ap-3-10.dvi Acta Polytechnica Vol. 50 No. 3/2010 Higher-Spin Triplet Fields and String Theory D. Sorokin Abstract We review basic properties of reducible higher-spin multiplets, called triplets, and demonstrate how they naturally appear as part of the spectrum of String Field Theory in the tensionless limit. We show how in the frame-like formulation the triplet fields are endowed with the geometrical meaning of being components of higher-spin vielbeins and connections and present actions describing their free dynamics. 1 Introduction It is a pleasure to write this contribution on the occa- sion of the anniversary of Jiri Niederle. As a topic I have chosen a subject to whichProfessorNiederle and his collaborators have contributed with several inter- esting papers [1, 2, 3]. It concerns higher spin field theory. In particular, I would like to discuss a sys- tem of higher spin fields which are called triplets. The physical states of triplet fields describe massless parti- cles of decreasing spins. In the bosonic case the phys- ical states of a triplet have spins s, s − 2, s − 4, . . . ,1 or 0 and the physical states of a fermionic triplet have spin s, s − 1, s − 2, . . . ,1/2. In other words the triplets form reducible Poincaré group multiplets of massless higher-spin particles. As we shall see, they naturally arise in a tensionless limit of String Field Theory [4, 5, 6, 7, 8, 9, 10] as sets of three tensor fields (and this is where their name comes from [8]). We shall also discuss basic group-theoretical and geo- metrical properties of the higher-spin triplets revealed in [11]. Let us startwith a brief generic discussion of prob- lems of Higher Spin Field Theory. The construction of a consistent interacting theory of higher-spin fields is one of the oldest long standing problems in theo- retical physics of particles and fields. Since the early 1930s, it has been addressed by many distinguished theorists including Majorana, Dirac, Fierz & Pauli, Rarita & Schwinger, Bargmann & Wigner, Fang & Fronsdal, Weinberg, Velo & Zwanziger, Aragon & Deser, Singh&Hagen, deWit&Freedman,Fradkin& Vasiliev and (later on) by many others1. The main problem of Higher Spin Field Theory is the construction of consistent interactions of higher- spin fields. Higher-spin interactions, e.g. with electro- magnetic field and gravity, must respect gauge sym- metries and be consistent with quantum mechanical principals such as unitarity, causality, etc. To solve the problem of higher-spin interactions one first looks for a most appropriate description of free higher-spin fields which would allow for general- ization to an interacting theory. Revealing the group- theoretical and geometrical structures underlying this theory is of great importance. Until now, the most successful general approach to the constructionof non- linear (i.e. interacting) classical higher-spinfield equa- tions has been the so-called unfolding techniques put forward by M. Vasiliev [18, 19] (see [12] for a review and references). Great efforts have been made in studying both, massive and massless, higher-spin field theories. Jiri Niederle has concentrated, in particular, on study- ing electro-magnetic interactions of massive higher- spin fields, which, actually, reveal major issues of the generic problem [2, 3]. It would be very interesting and very important for the further development of this subject to compare the results obtained in the construction of higher-spin interactions with the structure of a known example of consistent higher-spin field theory which is String Theory2. People have worked in this direction since the early years of String Theory, and this is how, for example, the system of higher spin triplet fields was first found [4]. So let usmake a very short overviewof the place of the higher-spin fields in String Theory. 2 String Theory as a theory of interacting massive higher-spin fields String excitations give rise to an infinite number of fields of increasing spin andmass, the mass of a string state being a linear function of its spin (Regge tra- jectory) with the proportionality coefficient being the string tension T . For instance, in open string theory we have M2s ∼ T(s −1) (2.1) The infinite tower of higher-spin string states plays a crucial role in ensuring a smooth ultraviolet behav- ior (or even UV finiteness) of superstring theory, thus making it a consistent theoryofquantumgravity, since gravity is an intrinsic part of String Theory. 1For reviews on various aspects of Higher Spin Field Theory and references see e.g. [12, 13, 14, 15, 16, 17]. 2Let us recall that StringTheory emerged as a proposal to explain (among other things) the mass-spin dependence (known as Regge trajectory) of higher-spin hadronic resonances observed in experiments in the 1960s. 48 Acta Polytechnica Vol. 50 No. 3/2010 In String Theory the higher-spin excitations are massive, but our experience in the quantum field the- ory of vector fields teaches us that for quantumconsis- tency themass of the vector fields should be generated as a result of spontaneous breaking of a symmetry of massless vector fields. If one tries to extrapolate this statement to String Theory, then the natural question arises whether String Theory can be a spontaneously broken phase of an underlying gauge theory of mass- less higher-spin fields. From eq. (2.1) we see that the mass tends to zero in the limit inwhich the string ten- sion goes to zero. So people have tried to answer this question by taking a tensionless limit of String The- ory (see, e.g. [20, 21, 22, 23] for various aspects of the tensionless string limit and higher-spin theory). One can try to deduce the structure of higher-spin field interactions from the action of String Field The- ory. E.g. the action of Open String Field Theory has the following schematic Chern-Simons-like form (see [24] for more details) Sopen string = 〈Φ|Q|Φ〉+ |Φ〉3 (2.2) where |Φ〉 is a string field and Q is a BRST operator associated with the symmetries of String Theory. Due to the complexity of the problem, only terms in the Lagrangiandescribing a system of free massless higher-spin fields have been obtained from the String Field Theory action so far. As we have already men- tioned, this systemofmassless fieldswasfirst obtained by S. Ouvry and J. Stern in 1986 [4]. The simplest triplet consists of a symmetric tensor or (in the case of half integer spins) tensor-spinor fields of rank s, s −1 and s −2 φm1···ms(x), Cm1···ms−1(x), Dm1···ms−2(x) (2.3) where the indices m = 0,1, . . . , D − 1 are indices of space-time of dimension D. The spectrumof the phys- ical statesof thesefields consistsofparticlesofdecreas- ing spin: s, s −2, s −4, . . . ,1 or 0 – in the case of bosons s, s −1, s −2, . . . ,1/2 – in the case of fermions. Since each value of spin corresponds to an irreducible representation of the Poincaré group, the triplet of the fields (2.3) describes a reducible multiplet of higher- spin states. 3 Massless higher-spin triplets from String Theory Let us consider how the equations of motion of the triplet fields (2.3) arise in the tensionless and free field limit of String Field Theory. In the tensionless limit T → 0, the nilpotent open bosonic string BRST oper- ator is (see e.g. [9]) Q = c0pmp m + ∑ k �=0 ( c−kpma m k − k 2 b0c−kck ) , (3.4) k = ±1, ±2, . . . , ±∞ Q2 =0 where pm is the momentum of string center-of-mass, amk are string creation (for k < 0) andannihilation (for k > 0) oscillator operators (with |k| = 1,2, . . . , ∞ la- belling the Regge trajectories), c0 and ck are string reparametrizaition ghosts, and b0 and bk are anti- ghosts. The (anti)commutator relations satisfied by the operators are [amk , a n l ] = k δl+k,0 η mn, {ck, bl} = δk+l,0 . (3.5) A string field is a sum of an infinite number of higher- spin fields generated by acting on the string vacuum with all possible combinations of creation operators |Φ〉 = ∑ φmn···p···q···(x) · (3.6) aman · · · ap · · · aq c · · · b · · ·b |0〉. Now, one may ask the question, do independent finite subsets of higher-spin fields exist inside (3.6) which satisfy (at least linearized) string field equations ofmotion that follow fromthe action (2.2)? The linear field equations are Q|Φ〉 =0 . (3.7) Because of the nilpotency of the BRST operator, they are invariant under the BRST gauge transformations δ|Φ〉 = Q|Λ〉 , Q2 =0 . (3.8) To extract from (3.6) a finite independent set of fields satisfying eq. (3.7) let us pick up string states corre- sponding to a single Regge trajectory (e.g. k=1) and cut this trajectory at a level of spin s. Then it turns out that the string field (3.6) reduces and splits into independent pieces each of which is the sum of three terms |Φ〉triplet = φm1···ms(x)a m1 −1 · · · a ms −1 |0〉+ (3.9) Cm1···ms−1(x)a m1 −1 · · · a ms−1 −1 c0b−1 |0〉+ Dm1···ms−2(x)a m1 −1 · · ·a ms−2 −1 c−1b−1 |0〉 . By independent we mean that each set of fields (3.9) independently satisfies the string field equation (3.7). Symmetric tensor fields φs(x), Cs−1(x) and Ds−2(x) form the simplest bosonic higher-spin triplet which satisfies entangled equations of motion that follow from (3.7), namely: � φm1···ms = s ∂(ms Cm1···ms−1) , � ≡ ∂n ∂ n Cm1···ms−1 = ∂ n φnm1···ms−1 − (s −1) · ∂(ms−1 Dm1···ms−2) , � Dm1···ms−2 = s ∂ n Cnm1···ms−2, (3.10) 49 Acta Polytechnica Vol. 50 No. 3/2010 where (m1 · · · mp) denotes the symmetrization of the indices with weight 1 p! . Equations (3.10) are invariant under the gauge transformations with a local symmetric parameter Λm1···ms−1(x) that follow from the BRST symmetry (3.8): δ φm1···ms = s ∂(ms Λm1···ms−1) , δ Cm1···ms−1 = �Λm1···ms−1 , δ Dm1···ms−2 = ∂ n Λnm1···ms−2. (3.11) From the second of eqs. (3.10) we see that Cs−1 is an auxiliary field which is expressed in terms of deriva- tives of φs and Ds−1, and from the form of the gauge variationof Ds−1 we see that this field is a pure gauge. Thus, all physical degrees of freedom are contained in the field φm1···ms(x). As we have already mentioned and as follows from the analysis of eqs. (3.10)–(3.11) [9], the physical degrees of freedom of φm1···ms(x) are particle states of spin s, s −2, s −4, . . . , 1 or 0 (de- pending whether s is even or odd). A questionwhich can nowbe addressed is whether the triplet fields may have some geometrical nature? 4 Geometrical meaning of HS triplet fields. Frame-like formulation Higher Spin Field Theory is a gauge theory. Un- derstanding its underlying group-theoretical and ge- ometrical structure is of great importance for making progress in solving the higher-spin interaction prob- lem. Understanding the geometrical nature of the triplets is part of this generic problem. As far as the physical field φm1···ms(x) is concerned, its group- theoretical properties resemble very much the met- ric field of General Relativity. The gravitational field gmn(x) is symmetric and it transformsunder linearized diffeomorphisms as a gauge field δgmn = ∂m ξn(x)+ ∂n ξm(x), which is similar to the transformationproperties (3.11) of φm1···ms(x). What about auxiliary fields Cs−1(x) and Ds−2(x)? Can they be related to other geometrical quantities, such as a generalizedChristoffel symbol, i.e. a higher- spin connection associated with the higher-spin local symmetry? The answer to this question is positive. The geometrical nature of the triplet fields manifests itself rather clearly in the frame-like formulation [11] which is the generalization to higher-spinfields [25, 26] of the Cartan-Tamm-Weyl formulation of gravity in terms of vielbeins and spin connections. In the frame formulation of gravity the gravita- tional field is described by a vielbein one-form ea = dxm em a(x), (4.12) which carries the tangent space Lorentz index. In this formulation, in addition to the local diffeomorphisms, the theoryof gravitypossesses local Lorentz symmetry δ ea(x) = eb ξb a(x) , (4.13) where ξab = −ξba is a parameter of local Lorentz transformation. The gauge field associated with local Lorentz symmetry is the spin connection ωab = dxm ωabm(x)= −ω ba, (4.14) which transforms under the infinitesimal local Lorentz transformations as follows δ ωab = d ξab − ξac ωcb + ξbc ωca. (4.15) The metric field gmn is composed of the vielbeins gmn = e a m e b n ηab. (4.16) In the linear approximation, in which em a = δam + ẽm a(x) and gmn = ηmn + g̃mn(x), eq. (4.16) reduces to g̃ab = ẽab + ẽba . (4.17) Note that in the linear approximation there is no dis- tinction between world indices m, n, . . . and tangent space indices a, b, . . . The Einstein gravity is characterized by the so- called torsion-free constraintwhich relates the vielbein and spin connection d ea + eb ωb a =0 . (4.18) By analogy with the frame formulation of gravity, in the frame-like formulationofhigher-spinfield theory [25, 26] one introduces ahigher-spinvielbein one-form, which is symmetric in the s −1 tangent-space indices, ea1···as−1 = dxm em a1···as−1 (4.19) and the higher-spin connection ωa1···as−1,b = dxm ωm a1···as−1,b (4.20) which is symmetric in the indices a1 · · · as−1 and sat- isfies the following properties: i) the symmetrization of all of its tangent-space indices a and b brings zero ω(a1···as−1,b) =0 , (4.21) and ii) the trace of the index b with any of the indices a is zero ωa1···as−1,b ηa1b =0 . (4.22) Theproperties (4.21)and(4.22)of thehigher-spincon- nection are somewhat analogous to the antisymmetry propertyof the spinconnection(4.20). Todescribe sin- gle higher-spinfields, one should impose on thehigher- spin vielbein and connection stronger trace-less condi- tions (see [12, 11] for a review and references) ea1a2···as−1 ηa1a2 = 0 , (4.23) ωa1a2···as−1b, ηa1a2 = 0 . 50 Acta Polytechnica Vol. 50 No. 3/2010 If eqs. (4.23) are satisfied, then the traceless condi- tion (4.22) for ω follows from (4.23) and (4.21). As has been shown in [11], the reducible triplet multi- plets are described by e and ω which are not subject to the traceless constraints (4.23) and ω satisfies a less restrictive constraint (4.22). As in the case of Einstein gravity, we also assume that the higher-spin vielbein and connection satisfy the torsion-free condition d ea1···as−1 − dxm ωa1···as−1,bηmb =0 . (4.24) The torsion-free condition is preserved under the fol- lowingvariationof e and ω, which are local gauge sym- metries of the frame-like formulation δea1···as−1 = d ξa1···as−1(x)− (s −1)ξa1···as−1,b(x)ηmb, δωa1···as−1,b = (4.25) d ξa1···as−1,b(x)− (s −2)ξa1···as−1,bc(x)ηmc , where the parameter ξa1···as−1 is symmetric, the pa- rameter ξa1···as−1,b(x) has the same symmetry proper- ties as ω and theparameter ξa1···as−1,bc is symmetric in the indices a1 · · · as−1 and (separately) in the indices bc. In addition, the symmetrization in ξa1···as−1,bc of either b, or c with all a gives zero, and the trace of b or c with any a also gives zero. The higher-spin triplet fields (3.9) turn out to be components of the higher-spin vielbein and connec- tion [11]. The relation is as follows φa1···as = em (a1···as−1 ηas)m , Da1···as−2 = em a1···as−2as−1 δmas−1 (4.26) Ca1···as−1 = (s −1)ωa1···as−1,mm + ∂ m em a1···as−1 . The relation (4.26) between the two sets of fields has been established in [11] by comparing their gauge transformations and equations ofmotion in the frame- like and metric-like formulation. In the frame-like formulation, for the higher-spin vielbein and connec- tion the equations of motion are the torsion-free con- dition (4.24), which allows one to express the com- ponents of ω in terms of derivatives of e, up to the Stueckelberg gauge transformations with the param- eter ξa1···as−1,bc(x) of eq. (4.25), and the following differential equation on ω δm(b ∂ cωd;n1···ns−2)[c,d] + ∂d ω(b;n1···ns−2) [m,d] + ∂(b ω d; n1···ns−2)[d, m] =0 , (4.27) where the indices separated by ‘;’ are world indices, i.e. they correspond to the index m in (4.20). This equation is the dynamical field equation of the higher- spin vielbein, when ω is expressed in terms of ∂ e in virtue of the torsion-free conditions. The equations of motion (4.24) and (4.27) can be obtained from the action which generalizes to higher- spin fields the action for gravity in the frame formula- tion. In any dimension D > 1 the gravity action can be written as an integral over the differential form S[ea, ωab] = 1 2 ∫ MD ea1 . . . eaD−2εa1...aD−2bc R bc = (4.28) 1 2 ∫ MD ea1 . . . eaD−2εa1...aD−2bc (d ω bc + ωbd ω dc) , or up to a total derivative S[ea, ωab] = 1 2 ∫ MD εa1...aD−3dbc e a1 . . . eaD−3 ·( (D −2)ded ωbc + ed ωbf ωf c ) = (4.29)D −2 2 ∫ MD εa1...aD−3bcd e a1 . . . eaD−3 · (deb ωcd − 1 2 ef ωbf)ω cd , where the wedge product of the differential forms is assumed. Thehigher-spingeneralizationof the linearizedgravity action (4.29) in Minkowski space is S = ∫ M D dxa1 . . . dxaD−3 εa1...aD−3pqr · (4.30) (d en1...ns−2p − s −1 2 dxm ω n1...ns−2p, m)ωn1...ns−2 q, r . This action is a straightforward generalization of the four-dimensional action of [25]. The torsion-free con- dition (4.24) and the dynamical equation (4.27) are obtained by varying eq. (4.30) with respect to ω and e, respectively. Let us recall that if in equations (4.24), (4.27) and (4.30) thehigher-spinvielbeinandconnection together with corresponding gauge symmetry parameters were subject to the additional traceless condition in their symmetric target-space indices (4.23), the frame-like formulation based on the action (4.30) would describe irreducible physical states corresponding to massless particles of single spin s (see [11] for more details). It is a certain relaxation of the trace constraints on the higher-spin vielbeins and connections which ex- tended the irreducible higher-spin system to the re- ducible (triplet) higher-spin multiplet. It should also be noted that because of the peculiarity of the form of the frame-like action (4.30), which is constructed of the wedge product of one-forms ω and (the differen- tial) of e, it describes the higher-spin triplets whose physical states have spins s, s − 1, . . . , 3 or 2. The scalar and the vector fields can nevertheless be added to this construction in a conventionalway (see [12, 11] for more discussion of this point). An analogous formulation has been constructed in [11] for the fermionic higher-spin triplets describing physical states of spin s, s − 1, . . . ,3/2. As in the bosonic case regarding the scalar and the vector states 51 Acta Polytechnica Vol. 50 No. 3/2010 of the triplets, the field of spin 1/2 can be included into this construction as an independent field. The action for the fermionic triplets is of the first order in derivatives of the fermionic one-form ψa1···as−1/2,α = dxm ψ a1···as−1/2,α m , (4.31) where α is the spinorial index and s is now a half- integer. The field ψ is symmetric in the indices a. In a D-dimensional flat space-time the fermionic triplet action in the frame-like formulation has a very simple form S = i ∫ M D dxa1 . . .dxaD−3 εa1...aD−3pqr · ( ψ̄d1...ds−3 2 γpqr dψ d1...ds−3 2 − (4.32) 6 ( s − 3 2 ) ψ̄d1...ds−5 2 pγq d ψ d1...ds−5 2 r ) , It is invariant under the following local transforma- tions of the field ψ δψa1...as− 3 2 = (4.33) dξa1...as− 3 2 − ( s − 3 2 ) dxb ξa1...as− 3 2 ,b with the tensor-spinor parameter ξαa1...as− 3 2 ,b having the following symmetry properties ξa1...as− 3 2 ,b = ξ(a1...as− 3 2 ) ,b , (4.34) ξ(a1...as− 3 2 ,b) = 0 , andbeing subject to the (gamma)-tracelessconstraints with respect to the index b γb ξ a1...as−3 2 ,b =0 ξ a1...as− 5 2 b, b =0 . (4.35) If the fermionic one-form ψ and the fermionic pa- rameters of the gauge transformations (4.33) were (in addition) gamma-traceless in the indices a, the action (4.32)would describe a single (irreducible) higher-spin field of spin s. The frame-like formulation of the bosonic and fermionic higher-spin triplet fields considered above can be generalized to describe these fields in the anti- deSitter space [11]. Inparticular, theuse of the frame- like formulation has allowed us to overcome technical difficulties encountered in [9, 27] and to obtain in [11] a description of the fermionic higher-spin triplets in AdS.This is afirst step towards the studyof consistent interactions of fermionic triplets, since, as has been known for a long time, a consistent theory of interact- ingmassless higher-spin fields should be formulated in a space-time background of constant curvature, like the AdS space. 5 Conclusion We have seen how a triplet system of fields of spin s, s − 2 . . . ,1 or 0 appears in a truncated action for StringFieldTheory. Using the frame-like formulation, we endowed these fields with a geometrical meaning of higher-spin vielbeins and connections subject to a torsion-free condition and transforming under higher- spin local symmetries. The frame-like actions for the bosonic and fermionic triplet fields have been con- structed in flat and AdS spaces. It is of great interest and importance to generalize these results by analyz- ingpossible interactionsof thehigher-spin tripletfields (see [17, 28] for a recent discussion of this issue and references). This would give a new insight into the structure of Higher Spin Field Theory, String Theory and AdS/CFT correspondence. Acknowledgement This work was partially supported by the INFN Spe- cial Initiative TV12, the INTAS Project Grant 05- 1000008-7928, and anExcellence Grant of Fondazione Cariparo. References [1] Kotecky, R., Niederle, J.: Conformally Covariant Field Equations II. First Order Massless Equa- tions, Rept. Math. Phys. 12 (1977) 237–249. [2] Niederle, J., Nikitin, A. G.: Relativistic wave equations for interacting massive particles with arbitrary half-integer spins, Phys. Rev. D64 (2001) 125013. [3] Niederle, J., Nikitin, A. G.: Relativistic Coulomb problem for particles with arbitrary half-integer spin, J. Phys. A39 (2006) 10931–10944, arXiv:hep-th/0412214. [4] Ouvry, S., Stern, J.: Gauge fields of any spin and symmetry, Phys. Lett. B177 (1986) 335. [5] Bengtsson, A. K. H.: A unified action for higher spin gauge bosons from covariant string theory, Phys. Lett. B182 (1986) 321. [6] Henneaux, M., Teitelboim, C.: First and second quantized point particles of any spin. In Santiago 1987, Proceedings, Quantum mechanics of fun- damental systems 2 113–152. (see Conference In- dex). [7] Pashnev, A. I.: Composite systems and field the- ory for a free Regge trajectory, Theor. Math. Phys. 78 (1989) 272–277. [8] Francia, D., Sagnotti, A.: On the geometry of higher-spin gauge fields, Class. Quant. Grav. 20 (2003) S473–S486. 52 Acta Polytechnica Vol. 50 No. 3/2010 [9] Sagnotti, A., Tsulaia, M.: On higher spins and the tensionless limit of string theory, Nucl. Phys. B682 (2004) 83–116. [10] Barnich, G., Bonelli, G., Grigoriev, M.: From BRST to light-cone description of higher spin gauge fields, arXiv:hep-th/0502232. [11] Sorokin,D. P., Vasiliev,M.A.: Reducible higher- spin multiplets in flat and AdS spaces and their geometric frame-like formulation, Nucl. Phys. B809 (2009) 110–157. [12] Bekaert, X., Cnockaert, S., Iazeolla, C., Vasi- liev,M.A.: Nonlinearhigher spin theories in vari- ous dimensions, arXiv:hep-th/0503128. [13] Vasiliev, M. A.: Higher spin gauge theories in various dimensions, Fortsch. Phys. 52 (2004) 702–717. [14] Sorokin, D.: Introduction to the classical the- ory of higher spins, AIP Conf. Proc. 767 (2005) 172–202, arXiv:hep-th/0405069. [15] Bouatta, N., Compere, G., Sagnotti, A.: An in- troduction to free higher-spin fields, arXiv:hep-th/0409068. [16] Sagnotti, A., Sezgin, E., Sundell, P.: On higher spins with a strong Sp(2,R) condition, arXiv:hep-th/0501156. [17] Fotopoulos, A., Tsulaia,M.: Gauge InvariantLa- grangians for Free and Interacting Higher Spin Fields. A Review of the BRST formulation, Int. J. Mod. Phys. A24 (2009) 1–60. [18] Vasiliev, M. A.: Equations of motion of interact- ingmassless fields of all spins as a free differential algebra, Phys. Lett. B209 (1988) 491–497. [19] Vasiliev, M. A.: Consistent equations for inter- actingmassless fields of all spins in the first order in curvatures, Annals Phys. 190 (1989) 59–106. [20] Gross, D. J.: High-Energy Symmetries of String Theory, Phys. Rev. Lett. 60 (1988) 1229. [21] Sundborg, B.: Stringy gravity, interacting ten- sionless strings and massless higher spins, Nucl. Phys. Proc. Suppl. 102 (2001) 113–119. [22] Lindstrom, U., Zabzine, M.: Tensionless strings, WZWmodels at critical level andmassless higher spin fields, Phys. Lett. B584 (2004) 178–185. [23] Bonelli, G.: On the tensionless limit of bosonic strings, infinite symmetries and higher spins, Nucl. Phys. B669 (2003) 159–172. [24] Witten, E.: InteractingFieldTheory ofOpen Su- perstrings, Nucl. Phys. B276 (1986) 291. [25] Vasiliev, M. A.: ‘Gauge’ form of description of massless fields with arbitrary spin. (in Russian), Yad. Fiz. 32 (1980) 855–861. [26] Aragone, C., Deser, S.: Higher spin vierbein gauge fermions and hypergravities, Nucl. Phys. B170 (1980) 329. [27] Buchbinder, I. L., Galajinsky, A. V., Krykh- tin, V. A.: Quartet unconstrained formulation for massless higher spin fields, Nucl. Phys. B779 (2007) 155–177. [28] Fotopoulos, A., Tsulaia, M.: Current Exchanges for Reducible Higher Spin Multiplets and Gauge Fixing, JHEP 10 (2009) 050. Dmitri Sorokin E-mail: dmitri.sorokin@pd.infn.it INFN, Sezione di Padova via F. Marzolo 8 35131 Padova, Italia 53