Extended Supersymmetry Gauge Dynamics: Scope and Structure

Four-dimensional $\mathcal N=2$ and $\mathcal N=4$ gauge theories have eight and sixteen real Poincaré supercharges, respectively. That extra symmetry is not merely a larger list of conserved charges: it relates fields that an $\mathcal N=1$ description would allow to have independent couplings, restricts vacuum geometry, and sharply limits quantum corrections. This page compares rigid theories in flat four-dimensional spacetime. Supergravity, twists, and the detailed evidence for exact S-duality lie outside this comparison.

Required background. Extended supersymmetry, R-symmetry, and central charges fixes the real-supercharge count and explains why a BPS bound is conditional on a charge sector. N=2 multiplets, Lagrangians, and vacuum branches supplies the vector- and hypermultiplet structure used below.

Helpful background. Extended superspace and off-shell limits distinguishes physical supersymmetry from the subgroup made manifest by a chosen formalism.

How the extra supercharges organize the fields

A massless gauge boson has two physical helicities. Supersymmetry requires the same number of on-shell bosonic and fermionic degrees of freedom in its multiplet. The minimal four-dimensional gauge multiplets therefore grow as follows:

multiplet	bosonic on-shell fields	bosonic degrees of freedom	fermionic on-shell fields
$\mathcal N=2$ vector	$A_\mu$ and one complex scalar $\phi$	$2+2=4$	two Weyl gaugini
$\mathcal N=2$ hyper	two complex scalars, equivalently four real scalars	$4$	two Weyl fermions
$\mathcal N=4$ vector	$A_\mu$ and six real scalars $X^I$	$2+6=8$	four Weyl gaugini

An $\mathcal N=4$ vector multiplet can consequently be regrouped in either of two useful ways,

$$\mathcal V_{\mathcal N=4} =\mathcal V_{\mathcal N=2}\oplus \mathcal H_{\mathcal N=2}^{\rm adj} =V_{\mathcal N=1}\oplus \Phi_1^{\rm adj}\oplus\Phi_2^{\rm adj}\oplus\Phi_3^{\rm adj}.$$

These are descriptions of the same fields, not three different theories. In $\mathcal N=1$ language, gauge invariance alone would permit several independent wave-function and superpotential couplings. The twelve hidden supercharges force the three chiral multiplets to be adjoint-valued and fix their cubic superpotential coupling relative to the gauge coupling. Dimensional reduction of ten-dimensional $\mathcal N=1$ Yang–Mills makes those relations manifest: the four-dimensional gauge field and six scalars are the ten components of one connection Brink, Schwarz, and Scherk 1977, §§2–3.

The R-symmetries also reveal the difference. The $\mathcal N=2$ algebra has $SU(2)_R\times U(1)_r$ at the classical level, while the $\mathcal N=4$ fields transform under

$$SU(4)_R\simeq Spin(6)_R.$$

The six $X^I$ form the vector $\mathbf6$ of $Spin(6)_R$, and the four left-handed gaugini form a chiral spinor $\mathbf4$. In a nonconformal $\mathcal N=2$ gauge theory the continuous $U(1)_r$ can be anomalous, so the algebraic or classical R-symmetry must not automatically be advertised as an exact quantum symmetry.

N=4 as a special N=2 gauge theory

Start with an $\mathcal N=2$ vector multiplet for a simple gauge algebra $\mathfrak g$ and add one massless adjoint hypermultiplet. The $\mathcal N=2$ Yukawa coupling is already tied to the gauge coupling. The adjoint matter and zero mass enlarge the symmetry to $\mathcal N=4$.

This embedding gives a quick perturbative check. In an $\mathcal N=2$ theory with full hypermultiplets in representations $R_h$, the one-loop coefficient is proportional to

$$b_0=2h^\vee-2\sum_h T(R_h)$$

Here $h^\vee$ is the dual Coxeter number. Normalize long roots to have length squared two and define the representation index by

$$\operatorname{tr}_{R}(T^aT^b)=T(R)\delta^{ab}.$$

Then $T({\rm adj})=h^\vee$; for example, $T(\mathbf N)=1/2$ for the fundamental of $SU(N)$. One adjoint hypermultiplet makes $b_0=0$. Equivalently, the $\mathcal N=1$ decomposition contains one vector and three adjoint chirals, giving $3T(G)-3T(G)=0$. This cancellation is an important consistency check, but a one-loop equation by itself is not an all-orders proof. The stronger perturbative and nonperturbative statements are separated carefully on finiteness, conformality, and the evidence ceiling.

The same embedding explains both the power and the limitations of lower-$\mathcal N$ notation. An $\mathcal N=1$ or $\mathcal N=2$ superspace action is often the most economical calculation tool, but any regulator, counterterm, or deformation must still respect the hidden supercharges if it is to describe the $\mathcal N=4$ theory.

Vacuum branches and their effective geometry

For a generic four-dimensional $\mathcal N=2$ gauge theory, the vacuum space can have distinct Coulomb, Higgs, and mixed branches.

• On a Coulomb branch, vector-multiplet scalars acquire expectation values. The local two-derivative Abelian action is governed by a holomorphic prepotential $\mathcal F(a)$, with

  $$\tau_{ij}(a)=\frac{\partial^2\mathcal F}{\partial a^i\partial a^j}.$$

  One-loop and instanton effects can make $\tau_{ij}$ vary, and electric–magnetic monodromy prevents one electric coordinate system from covering the entire branch.

• On a Higgs branch, hypermultiplet scalars solve a triplet of moment-map equations and are divided by the gauge group. The resulting hyperkähler geometry is not encoded by the Coulomb-branch prepotential.

• On a mixed branch, both types of expectation value survive, and the unbroken low-energy theory contains both vector- and hypermultiplet directions.

The pure $SU(2)$ Seiberg–Witten solution is the canonical example in which the local $\mathcal N=2$ prepotential must be assembled into global period and monodromy data Seiberg and Witten 1994, §§2–4.

For $\mathcal N=4$ SYM, the scalar potential vanishes exactly when

$$[X^I,X^J]=0, \qquad I,J=1,\ldots,6.$$

The six Hermitian scalars can then be conjugated simultaneously into a Cartan subalgebra $\mathfrak t$. Dividing by the remaining Weyl transformations gives one $Spin(6)_R$-covariant moduli space,

$$\mathcal M_{\rm vac} =\frac{\mathbb R^6\otimes\mathfrak t}{W}.$$

For $\mathfrak{su}(2)$ this is $\mathbb R^6/\mathbb Z_2$: a six-vector $\vec v$ and $-\vec v$ represent the same vacuum. Away from the origin the gauge algebra is $\mathfrak u(1)$; at the origin the off-diagonal vector multiplets become massless and restore $\mathfrak{su}(2)$. Choosing an $\mathcal N=2$ subalgebra selects particular complex and hypermultiplet coordinates inside this single space, but that bookkeeping choice does not split the full $\mathcal N=4$ vacuum space into unrelated quantum branches.

A comparison with the hypotheses visible

The following table concerns four-dimensional Lagrangian gauge theories with compact semisimple gauge algebra. It is not a classification of every non-Lagrangian SCFT with eight or sixteen supercharges.

question	$\mathcal N=2$ gauge theory	$\mathcal N=4$ SYM
local multiplets	vectors plus full or, when allowed, half hypermultiplets	one maximally supersymmetric vector multiplet
continuous local couplings	gauge couplings; masses and superpotential data constrained by $\mathcal N=2$	one complex gauge coupling for each simple factor; relative Yukawa and scalar couplings fixed
R-symmetry	$SU(2)_R\times U(1)_r$ classically; $U(1)_r$ may be anomalous	$SU(4)_R\simeq Spin(6)_R$
vacuum geometry	Coulomb, Higgs, and mixed branches	commuting-scalar Weyl quotient, with different lower-$\mathcal N$ slices
two-derivative Coulomb data	generally curved special Kähler geometry with quantum corrections	constant Abelian coupling and a flat metric on each smooth covering chart, followed by the Weyl quotient
beta function	the holomorphic Wilsonian gauge coupling has no perturbative corrections beyond one loop, but the coefficient need not vanish	one-loop cancellation plus much stronger all-orders constraints
protected information	prepotential data, Higgs-branch data, BPS indices, and selected chiral sectors, each with hypotheses	larger BPS sectors and stronger superconformal Ward identities, but not protection of every observable
electromagnetic duality	often a transition between infrared Abelian frames over one moduli space	a proposed equivalence between globally specified ultraviolet conformal theories

The last row is the conceptual turning point. In an asymptotically free $\mathcal N=2$ theory, a magnetic coordinate may become weakly coupled only near a singular point of the low-energy moduli space. Montonen–Olive duality instead proposes that electrically charged gauge particles and magnetically charged states belong to equivalent descriptions of a complete theory Montonen and Olive 1977, pp. 117–120. The latter claim is much stronger and requires global theory data.

Global data and protected data remain distinct

Extended supersymmetry fixes local interactions but does not determine the global theory. The gauge algebra $\mathfrak g$ specifies the adjoint fields. A compact global form, the allowed bundles, a maximal mutually local spectrum of genuine Wilson–’t Hooft lines, and any discrete theta datum determine which nonlocal probes and topological sectors exist. The N=4 theory card fixes the local normalization; line operators, global forms, and discrete theta data complete the global specification.

Protection is another independent question. Supersymmetry can fix a BPS mass after the central charge and an existing stable state have been specified. Superconformal shortening can fix a local operator dimension while leaving its four-point function dynamical. Neither statement implies that every formal charge is populated, every short state is stable, or every correlator is coupling independent.

These distinctions prevent three common overextensions:

1. a local Lagrangian does not determine the genuine-line spectrum;
2. a BPS inequality does not prove existence or stability of a state; and
3. a protected operator dimension does not fix unprotected OPE data.

Off-shell descriptions and duality frames

No conventional formulation is known with a finite number of auxiliary fields that is simultaneously local, Lorentz covariant, and makes all sixteen $\mathcal N=4$ supersymmetries close off shell. Siegel and Roček’s counting argument excludes a broad class of such finite completions under explicit component-field assumptions Siegel and Roček 1981, pp. 275–277. It does not exclude infinite auxiliary towers, extra bosonic coordinates, light-cone methods, or a formulation that makes only a subgroup manifest.

Likewise, an electric Lagrangian cannot display electric–magnetic duality as an ordinary local redefinition of the same potential. A magnetic potential becomes local only after changing polarization or dualizing an Abelian description. Boundary conditions and genuine lines remember which polarization was chosen.

Common pitfalls

Counting manifest rather than physical supersymmetry. An $\mathcal N=1$ superspace presentation of $\mathcal N=4$ SYM still describes sixteen physical supercharges only when the couplings and transformations obey the hidden-supersymmetry relations.

Calling every N=2 slice a separate N=4 branch. A chosen $\mathcal N=2$ subalgebra decomposes the six scalars into vector- and hypermultiplet coordinates. The full $Spin(6)_R$ rotates those choices inside one commuting-scalar quotient.

Replacing an exact statement by a slogan. “N=4 is finite” must be unpacked into perturbative beta-function, protected-sector, conformal, and nonperturbative claims with different evidence.

Exercises

1. Decompose the maximally supersymmetric multiplet

Starting from $A_\mu$ and six real adjoint scalars, recover the bosonic content of one $\mathcal N=2$ vector plus one adjoint hypermultiplet. Check the on-shell bosonic degree count.

Solution

Choose two real scalars and combine them into the complex scalar $\phi$ of the $\mathcal N=2$ vector multiplet. Together with the two helicities of $A_\mu$, this gives four bosonic degrees of freedom. The remaining four real scalars form the adjoint hypermultiplet, also with four bosonic degrees of freedom. Their sum is the $2+6=8$ bosonic degrees of the $\mathcal N=4$ vector multiplet.

2. Check the adjoint-hypermultiplet cancellation

Use $b_0=2h^\vee-2\sum_hT(R_h)$ and $T({\rm adj})=h^\vee$ to evaluate the $\mathcal N=2$ one-loop coefficient after adding one adjoint hypermultiplet. Why is the result not, by itself, an all-orders proof?

Solution

The vector contributes $2h^\vee$ and the adjoint hyper contributes $-2h^\vee$, so $b_0=0$. A one-loop coefficient constrains only the first perturbative term unless additional Ward identities and counterterm restrictions are supplied. The full $\mathcal N=4$ conclusion uses all sixteen supercharges, not merely this arithmetic cancellation.

3. Locate the singular point of the SU(2) quotient

Write an $\mathfrak{su}(2)$ vacuum as $X^I=v^I\sigma_3/2$. Show that the Weyl quotient identifies $\vec v\sim-\vec v$ and determine where the off-diagonal gauge bosons become massless.

Solution

The nontrivial Weyl reflection exchanges the two eigenvalues of every $X^I$, sending $\sigma_3\mapsto-\sigma_3$ and hence $\vec v\mapsto-\vec v$. An off-diagonal field has mass squared proportional to $\sum_I|\alpha(X^I)|^2=|\vec v|^2$ in the normalization used here. It is massive for $\vec v\neq0$ and becomes massless only at the orbifold fixed point $\vec v=0$, where the full $\mathfrak{su}(2)$ algebra is restored.

References

• Brink, Lars, John H. Schwarz, and Joël Scherk. “Supersymmetric Yang–Mills Theories.” Nuclear Physics B 121 (1977): 77–92. doi:10.1016/0550-3213(77)90328-5.
• Montonen, Claus, and David Olive. “Magnetic Monopoles as Gauge Particles?” Physics Letters B 72 (1977): 117–120. doi:10.1016/0370-2693(77)90076-4.
• Seiberg, Nathan, and Edward Witten. “Electric–Magnetic Duality, Monopole Condensation, and Confinement in $\mathcal N=2$ Supersymmetric Yang–Mills Theory.” Nuclear Physics B 426 (1994): 19–52; erratum 430 (1994): 485–486. doi:10.1016/0550-3213(94)90124-4.
• Siegel, Warren, and Martin Roček. “On Off-Shell Supermultiplets.” Physics Letters B 105 (1981): 275–277. doi:10.1016/0370-2693(81)90887-X.