Graded Spacetime Symmetry and the Supersymmetry Theorems

Supersymmetry is the nontrivial finite-dimensional graded extension of relativistic spacetime symmetry allowed by the standard particle S-matrix assumptions. In four-dimensional Lorentzian QFT, its odd generators are spinors $Q^I_\alpha$ and their adjoints; their anticommutator produces momentum and, for extended supersymmetry, scalar central charges. This is a classification of possible symmetry algebras—not an existence theorem for an interacting QFT, and not a statement about theories outside the assumptions.

Required background. Lie groups, Lie algebras, and adjoint maps supplies Jacobi identities and representations, while Lorentz fields and Poincaré particles supplies the Lorentz and little-group classification used below.

Helpful background. Projective actions and central extensions distinguishes a central extension from an ordinary internal generator. Poincaré covariance and the spectrum condition explains the positive-energy input.

Graded symmetry and its Jacobi identity

A Lie superalgebra is a $\mathbb Z_2$-graded vector space $\mathfrak g=\mathfrak g_{\bar0}\oplus\mathfrak g_{\bar1}$ with bracket

$$[X,Y\}=XY-(-1)^{\lvert X\rvert\lvert Y\rvert}YX.$$

The bracket is a commutator unless both entries are odd, in which case it is an anticommutator. Associativity implies the graded Jacobi identity

$$(-1)^{\lvert X\rvert\lvert Z\rvert}[X,[Y,Z\}\} +(-1)^{\lvert Y\rvert\lvert X\rvert}[Y,[Z,X\}\} +(-1)^{\lvert Z\rvert\lvert Y\rvert}[Z,[X,Y\}\}=0.$$

The even part contains the Poincaré algebra and any internal symmetries. Odd generators reverse fermion parity. Lorentz covariance requires the odd subspace to be a sum of finite-dimensional Lorentz representations. The decisive point is that positivity constrains $\{Q,Q^\dagger\}$, while the graded Jacobi identities constrain every candidate term in the algebra. A detailed construction from graded group parameters is given in Weinberg 2000, § 25.1, pp. 25–29.

For example, once translations commute with the odd generators,

$$[P_\mu,Q]=0,$$

the Jacobi identity for $(Q,Q^\dagger,P_\mu)$ forbids a Lorentz generator $M_{\rho\sigma}$ on the right-hand side of $\{Q,Q^\dagger\}$: $M_{\rho\sigma}$ does not commute with $P_\mu$. Lorentz covariance then leaves momentum and permitted internal or topological charges in the relevant spinor bilinears.

What the extension theorems assume

The Coleman–Mandula theorem classifies ordinary bosonic symmetries of an S-matrix. The Haag–Łopuszański–Sohnius analysis repeats the problem for a $\mathbb Z_2$-graded algebra and identifies supersymmetry as the surviving nontrivial option. The original statements are technical; the following table records the assumptions that matter when applying them. Coleman and Mandula formulate the connected S-matrix result explicitly in Coleman and Mandula 1967, pp. 1251–1256, and the graded extension is developed in Haag, Łopuszański, and Sohnius 1975, pp. 257–274.

Input	Role in the conclusion	What fails without it
Lorentzian Poincaré invariance	Classifies generators by finite-dimensional Lorentz representations	Curved backgrounds, nonrelativistic systems, and conformal defects have different symmetry algebras
A particle S-matrix with stable asymptotic states	Makes “symmetry” an operator commuting with scattering	A CFT without an ordinary particle S-matrix is not covered
Nontrivial scattering at generic kinematics	Excludes a free or kinematically degenerate S-matrix with accidental higher symmetries	Free theories can carry much larger symmetry algebras
Analyticity of elastic amplitudes in the required domain	Turns momentum dependence into strong algebraic constraints	Integrable or singular kinematics can evade the proof
Finitely many particle types below any fixed mass	Excludes uncontrolled infinite towers	String-like spectra and some higher-spin limits lie outside the argument
Additive action on multiparticle states and sufficiently regular momentum-space kernels	Connects one-particle generators to scattering	Nonlocal, higher-form, and non-additive actions need a different theorem
Positive-energy unitary Hilbert space	Makes $\{Q,Q^\dagger\}$ positive and fixes the momentum coefficient	Indefinite-metric or nonunitary representations need not obey the same bound

A mass gap is a common sufficient setting for the particle analysis, but it should not be recited as a universal hypothesis. Massless theories require separate infrared care: even the usual charged-particle S-matrix can fail to exist without inclusive or dressed states. Weinberg gives a hypothesis-conscious proof and discusses the massless limitations in Weinberg 2000, § 24.1 and Appendix 24.B, pp. 1–4 and 12–22.

From Lorentz covariance to super-Poincaré form

Work first in four-dimensional Minkowski space. A finite set of odd charges decomposes into Lorentz representations $(j_L,j_R)$. Positivity makes the anticommutator of a nonzero component with its adjoint nonzero. The even generators allowed by the bosonic theorem transform only as

• a vector $(\tfrac12,\tfrac12)$ for $P_\mu$;
• $(1,0)\oplus(0,1)$ for $M_{\mu\nu}$; and
• scalars $(0,0)$ for internal charges.

If an odd generator transformed as $(j_L,j_R)$, its product with its adjoint would contain $(j_L+j_R,j_L+j_R)$. Matching the allowed even representations forces $j_L+j_R\leq\tfrac12$. A Lorentz-scalar odd generator would map bosonic and fermionic one-particle states without changing their spin, contradicting the spin–statistics assignment under the theorem’s local, unitary particle assumptions. The surviving possibilities are therefore $(\tfrac12,0)$ and $(0,\tfrac12)$. This is the key representation-theoretic step in Weinberg 2000, § 25.2, pp. 29–35.

After a positive change of basis among $\mathcal N$ copies, the point-particle algebra has the schematic form

$$\begin{aligned} \{Q^I_\alpha,\bar Q_{\dot\beta J}\} &=2\sigma^\mu_{\alpha\dot\beta}P_\mu\,\delta^I{}_J,\\ \{Q^I_\alpha,Q^J_\beta\} &=2\epsilon_{\alpha\beta}Z^{IJ},\\ [P_\mu,Q^I_\alpha]&=0, \end{aligned}$$

where $Z^{IJ}=-Z^{JI}$. The graded Jacobi identities imply that the scalar $Z^{IJ}$ commutes with the supertranslations. Internal automorphisms can rotate the matrix $Z^{IJ}$, so a fixed nonzero charge sector retains only its stabilizer; a generator that is central in the entire enlarged algebra must also be invariant under that action. The exact two-component convention and Hermitian conjugation are fixed on the four-dimensional N=1 algebra; extended and tensorial charges are separated on the extended algebra page.

The same logic works in other Lorentzian dimensions, but the available spinor reality conditions and bilinears change. Nahm’s dimension-independent classification is the structural source Nahm 1978, pp. 149–166; the practical translation is developed on supersymmetry across dimensions and signatures.

Boundaries and genuine loopholes

The theorem’s conclusion should be weakened exactly when its inputs are weakened:

Conformal symmetry. A conformal theory has $K_\mu$ and $D$ in addition to $P_\mu$ and $M_{\mu\nu}$. Odd conformal charges $S$ can therefore accompany $Q$. This produces a superconformal algebra, not a counterexample to the S-matrix theorem; the bounded handoff is superconformal algebras and shortening.

Extended objects and boundaries. Strings, walls, and branes can carry tensorial surface charges. These commute with translations but transform under Lorentz transformations, so they are not central in the full super-Poincaré algebra. Their presence violates the point-particle/additivity setup rather than the graded Jacobi identity.

Two-dimensional integrability and higher-spin symmetry. Infinitely many conserved charges, special kinematics, or an infinite particle tower violate finiteness or generic-scattering assumptions. Such systems require their own classification.

Gauge redundancy and spontaneous breaking. Gauge transformations are redundancies, not ordinary global S-matrix generators. A spontaneously broken supersymmetry may still be an exact algebraic symmetry of the theory even though the vacuum is not annihilated by $Q$; the representation is then not organized into degenerate vacuum multiplets.

Algebra versus QFT existence. Passing Lorentz covariance and Jacobi identities proves only that a candidate algebra is allowed. One must still construct operators on a common domain, a positive-energy representation, local observables, and—if dynamics is claimed—an anomaly-free interacting QFT. The theorem-first status of stronger existence claims belongs to theorem-first claim records.

A compact consistency check

Given a proposed odd charge $\mathcal Q$, proceed in this order:

1. Name the spacetime dimension, signature, asymptotic-state setting, and real form.
2. Determine the Lorentz representation and adjoint of $\mathcal Q$.
3. Decompose the symmetric spinor product into Lorentz tensors.
4. Retain only charges admitted by the physical setting: momentum, scalar central charges, or explicitly justified tensorial surface charges.
5. Check every graded Jacobi identity, especially $(P,Q,Q)$, $(M,Q,Q)$, and $(T,Q,Q)$ for an internal generator $T$.
6. Evaluate $\langle\psi|\{A,A^\dagger\}|\psi\rangle$ for arbitrary linear combinations $A$ of supercharges. A negative value rules out a unitary positive-energy representation.
7. State the theorem ceiling: “allowed algebra,” “represented on states,” and “realized by an interacting local QFT” are three different conclusions.

Common pitfalls

Calling supersymmetry an exception with no qualifications. It is the graded extension allowed after replacing an ordinary Lie algebra by a Lie superalgebra under a related assumption set. It does not invalidate Coleman–Mandula’s conclusion about ordinary bosonic generators.

Treating every extra term as a central charge. A true central charge commutes with Lorentz transformations. A $p$-form brane charge transforms as a tensor and is central only in the narrower supertranslation sense.

Inferring dynamics from closure. A symbolic Jacobi check cannot establish locality, unitarity, anomaly freedom, or the existence of an interacting S-matrix.

Check your understanding

Why can $M_{\mu\nu}$ not appear in $\{Q,Q^\dagger\}$ once $[P_\rho,Q]=0$?

Answer

Apply the graded Jacobi identity to $(P_\rho,Q,Q^\dagger)$. Its two terms containing $[P,Q]$ vanish, so $[P_\rho,\{Q,Q^\dagger\}]=0$. Momentum commutes with itself and with scalar central charges, but $[P_\rho,M_{\mu\nu}]\neq0$. Therefore a Lorentz-generator term is forbidden.

References

• Sidney Coleman and Jeffrey Mandula, “All Possible Symmetries of the S Matrix,” Physical Review 159 (1967), 1251–1256, DOI.
• Rudolf Haag, Jan T. Łopuszański, and Martin Sohnius, “All Possible Generators of Supersymmetries of the S-Matrix,” Nuclear Physics B 88 (1975), 257–274, DOI.
• Werner Nahm, “Supersymmetries and Their Representations,” Nuclear Physics B 135 (1978), 149–166, DOI.
• Steven Weinberg, The Quantum Theory of Fields, Volume III: Supersymmetry, Cambridge University Press (2000), Chapters 24–25 and 32, DOI.