Abstract
The Schläfli identity, which is important in Regge calculus and loop quantum gravity, is examined from a symplectic and semiclassical standpoint in the special case of flat, three-dimensional space. In this case a proof is given, based on symplectic geometry. A series of symplectic and Lagrangian manifolds related to the Schläfli identity, including several versions of a Lagrangian manifold of tetrahedra, are discussed. Semiclassical interpretations of the various steps are provided. Possible generalizations to three-dimensional spaces of constant (nonzero) curvature, involving Poisson–Lie groups and q-deformed spin networks, are discussed.
Export citation and abstract BibTeX RIS
1. Introduction
The Schläfli identity, which is familiar in applications of Regge calculus in general relativity (Regge 1961, Misner et al 1973, Regge and Williams 2000), is a differential relation connecting the volume of a polyhedron in an n-dimensional space of constant curvature with the -volumes and dihedral angles of its -dimensional faces. In this article we deal with the special case of a tetrahedron in Euclidean , for which the identity itself is given by (4) below. In this case we provide an apparently new proof of the Schläfli identity, one based on symplectic geometry. We also discuss some geometrical constructions related to the proof, including a pair of symplectic reductions that take us from a 48-dimensional symplectic manifold in which the proof is set to a 12-dimensional symplectic manifold in which the space of tetrahedra is realized as a Lagrangian submanifold. The inspiration for our proof comes from a semiclassical or asymptotic analysis of the Wigner -symbol (Wigner 1959, Edmonds 1960, Ponzano and Regge 1968, Schulten and Gordon 1975a, 1975b, Biedenharn and Louck 1981, Roberts 1999, Taylor and Woodward 2005, Aquilanti et al 2012). As our analysis proceeds we provide semiclassical interpretations of many of the steps.
The Schläfli identity is useful for obtaining formulas for the volume of polyhedra in spaces of constant curvature. Schläfli's (1858) original derivation concerned spaces of positive curvature (we note that several of the citations to this article in the recent literature are incorrect). The result was generalized to spaces of negative curvature by Sforza (1907) and new proofs given by Kneser (1936). More recent treatments include Milnor (1994), Alekseevskij et al (1993) and Yakut et al (2009).
Another approach to the Schläfli identity has been explored recently (Rivin and Schlenker 2000, Souam 2004). It begins with a formula which is valid for an arbitrary hypersurface embedded in an Einstein manifold (a manifold of constant curvature whose metric satisfies the Einstein equations with a cosmological constant). The formula relates the variation of the volume enclosed by the hypersurface to variations of the extrinsic curvature and induced metric on the hypersurface. We have shown that the same formula can be obtained by varying the Einstein–Hilbert action (the integral of the curvature scalar) evaluated on the enclosed region and relating that variation to a surface integral over the boundary using manipulations similar to those employed in the ADM formalism in general relativity (Misner et al 1973, Thiemann 2007). For a polyhedron, the extrinsic curvature is concentrated in the manner of a delta function on the codimension two faces of the polyhedron, where its integral is related to the dihedral angles. When applied to the polyhedron, the formula thus reduces to the Schläfli identity.
Although our (symplectic) proof of the Schläfli identity stands on its own, the applications that we have in mind are related to Regge calculus, an approach to discretizing general relativity, and to loop quantum gravity (Rovelli 2004, Thiemann 2007), which also involves a discretization of the degrees of freedom of the gravitational field. See, for example, Barrett and Steele (2003), Livine and Oriti (2003), Dittrich et al (2007), and Bahr and Dittrich (2009, 2010), where the Schläfli identity plays a crucial role. See also Carfora and Marzuoli (2012) for a modern, comprehensive treatment of simplicial methods in quantum gravity and other fields, including the role of state sums and their regularizations. We also note the application of the 6j-symbol and other spin networks in quantum computation (Marzuoli and Rasetti 2002, 2005a, 2005b, Garnerone et al 2006, 2007), and in chemical physics (Aquilante et al 2008, 2009, Anderson et al 2009, Carfora et al 2009). We suspect that our symplectic approach to the Schläfli identity may be especially relevant in loop quantum gravity, where spin networks such as the -symbol play an important role and where the classical phase space (Freidel and Speziale 2010, Livine and Tambornino 2011) has been identified with the same symplectic manifolds that appear in our analysis.
We begin in section 2 by explaining the shape space of tetrahedra and by presenting an ad hoc construction of a symplectic manifold and a submanifold thereof that can be identified with the space of tetrahedra. This submanifold is Lagrangian by virtue of the Schläfli identity, with a generating function (7) that we call the 'Ponzano–Regge phase'. In section 3 we present an integral representation of the Ponzano–Regge phase, essentially a derivation of the phase of the asymptotic expression for the Wigner -symbol, following the method of Roberts (1999). In section 4 we present a proof of the Schläfli identity using this integral representation. The method of proof involves integrating the symplectic form over a certain surface in phase space and then using Stokes' theorem, in a manner common to several basic proofs in classical mechanics (Arnold 1989). Then in sections 5 and 6 we carry out a sequence of two symplectic reductions on the symplectic manifold in which the proof of the Schläfli identity is set, recovering at the end the symplectic manifold and Lagrangian submanifold of tetrahedra that we started with in section 2. In section 7 we present various remarks and conclusions, including an outline of some of the features of the generalization of this work to the q-deformed -symbol and the Schläfli identity in spaces of constant, nonzero curvature.
2. The shape space and Lagrangian manifold of tetrahedra
2.1. Tetrahedra and their shapes
A tetrahedron may be defined as a subset of Euclidean in terms of an ordered set of four points, which are the vertices. The six edges and their lengths are defined in terms of the vertices. As special cases we allow the tetrahedron to be flat (the four vertices may lie in a plane) or of lower dimensionality (the vertices may lie in a line or they may all coincide). Any or all of the vertices are allowed to coincide, so that some or all of the edge lengths may be zero.
By this definition the space of tetrahedra is . If we consider two tetrahedra equivalent that are related by translations, then the space reduces to , in which the three vectors can be taken as the edge vectors emanating from a given vertex. If we expand the equivalence classes to include proper rotations, then the space of tetrahedra reduces to what we will call the 'shape space of tetrahedra'
where the action of is the diagonal action on all three copies of (that is, it is a rigid, proper rotation of the tetrahedron) and where ≅ means 'is diffeomorphic to'. This is shown by Narasimhan and Ramadas (1979) and discussed further by Littlejohn and Reinsch (1995). In the following two tetrahedra will be considered to have the same shape if they are related by a translation and a proper rotation.
Then it turns out that the space of flat tetrahedra (those whose vertices lie in a plane in ) constitute a subspace , which we will call the 'shape space of flat tetrahedra'. If we define the volume V of the tetrahedron as the triple product of three vectors emanating from a given vertex, then the subspace of flat tetrahedra divides shape space into three subsets, those on which , and V = 0 (the last being the subspace of flat tetrahedra itself). Furthermore, performing a spatial inversion of a tetrahedron in causes the point of to be reflected in the hyperplane of flat tetrahedra. In particular, the shape of a tetrahedron is invariant under spatial inversion if and only if it is flat.
It is also shown in the references cited that the six edge lengths of the tetrahedron form a coordinate system on the regions and of , that is, there is a one-to-one map from these regions of to a region of the six-dimensional space with coordinates Jr, , where Jr is the length of edge r. This map is not onto, however, since there are values of the Jr that do not correspond to any tetrahedron. In the first place these lengths obviously must satisfy and the four triangle inequalities for the four faces of the tetrahedron; in addition, there is a further requirement that the faces can be assembled into a tetrahedron. All these conditions can be expressed in terms of the minors of the Cayley–Menger determinant (Ponzano and Regge 1968) or of an associated Gram matrix (Littlejohn and Yu 2009). Spatial inversion maps a tetrahedron with volume V into one with volume , without changing the edge lengths Jr; therefore, given edge lengths such that a tetrahedron exists, the shape of the tetrahedron is determined to within a spatial inversion (hence uniquely, for a flat tetrahedron).
We now define the dihedral angle associated with edge r for a tetrahedron of a given shape, first for the case . A given edge is the intersection of two adjacent faces; the dihedral angle is not defined unless the areas of the two faces are nonzero, so we assume this. (If a face has zero area, then we will refer to it as 'degenerate.') Then we define the dihedral angle as the angle between the outward pointing normals to the two faces. This gives . Finally, if , we define as the negative of the angle for the spatially inverted shape (which has ). With these conventions, the dihedral angles lie in the range , and all dihedral angles change sign under spatial inversion, modulo .
The subset of shape space on which one or more dihedral angles are not defined consists of tetrahedra with one or more degenerate faces. All such tetrahedra have zero volume, so they form a subset of the shape space of flat tetrahedra. This subset has codimension 1 inside (more precisely, it is the union of smooth manifolds, of which the maximum dimensionality is 4). If the dihedral angles are defined for a flat tetrahedron, then they are either 0 or π, and are constant inside connected regions of the space of flat tetrahedra; these regions are separated by the codimension 1 subset upon which some face is degenerate. We will denote the subset of shape space upon which the dihedral angles are defined by ; it is minus the tetrahedra with one or more degenerate faces.
Our definition of the dihedral angles differs from the usual one used in discussions of the Schläfli identity, which is the absolute value of the definition given here. Our definition has the advantage that the dihedral angles are smooth functions along a smooth curve crossing the subspace of flat tetrahedra, modulo , if we avoid shapes with degenerate faces. More precisely, all dihedral angles are either 0 or π on the subspace of flat tetrahedra, as long as we avoid degenerate faces; as we pass from the region to the region through this subspace, an angle which is 0 on the flat tetrahedron passes smoothly from positive to negative values, while an angle that is π on the flat tetrahedron jumps discontinuously from to . In both cases, the differential is smooth. As an example of such a motion we may rotate two adjacent faces relative to one another about their common edge, so that one face passes through the plane of the other. Along this motion the lengths of all the edges are constant except for the one opposite the edge common to the two faces.
We remark that the inclusion of negative angles in the definition of the dihedral angles is merely a convenience in the case of the asymptotics of the -symbol, but such an extension to a full range of in the dihedral angles is necessary for more complex spin networks, such as the -symbol. The definition of dihedral angles given here is equivalent to the general definition given by Haggard and Littlejohn (2010). In addition, negative dihedral angles emerge naturally at the end of the sequence of symplectic reductions carried out in this paper (see section 6).
If a tetrahedron has one or more edge lengths that are zero (that is, if two or more vertices coincide), then two or more faces are degenerate; these tetrahedra form a set of shapes that is a subset of the tetrahedra with degenerate faces. It has codimension 2 inside the space of flat tetrahedra.
2.2. Lagrangian interpretation of the Schläfli identity
For the time being we concentrate on the region , where the dihedral angles are functions of the edge lengths Jr. Since the do not change if the edge lengths are scaled by some positive factor, Euler's theorem on homogeneous functions implies
The Schläfli identity has a similar appearance; it is
which, after multiplying by dJs and summing over s, can be written
See Luo (2008) for other identities involving the matrix , including the case of tetrahedra in spaces of constant (nonzero) curvature.
If we differentiate (3) with respect to Jk and antisymmetrize in k and s, we obtain
that is, the matrix is symmetric. Thus the Schläfli identity (3) implies the Euler identity (2). It also implies
where S is given by
We refer to S as the 'Ponzano–Regge phase' since it appears in the factor in the asymptotic formula for the Wigner -symbol due to Ponzano and Regge (1968).
These relations imply that the space of tetrahedra is a Lagrangian submanifold of a 12-dimensional space with coordinates , of which a schematic illustration is given in figure 1. The six-dimensional manifold L is the graph of the functions (6); because it is the graph of a gradient, L is Lagrangian with respect to the symplectic form
The figure shows a point x of the six-dimensional space of edge lengths (call it 'J-space') and the vertical manifold of constant edge lengths above it which intersects L in the point y, which in turn projects onto the space of angles (call it 'ψ-space') at point a. Due to the homogeneity of the functions , the point a does not move if x is moved along the radial line R in J-space, that is, if the edge lengths are scaled by a common factor. This means that the line a–y is everywhere tangent to L, and that L is vertical in one dimension over ψ-space. Manifold L, which is six-dimensional and which has a nonsingular projection onto J-space, has only a five-dimensional projection onto ψ-space. That is, points of ψ-space that can be realized as dihedral angles of a tetrahedron are first order caustic points of the projection from L.
The construction given has applied in the region . The same construction works in the region , providing a second branch to the manifold illustrated in figure 1, obtained from the first by . On the subset V = 0, excluding shapes with degenerate faces, the main problem is the angles that jump discontinuously from π to −π. We can avoid this by speaking of a trivial T6-bundle over . Then what we are calling the functions becomes a smooth section of this bundle, that is, the surface L, extended to flat shapes with nondegenerate faces. Then the differential forms and the combination seen in (4) are well defined everywhere in the bundle, and the Schläfli identity in the form (4) holds everywhere on L. Also, the Ponzano–Regge phase S, which involves the angles themselves (not their differentials), is a function on L; but it is discontinuous at the flat shapes.
In our next step we exploit a certain integral representation for the Ponzano–Regge phase S. It turns out that S can be expressed as a line integral of a symplectic one-form in a certain 48-dimensional symplectic manifold along a path running along one 24-dimensional Lagrangian submanifold and then back along another, in the type of geometry that arises in the semiclassical analysis of scalar products in quantum mechanics. This is explained in section 3. This representation is then used in section 4 to prove the Schläfli identity. Then in subsequent sections it is shown that the 12-dimensional symplectic manifold illustrated in figure 1 ('J–ψ-space' or the T6 bundle just described) can be obtained from the 48-dimensional one in which the integral representation is expressed by means of two symplectic reductions. This process reveals information about the symplectic manifold in figure 1 that is not apparent from the presentation so far.
3. An integral representation of the Ponzano–Regge phase
In this section we present a derivation of the Ponzano–Regge phase as the principal contribution to the phase of the asymptotic expression for the Wigner -symbol. The derivation is a reformulation of that given by Roberts (1999), which as far as we know is the most symmetrical and elegant available. We express it, however, in a somewhat different geometrical language than Roberts, using real Lagrangian manifolds that are parameterized as level sets of various momentum maps and an arbitrary real representation of the wave functions, rather than a complex or coherent state representation. We invoke the semiclassical treatment of the -symbol merely to motivate the geometry behind the representation of the Ponzano–Regge phase as a certain line integral along certain Lagrangian manifolds; this is what we will need for our proof of the Schläfli identity in section 4. For the purposes of this paper the manifolds need not be quantized.
Our brief discussion of the semiclassical mechanics of the -symbol involves what we call the '-model' of the -symbol, which is due to Roberts (1999) and which is explained in terms of spin networks in section 3 of Aquilanti et al (2012). In this model, the -symbol is expressed as the scalar product of two vectors and in a certain Hilbert space, where the labels A and B stand for two complete sets of commuting observables of which the vectors are eigenvectors, with certain normalization and phase conventions. The two complete sets involve angular momentum operators whose components do not commute except on the subspace where the angular momenta vanish, which is precisely the subspace in terms of which the -symbol is defined. These two sets of commuting operators correspond classically to two sets of Poisson commuting functions on a certain phase space. Again, the sets involve various (now classical) angular momenta whose components do not Poisson commute except on the level set upon which the angular momenta vanish, which are precisely the Lagrangian manifolds relevant to the semiclassical evaluation of the scalar product .
3.1. Phase spaces
We now explain the phase space in which these manifolds live. The notation is the same as in Aquilanti et al (2007, 2012), with minor modifications which are noted. We begin with the phase space or symplectic manifold Φ of a two-dimensional, isotropic harmonic oscillator with unit mass and frequency. This is a classical version of the harmonic oscillator system used by Schwinger (1952) and Bargmann (1962) in their treatment of the representation theory of , which anticipated many of the features of the modern theory of geometric quantization (Kirillov 1976, Bates and Weinstein 1997, Simms and Woodhouse 1977, Echeverría-Enríquez et al 1999). Coordinates on Φ are , so that as a symplectic manifold , where means . We use indices μ, ν, etc to run over 1,2, indexing the two harmonic oscillators; we often suppress these indices with an implied summation. We introduce complex coordinates , on Φ, so that Φ can be seen as , where z (without the μ subscript) is seen as a column vector or spinor , where is seen as a row spinor and where again a summation over μ is implied. We denote the symplectic two-form on Φ by , where θ can be taken to be . Although θ is complex, we will be only interested in integrals of it along closed loops, for which the integrals are real.
The phase space Φ has several functions defined on it, including
where , are the Pauli matrices and where spinor contractions are implied. We note that , where H is the harmonic oscillator . These functions satisfy , where we use bold face for the three-vector with components Ji. There are also the Poisson bracket relations , . The definition of Ji in (9) implies a map , which we call the Hopf projection because when restricted to the three-sphere it is the projection map of the Hopf fibration of S3 over S2 (Cushman and Bates 1997, Holm 2011). Here is the space with coordinates , which we call 'angular momentum space' (it is the dual of the Lie algebra of ).
We construct larger symplectic manifolds by taking products of Φ. First we construct the symplectic manifold
where spinors z and z' are coordinates in the first and second copies of . For brevity we will write this as or , with the symplectic form in (10) understood. We introduce primed versions of (9) to define I' and , so that has functions I, , I' and defined on it.
On I and I' generate the actions, and , respectively, where α is the angle conjugate to the Hamiltonian function I or I'. Similarly, and generate actions, that is, Hamiltonian function , where is a unit vector, generates the action
where α is the angle conjugate to H and where
represents an element of in axis-angle form. Similarly, generates the action
Finally we define the symplectic manifold
with coordinates zr and , , where r labels the factors in (14). The symplectic one-form and two-form on are
and
By adding an r-subscript to the functions defined in (9), with or without a prime, we obtain functions Ir, , , , on . By extending to all twelve copies of we obtain a projection , where the latter space is the angular momentum space for all twelve angular momenta, , , .
3.2. The A-and B-manifolds
Now we introduce two Lagrangian manifolds in , specified as the level sets of collections of functions that Poisson commute on the manifolds. We call these the A- and B-manifolds. The functions defining the A-manifold and their contour values are given by
where the first row contains the functions and the second row contains the contour values, and where , , and . Notice that the contour values of the primed are the same as those of the unprimed Ir, that is, Jr. It is assumed in the following that the Jr are given numbers such that a tetrahedron exists with edge lengths Jr, and that the faces are nondegenerate so that the dihedral angles are defined. In particular, this means that , . There are 24 functions in the A-list (six I's, six I''s and components of angular momenta). For the given values of Jr these functions are independent and define a 24-dimensional submanifold of (the A-manifold). Also, since the functions Poisson commute with each other on the A-manifold, it is Lagrangian (see the discussion in section 4.3 of Aquilanti et al (2012)).
The vanishing of the four sums of three angular momenta indicated by (17) represents four triangle conditions among the twelve angular momenta and defines four triangles in a single copy of , if all twelve vectors are plotted in this space. The choice of the particular sets of three angular momenta to form triangles is governed by the operators of which the vector is an eigenvector, as explained in section 3 of Aquilanti et al (2012). It is also a consequence of the rules explained in Haggard and Littlejohn (2010) for translating a spin network into a stationary phase condition involving collections of angular momenta.
The functions defining the B-manifold and their contour values are given by
where etc. Notice that the contour values of the functions Ir and are the same as on the A-manifold (17). Now there are twelve conditions on Ir and and components of angular momentum vectors, or 30 conditions altogether; but these are not independent, since implies . Thus the conditions are superfluous and can be dropped from the list, giving 24 independent conditions. Therefore the B-manifold is a 24-dimensional submanifold of . It is also Lagrangian, for the same reason as the A-manifold. We call the conditions the 'diangle conditions'; when they are satisfied, and are equal and opposite.
3.3. Intersections and tetrahedra
If the A- and B-manifolds intersect, then both the diangle and triangle conditions on the twelve angular momenta hold, and the angular momenta define a tetrahedron, as illustrated in figure 2. Each edge of the tetrahedron is represented by a pair of oppositely pointing vectors and . Conversely, if a tetrahedron exists with the given edge lengths Jr, , then the A- and B-manifolds intersect. Since we are assuming that the values of the Jr are such that a tetrahedron does exist, we are assured that the A and B manifolds do intersect.
Download figure:
Standard image High-resolution imageGenerically two 24-dimensional manifolds in a 48-dimensional space intersect in a discrete set of points (a zero-dimensional set), but the intersections of the A and B manifolds (for the assumed values of the Jr) are 15-dimensional. The intersection has a nongeneric dimensionality because of a 'common symmetry group' between the A- and B-manifolds, as explained in somewhat general terms in section 6.2 of Aquilanti et al (2012). The basic idea is the following. The A-list of functions (17) is the momentum map (Abraham and Marsden 1978, Marsden and Ratiu 1999) of a group Ga whose action on is generated by the A-list of functions. This group is , where the factors are generated by the Ir and , and the factors are generated by the four partial sums of angular momenta, , , and . The action generated by one of the Ir or is the changing of the overall phase of one of the spinors zr or ; for example, the Hamiltonian flow generated by Ir is
where α is the parameter of the flow (it is the angle conjugate to Ir). Under this flow all unprimed spinors zs for and all primed spinors are not affected. The action generated by is similar (it only affects ). The flow (19) is a motion along the fiber (the 'Hopf circle') of the Hopf fibration for the particular copy of on which zr is a coordinate; the period of the circle is . All the functions Ir, , and are invariant along the flow (19). The action generated by one of the triangle sums of angular momenta, for example, is the multiplication of the selected spinors (z1, z2 and z3 in this case) by an element , which causes the corresponding vectors (, and in this case) to be multiplied by , where
This is the standard projection from to . Under this action, the other vectors are not affected, nor are any of the Ir or . Similarly the B-list of functions is the momentum map for a group , where the factors are the same as in Ga, while each of the six copies of is generated by and affects a pair for a given r (and thus the vectors , are rotated by the same element ). Because the A- and B-manifolds are Lagrangian, these manifolds are not only level sets of the corresponding momentum maps, but also the orbits of the corresponding groups.
The group Gc common to Ga and Gb has an action on that consists of symplectic transformations common to the A- and B-actions. It is generated by Ir, and by
and thus it is the 15-dimensional group , where the single factor of multiplies all spinors zr, by a common element of . The 15-dimensional orbits of Gc are common to both the A- and B-manifolds, and thus lie in their intersection. In fact the intersection is precisely 15-dimensional (for the given values of Jr).
If the volume of the tetrahedron is nonzero, however, then the intersection of the A- and B-manifolds is not connected. This is because both the tetrahedron and its image under spatial inversion simultaneously satisfy the A- and B-conditions, but the group Gc contains only proper rotations, and cannot map a tetrahedron of nonzero volume into its image under inversion. These facts lie behind the schematic illustration of the A- and B-manifolds and their intersections shown in figure 3, where the two connected pieces of the intersection are labeled T1 and T2 (for 'tetrahedra', since these intersections are the places in where the vectors and form a tetrahedron). That is, figure 3 is drawn for the case , for which the manifolds T1 and T2 correspond to two points of shape space , related by spatial inversion. As we move along T1 or T2, the only things that change are the overall phases of the 12 spinors and the overall orientation of the tetrahedron, neither of which changes the shape.
Download figure:
Standard image High-resolution image3.4. A path for the Ponzano–Regge phase
The branches of the semiclassical approximation of a scalar product such as correspond to intersections of the Lagrangian manifolds corresponding to vectors and , and the relative phase between two branches is the integral of the symplectic one-form along a path starting on one intersection, running along one of the Lagrangian manifolds to the other intersection, and then back to the initial point along the other Lagrangian manifold. This is discussed in Littlejohn (1990) and in Aquilanti et al (2007, 2012). In the case of the Ponzano–Regge formula, the relative phase between the two branches contained in is (dropping the which is irrelevant), which therefore is the integral of θ in (15) along a path starting on T1, running along the B-manifold to T2 and then back along the A-manifold to the same point of T1. Obviously if the path is traversed in the opposite direction, we obtain a change of sign in the value of the integral. These facts are not needed for the proof of the Schläfli identity presented in section 4, but they motivate the general approach.
Such a path can be constructed out of group actions in Ga and Gb. We start at a point , as illustrated in figure 3, which corresponds to a definite tetrahedron in , as illustrated in figure 2. We write , , etc for the unit outward pointing normals to the faces, evaluated at P. For each edge labeled by r, one of the faces meeting at the edge contains and the other contains . We take the unit normal to the face with and dot it into and sum over all the edges. This gives a Hamiltonian
which is a linear combination of the functions in the B-list, and so generates a motion along the B-manifold. Call the parameter of the flow generated by this Hamiltonian α. We follow the flow for an elapsed parameter of , which rotates each vector , by an angle π about an axis orthogonal to itself, thereby inverting all twelve vectors , . This maps the tetrahedron into its inverted image, taking us along a path such as in figure 3 to the other branch of the intersection T2. At intermediate points along this path we do not have a tetrahedron or even faces, that is, the four triangle conditions for the faces are not maintained, but we do have . At point Q the inverted tetrahedron appears. Under this flow the twelve spinors are multiplied by elements of , for example, z1 and are transformed according to
In addition, the action integral of the symplectic one-form θ in (15) along the path is , the value of the Hamiltonian along the flow times the elapsed parameter (this applies to any Hamiltonian that is bilinear in the z's and 's). But since along the B-manifold, the Hamiltonian is zero, and the accumulated action is zero.
Next we follow a path from T2 along the A-manifold to a point on T1, as illustrated in figure 3. The path follows the Hamiltonian flow generated by
that is, H is a sum over all the faces of the normals times the partial sums of vectors in the triangle conditions defining the faces. The normals are still evaluated at P and thus are constant; since these are outward pointing normals at P, they are inward pointing at Q. We follow the flow for an elapsed parameter of , which causes each face to rotate by an angle π about its normal, thereby inverting all vectors , and returning us to the original tetrahedron (at the point P' in figure 3). At intermediate stages along the path the diangle conditions are violated, but the triangle conditions continue to hold so the faces are well defined. On reaching P', the diangle conditions hold once again, and the tetrahedron is reassembled. Again, the action integral vanishes, since the face vectors etc are all zero along the flow. As for the action on the spinors, in the case of z1 and it is
The overall effect of these two flows on the primed spinors is the identity, for example, (24) and (27) imply , while the unprimed spinors suffer a change of overall phase, for example, the effect of (23) and (26) on z1 is
where and . But
where and is the dihedral angle at edge 1, as defined previously. (We have used and .) But since
the overall effect on spinor z1 is
and similarly for the other spinors. Under the two flows the tetrahedron in returns to itself but the unprimed spinors do not, instead they have moved to a different point on their Hopf circles, which is why the point P' is indicated as different from the starting point P in figure 3. This motion (along the Hopf circles) can be regarded as a holonomy or geometric phase (Berry 1984) in a bundle (a Hopf bundle) over angular momentum space .
We close the loop along a path as in figure 3 in six steps, with each step moving us along the Hopf circle of one of the spinors zr. The Hamiltonian for the r-th step is Ir, with an elapsed parameter given by . For example, in the first step the Hamiltonian flow of I1 with parameter has the effect
thereby cancelling the phase in (31) and returning z1 to its original value. The value of here is the one at the initial condition P, but that is the same value appearing in (31), and it is constant on T1 (and therefore it is the same as at P' and along the I1-flow). The action along this I1 flow is . When all six flows under the Ir are carried out, the total action is twice the Ponzano–Regge phase (7).
In summary, the action integrals along the first two legs of the path vanish, while that along the third gives twice the Ponzano–Regge phase. This method of deriving the phase, a part of the Ponzano–Regge formula, is basically due to Roberts (1999), and it is much simpler than the derivation we gave in the asymmetrical -model of the -symbol presented in Aquilanti et al (2012). The main advantage of the -model seems to be its close connection with the spherical phase space of the -symbol, which leads to an easy derivation of the amplitude of the Ponzano–Regge formula (due originally to Wigner (1959)). The part of this derivation that we need for this paper is the representation of the Ponzano–Regge phase as (one half of) the integral of the symplectic one-form along the path .
4. Proof of the Schläfli identity
The Schläfli identity involves variations in the edge lengths Jr, so we must consider families of A- and B-manifolds and their intersections in which the values of Ir and are variable. To do this we replace the conditions with simply , thereby maintaining the equal lengths of vectors and . The A-manifold thereby enlarges into a manifold we call , defined by , plus the four triangle conditions, which constitute independent conditions and define a 30-dimensional manifold. Similarly, the B-manifold enlarges to a manifold , defined by the six diangle conditions . The conditions are not independent of these and need not be added. Thus there are independent conditions, and is also 30-dimensional. Just as the A- and B-manifolds are Lagrangian, the - and -manifolds are coisotropic, since the functions defining the level sets have vanishing Poisson brackets among themselves on the level sets.
As for the 15-dimensional intersection manifolds T1 and T2, we have been assuming that the values of Jr were such that a tetrahedron exists with nonzero volume, but when we allow the Jr to be variable their values will include tetrahedra of zero volume. These (flat) tetrahedra are equal to their own images under spatial inversion (modulo a proper rotation), so manifolds T1 and T2 merge at such values of the edge lengths. Therefore we define a manifold ; it is connected. Manifold is specified by the union of the conditions for and , which are simply the triangle and diangle conditions taken together (again, follows from the diangle conditions). These are not independent, because both the triangle and diangle conditions imply the vanishing of the total angular momentum (21); so the number of independent conditions defining is , and the manifold is 21-dimensional (since ).
We now take the point P of figure 3 and allow it to sweep out a curve lying inside for , where the original P is now . For simplicity we assume this curve does not cross any flat tetrahedra (those with zero volume). Each serves as the initial point for a contour with three legs, first running along the B-manifold to a point , representing an inverted tetrahedron, then back along the A-manifold to a point , to a tetrahedron with the original shape, and then along Hopf circles to return to the original point . The values of the Jr are fixed by the initial point , and are constant along the contour; they are therefore functions of λ. So are the dihedral angles and the action function S, which is the integral of θ along the path. Starting from points , these paths sweep out a two-dimensional surface illustrated in figure 4, in which the three segments of the wall, swept out by the three legs of the paths, are illustrated.
Download figure:
Standard image High-resolution imageWe now integrate the symplectic form ω along the walls of this surface. On the first () segment we use λ and α as coordinates, where α is the evolution parameter of the Hamiltonian (22). The integral is
But is the Hamiltonian vector field which we write as XH, so the integrand can be written
where we use Hamilton's equations in the form
and we write the final result as (as an ordinary derivative) because H restricted to the first segment of the wall is a function only of λ (it is independent of α). In fact, H in (22) vanishes on the first segment, so and the integral of ω is zero on this segment. A similar argument shows that the integral of ω also vanishes along the second segment, where the Hamiltonian is (25).
Thus the only contribution to the integral of ω is from the third segment
a result we will use later. As for the third segment, we break it into six steps as in section 3, where the r-th step uses the Hamiltonian and the elapsed parameter . For example, in the first step we have
where we have transformed the integrand as before. Doing this for all six steps (for all ), we finally obtain
Now we apply Stokes' theorem to the integral of ω over the walls, which gives
where the loop integrals of θ along the top and bottom edges of the walls give the left-hand side and where we have cancelled a factor of 2. One can check that the sign conventions of Stokes' theorem are correct in (39). Finally, differentiating this with respect to and using (7) give the Schläfli identity in the form
This is our proof of the Schläfli identity in Euclidean .
5. The first reduction
This section and the next apply symplectic reduction, the geometrical method of passing from a symplectic manifold with a symmetry to a smaller symplectic manifold in which the symmetry has been eliminated. Symplectic reduction is described in Abraham and Marsden (1978), Cushman and Bates (1997), Marsden and Ratiu (1999) and Holm (2011). Basic elements of symplectic reduction that are called upon below are the symplectic action of a group, the momentum map, and the formation of the reduced symplectic manifold as the quotient of a level set of the momentum map by the 'isotropy subgroup', which in all examples in this paper is the entire group.
We now apply symplectic reduction to the 48-dimensional phase space , producing a 36-dimensional phase space that is almost diffeomorphic to (a certain zero level set must be handled specially, but except for this subset of measure zero the quotient space is a power of ). Both manifolds and of section 3.2 (and hence the A- and B- manifolds for all values of the Jr) are submanifolds of the 42-dimensional set given by , . This is a level set of the momentum map with components , whose associated group is . Under the symplectic reduction of the given (zero) level set of the momentum map by the Lagrangian A- and B-manifolds project onto Lagrangian manifolds in the quotient space, as does the entire construction presented in section 4 and illustrated in figure 4. For example, the contour for the integral representation of the Ponzano–Regge phase projects onto a contour in the reduced symplectic manifold giving another (lower dimensional) integral representation for the same phase. Also, the surface swept out when this contour is allowed to move by varying the Jr projects onto the reduced space, as does the proof of the Schläfli identity based on it.
5.1. Reduction of by
Each factor of acts on a single factor of for a given value of r, so to study the reduction we restrict attention to single value of r and a single copy of . When we are done we just take the six-fold product of the reduction of by to get that of by . It does not matter which copy of we work with so in the following we drop the r index. This same reduction was discussed by Freidel and Speziale (2010) in connection with the classical phase space associated with spin networks in loop quantum gravity. The following treatment of this reduction differs from that of Freidel and Speziale in several particulars, some of which will be pointed out as we proceed.
On the eight-dimensional phase space the symplectic one-form and two-form are
and
On this space the Hamiltonian function generates the action
where α is conjugate to H. We are interested in the seven-dimensional level set L given by . This condition can also be written , so L has the structure of a cone (it is the union of a set of vector spaces passing through origin of ). Except on the subset (the origin, a single point), the orbits of the action (43) on L are circles of period in the coordinate α, while the orbit through the origin is just the origin itself (a single point).
Because of the change in the dimension of the orbit, the quotient space is stratified. Let us define as L with the origin removed
so that on the orbits are circles, and . As for the quotient space, let us define
so that as a set the entire quotient space is . In the following we will concentrate on and , the latter of which is a symplectic manifold of dimension 6. Our aim will be to identify topologically, to find convenient coordinates on it, and to find its symplectic form.
It turns out that is symplectomorphic to with a certain subset removed. Because of the stratification of the quotient space, Q is not diffeomorphic to the full space . This situation is handled more completely in the theory of 'symplectic implosion' (Guillemin et al 2002).
It is somewhat awkward to demonstrate the relation between and in the coordinates and in terms of the symplectic structure given so far. Instead it is convenient to use a symplectic map which we call K2, taking us from to another symplectic manifold which we call , and to carry out the symplectic reduction on . When this is done, we can use K2 to pull the reduction back to . As a manifold, is with coordinates , just as with , but the symplectic forms on are
and
that is, with a change in sign in the primed term relative to the symplectic forms (41) and (42) on . (The notation is slightly illogical, since the sign is changed only in the primed term of the symplectic form.) Functions I, I', and are defined on by the same formulas (9) as on , but because of the change in sign in the symplectic structure we have the Poisson bracket relations on . The other Poisson bracket relations on are the same as on , namely, and . Also, any function of Poisson commutes with any function of .
We now make a digression to explain some maps that will be used to construct the map .
5.2. Some useful maps
In this section we consider the phase spaces and The two-component spinor z is a coordinate on both Φ and , but the symplectic forms are of opposite sign. Space Φ is the same one introduced in section 3.1. Notice that the phase spaces discussed in section 5.1 can be written and , where it is understood that symplectic forms add under the Cartesian product.
For motivational reasons we explain the semiclassical significance of Φ and . Space Φ is the phase space corresponding to the Hilbert space . We view as the space of wave functions of the quantum two-dimensional harmonic oscillator, which is used in the Schwinger (1952) and Bargmann (1962) treatment of the representations of . We also view as a space of Dirac ket vectors. To say that corresponds to Φ means at least two things. One is that linear operators from to itself are mapped into functions on Φ by means of the Weyl correspondence (Berry 1977, Balazs and Jennings 1984, Ozorio de Almeida 1998). Another is that wave functions in are represented by Lagrangian manifolds in Φ; the phase of the semiclassical approximation to the wave function is the integral of the symplectic one-form (which differs from by an exact differential) along the Lagrangian manifold, times .
We define as the space dual to , that is, contains bra vectors or complex valued linear functionals on . We can also think of as containing wave functions , that is, a wave function in is the image under the metric of some wave function in . The metric is an antilinear map. If corresponds to a Lagrangian manifold in Φ, then corresponds to the same Lagrangian manifold in . However, since the symplectic one-form in is the opposite of that in Φ, its integral along the Lagrangian manifold in is opposite the corresponding integral in Φ. This inverts the phase of the semiclassical approximation to the wave function, which is just what we want for the complex conjugated wave function.
These considerations motivate the following definition of the classical map G corresponding to the metric :
In other words, G is the identity as far as the coordinates z are concerned, but as a map it connects different spaces. Moreover, it is antisymplectic, since the push-forward of the symplectic form on Φ under G is minus the symplectic form on .
Another map between Φ and is defined by
where is the complex conjugated column spinor and where
Under K the symplectic form on Φ is pushed forward into the form on which is the same as the symplectic form on so the map K is symplectic. It is the classical analog of the quantum map K discussed in Aquilanti et al (2012), modulo phase conventions that we will not go into here. Suffice it to say that the quantum map K is a linear map between and , unlike , which is antilinear. We see that linear maps in quantum mechanics correspond to symplectic maps classically, and quantum antilinear maps to classical antisymplectic maps.
Under K the functions Ji on Φ are pushed forward into the functions on , while I becomes I. Since K is symplectic the formation of the Poisson bracket commutes with the push-forward; for example, computing on Φ gives J3, which pushes forward into ; but J1 and J2 push forward into and whose Poisson bracket on is computed according to , where the final minus sign comes from the inverted symplectic structure on . The answers are the same.
A third map is time-reversal, defined by
In the coordinates z, Θ is defined by the same equation as K, but unlike K it maps Φ to Φ so it is antisymplectic. We call Θ time-reversal because it is the classical analog of the quantum–time-reversal operator (Messiah 1966), an antilinear operator that inverts the direction of angular momenta. The action shown in (51) is the same as that of the quantum time-reversal operator on the spinor of a spin- particle. Under Θ, the functions Ji go into , while I goes into I, just as under K; but now these function are on Φ, not . We note that the three maps introduced are related by
Considered as a map from to , time-reversal is an antiunitary operator satisfying . It has the property
From this follows the cited transformation properties of I and . For example, using an obvious notation for the scalar product of two two-component spinors and the properties of antilinear operators we have
where we have used the fact that I is real. Similarly, we have
since Ji is real. equation (53) also implies that time-reversal commutes with rotations, that is
for all . This follows if we write g in the axis-angle form (12) since the antilinear Θ inverts the signs of both and i.
5.3. Reduction of by
We return now to the discussion interrupted at the end of section 5.1. We define the symplectic map
in which the K map is applied only to the second argument. The 2 subscript on K2 is a reminder of this fact. Note that may be regarded as the phase space corresponding to vectors , where and are state vectors for the two-dimensional harmonic oscillator, while may be regarded as the phase space corresponding to vectors .
We will carry out the reduction of by described in section 5.1 by mapping the entire construction over to via K2. Under the symplectic map K2 symplectic group actions, momentum maps and level sets in are mapped into the same types of objects in . For example, the function on maps into on , whose Hamiltonian flow on is given by
with a change in sign on the second half as compared with (43) due to the change in sign in the second term of the symplectic form. This is the action on by which we carry out the reduction. The fact that both spinors z and z' transform in the same way under this action is the main fact that makes the reduction on more transparent than that on .
The level set is the same seven-dimensional cone discussed in section 5.1. We use the same symbol L for it on as on , and we define and exactly as in (44) and (45). Obvious functions on that are constant on the orbits (58) and which therefore can be projected onto the quotient space are quantities bilinear in and z or in and z', including I, I', and .
Another set of functions that can be projected onto the quotient space is constructed as follows. We begin by defining a map from to 2 × 2 complex matrices. If , then the map is
where means the matrix whose two columns are z and . We note that so if then M(z) has an inverse, and, in fact
Thus M(z) can be identified with a quaternion, that is, it can be written , where is a real four-vector.
Now let (so that ), and let g be an element of such that
Then by (56) we have , and
or,
Thus, for , there is a unique such that , given as an explicit function of z and z' by (63). Moreover, the matrix g, that is, its components, are constant along the orbits (58). This is obvious since g in remains the same if both z and z' are multiplied by the same phase.
We remark that the map can be used to show the equivalence of the Hopf circles as defined by us, that is, , following the Hamiltonian flow of I on , with the definition used by Freidel and Speziale (2010) and other authors, which is
where is a spin- rotation of angle α about the z-axis and where we use the antilinearity of Θ. In the latter definition the Hopf circles are identified with the cosets in with respect to the subgroup of rotations about the z-axis. The two definitions of the Hopf circles are the same, apart from a switch in the direction of traversal.
We use these results to construct global coordinates on . A point of can be identified with a pair such that . But in view of the uniqueness of such that , such a point can also be uniquely identified by the pair (z,g) or the pair , where and where and . Moreover, any such pair, in either version, corresponds to a unique point of , and either pair can be taken as coordinates on . Defining
we see that is diffeomorphic to and to via the two coordinate systems (z,g) and .
When we subject a point of with coordinates (z,g) to the action (58), we have , ; or, in coordinates , we have , . In either case, the action is a motion along the Hopf circle in one of the two (z or z') factors, while g is left invariant. So the quotient space is obtained by dividing the factor in the two factorizations of by the Hopf action, while leaving the factor alone. The result is
where
In the final quotient operation the projection is the Hopf map discussed below (9), so that coordinates on are either or , where and . Vectors and belong to and are related by
where R(g) is defined by (20). This can be proved with the relation,
essentially a statement of the coadjoint representation of .
The cotangent bundle is diffeomorphic to or , where is the dual of the Lie algebra . This is proved by taking an element of , which is a one-form at a point , and pulling it back to the identity by either left or right translations. The pulled-back form is then expressed in some basis in . In view of (66) our space is diffeomorphic to minus the zero section, which we write as
Having found the topology of our quotient space and convenient coordinates on it, we now work out the symplectic form and show that it is identical to the natural symplectic form on .
5.4. Symplectic structure on
The symplectic structure on can be obtained simply by working out the Poisson brackets of the coordinates or on , where g means the matrix with components (a matrix in , not to be confused with a metric). These Poisson brackets can be computed on using the symplectic structure (47), and they survive unaltered when projected onto . The brackets involving and are not difficult, since these vectors are the generators of actions on z and z', and the brackets can be worked out from the explicit expressions (63). But the latter calculation is lengthy and uninspiring. A better approach is to work with the symplectic form.
The symplectic two-form (47) can be restricted to , where it is no longer symplectic but it is the pull-back under the projection map of the symplectic form on . This can be used to find the symplectic form on . As a practical matter, this means using the defining relation of L (that is, ) to restrict ω to , and then expressing the result in terms of the coordinates on and their differentials. Since we have two coordinate systems on , this is to be done in two different ways.
Actually this process can be carried out on the symplectic one-form (46). Since implies for , on we can set and , or
where we have used
In (72) I is the function defined in (9), not the identity matrix; multiplication of I by the identity matrix is understood. However, since we have and , or
and
These imply
Similarly eliminating z in favor of g and z' gives
These are four expressions for the symplectic one-form on . Freidel and Speziale (2010) have derived similar expressions.
5.5. Symplectic structure on
In the following we regard as a three-dimensional submanifold of 2 × 2 complex matrix space, that is, . At some risk of confusion we use g to denote a point of as well as a complex matrix with components . At the identity element the tangent space (the Lie algebra) is spanned by three tangent vectors , defined by
In these equations g can be thought of as a matrix of complex functions (the components ) defined on matrix space or on the submanifold thereof; and ei can be thought of either as vectors in at the identity or vectors in the tangent space to matrix space at the same point. In the following we prefer to write these equations in a slightly different form
where now dg means the matrix of differential forms, the differentials of the functions just introduced, either on matrix space or restricted to the submanifold, and where the vector ei acts on a differential form in the usual way. These equations define ei, which can also be written
expressing them as linear combinations of tangent vectors to matrix space. These linear combinations are tangent to the submanifold at the identity.
Now transporting ei to an arbitrary point by left and right translations, we obtain the left- and right-invariant vector fields on , defined by their actions on the matrices of differential forms dg and
where the vector fields are evaluated at (the same g that appears on the right-hand sides). These definitions imply
and
showing that the Lie algebra in the differential geometric sense agrees with the matrix Lie algebra of the matrices .
Now we define one-forms on
These could also be thought of as one-forms on matrix space, but the two different versions given are only equal when restricted to . Then by a direct calculation we find
and similarly we find . Therefore and are respectively the left- and right-invariant one-forms on . Then (81) implies
The definitions (83) were motivated by the expressions (75) and (76), but note that and are forms on , while the forms in (75) and (76) are on .
We are also interested in the symplectic one-form on , which is constructed according to the procedure described by Arnold (1989), p. 202. On any manifold M we let , so that α is a one-form at some point . Then we define θ, the symplectic one-form on , by , where is the bundle projection. To apply this procedure to we let so that α is a one-form on at a point , and we express α in terms of its components with respect to the bases and at g
Then and are two coordinate systems on . In terms of these coordinates the symplectic one-form is
that is, the same formula but now with a reinterpretation of and , which have been pulled back from to become forms on .
5.6. Identifying and
We now have two coordinate systems on , and ; and two on , and . Comparing the forms (75) and (87) and using the definitions (83), we see that if coordinates on are identified with on then the symplectic one-forms agree. That is, if we just set , then the identification between and is a symplectomorphism. Similarly, setting produces another symplectomorphism between and . These are the same symplectomorphisms, because on the left- and right- components of a form are related by the coadjoint representation
where both forms are evaluated at the group element g. But under our coordinate transformations this is the same as (68) on . With these identifications, we can write the symplectic one-form on as
This is valid on the version of that is derived from by symplectic reduction. There is another version of that we now describe.
5.7. Pulling the reduction back to
We now pull back the reduction that we have just carried out on to via the symplectic map K2, defined in (57). The relation defining the coordinates (z,g) or on , which is on , becomes on . When this is used to express the symplectic form (41) on and thence on , we obtain
equation (89), that is, with a minus sign in the primed expressions relative to (76). This sign is easily understood as the effect of K2 on the functions (that is, they are mapped into ). equation (89) in turn implies that when reducing from , is identified with by and (with a minus sign in the primed expression). Thus the symplectic one-form on , when obtained by reduction from , is
Similarly, the relation (68), which applies to when derived from , becomes
when is obtained from .
From (91) we obtain the symplectic two-form
where we have used (85). The matrix of the components of ω in the bases along the rows and along the columns is
where . Inverting this we obtain the matrix of the Poisson tensor in the bases along the rows and along the columns
This in turn implies that the Poisson bracket of two functions on , when obtained by reduction from , is
Carrying out the same procedure on the primed version of the symplectic form, we obtain
and
To check these calculations (especially the signs) we can use the Poisson brackets (96) or (98) to study various group actions on . For example, the Hamiltonian , where is a unit vector, produces the flow
where we use (96) and (80) and where α is the angle conjugate to H. These have the solutions
where is defined by (12) and where R and u are related by (20). Similarly, the flow generated by is
with solutions
These are the expected actions of (left and right) on , given the actions (11) and (13) on . That is, one set is mapped into the other through the relation .
Another Poisson bracket of importance is
one that is easiest to derive back on , where is a function of z and , and is a function of z' and . On the vanishing of this Poisson bracket is a reflection of the fact that left and right translations commute. This Poisson bracket implies that is constant along the -flow (101), and is constant along the -flow (104).
Another Hamiltonian function is of interest, namely, . If α is the conjugate angle, then Hamilton's equations are
where the two forms of the g-equation come from the two versions of the Poisson bracket (96) or (98), or through the use of (69) and (92); and the - and - equations follow from (105). These equations have the solution
where and and where and are constant. The orbit in is a circle upon which and are constant; the group is .
Finally, let us consider the Hamiltonian function . Since and commute, the action is the product of the left and right actions shown in (101) and (104), that is
5.8. Manifolds on the quotient space
The A- and B-manifolds, defined as submanifolds of by (17) and (18), are subsets of the level set L of the momentum map of the first reduction, that is, the conditions , hold on both the A- and B-manifolds. Moreover, the A- and B-manifolds are both invariant under the group action of that reduction, that is, they are composed of a union of orbits of this group. This means that they survive the projection onto the quotient space , so long as none of the edge lengths is allowed to be zero (the condition was understood in our original formulation of these manifolds). The same is true of the larger - and -manifolds, as long as we exclude zero edge lengths. On a single copy of , the condition is automatic, so on the A-manifold is defined as the level set Ir = Jr, , plus the four triangle conditions seen in (17), a total of independent conditions that specify an 18-dimensional Lagrangian manifold in .
As for the B-manifold in , it is the level set , plus the six diangle conditions, , , nominally conditions. However, these are not all independent. We examine the question of independence in a single copy of , in which we consider the submanifold defined by and , where is given. We call this submanifold . Let (1 means the identity in ) be a group element, and consider the intersection of the fiber over g with . Points of this fiber are covectors at g, whose components with respect to the right- and left-invariant bases are and . On the intersection these components satisfy as well as(92). Since we are assuming , (92) can only be satisfied if is parallel or antiparallel to the axis of the group element g. That is, if we write g in axis-angle form, , where is a unit vector, then we must have
We exclude because only for is the axis a unique unit vector in . Therefore over the manifold consists of two sections of the bundle , so it is three-dimensional. It is also Lagrangian, being the level set of functions that commute on the level set. As for the points , here (the identity 3 × 3 rotation) and the intersection of with is the two-sphere . This means that the Lagrangian manifold is vertical in two dimensions over (these points are second order caustics). The two branches of over merge together at , and is connected. The manifold for is illustrated in figure 5. Finally, although the set J = 0 (that is, ) is not a part of it does belong to ; it is just the zero section of the bundle, which is diffeomorphic to itself, and is therefore three-dimensional. The manifolds for form a one-parameter family of three-dimensional Lagrangian manifolds in .
Download figure:
Standard image High-resolution imageFinally, the B-manifold is the six-fold product ; it is an 18-dimensional Lagrangian submanifold of the quotient space .
As for the enlarged manifolds and , if we exclude zero edge lengths then these project onto 24-dimensional submanifolds of . Manifold is specified simply by the four triangle conditions, that is scalar conditions. As for manifold , on a single copy of let us define
which we can extend to on when relevant. Then Λ is a four-dimensional submanifold of or , and the manifold is the 24-dimensional subset of given by . It is also the level set of the diangle conditions , ; these are conditions, the same ones that define as a subset of , but only 12 are independent on , as indicated by the fact that is 24-dimensional.
Now we return to figure 3 and consider the projection of the manifolds illustrated there onto the quotient space . The projections of the A- and B-manifolds have just been described; each loses six-dimensions upon projection, becoming 18-dimensional, because each is a union of six-dimensional orbits of the symmetry group . The same applies to their intersections T1 and T2, which drop from 15-dimensions to nine. The (projected) nine-dimensional versions of T1 and T2 on are the orbits of a nine-dimensional common group, which is worth describing.
The momentum map for the common group consists of the six functions on , as well as the three functions in , defined by (21). The group itself is . As for the Ir, the flow of each one acts on only one copy of , and that action was presented in (108) and below. Thus the set of the six Ir generates an action of with orbits that are six-tori, upon which the vectors and are constant. Since the diangle and triangle conditions depend only on these vectors, if they are satisfied at one point on the six-torus, they are satisfied at all points on it; thus, the projected T1 and T2 manifolds in are unions of such orbits. As for , this is a sum over r of vectors of the form for each r; the flow generated by each one of these is given by (109). The effect of the group action is thus to conjugate each gr by the same element of , while rotating all of the vectors and by the corresponding rotation in . The latter leaves the diangle and triangle conditions invariant, thus maintaining the conditions for a tetrahedron. Thus, the projected manifolds T1 and T2 are invariant under this action. Assembling these facts, one can show that the manifolds T1 and T2 in consist of precisely one nine-dimensional orbit of the group (for the assumed values of the Jr).
Finally, the contour illustrated in figure 3 projects onto a similar contour in . Also, since the symplectic one-form on is the pull-back of that on , the integrals of the symplectic one-forms along the respective contours are the same. (In general in symplectic reduction, it is only the symplectic two-forms that are guaranteed to be related by the pull-back, but in this example it works for the one-forms.) This means that the integral representation of the Ponzano–Regge phase presented in section 3.4) projects onto .
Similarly, the manifolds illustrated in figure 4 also project onto , losing six-dimensions in the process. The proof of the Schläfli identity given in section 4 and based on these manifolds also carries over to the quotient space with just a change of context. One of the manifolds used in this proof is , which is the union of all manifolds T1 and T2 for all values of the Jr (we may exclude Jr = 0). This is the manifold of tetrahedra, and being a six-parameter family of nine-dimensional manifolds, it is 15-dimensional.
5.9. Semiclassical interpretation of
We make some final remarks of a semiclassical nature concerning the Lagrangian manifolds . Whenever a Lagrangian manifold occurs in a classical context it is useful to ask what its semiclassical significance is. In the case of the it is easiest to answer this question with respect to the version of obtained from rather than from . In the following we will proceed somewhat intuitively, stating several plausible facts without proof and switching back and forth between the classical geometry and the corresponding objects and operations in the linear algebra of the corresponding quantum systems.
For a given J, is the three-dimensional level , . Let us denote its preimage under the projection map: by MJ, which is the level set I = J, inside . Then , and MJ is a four-dimensional Lagrangian manifold in . It will be easiest to explain the semiclassical significance of MJ first, and then to turn to .
We denote the Hilbert space of two harmonic oscillators by as above. As explained by Schwinger (1952) and Bargmann (1962), carries a unitary representation of which we denote by , and under this action decomposes into one carrier space of irrep j for each j
A convenient basis on is , where '' means all j and m allowed by the usual restrictions on angular momentum quantum numbers. A basis in is the restricted set .
The Hilbert space corresponding to is , that is, it consists of linear combinations of vectors of the form , in Dirac notation. These vectors are otherwise linear operators . A convenient basis on is the set . Now the subset of indicated by , that is, the set , corresponds to the subspace of on which . Let us call this space . Space is spanned by , and it contains the operators that leave each irreducible subspace invariant. That is, the operators in are are block diagonal in the basis. In summary, is the space of operators corresponding semiclassically to the manifold L.
Restricting the level set further by requiring that for a given J, we obtain a classical submanifold of that corresponds to operators that are linear combinations of the set , that is, operators in with a fixed value of j. These operators map into itself for the given j, while annihilating all vectors in for . Let us call this space ; it has dimensionality .
Restricting the level set further by adding , we obtain the classical manifold MJ. On the Hilbert space the operators generate an action of given by , where is any operator. Also, the condition specifies a subspace of operators for which , that is, operators that are invariant under rotations. But the only operator in that is invariant under rotations is the projection operator onto subspace
according to Schur's lemma. This is the operator or vector in that corresponds to the classical Lagrangian manifold MJ.
Now we turn to the projection of L onto and of onto , and explain the corresponding linear algebra. The symplectic manifold is the cotangent bundle of , so it should be the phase space corresponding to , the Hilbert space of wave functions on . An obvious basis on this Hilbert space is the set of components of the rotation matrices, , which are orthonormal with respect to the Haar measure on the group. Actually, it is better to use the complex conjugates of these functions on , , which, when regarded as wave functions of a rigid body, have the correct transformation properties under rotations (Littlejohn and Reinsch 1997). In view of the quantum numbers, these wave functions are obviously the images of the vectors under the quantum analog of the classical projection map, and it would appear that we have a map .
Explicitly, this map is
where . In particular, the basis vector is mapped into
as expected. As for the projection operator , its wave function on is the character
where actually depends only on the conjugacy class, and where we have used the fact that for the characters are real. Thus we see that the wave function on corresponding to the Lagrangian manifold is the character .
The trace in (114) is the scalar product in , so semiclassically its stationary phase points should be the intersections of the Lagrangian manifolds in corresponding to operators X and U(g). As for U(g), its Lagrangian manifold is , precisely the coordinate transformation we used in section 5.3 on passing from to . One way to see this is to note that the graph of a symplectic map in is a Lagrangian manifold in the latter space. In this case, , which is the obvious symplectic map corresponding to U(g), specifies a Lagrangian manifold in that is the obvious candidate for the manifold supporting the operator U(g). The fact that it is Lagrangian can be verified directly by computing the Poisson brackets among the components of and in the symplectic structure (47). These Poisson brackets all vanish. The Lagrangian manifold in is actually a plane, which moreover is a subset of L.
Another point of view is the following. The operator U(g) belongs to the space . Let us call operators in this space 'ordinary operators,' while linear operators mapping this space into itself we will call 'superoperators' (terminology without relation to supersymmetry). Then an ordinary operator can be associated with a superoperator by either left or right multiplication. For example, the ordinary operator , the usual annihilation operator, becomes a superoperator by either left or right multiplication, or , where X is an ordinary operator. Then on the phase space the superoperator, left multiplying by , has Weyl symbol , while the superoperator, right multiplying by , has Weyl symbol , in the coordinates we have been using on . These operators satisfy the relation
which can be expressed by saying that U(g) is an eigenoperator of the commuting superoperators whose symbols are the components of with eigenvalues 0. Thus U(g) is supported semiclassically by the Lagrangian manifold .
The Lagrangian manifold in is a subset of L and moreover consists of orbits of . Thus it survives the projection onto the quotient space, where in fact it becomes simply on . That is, it is the fiber over the point g in the cotangent bundle. This fiber is naturally Lagrangian. So the intersections of Lagrangian manifolds implied by (114) projects onto an intersection of Lagrangian manifolds in , one being and the other the fiber over g in . This is the expected geometry for a wave function on .
Finally we note that the characters in are given explicitly by
where we write g in axis-angle form, . By tracking the Lagrangian manifolds and the densities on them through the projection process we have just described it can be shown that the denominator can be interpreted as an amplitude determinant in the semiclassical expression for the wave function. The amplitude diverges at and , that is, at the group elements , where as we have noted the Lagrangian manifold has caustics when projected onto the base space . The numerator is the sum of two branches of a WKB wave function, corresponding to the two branches of seen in figure 5 or in figure 7 below.
Download figure:
Standard image High-resolution imageDownload figure:
Standard image High-resolution imageIn the case of the character formula the semiclassical treatment is exact. This can be seen as an example of the Duistermaat–Heckman theorem, which also connects this discussion with mathematical work on the asymptotics of group structures. See Duistermaat and Heckman (1982), Blau and Thompson (1995), Stone (1989) and Ben Geloun and Gurau (2011).
6. The second reduction
We now carry out a second reduction, this one taking us from or to a new quotient space we denote by or Σ. It will turn out that Σ is essentially the symplectic manifold found in an ad hoc manner in section 2 and illustrated in figure 1. The symmetry group for this reduction is , and its momentum map and the level set to be used are given by , .
Before we develop this further, however, there is one obvious problem, if we intend to project the contour for the Ponzano–Regge phase, illustrated in figure 3, onto . That is, although the B-manifold is a subset of the level set , , as required if it is to survive the projection onto , the same does not apply to the A-manifold. Therefore we do not seem to have a contour for the Ponzano–Regge phase on . We address this problem before studying the second reduction further.
6.1. A new contour for the Ponzano–Regge phase
Consider the contour illustrated in figure 3. It was explained in section 3 that the integral of the symplectic one-form θ around this contour is , where S is the Ponzano–Regge phase given by (7). In this section we interpret this contour as belonging to either or and θ as given by either (15) or (91). However, it was also pointed out in section 3 that the integrals of θ along the legs and vanish, so in fact the closed contour can be restricted to just the open contour without changing the result. The latter contour is an open contour contained in the intersection set T1.
Next, in section 4 a family of such contours, parameterized by λ, was considered, as illustrated in figure 4. These sweep out a two-dimensional surface illustrated in the figure. Applying Stokes' theorem to the integral of over this surface (the 'walls'), we find that this integral is . On the other hand, it was pointed out in section 4 that the integrals of ω over the first and second legs of the surface vanish, as indicated by (36). Now applying Stokes' theorem to the integral over the third leg only, we obtain
where C is the closed contour which may be seen in figure 4 and which is illustrated separately in figure 6. In fact, the integral along the top and bottom edges of C give the result shown in (119), so the integrals of θ along the vertical segments in figure 6 must cancel each other.
For our purposes the main point is that this new contour C lies entirely in the space of tetrahedra. This is a subset of , which survives the projection onto , so the contour C illustrated in figure 6 does also, and gives us an integral representation of the Ponzano–Regge phase (actually the difference seen in (119)) on the quotient space .
Before leaving this contour we will specialize the initial conditions for future convenience. The two-dimensional surface illustrated in figure 4 is swept out by curves . All we require of the initial point is that it lie in the space of tetrahedra, that is, that the triangle and diangle conditions be satisfied. Supposing this to be true, there still remains the question of the initial phases of the spinors . Let us now assume that at , , where is a spinor such that . Let us also assume that at , . This implies , as required of the initial conditions.
Then the procedure of section 3 shows that at point , we have (the same as at ), and , where is the dihedral angle of edge r of the initial tetrahedron. Thus, one can say that the contour C illustrated in figure 6 has the dihedral angles built into it.
These choices are convenient, because they imply that at (and therefore all along the segment in figure 6) we have gr = 1, . As for point , we have
because this element, when applied to , brings out the phase . Here .
6.2. Back to the second reduction
We now carry out the second reduction, in which the group is , generated by , . We work on the level set , which is otherwise . For simplicity we work with the space rather than . As before the group action is a product of an action on each factor, in this case acting on ; the latter action is given explicitly by (109). We drop the r index when considering a single factor.
The zero level set in the six-dimensional space is Λ, defined by (111); it is four-dimensional. When considering the action of on Λ it is convenient to decompose Λ into the family of three-dimensional submanifolds , as indicated by (111), since each is invariant under the same action. See figure 5. To show this explicitly, consider a point as in the figure, so that the action (109) implies
where . That is, the axis of the rotation transforms the same as and , so the conditions , are maintained by the action (109). As for the points , the two-spheres over these points (sets and in the diagram) are invariant under the action (109). It is also easy to see that for J = 0 is invariant under the action. Since all sets are invariant under the action, so is Λ, and the isotropy subgroup is the entire group .
As for the orbits of the action, consider first and . Then the action (109) causes the point g to sweep out a conjugacy class in , which is a two-sphere; and the point on over g on either (upper or lower) section sweeps out the portion of the section over the conjugacy class. But since proper rotations cannot connect with , the orbits of the action (109) lie entirely on the upper or lower sections, and the two branches are not connected by the action. Thus the orbits are copies of the conjugacy class over which they lie, and are two-spheres. Similarly, over the points the orbits are two-spheres, so all orbits for are two-spheres. As for the set for J = 0, the orbits are the conjugacy classes, which are two-spheres except at where they are single points.
To parameterize the conjugacy classes of we write so that ϕ in the range is a coordinate of the conjugacy classes, with meaning g = 1 and meaning . Then to construct the quotient space we take first the case , for which is three-dimensional and all orbits are two-spheres, so that the quotient space is one-dimensional. It consists of two branches over , corresponding to , which join together in a single branch over and , as illustrated in figure 7. The quotient space for is a circle. We introduce a coordinate τ on this circle by defining on the upper branch where , and on the lower branch where . Then we can take τ in the range . This takes care of the case .
Forming the union of these circles for all , we obtain a cylinder with coordinates , . This is the quotient manifold under the second symplectic reduction applied to . At J = 0 the set is just itself, and the quotient space is the one-dimensional interval , the set of conjugacy classes. Therefore we can pinch off the cylinder at J = 0 into this interval to form the entire quotient space. Alternatively, since the set J = 0 was a single point of the level set in the first reduction and the set for J = 0 is not a part of the image of the first reduction, it is reasonable to take care of the case J = 0 in an ad hoc manner by pinching off the cylinder at J = 0 into a point. The resulting set is topologically , with polar coordinates with . Recall also that the case J = 0 means a tetrahedron for which the dihedral angles are not defined, so it plays no role in the Schläfli identity.
Extending this construction to all values of , we can define the quotient space as the sixth power of the cylinder with coordinates just described; and similarly, we define Σ as the sixth power of the plane with polar coordinates .
6.3. The symplectic form on Σ or
To compute the symplectic form on the cylinder we first restrict the form (90) to the level set, upon which . This gives
But , so
Also
Now upon substituting (124) into (122) the term involving vanishes, since ( is a unit vector) and since . As for the term in , it gives
after a bit of algebra, where the ± sign refers to the upper or lower branches. But this can be expressed in terms of the coordinate τ on the circles, giving
This is the symplectic form on the cylinder, or the plane , if we extend it to J = 0 in the manner suggested above. It looks like the symplectic form for the harmonic oscillator in action-angle variables, except that the angle τ has a range of .
Taking the sixth power of this, we obtain the symplectic one-form on Σ or , which is
6.4. The contour and Lagrangian manifold on
Now let us project the contour illustrated in figure 6 onto the reduced space. The contour contains a path , along which and are functions of λ, defining a family of tetrahedra for which (by our assumptions) ; therefore, along this path, Jr and are functions of λ, with . Also, according to the conventions introduced in section 6.1, we have gr = 1 along this path, and hence . Next, along the path , Jr is constant while the group element gr has axis and an angle (or conjugacy class) ϕ that goes from 0 to , as indicated by (120). This means that , so we are on the lower branch of figure 7, and goes from 0 to .The path is similar; Jr is constant along this path, and goes from to 0. Finally, along the path , Jr and are functions of λ, with . This projected contour lies in the space or Σ; when it is projected onto one of the spaces for a fixed r, we obtain figure 8.
Download figure:
Standard image High-resolution imageFinally, the manifold on given by , where the are functions of the six Jr's, is a Lagrangian manifold, according to the logic of section 2. In fact, is essentially the symplectic manifold constructed in section 2, apart from signs and factors of 2. A surprising aspect of this result is that the dihedral angles of the tetrahedron emerge in the end as labels of conjugacy classes in .
7. Conclusions
To conclude we will discuss some aspects of this calculation that have not been mentioned so far. First, it would be possible to carry out the two symplectic reductions of this paper in a single step, that is, by taking the momentum map and level set to be , specifying the first and only reduction. This condition implies , so it includes the first reduction as carried out in this paper. We have proceeded in two steps because it is easier to see how the reductions work this way, and because the intermediate stage, which involves , is interesting in its own right, with potential applications in quantum gravity.
Another possibility is to reduce the original problem by , which means a set of twelve copies of the Hopf fibration, resulting in a reduced phase space which is , in which the twelve angular momentum vectors and , with fixed magnitudes, are coordinates. In this case the Ponzano–Regge phase can be expressed as a contour along 12-dimensional Lagrangian manifolds in this 24-dimensional space, and the dihedral angles appear as solid angles of wedges on spheres. This phase space can then be reduced again by , which produces a phase space that is closely related to the shape spaces, and .
This final space is not, however, the symplectic manifold S2 of Kapovich and Millson (1995, 1996), which is discussed in terms of the asymptotics of the -symbol by Aquilanti et al (2012). That is the shape space of a quadrilateral in with fixed lengths of its four edges. The relation between the space of this paper and the shape space of quadrilaterals is discussed in the theory of 'remodeling' of spin networks, which is the topic of the dissertation of one of us (AH). A paper describing this theory is currently in preparation.
The Schläfli identity is valid for arbitrary polyhedra in (not just tetrahedra). This follows trivially by dividing the polyhedron into tetrahedra, and summing the (tetrahedral) Schläfli identity over those. This process involves a triangulation of the surface of the polyhedron. Conversely, a triangulated surface in without boundary (a two-cycle), which is automatically the boundary of a three-dimensional region, can be realized as one of the stationary phase geometries of some spin network. This follows from the rules for translating a spin network into a stationary phase condition, expressed in terms of angular momentum vectors in , as discussed in Haggard and Littlejohn (2010). For example, a triangular bipyramid, which can be formed by placing two tetrahedra with two congruent faces adjacent to each other, is one of the stationary phase geometries associated with the trivial -symbol Yutsis et al 1962, Varshalovich et al 1981). This spin network has altogether four stationary phase geometries, obtained by subjecting one or the other of the tetrahedra to time-reversal. The spin network itself is the product of two -symbols, and the four stationary phase geometries correspond to the four exponential terms in the product of the two cosines of the Ponzano–Regge formula for each -symbol. More generally, a spin network, one of whose stationary phase geometries is a triangulated two-cycle in , can be expressed as a product of -symbols (a single term, with no summation), where each -symbol corresponds to one of the tetrahedra of the triangulation of the interior. In this (fairly trivial) way, the proof of the Schläfli identity given in this paper generalizes to arbitrary polyhedra in .
Our symplectic approach to the Schläfli identity suggests generalizations apparently not previously contemplated. These arise from the asymptotics of spin networks whose stationary phase geometries are not two-cycles in , and which therefore do not enclose a volume. An example is the nontrivial -symbol, discussed in Haggard and Littlejohn (2010), in which the stationary phase geometries can be viewed as one cell in a tessalation of a subset of homeomorphic to , or, perhaps better, as a triangulation of . The nontrivial -symbol cannot be written as a product of -symbols, nor can the stationary phase surface be represented as the boundary of a collection of tetrahedra. The edges of this surface do, however, have lengths and dihedral angles. Not only is the obvious generalization of the Schläfli identity valid for these lengths and angles, but the symplectic proof presented in this paper carries over with trivial modifications.
Thus far we have restricted attention to the Schläfli identity in Euclidean . Schläfli's original proof treated the case of the constantly curved three-sphere S3, Schläfli (1858). In fact, similar identities hold in each of the three-dimensional constant curvature spaces, including hyperbolic three-space , as well as in arbitrary dimensions; the metrical proofs of Milnor (1994) and Alekseevskij et al (1993) already mentioned treat the general case where the sectional curvature κ is variable. If we parameterize this curvature with for , and S3 respectively, then all three identities can be summarized in a single formula
We believe that the symplectic method of proof described in this paper can be generalized to an arbitrary constant curvature three space. This is briefly outlined in the following paragraphs and will be described more fully in a future publication.
Once again, we define a tetrahedron in each of the constant curvature spaces and S3 through an ordered set of four vertices. We connect the vertices by geodesic arcs and view these arcs as subsets of completely geodesic surfaces that make up the faces of the tetrahedron. The curved space versions of the Schläfli identity apply to these curved tetrahedra.
Just as a Euclidean tetrahedron emerges from the asymptotics of the ordinary Wigner -symbol, a tetrahedron in spaces of constant (positive or negative) curvature emerges from the asymptotics of the q-deformed -symbol, an invariant in the representation theory of the quantum group or Hopf algebra (Chari and Pressley 1994). This much is clear already from the work of Mizoguchi and Tada (1992) and of Taylor and Woodward (2005), both of whom used one-dimensional WKB techniques along the lines of Schulten and Gordon (1975a, 1975b).
These asymptotics are important in quantum gravity. A number of works have drawn connections between the q-deformed symbols and discretizations of general relativity that include a cosmological constant. In connection with loop quantum gravity Han (2011) and Fairbairn and Meusburger (2012) have developed q-deformed spin foam models and Dupuis and Girelli (2013), Bonzom et al (2014a, 2014b), and Dupuis et al (2014) have developed further connections between the geometry of q-deformation and the cosmological constant.
A fully symplectic derivation of these asymptotics has yet to appear and this is part of what we would like to achieve in our future work. We will now outline some progress in this direction, speaking of the case of real , which corresponds to tetrahedral geometry in the hyperbolic space .
The Hopf algebra is generated by three operators, Jz and , satisfying the commutation relations
where in comparison to the usual formulas we have inserted factors of ℏ, assuming that Jz, have dimensions of angular momentum. This algebra can be realized in terms of the raising and lowering operators of a q-deformed harmonic oscillator, in a generalization of the Schwinger–Bargmann representation of , as shown by Biedenharn (1989) and Macfarlane (1989). We follow Biedenharn's convention for q in (129). The q-deformed -symbol is constructed out of irreducible representations of this algebra, labeled by six values of the quantum number j, in a manner similar to that used to construct the ordinary -symbol out of six irreps of . In fact, the q-deformed symbol approaches the ordinary one as , just as the Hopf algebra (129) approaches the ordinary Lie algebra in the same limit.
To study the asymptotics of the representations of the algebra (129) we wish to take the limit . This limit does not exist if q is held fixed, because of the terms . To obtain a well defined limit we follow Mizoguchi and Tada (1992) and Taylor and Woodward (2005) by scaling q along with ℏ. We do this by requiring that the ratio
be held fixed in the limit . With this understanding the limit converts the operator algebra (129) into the Poisson algebra
where . In these Poisson bracket relations the operators have been 'dequantized,' that is, they are no longer linear operators but rather functions on the phase space Φ of the two-dimensional q-deformed harmonic oscillator. We do not give the definitions of these functions here but they are the dequantization of the formulas given by Biedenharn (1989) and Macfarlane (1989) for the J-operators in terms of the q-deformed raising and lowering operators. As for J0, it plays the role of the curvature of the hyperbolic space . If we like we can set , corresponding to the standard case in (128), or we can keep J0 which is useful for studying the limit which is the nondeformed case q = 1.
The definitions of provide a map from the phase space Φ to , in which the J's are coordinates (this is a generalization of the Hopf map discussed below (9)). The Poisson bracket relations (131) mean that this version of is a Poisson manifold. In the nondeformed case (q = 1), the resulting space is , the dual of the Lie algebra of , otherwise 'angular momentum space'. This is a vector space, hence an Abelian group under addition. In the deformed case, the space with as coodinates is a non-Abelian group, isomorphic to the subgroup of consisting of upper triangular matrices with real diagonal elements. We call this group B; the identification of angular momentum space with matrices in B is specified by the matrix
where we have set . Since these are elements of they correspond to Lorentz transformations and thus have an action on the unit mass shell in Minkowski , which otherwise is a realization of the hyperbolic space . By this action 'angular momentum space' in the deformed case, previously identified with , is seen to be diffeomorphic to , which is a more useful identification. The group plays a role in the Iwasawa decomposition of (Helgason 1978), in which an arbitrary element of is factored uniquely into a product bu, where and . Elements can be used to identify cosets in the space , which otherwise is the hyperbolic space . The group is by (131) and (132) not only a group but a Poisson manifold. In this manner there appear the elements of a Poisson–Lie group (Chari and Pressley 1994, Kosman-Schwarzbach 2004).
The comultiplication rule of the Hopf algebra (129) can be written in physicist's language as
where we have set . This is a generalization of the addition of angular momenta, , in the nondeformed case. The significance of (133) is that if and satisfy the Hopf algebra (129), then so do , so that products of representations of the Hopf algebra are also representations. When (133) is dequantized, we obtain
where we have set . This version of the comultiplication rule has the property that if and satisfy the Poisson algebra (131), then so does . It also has the property that if and correspond to group elements , according to (132), then corresponds to group element . In other words, the comultiplication rule (134) is the multiplication law for the group B in coordinates on B.
In particular, if in (134), then , or . By letting elements of B act on the origin in those elements can be used to label points of , and a relation such as can be interpreted geometrically as the tracing and retracing a geodesic line segment in . That is, it becomes a q-deformed version of a diangle condition. (Our notation here is slightly different from that used in section 3.2 for the diangle condition: and here correspond to and there.) The length of the line segment can be interpreted as the value of the q-deformed version of the Casimir function, the generalization of the function I in (9), for either or . Combining the diangle condition with the Casimir conditions gives the specification of a Lagrangian manifold in the q-deformed version of the phase space , and taking the six-fold product of this gives us the q-deformed version of the B-manifold, a Lagrangian submanifold of .
Comultiplication is associative but not commutative. Taking the coproduct of three copies of the Poisson algebra (131) gives
Now the condition corresponds to when , and are mapped into group elements by (132), and this in turn can be interpreted geometrically as a triangle condition in . When the values of the three Casimirs (that is, the lengths of the edges of the triangle) are fixed, we obtain a Lagrangian manifold in the product space . And when four triangle conditions are combined in we obtain the q-deformed version of the A-manifold, a Lagrangian submanifold of .
Thus we obtain the basic geometrical picture for the asymptotics of the q-deformed -symbol. The A- and B- manifolds have intersections which correspond geometrically to a tetrahedron in , and the phase of the asymptotic expression is the integral of the symplectic form from one component of the intersection, along the B-manifold to the other component, and then back along the A-manifold to the first component, with a final motion along the intersection to the initial point. The details are more complicated than in the nondeformed case mainly because the group B is non-Abelian, but they follow the outline presented in this paper for the nondeformed case.
Generalizations to dimensions other than three (where the Schläfli identity is known to hold) are more speculative, since our approach relies on the three-dimensional nature of . Perhaps or could be used in four dimensions, in which the Killing form of the Lie algebra could be understood as a metric (Euclidean or Minkowskian) on the space of two-forms. For this the Plebanski (1977) formalism may be relevant.
We will report on these various generalizations of this work in future publications.
Acknowledgments
We are grateful to Chris Manon for explaining the theory of symplectic implosion to us. HMH acknowledges support from the National Science Foundation (NSF) International Research Fellowship Program (IRFP) under Grant No. OISE-1159218.