Skip to content

Superspace and Supertranslations

Superspace realizes supersymmetry as ordinary geometry on a space with commuting coordinates xμx^\mu and anticommuting coordinates θα,θˉα˙\theta^\alpha,\bar\theta^{\dot\alpha}. The mixed term in a finite supertranslation is forced by the anticommutator {Qα,Qˉα˙}∝Pμ\{Q_\alpha,\bar Q_{\dot\alpha}\}\propto P_\mu: two fermionic translations differ by a spacetime translation. Deriving that group law fixes the differential generators and prevents the sign table for QQ, Qˉ\bar Q, DD, and Dˉ\bar D from becoming a collection of unrelated conventions.

Required background. The Four-Dimensional N=1 Super-Poincaré Algebra supplies the algebra. Graded Algebra, Grassmann Variables, and Berezin Integration supplies left differentiation and the graded product rule.

Helpful background. Lie Groups, Lie Algebras, the Exponential Map, and the Adjoint Action supplies the Baker–Campbell–Hausdorff viewpoint used for the finite composition law.

Supertranslation coordinates and their dimensions

Section titled “Supertranslation coordinates and their dimensions”

Work in four-dimensional Lorentzian spacetime with

σμ=(1,σ),σˉμ=(1,−σ),Pμ=i∂μ.\sigma^\mu=(\mathbf1,\boldsymbol\sigma), \qquad \bar\sigma^\mu=(\mathbf1,-\boldsymbol\sigma), \qquad P_\mu=i\partial_\mu.

The superspace point is

zM=(xμ,θα,θˉα˙),z^M=(x^\mu,\theta^\alpha,\bar\theta^{\dot\alpha}),

with Grassmann parities

∣xμ∣=0,∣θα∣=∣θˉα˙∣=1.\lvert x^\mu\rvert=0, \qquad \lvert\theta^\alpha\rvert =\lvert\bar\theta^{\dot\alpha}\rvert=1.

Because [Pμ]=1[P_\mu]=1 and [Qα]=1/2[Q_\alpha]=1/2,

[xμ]=−1,[θα]=[θˉα˙]=−12.[x^\mu]=-1, \qquad [\theta^\alpha]=[\bar\theta^{\dot\alpha}]=-\frac12.

Barred coordinates are Lorentzian conjugates of unbarred coordinates on real Minkowski superspace. They remain independent generators while doing Grassmann algebra; conjugation reverses the order of odd factors.

Throughout this page the odd derivatives act from the left. For homogeneous superfields FF and GG,

∂α(FG)=(∂αF)G+(−1)∣F∣F(∂αG),∂ˉα˙(FG)=(∂ˉα˙F)G+(−1)∣F∣F(∂ˉα˙G),\begin{aligned} \partial_\alpha(FG) &=(\partial_\alpha F)G +(-1)^{\lvert F\rvert}F(\partial_\alpha G),\\ \bar\partial_{\dot\alpha}(FG) &=(\bar\partial_{\dot\alpha}F)G +(-1)^{\lvert F\rvert}F(\bar\partial_{\dot\alpha}G), \end{aligned}

where ∣F∣∈{0,1}\lvert F\rvert\in\{0,1\} is the Grassmann parity. In particular,

∂α(θβF)=δαβF−θβ∂αF.\partial_\alpha(\theta^\beta F) =\delta_\alpha{}^\beta F -\theta^\beta\partial_\alpha F.

The minus sign comes from moving the odd derivative past the odd factor θβ\theta^\beta; it occurs whether FF is even or odd. By contrast, the ordinary spacetime derivative is even and obeys the ungraded product rule. Switching the side of the odd derivatives changes a package of signs, not just this example.

Choose the active left action of a supertranslation (a,ϵ,ϵˉ)(a,\epsilon,\bar\epsilon) on a point to be

θ′=θ+ϵ,θˉ′=θˉ+ϵˉ,x′μ=xμ+aμ−iϵσμθˉ+iθσμϵˉ.\begin{aligned} \theta'&=\theta+\epsilon,\\ \bar\theta'&=\bar\theta+\bar\epsilon,\\ x'^\mu &=x^\mu+a^\mu -i\epsilon\sigma^\mu\bar\theta +i\theta\sigma^\mu\bar\epsilon. \end{aligned}

The mixed shift is bilinear because an odd parameter cannot shift an even coordinate by itself. Apply first (a1,ϵ1,ϵˉ1)(a_1,\epsilon_1,\bar\epsilon_1) and then (a2,ϵ2,ϵˉ2)(a_2,\epsilon_2,\bar\epsilon_2). Direct substitution gives

ϵ21=ϵ1+ϵ2,ϵˉ21=ϵˉ1+ϵˉ2,a21μ=a1μ+a2μ+i(ϵ1σμϵˉ2−ϵ2σμϵˉ1).\begin{aligned} \epsilon_{21}&=\epsilon_1+\epsilon_2,\\ \bar\epsilon_{21}&=\bar\epsilon_1+\bar\epsilon_2,\\ a_{21}^\mu &=a_1^\mu+a_2^\mu +i\left( \epsilon_1\sigma^\mu\bar\epsilon_2 -\epsilon_2\sigma^\mu\bar\epsilon_1 \right). \end{aligned}

Reversing the two transformations changes the sign of the bilinear term. Subtracting the 1-then-2 ordering from the 2-then-1 ordering, as required by the field convention [δ1,δ2]=δ1δ2−δ2δ1[\delta_1,\delta_2]=\delta_1\delta_2-\delta_2\delta_1, gives the translation parameter

ξμ=2i(ϵ2σμϵˉ1−ϵ1σμϵˉ2).\xi^\mu =2i\left( \epsilon_2\sigma^\mu\bar\epsilon_1 -\epsilon_1\sigma^\mu\bar\epsilon_2 \right).

This is the finite-coordinate form of

{Qα,Qˉα˙}=2σαα˙μPμ.\{Q_\alpha,\bar Q_{\dot\alpha}\} =2\sigma^\mu_{\alpha\dot\alpha}P_\mu.

The truncation of the Baker–Campbell–Hausdorff series is exact: the odd-odd bracket is the central translation within the supertranslation subgroup, and further nested brackets vanish. Salam and Strathdee introduced this systematic superfield realization in Salam and Strathdee 1974, pp. 477–482. In four-component notation, Weinberg 2000, § 26.2, pp. 60–61 constructs the corresponding infinitesimal superspace differential generator and checks its algebra; the finite law above follows by composing the declared coordinate action.

Differential supercharges from the coordinate action

Section titled “Differential supercharges from the coordinate action”

Let F(x,θ,θˉ)\mathscr F(x,\theta,\bar\theta) be a scalar superfield. Expanding its pullback under the infinitesimal transformation at fixed original coordinates yields

δF=(ϵαQα+ϵˉα˙Qˉα˙+aμ∂μ)F,\delta\mathscr F = \left( \epsilon^\alpha Q_\alpha +\bar\epsilon_{\dot\alpha}\bar Q^{\dot\alpha} +a^\mu\partial_\mu \right)\mathscr F,

with

Qα=∂∂θα−iσαα˙μθˉα˙∂μ,Qˉα˙=−∂∂θˉα˙+iθασαα˙μ∂μ.\begin{aligned} Q_\alpha &= \frac{\partial}{\partial\theta^\alpha} -i\sigma^\mu_{\alpha\dot\alpha} \bar\theta^{\dot\alpha}\partial_\mu,\\ \bar Q_{\dot\alpha} &= -\frac{\partial}{\partial\bar\theta^{\dot\alpha}} +i\theta^\alpha\sigma^\mu_{\alpha\dot\alpha}\partial_\mu. \end{aligned}

The minus sign in the leading term of Qˉ\bar Q implements Lorentzian conjugation together with left differentiation. It does not reverse the desired shift of the barred coordinate. Raising the operator index with ϵα˙β˙\epsilon^{\dot\alpha\dot\beta} gives

ϵˉα˙Qˉα˙θˉβ˙=−ϵˉα˙ϵα˙β˙=+ϵˉβ˙.\bar\epsilon_{\dot\alpha} \bar Q^{\dot\alpha}\bar\theta^{\dot\beta} =-\bar\epsilon_{\dot\alpha} \epsilon^{\dot\alpha\dot\beta} =+\bar\epsilon^{\dot\beta}.

The final plus sign uses ϵˉα˙ϵα˙β˙=−ϵˉβ˙\bar\epsilon_{\dot\alpha}\epsilon^{\dot\alpha\dot\beta} =-\bar\epsilon^{\dot\beta}. Thus the leading minus in Qˉα˙\bar Q_{\dot\alpha} and the antisymmetry of the spinor metric are both required by the active coordinate shift δθˉ=ϵˉ\delta\bar\theta=\bar\epsilon. Acting on all three coordinates checks the construction:

xμθβθˉβ˙Qα−iσαγ˙μθˉγ˙δαβ0Qˉα˙+iθγσγα˙μ0−δα˙β˙\begin{array}{c|ccc} &x^\mu&\theta^\beta&\bar\theta^{\dot\beta}\\ \hline Q_\alpha& -i\sigma^\mu_{\alpha\dot\gamma}\bar\theta^{\dot\gamma}& \delta_\alpha{}^\beta&0\\ \bar Q_{\dot\alpha}& +i\theta^\gamma\sigma^\mu_{\gamma\dot\alpha}& 0&-\delta_{\dot\alpha}{}^{\dot\beta} \end{array}

Using the graded product rule gives

{Qα,Qβ}=0,{Qˉα˙,Qˉβ˙}=0,{Qα,Qˉα˙}=2iσαα˙μ∂μ=2σαα˙μPμ.\begin{aligned} \{Q_\alpha,Q_\beta\}&=0,\\ \{\bar Q_{\dot\alpha},\bar Q_{\dot\beta}\}&=0,\\ \{Q_\alpha,\bar Q_{\dot\alpha}\} &=2i\sigma^\mu_{\alpha\dot\alpha}\partial_\mu =2\sigma^\mu_{\alpha\dot\alpha}P_\mu. \end{aligned}

Terms containing θθˉ ∂μ∂ν\theta\bar\theta\,\partial_\mu\partial_\nu cancel because their Grassmann coefficient is antisymmetric while the spacetime derivatives commute. This explicit check is more reliable than importing one generator from a source with a different PμP_\mu convention.

The supertranslation group acts on itself from both sides. Infinitesimal left-invariant vector fields generate one copy of the algebra; right-invariant vector fields commute with that action and become the supercovariant derivatives on the next page. Calling both sets “fermionic translations” without specifying the side obscures their crucial relative sign.

For a field transformation there is also an active-versus-passive choice. This page uses an active coordinate action and the induced pullback written above. A passive coordinate change reverses the sign in δF\delta\mathscr F while leaving the invariant algebra unchanged. A convention translation must change the coordinate shift and differential operators together.

Exterior calculus on a supermanifold needs one more sign convention. We use a bigrading by form degree pp and Grassmann parity ∣⋅∣\lvert\cdot\rvert, with d\mathrm d raising form degree and preserving Grassmann parity. Homogeneous forms obey

αp∧βq=(−1)pq+∣α∣∣β∣βq∧αp,d(αp∧β)=dαp∧β+(−1)pαp∧dβ.\alpha_p\wedge\beta_q =(-1)^{pq+\lvert\alpha\rvert\lvert\beta\rvert} \beta_q\wedge\alpha_p, \qquad \mathrm d(\alpha_p\wedge\beta) =\mathrm d\alpha_p\wedge\beta +(-1)^p\alpha_p\wedge\mathrm d\beta.

Consequently,

dzM∧dzN=−(−1)∣zM∣∣zN∣dzN∧dzM.\mathrm dz^M\wedge\mathrm dz^N =-(-1)^{\lvert z^M\rvert\lvert z^N\rvert} \mathrm dz^N\wedge\mathrm dz^M.

In particular, two dx\mathrm dx forms anticommute, whereas dθ\mathrm d\theta and dθˉ\mathrm d\bar\theta commute under the wedge product. This last fact prevents the flat-superspace torsion below from canceling by mistake.

The one-forms

Eα=dθα,Eˉα˙=dθˉα˙,E^\alpha=\mathrm d\theta^\alpha, \qquad \bar E^{\dot\alpha}=\mathrm d\bar\theta^{\dot\alpha},

are manifestly invariant under constant supertranslations. The invariant bosonic form is

Eμ=dxμ+iθσμdθˉ−idθσμθˉ.E^\mu = \mathrm d x^\mu +i\theta\sigma^\mu\mathrm d\bar\theta -i\mathrm d\theta\sigma^\mu\bar\theta.

Indeed,

δ(dxμ)=−iϵσμdθˉ+idθσμϵˉ,\delta(\mathrm d x^\mu) =-i\epsilon\sigma^\mu\mathrm d\bar\theta +i\mathrm d\theta\sigma^\mu\bar\epsilon,

which is canceled by the variation of the remaining two terms. The exterior derivative is not merely quoted: the form-degree Leibniz rule gives

dEμ=i d(θσμdθˉ)−i d(dθσμθˉ)=i dθσμ∧dθˉ+i dθσμ∧dθˉ=2i dθσμ∧dθˉ.\begin{aligned} \mathrm dE^\mu &=i\,\mathrm d(\theta\sigma^\mu\mathrm d\bar\theta) -i\,\mathrm d(\mathrm d\theta\sigma^\mu\bar\theta)\\ &=i\,\mathrm d\theta\sigma^\mu\wedge\mathrm d\bar\theta +i\,\mathrm d\theta\sigma^\mu\wedge\mathrm d\bar\theta\\ &=2i\,\mathrm d\theta\sigma^\mu\wedge\mathrm d\bar\theta. \end{aligned}

In the second term, d(dθ θˉ)=−dθ∧dθˉ\mathrm d(\mathrm d\theta\,\bar\theta)=-\mathrm d\theta\wedge\mathrm d\bar\theta because dθ\mathrm d\theta has form degree one. The resulting torsion form encodes the odd-odd translation bracket. Thus flat superspace is not an ordinary direct product with trivial graded geometry, even though its underlying coordinates look like R4∣4\mathbb R^{4|4}.

The geometric construction and the relation between generators, covariant derivatives, and flat supervielbeins are developed in Gates, Grisaru, Roček, and Siegel 1983, §§ 3.2–3.4, pp. 62–88.

The component-constraint atlas follows this geometry through covariant derivatives, standard constraints, gauge equivalences, and component content.

Define

yμ=xμ+iθσμθˉ,yˉμ=xμ−iθσμθˉ.y^\mu=x^\mu+i\theta\sigma^\mu\bar\theta, \qquad \bar y^\mu=x^\mu-i\theta\sigma^\mu\bar\theta.

Under the finite supertranslation,

y′μ=yμ+aμ+2iθσμϵˉ+iϵσμϵˉ,yˉ′μ=yˉμ+aμ−2iϵσμθˉ−iϵσμϵˉ.\begin{aligned} y'^\mu &=y^\mu+a^\mu+2i\theta\sigma^\mu\bar\epsilon +i\epsilon\sigma^\mu\bar\epsilon,\\ \bar y'^\mu &=\bar y^\mu+a^\mu-2i\epsilon\sigma^\mu\bar\theta -i\epsilon\sigma^\mu\bar\epsilon. \end{aligned}

The parameter-only quadratic term is required for a finite transformation; at infinitesimal order it drops out. Most importantly, the transformed (y,θ)(y,\theta) depend on (y,θ)(y,\theta) and the parameters but not on the old θˉ\bar\theta as an independent coordinate. This makes the chiral subspace stable under supertranslations and anticipates the solution of Dˉα˙Φ=0\bar D_{\dot\alpha}\Phi=0.

A round-trip check is immediate. Substitute x=y−iθσθˉx=y-i\theta\sigma\bar\theta into the original transformation, calculate y′y', and then reconstruct x′=y′−iθ′σθˉ′x'=y'-i\theta'\sigma\bar\theta'. The original xx shift is recovered exactly.

Odd parameters do not commute through one another. Treating ϵ1ϵ2\epsilon_1\epsilon_2 as an ordinary commuting product erases the translation in the group commutator.

The sign of one generator is not independent. Changing PμP_\mu, the active/passive convention, derivative side, or the sign in yμy^\mu changes a package of formulas. Verify the anticommutator after every translation.

Chiral coordinates are not extra dimensions. They are an adapted coordinate system on the same superspace. A chiral superfield still has ordinary spacetime dependence.

Set a1=a2=0a_1=a_2=0. Compute both orders and subtract the 1-then-2 shift from the 2-then-1 shift.

Solution

The shifts of θ\theta and θˉ\bar\theta cancel. The two orderings differ by

Δxμ=2i(ϵ2σμϵˉ1−ϵ1σμϵˉ2),\Delta x^\mu =2i( \epsilon_2\sigma^\mu\bar\epsilon_1 -\epsilon_1\sigma^\mu\bar\epsilon_2),

so the field commutator is the translation ξμ∂μ\xi^\mu\partial_\mu.

Use the leading term of Qˉα˙\bar Q_{\dot\alpha} to show that ϵˉα˙Qˉα˙\bar\epsilon_{\dot\alpha}\bar Q^{\dot\alpha} shifts θˉβ˙\bar\theta^{\dot\beta} by +ϵˉβ˙+\bar\epsilon^{\dot\beta}, not by its negative.

Solution

Because Qˉγ˙θˉβ˙=−δγ˙β˙\bar Q_{\dot\gamma}\bar\theta^{\dot\beta} =-\delta_{\dot\gamma}{}^{\dot\beta},

ϵˉα˙Qˉα˙θˉβ˙=−ϵˉα˙ϵα˙β˙=+ϵˉβ˙.\bar\epsilon_{\dot\alpha}\bar Q^{\dot\alpha} \bar\theta^{\dot\beta} =-\bar\epsilon_{\dot\alpha} \epsilon^{\dot\alpha\dot\beta} =+\bar\epsilon^{\dot\beta}.

The two minus signs have different origins: one is built into the left-derivative representation of Qˉ\bar Q, and the other follows from reversing the indices of the antisymmetric epsilon tensor.

Show directly that δEμ=0\delta E^\mu=0.

Solution

Using constant parameters,

δdxμ=−iϵσμdθˉ+idθσμϵˉ.\delta\mathrm d x^\mu =-i\epsilon\sigma^\mu\mathrm d\bar\theta +i\mathrm d\theta\sigma^\mu\bar\epsilon.

The variations of iθσμdθˉi\theta\sigma^\mu\mathrm d\bar\theta and −idθσμθˉ-i\mathrm d\theta\sigma^\mu\bar\theta are respectively +iϵσμdθˉ+i\epsilon\sigma^\mu\mathrm d\bar\theta and −idθσμϵˉ-i\mathrm d\theta\sigma^\mu\bar\epsilon. The sum vanishes.

Why does y′μy'^\mu contain iϵσμϵˉi\epsilon\sigma^\mu\bar\epsilon?

Solution

It arises from substituting both shifted odd coordinates into iθ′σμθˉ′i\theta'\sigma^\mu\bar\theta'. It is second order in the finite odd parameters, so it is invisible in an infinitesimal formula but necessary for exact group composition.

Supercovariant Derivatives, Chirality, and Integrability constructs the commuting right-action operators and solves the chiral constraint. Chiral, Vector, Linear, and Field-Strength Superfields then turns those constraints into standard multiplets.

  • Gates, S. James, Jr., Marcus T. Grisaru, Martin Roček, and Warren Siegel. Superspace, or One Thousand and One Lessons in Supersymmetry. Frontiers in Physics 58. Reading, MA: Benjamin/Cummings, 1983. Corrected open edition, 2001. arXiv:hep-th/0108200.

  • Salam, Abdus, and John Strathdee. “Super-Gauge Transformations.” Nuclear Physics B 76 (1974): 477–482. DOI.

  • Weinberg, Steven. The Quantum Theory of Fields. Volume III: Supersymmetry. Cambridge: Cambridge University Press, 2000, § 26.2. DOI.

Original QFT.org content:CC BY 4.0, unless an item supplies different terms. Third-party material retains its own terms.