with these constraints.

The resulting physical Hilbert space is crucial for defining matter–gravity couplings, ensuring diffeomorphism invariance, and providing a foundation for the quantization of linearized gravity and higher-spin fields in de Sitter space. The method also ensures that observables and physical operators act within a well-defined invariant sector with a calculable inner product.

7. Extensions and Related Frameworks

The approach to group averaging described is robust and extends to free vector and tensor gauge fields in de Sitter space, with technical modifications to account for gauge constraints and representation types. Furthermore, the analytic and algebraic machinery—especially the explicit use of representation theory and harmonic analysis—is broadly applicable to other spacetimes with noncompact symmetry groups (e.g., anti-de Sitter, Minkowski) and underpins the algebraic classification of physical states, normal forms, and operator algebras in high-energy and mathematical physics.

---

In summary, the group averaging procedure for de Sitter free fields constructs the physical Hilbert space of invariant states by explicitly projecting Fock space states onto the de Sitter-invariant subspace using integration over the group. This process relies crucially on the analytic structure of unitary representations, the decay of boost matrix elements, and the spectral properties of de Sitter harmonics. The generalized framework ensures the correct imposition of linearization-stability constraints in quantum gravity and enables the consistent quantization of fields in curved spacetime backgrounds (0810.5163).