Localization and Protected Observables

Supersymmetry can make selected quantities exactly calculable even when a theory is interacting. Localization is one way this happens: a symmetry allows us to deform an integral while keeping its value fixed, until its contributing configurations and fluctuations become manageable. The symmetry, the allowed deformation and the definition of the integral must all be specified.

This volume begins with an ordinary finite integral. Its opening method applies to readers approaching either three- or four-dimensional theories. Curved-space spinors, matrix models and special functions are not prerequisites for that first calculation.

Start with an integral you can check completely

A finite-dimensional localization example uses two real variables and two Grassmann generators. The preparation is Grassmann calculus: ordered products, left derivatives and a stated integration measure.

Follow the odd operation until its square becomes a rotation. Construct an invariant deformation, differentiate the integral, and calculate the boundary term. Then evaluate the answer both directly and from the Gaussian neighborhood that survives a large deformation. The two calculations agree, including the fermionic sign and the normalization.

Your first checkpoint: explain why adding only the positive bosonic part of the deformation changes the answer. Calculate the error caused by a finite radial cutoff, and identify which completed insertion has a protected expectation value.

Separate the method from the theory-specific calculation

In a field theory, the integration variables become fields. The chosen supercharge can square to a spacetime symmetry together with internal transformations. Its action must be compatible with the background, boundary conditions, gauge fixing and regulator.

A calculation then needs both the contributing field configurations and the fluctuations around them. A continuous family of configurations leaves an integral over that family; a single saddle cannot be assumed in advance. The resulting determinants also need their regularization and phase conventions. Deformation independence is the reason a useful limit computes the original protected quantity, rather than merely an approximation to it.

Localization in a field theory derives the regulated Ward identity and separates the saddle configurations from their fluctuations. Its optional sphere section exhibits an exact deformation, its integration contour and a complete finite set of field modes.

The free chiral determinant completes that sphere calculation. Derive the scalar and spinor multiplicities, cancel their paired factors, and specify the finite subtraction that gives the normalized answer. Opposite real masses check its magnitude while leaving the individual phases visible.

A first gauge-theory matrix integral combines two charged determinants with the remaining gauge scalar and the FI source. Derive the one-flavor SQED measure, evaluate a convergent trial-charge example, and check why a finite matter-mode cutoff cannot be integrated over the whole scalar line.

The three-dimensional action lessons supply concrete matter and gauge actions for the sphere examples. The four-dimensional foundations provide an independent theory entrance. Use the additional action preparation when specializing the general method to those examples.

Know what the observable asks

A partition function depends on its manifold, sources and quantum prescription. Differentiating it with respect to a source can probe a current response, but explicit second source derivatives and local contact terms can contribute. Sources and currents introduces that distinction before supersymmetric refinements.

R-mixing and F-maximization uses the sphere free energy to select a superconformal current. Begin with a free chiral, then vary both flavor directions of SQED and compare its integral with the XYZ description. A disappearing interaction illustrates why an infrared symmetry can invalidate a restricted ultraviolet trial family.

Source derivatives and contact terms follows both source insertions through a second derivative. Calculate a compact background counterterm, identify the response that survives its integer ambiguity, and track the complex phase in a SQED–XYZ comparison.

Superconformal currents and two-point functions fixes the separated-point normalization. Derive the conserved tensor, translate it into a source response, and check its coefficient in a free chiral theory before connecting it to the stress tensor.

What squashing the sphere measures then follows the complete supersymmetric source family. Evaluate the integrated current correlator, perform its required subtraction and check the same response through a convergent momentum integral.

Reproduce the free-chiral response computes the same coefficient from the matter determinant. Track the physical R-charge as the sphere changes, evaluate the round-point curvature and check the finite subtraction with a separate spectral sum.

Universal observables and scheme dependence makes the comparison precise: identify which local terms can change the finite answer and which response survives them. A normalized loop and an unnormalized sphere partition function need different checks.

Count protected states with an index

What survives in a supersymmetric index? asks a different question: it is a trace with signs and selected commuting charges. Choose the radial supercharge, derive the allowed fugacities and follow a positive pair through its cancellation. Its protection does not make it an unsigned count of every state. Likewise, an exact value for a chosen supersymmetric observable does not determine every correlation function or every coupling of a theory.

Count a free chiral multiplet by hand makes the signed trace concrete. Derive its scalar and conjugate-fermion modes, list their first products, and explain how a zero can coexist with a vacuum.

Gauge projection and an index check of SQED–XYZ then imposes electric neutrality in every magnetic sector. Derive the vacuum charges, recover the meson and monopoles, and change a fixed background flux to test the source prescription.

Keep these distinctions visible as the calculations become more elaborate. For each result, be able to name the protected quantity, the symmetry that protects it, the parameters it may depend on, and the physical information that remains outside its scope. The guided learning paths connect these methods to the sphere, duality and four-dimensional examples.