Skip to content

Witten Index

The Witten index is a fermion-parity-graded supersymmetric trace that cancels paired positive-energy states and counts zero modes with signs.

Version
v1 · 2026-10-03 · History
Domain-specific #
13697
Domain group
Natural Sciences
Origin domain
Physics
Subdomain
Supersymmetric Quantum Mechanics → Physics
Aliases
Supersymmetric Index

Core Idea

The Witten index is the graded trace I(β)=Tr[(-1)^F e^{-βH}] in a supersymmetric quantum system. The factor (-1)^F gives bosonic states a plus sign and fermionic states a minus sign. Supersymmetry pairs positive-energy states of opposite parity at the same energy, so each pair contributes e^{-βE}−e^{-βE}=0. When the trace and spectrum permit this cancellation, only zero-energy states remain: I=n_B(0)−n_F(0). This makes the result independent of β in the standard discrete setting.[1][2]

The conclusion is asymmetric. A nonzero index guarantees at least one zero-energy state and obstructs complete supersymmetry breaking under the stated assumptions. An index of zero is inconclusive: it could mean no zero modes, or equally many bosonic and fermionic zero modes. Treating it as a headcount rather than a signed difference reverses that logic.[1][2]

Structural Signature

Sig role-phrases:

  • Supersymmetric Hilbert space: carries a Hamiltonian and supercharges relating sectors.
  • Fermion-parity grading: weights bosonic and fermionic states oppositely.
  • Positive-energy pairing: makes excited-state contributions cancel level by level.
  • Zero-energy sector: unpaired or differently counted ground states supply the remaining integer.
  • Spectral/trace conditions: discrete, well-controlled cases justify the formal supertrace; continuum, boundaries or limiting volumes need care.[1][2]

Condensed: graded trace + supercharge pairing → excited-state cancellation → signed zero-mode count.

What It Is Not

It is not the ordinary thermal partition function Tr(e^{-βH}), which adds bosonic and fermionic contributions rather than subtracting them. It is not the total number of supersymmetric vacua, because opposite-parity ground states can cancel. It is not a universal yes/no detector for supersymmetry breaking: nonzero is decisive in one direction, zero is not. A formal sum over an uncontrolled continuous spectrum should not be called a protected index without specifying boundary and regularization conditions.[2][1]

Scope of Application

In elementary supersymmetric quantum mechanics, the supercharge maps positive-energy bosonic states to fermionic partners and vice versa. Tong's derivation turns the graded trace into the difference of zero-energy dimensions. A tiny paired-spectrum calculation already shows the logic and why finite-temperature-looking β drops out.[2]

In a geometric supersymmetric sigma model whose states are differential forms on a compact manifold, fermion parity corresponds to even or odd form degree. The index becomes the alternating sum of Betti numbers, the Euler characteristic. Tong works out spheres and tori: S² has b₀=b₂=1 and index 2, while S¹ has b₀=b₁=1 and index 0 despite having zero modes of both parities. This geometric case demonstrates why the inverse implication fails.[3]

Witten's original work applies index reasoning to supersymmetry breaking in field theories, but the present entry does not infer a vacuum count for an arbitrary gauge group or extrapolate compact quantum-mechanical proofs to infinite-volume field theories without additional conditions.[1]

Clarity

The β in the formula resembles inverse temperature because the trace includes e^{-βH}. The minus sign from parity changes its purpose: it is a cancellation device, not a positive statistical weight. At E>0, supersymmetry supplies equal-energy opposite-parity partners and the β dependence disappears pairwise. At E=0, supercharges need not generate a distinct partner, leaving the signed count. “Index = 0” therefore means net cancellation, not “empty ground-state sector.”[2]

Manages Complexity

Directly enumerating all energy eigenstates can be impossible. The index throws away paired excitations and focuses attention on the exceptional zero sector. It may also survive suitable continuous deformations even when individual excited levels move, because their opposite signs continue to cancel. This compression has a price: it intentionally discards the total number of balanced opposite-parity zero states. One must use other information to distinguish those possibilities and check that states do not escape through a continuum or boundary during a deformation.[1][2]

Abstract Reasoning

Consider a formal discrete supersymmetric spectrum with one bosonic state at E=0 and, at each positive energy E_j, one boson and one fermion. Its index is +1 + Σ_j(e^{-βE_j}−e^{-βE_j}) = 1. The conclusion is not that the excited spectrum is absent, but that its graded contribution is zero. Replace the lone zero boson with one zero boson and one zero fermion: the index becomes zero while two supersymmetric ground states remain. This transparent constructed example isolates the theorem without pretending to be a laboratory spectrum.[2]

For the geometric model, even-degree and odd-degree harmonic forms take the roles of bosonic and fermionic zero modes. On S², degree zero and degree two each contribute +1, yielding 2. On S¹, degree zero contributes +1 and degree one −1, yielding 0. Both manifolds have supersymmetric zero states, yet their indices differ because their parity balances differ.[3]

Knowledge Transfer

The graded-cancellation argument travels from a simple discrete spectrum to supersymmetric geometry and certain quantum field theories. The transfer requires an actual fermion-parity grading, supercharge pairing and a meaningful trace; merely writing (-1)^F next to an unrelated Hamiltonian does not establish an invariant. The Euler-characteristic identification depends on the specific de Rham sigma-model construction, not on every supersymmetric system. A gauge-theory index can have further global and boundary subtleties beyond these examples.[1][3]

Examples

Constructed discrete paired spectrum

Let one bosonic zero state coexist with one boson–fermion pair at energy E=1 and another at E=2. Then I(β)=1+(e^{-β}−e^{-β})+(e^{-2β}−e^{-2β})=1 for every β in this finite model. Adding a fermionic zero state would make I=0 without eliminating either zero state. This is a worked algebraic instance of Tong's cancellation, not a source-reported physical system.[2]

Mapped back: the listed states are the supersymmetric Hilbert-space model; plus/minus signs are fermion parity; the E=1,2 pairs cancel; the zero-sector imbalance determines I; finiteness supplies unproblematic trace conditions.

S² and S¹ de Rham sigma models

In Tong's geometric construction, the index equals Euler characteristic. S² has one harmonic 0-form and one harmonic 2-form, both even, so the index is 2. S¹ has one harmonic 0-form and one harmonic 1-form, so the index is 1−1=0. The circle is the decisive counterexample to interpreting zero index as no supersymmetric ground state.[3]

Mapped back: differential forms provide the supersymmetric states; form-degree parity supplies the sign; positive modes cancel; harmonic forms comprise the zero sector; compactness keeps this example within controlled discrete-spectrum conditions.

Structural Tensions

No intrinsic opposed-cost tension is established within the defined Witten index. Signed cancellation is the invariant's mechanism and an interpretation limit, not a choice that trades away one independent design goal: zero may mean balanced bosonic and fermionic vacua, as in the S¹ model, rather than no vacua. Spectral escape is a hypothesis boundary: a finite paired-level argument cannot simply be carried into an infinite-volume or continuous-spectrum model without checking trace and boundary conditions. Diagnostic: for a zero value, are even and odd zero modes separately known? Diagnostic: are discreteness, boundary behavior and regularization controlled along the proposed deformation?[1][2][3]

Structural–Framed Character

The index sits near the structural end: graded Hilbert space, supersymmetric pairing and zero-energy difference determine the mathematics. Its evaluative weight concerns what a physicist may infer—nonzero certifies a supersymmetric ground state under assumptions, zero does not certify breaking. Human theory-building chooses a Hamiltonian, boundary conditions and regulator; no institution creates the cancellation, though the eponym and notation stabilize its vocabulary. The term travels literally from quantum mechanics into certain geometric and field-theoretic models when the same graded trace is defined. Importing it to any signed statistic without supercharge pairing is analogy, not recognition. Its character: a protected but deliberately incomplete supersymmetric zero-mode invariant whose power depends on spectral conditions.

Structural Core vs. Domain Accent

The skeletal relation is paired opposite-sign contributions cancel, leaving an exceptional residual count. No verified live prime for that broader relation has been established; it would require non-SUSY admissions and its own boundaries. The domain-bound mechanism is fermion parity, supercharges, Hamiltonian zero energy and trace regularization. Witten Index fails the prime bar because these physical/mathematical structures, not a generic “cancellation” motif, determine when the integer exists and what it proves.

This entry is a kind of Mathematical Invariant.

Mathematical Invariant is the strict parent under the stated trace-class and admissible-deformation conditions: the Witten index is preserved under the qualifying deformations. Euler characteristic is the value of the index in the specified sigma model, not a blanket synonym. Supersymmetry breaking is an inference target, not the index's definition.

Relationships to Other Abstractions

Local relationship map for Witten IndexParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Witten IndexDOMAINDomain-specific abstraction: Mathematical Invariant — is a kind ofMathematicalInvariantDOMAIN

Current abstraction Witten Index Domain-specific

Parents (1) — more general patterns this builds on

  • Witten Index is a kind of Mathematical Invariant Domain-specific

    The defined supersymmetric graded trace is a value invariant under specified admissible deformations.

Hierarchy path (1) — routes to 1 parentless root

Neighborhood in Abstraction Space

Witten Index sits in a sparse region of the domain-specific corpus (67th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Statistical Mechanics & Particle Phenomena (15 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With

An ordinary partition function has no parity subtraction. A raw ground-state count has no sign. A zero index is compatible with unbroken supersymmetry. Other indices, including mathematical operator indices and refined field-theoretic supersymmetric indices, may be related but require their own definitions and conditions; this entry does not merge them.

References

[1] Edward Witten, “Constraints on Supersymmetry Breaking” (1982), original index argument, course-hosted copy. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h

[2] David Tong, Supersymmetric Quantum Mechanics, lecture notes, §1.2.2 (printed p. 14), graded trace and positive-energy cancellation. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j

[3] David Tong, Supersymmetric Quantum Mechanics, lecture notes, §3.1.3 (printed pp. 80–82; chapter extract of the same notes), sigma-model Euler characteristic and sphere/torus examples. registry ↩a ↩b ↩c ↩d ↩e