You are on page 1of 7

. .

u . .



@ by Springer-Verlag 1979

value of H(S) (2.6), we can

:.ot~plirzysare for scattering ~tlidt operator Ices.

Geometrization of Quantum Mechanics

T. W. B. Kibble
Blackett Laboratory, Imperial College, London SW7 2BZ, England

;tablishes the

.ouragement and

Mech. Anal. 59, Funct. Anal. 29,


Abstract. Quantum mechanics is cast into a classical Hamiltonian form in terms of a symplectic structure, not on the Hilbert space of state-vectors but on the more physically relevant infinite-dimensional manifold of instantaneous pure states. This geometrical structure can accommodate generalizations of quantum mechanics, including the nonlinear relativistic models recently proposed. It is shown that any such generalization satisfying a few physically reasonable conditions would reduce to ordinary quantum mechanics for states that are "near" the vacuum. In particular the origin of complex structure is described.

operators with ticle Schrodinger 710-713 (1961) i, 684-689 (1962) ticle Schrodinger and four particle lath. Anal. Appl. rsis of operators.

1. Introduction Geometrical ideas, especially symplectic structures, have come to play an increasingly important role in classical mechanics [I, 21. The geometry of the classical phase space r also underlies the geometrical quantization programme of Souriau [3,4] (see also Kostant [5]). Moreover it is known that quantum dynamics can be expressed in terms of a Hamiltonian structure on the Hilbert space X of statevectors, where the imaginary part of the scalar product defines a symplectic structure [6]. However if one is seeking an axiomatic basis for quantum mechanics, it seems better to start from structures of direct physical significance, as in the operational approach of Haag and Kastler [7] and others [8,9] or the work on the geometry of quantum logics [lo, 111. It is pointed out in Sect. 2 that this can be achieved by a slight modification of the Hamiltonian formalism. We have to consider not the Hilbert space X itself but the manifold C of "instantaneous pure states" which is (essentially but not quite) a projective Hilbert space. This formalism is closely akin to the work of Mielnik [I21 on the geometry of the space of quantum states. It provides a convenient framework within *which to discuss possible generalizations of quantum mechanics. In particular it can readily accommodate the relativistic model theories proposed in a recent paper [13], or at least a large class of them. It is


math. Phys. 27.

)tentials and the '73) ems. Math. Ann. em. (to appear)


T. W. B. Kibble

worth noting that by formulatitig the theory on C rather than 2 we automatically ensure that it satisfies the scaling property shown in 1131 to be necessary for a consistent mcasuremcnt tlicory (scc also [ I 41). From tliis point of view, thc csscntial diffcrcncc bctwccn classical and quantum mechanics lies not in the set of states (save for the infinite dimensionality) nor in the dynamic evolution, but rather in the choice of the class of observables, which is far more restricted in quantum than in cl:~ssicalmcchan~cs. One motivation for tlie present work is the possibility that it might be of use in the unification of quantum mechanics with general relativity. The idea that this unification must bc an essentially geometric one, so long clianipioned by Einstein in Iiis search for a unified theory, has recently been coming back into favour. It seems natural tliat as a prerequisite quantum mechanics itself should be cast in geometrical language. Moreovcr there are good reasons for seeking to generalize it, to free it from tlie restrictions of linearity just as general relativity has freed space-time from tlic limitations of flatness 112, 151. The geometrical structure described in the present paper can easily be generalized to allow tlie space of quantum statcs to be an arbitrary infinite-dimensional symplcctic manifold. Obviously, any viable generalization of quantum mechanics must reduce to tliat theory for a widc rangc of phenomena. Onc of tlie main aims of this paper is to discuss how t h ~ might liappcn. I shall show, on tlie basis of some ratlier natural s physical assuniptions, tliat all thc main features of conventional quantum mechanics would cmerge naturally for states that are in a suitable sensc near tlie vacuum, near enough to be represented by vectors in the tangent space. I n Sects. 2 and 3 we recall the main features of symplectic structures and Hamiltonian dynamics, and introduce the quantum phase space C. For conventional quantum mechanics, C is essentially a projective conlplex Hilbert space; more precisely, a dense subspace thereof. However the formalism can be applied to a much more general case, in which C is a real infi~iite-dirne~isio~ial manifold with (weak) symplectic structure. Tlie dynamics is discussed in Sect. 3 in terms of a Hamiltonian function E on C. For ordinary quantum mechanics E is the expectation value of tlie Hamiltonian operator, and the evolution equation reproduces Sclirodinger's equation. The concept of a symmetry is examined in Sect. 4 witli particular emphasis on the space-time symmetries associated with the Poincare group. Although an eventual aim is tlie unificatio~i quantum theory with gravity, for the present a of flat space-timc is assumcd. So too is the existence of a unique vacuum state u. The tangent space T, to C at the vacuum plays an important role. The aim of Sect. 5 is to show tliat for states close enough to the vacuum to be represented by vectors in the tangent space, the formalism necessarily reduces to ordinary linear quantum mechanics. In particular, I shall show how, although C is a rcal manifold, the complcx structure of quantum mcclianics would naturally appear on the tangent space. The conclusions are summarized in Sect. 6 and a number of unresolved questions described.
2. The Quantum Phase Space Axiomatic treatments of quantutn mecha~iics often begin with the set of all mixed . . ,, -- . . -- . .

elements of this convex set. However we shall deal exclusively with the set . of Y pure states, and assume at the outset that all others call be expl-cssed as (presumably countable) mixturcs ol' tlicm. Morcovcr instcad of tlic I-lciscnbcrg-picture approach wc shall usc the Schrodinger picture. We assume that, with respect to a given choice of time axis each pure state can be represented by a path, or "history" in a space C of ~nslnnlancouspure states". For convcnicllcc, wc rcscl-vc thc tcrni slale for an

element of C; elements of 5" will be called histo~.ies. Of course, tlie dynamics establishes a one-to-one correspondence between clemcnts of C at a spccificd time and elements of Y. Although tlie formalism will be more general, i t will be useful to begin by discussing the structure of 2' in the special case of standard quantum mechanics. Indeed this is the only space we shall actually need in this paper because the generalized ~nodels considered use the same set of instantaneous states and differ from the standard theory only in tlic dynamics. Later, however, more general theories will be considered. In classical mechanics, C is of course tlie phasc space f, normally a finitedimensional manifold. but in quantum mechanics it is infinite-climcnsional. Essentially it is a projective Hilbcrt space, thc set of rays in a complex Hilbcrt space .Yf of statc vectors. I-Iowevcr tlicrc arc technical problc~nsassociated with the fact thal the Hamiltonian operator H is unbounded. We have to choosc between two alternative fornialisms, each with its peculiar advantages and disadvantages. 01ic is to work with the rays of .Y? itsclf and accept that the Hamiltonian vcctor field which spccifics tlie dynamics is defined only on a dense subspace [6]. Tlie othcr, which we shall use here, is to work fro~n start with a the dense subspace .Xof .Ye equipped witli a finer topology that makes H continuous. The chief drawback of this second alternative is that C possesses only a weak symplectic structure (see below). I f we werc trying to prove existence theorems for solutions of the time-evolution equation, this might be a major defect, but for our present purposes it is not i~ilportant. Let us assume then that there is a dense linear subspacc .X of .Yf which forms a common invariant donlain for all the operators we wish to consider, including in particular the Hamiltonian and otl~crsymmetry generators. For example, in a nonrelativistic ~iiany-bodytheory we may tale .X to be the subspace of states whose wave functions belong to sollie test-function space, say the Schwartz space Y (see for instance [18]). I n a field theory wc miglit take it to be tlie subspace generated by applying some algebra of observables to the vacuuni state. Physically, it is only states in this subspacc that we can actually prepare, so ,;//' is of .K more direct physical relevance tha~i The space .X is assumed to bc cquippcd with a topology fincr than that ol' .w; defined for example by a countable family of norlns [18]. whicli makes H and the other sym~iietry generators continuous on .YL I t is possible to do this in such a way that .X becoliles a Frtchet-Schwartz space. Now let .Tobe the set of all nonzero vectors in .;//. and let C" be tlic multiplicative group of all nonzero complex numbcrs. '1-Iicn we chuosc as our standard quantum phase space the projective sl7acc C = . f l o i ( r O . whicli \aT(, c h : ~ l l

T. W. B. Kibble



For the general theory, all we shall assume is that C is a real infinitedimensional paracompact manifold modelled on some FrCchet-Schwartz space K What this means [L9J is that C is a Hausdorff topological space that can be covered by a countable family of open sets U , on each of which a homeo~norphism 4, is defined to an open set in !; and that wlie~icverU,nUB+O, the map 4,4l,' restricted to +p(U,nU,,) is twice continuously differentiable. I n a general infinitedimensional non-Banacl~space, the definition of differentiability is quite problematic [20]. However in the particular case of Frechet-Schwartz spaces, there is a perfectly serviccablc dcfinition [21] which allows the construction of Ck manifolds (see also [22. 231). It might not be unreasonable to require that C be a C" manifold, but we shall not need that assu~nptionhere. I n addition to its manifold structure, C is required to have a weak symplectic structure. This mcans [6] that there is defincd on C a closed two-form w wliich is weakly tion-degenerate in the sense that if o,(X, Y)=O for all tangent vectors YE T,C,then X =O. This for111can be used to "lower" indices [I]. It defines a map X+.Xb from the tangent bundle T C to the cotangent bundle P C : if X is any tangent vector (or vector field) the corresponding cotangent vector (or differential one-form) is Xb=ixo, where ix is the interior product (evaluation function) defined by ixw(Y)=w(X, Y) . (Note that therc is a difference of a factor 2 between co~iventions used here and in of references [I, 2, 61. 1 follow the co~iventions Choquet-Bruhat et al. [24]). The map X+XQs injective. However, w is only weakly nondegencrate in the sense that X+.XDis not in general bijective. There may be one-forms that are not images of any vector field. I n the special case of ordinary quantum mechanics, w is defined in terms of the inner product. (As well its finer topology .T retains the inner product ( ., .) of 2.) To define o we need to introduce a local coordinate system in C. Let IT be the canonical projection of ,To onto C. Thus if u is any nonzero vector in X, nu=u denotes the corresponding state. Let v be any normalized vector in .j/L Then in a neighbourhood of n v ~ C can represent states uniquely by vectors u in the we hyperplane Thus we have a ho~neomorphism from this neighbourhood to an open set in ZV. 4 We then define o) at nv by (2) c.,(X, Y)= 2 Im ((I)&, d)* Y) for any X, YET,,C, or more generally at any point in this coordinate patch by (1)

This two-form w is obviously skew-symmetric and wcakly nondegcnerate, But while w is readily seen to be closed, it is not exact. (1t is easy to construct a closed surface over which its integral is nonzero, although locally ( I ) = - d O with

This is another interesting difference between quantum and classical mechanics. Classically, the phase space has the structure of a cotangent bundle and c u = -dO where O is the canonical one-fonn O=Cpdy. No such form exists globally on the quantum phase space C.
3. Dynamics on the Quantum P l ~ a s c Space

Let us now consider the specification of dynaniics on C , assumed to be a real CZ Frichet-Schwartz manifold equipped with a weak symplectic structure. The dynamics is described by a Hamilionian flow, that is to say a oneparameter group of diffeomorphis~ns :C - t C (with s,, = 1 and s,+, = s, +s,) which s, preserve the symplectic structure, i.e.
(4) The z, depend continuously and differentiably [21] on r and can thus bc written


z , = c x p ~ K, where K is a C' vector field on C , which of coursc also leaves invariant the symplectic structure :

where L, is the corresponding Lie derivative. (Had we adopted the alternative in approach ~nentioncd Sect. 2 we could not have required T, to be differentiable in t and would have had to allow K to be defined only on a dense subset of C, see [61.) A history, i.e. an element of .C/: is an integral curvc of this flow in C, a curvc c everywhere tangent to K :

dt - K c , , , ,

or equivalently

By virtue of the identity L,=i,d+di, from (5) and the definition (1) illat d(Kb)=O,

and the fact that w is closed, it follows


where X, YE T,Z and P,, is the orthogonal projector onto the subspace normal to u.

i.e. K b is closed.


?'. W. B. Kibble



In standard quantum mechanics, the space C defined in Sect. 2 is a simply connected paracompact manifold, and so its first de Rhain col~omology group is in trivial. (De Rliam's tlieorc~ii general is hard to prove but here we need only the comparatively straightforward result that a closed form in such a space is exact.) This is an important fcaturc, bccausc i t mcans that from (8) we can conclude that K" is exact, i.e. that a Hamiltonian function cxists. We shall assume that tlie same is true in any gcncralizcd quantum theory wc consider, though it might iiidccd be intriguing to consider a theory witli a non-simply-conliected quantum phase spacc, a toroidal space say. With this assumptio~lit follows from (8) that there exists a Hamiltoiiiali function E, ~ulique to a constant, such that up
Kb=dE .
I t follows that any liistory is a solution of Halllilton's equations

model of a scalar field 4. Let H be the Hamiltonia~l operator for the usual linear quantum field theory with d4 i~~teraction. us take Let (13) in placc of ( I I), whcrc :(I)': is tlcfiricd h y suhtr;~ctiiiga11 (infil~ilc) constant, as in [13]. Thcn instead of (12) we obtain thc Sclirodingcr cquation E(u) = (H)"

$ 3 L S



(We write (. . . ) , for (. . ., )) n . Except for the fact that the arbitrary timedepelideiit factor is made explicit, (14) is precisely the evolutio~i equation of one of the models proposed in [13]. (There is also a difference of a factor of 2 in 1.)This model can be thought of as one in which the particlc mass bccomcs statcdcpcndcnt and position-dcpcndc~~t,

Of course the function E uniquely determines tlie vector field K and therefore the one-parameter group z,. What is I I I U C ~ less obvious is whether, given E, any such flow exists, or ill other words to prove an existence theorem for solutions of the differential Eq. (10). Wc shall not attempt to treat this problem here (see [6]). 111ordinary quantum mechanics, the functio~iE is identified with the expectation valuc of thc Hamiltonian opcrator, E(u) = (H),

It is clear that many other generalized models can be constructed in the same way. For example, if we add to (13) the extra term

we obtain a model in whicli both n12 and the (64 coupling constant g acquire a state-dependence,

(44 H4u) ( 4 ~ 7$10

It is easy to verify that witli the definition (3) for o this reproduces Schrodinger's equation. We seek a curve ~p ,X such that noy=c. Equation (10) holds as an in equation for one-forms on C is. for arbitrary variations of c, but it can be pulled i.e. back by n to yield an equation for one-forms on .To, for arbitrary variations of y. The equation in fact itlvolves only transverse variations, and thus yields

Not all tlie models of [I31 call be expressed in this way because some of them do not possess a conserved energy functional that can serve as our Hamiltonian function E. However it seems ~iatural require the existence of such a function, so to this is not a severe restriction. As we noted in the introduction, by formulating the tlieory 011 C rather than we automatically eiisure the invariance under scaling transformations ybi,rll which was show11 in 1131 to be a necessary prerequisite for a consistent measurement theory (see also [14]).
4. Symmetries

where P is the orthogonal projector defined in Sect. 2. Equivalently

where ~ ( tis) an undetermined complcx function. Its presence signifies our freedom to multiply tp by an arbitrary time-varying factor without changingc. Normally of course we choose Ima=O to preserve normalization and fix Recx (equivalent to a variable zcro-point of cncrgy) by somc othcr convention. A large class of generalized models can be described by this same formalism, using the same set C and the same symplectic structure but with a more geiieral choice for the Hamiltonian function E. For example, let us consider as in [13] a

As a preliminary to the discussion of how conventional quantum mechanics might emerge as a linear approximation to a more geiieral theory, it will be useful to examine the representation of symmetries. A one-parameter group of (time-indcpcndent) symmctrics is a group of diffeomorphis~iis4, of C onto itsclf whicli Icavc invariant both tlic symplcctic structurc and the Hamiltonian function, It follows that 4, commutes witli the time evolution z,. The generator of tlie group is a C' vector field X on C which also leaves invariant tlie symplectic structurc and

commutes with the generator of time evolution [X,K]=O. (16)

. .

Just as for K itself, it follows that X bis closed, llence exact, so that there exists a generating function G,, unique up to an additive constant, such that Xb=dG, . (17)

The uniqueness is not essential. We could easily accommodate a discretely

degenerate vacuum state. A continuous degeneracy, however, would be unacceptable. Because u is unique, it must be invariant under any sylnmetry. For any oncparameter group 6, of symmetries, iP5u = u, or if X is the corresponding generator

If X and Y are symmetry generators, the effect of one on the generating function of the other is

The generating function G = G, satisfies in the scnsc that thc Icft hand side is a possible generating function for [X, Y ] i.e. this equation holds up to an arbitrary constant. We can easily generalize the discussio~l include time-dependent symmetries, to generated by vector fields that do not satisfy (16). Equation (18) is still valid if one or other vector field is K . Wc shnll bc intcrcslcd in ~>a~.ticul:lr spacc-time symmetries. We shall assume in that there is a realization by sy~nrnclries
(l4 4+>%,>,1

It is convenient to fix the arbitrary constants in generating functions by the condition G(v)= 0 . In particular
. .

of the connected component of the Poincare group, or rather1, if we wish to accon~n~odate fermions, of its two-fold universal coveri~lg group P. There is then a Let corresponding realization by vector fields of its Lie algebra 9. us denote by K , and R,,. the generators of translatioos and rotations, obeying the usual commutation rules, for example

Note that these conditions are consistent with the validity of Eq. (19) without extra constant terms; An important role will be played by the tangent space T,=T,Z to I at the vacuum state. Because of (22) any symmetry generator X induces on T, a linear transformation X: which of course preserves the symplectic structure: x;*w=o. Because of (23) and (24), there is a quadratic approximation to G on the tangent space. We may define the second derivative G' as a symmetric bilinear function of : vectors in T,. If c is any curve through v, with

where i l l , is the Minkowski-space metric tensor. Of course KOis precisely the timetranslat~onvector field K introduced earlier. The associated generating functions are the energy-momentum and angular momentum functions, PI1 GK, , =
J@\'= GRPY .


Po is the Hamiltonian function E. By virtue of (18) these functions transform in the correct way under translations and rotations. For example, L,,P,

From (17) there follows a relation between X i and G, namely :

LR,,PI, = ?rlpPv Y A,$', . Next, let us assume the existence of a unique vacuum state UEC invariant under all operations of the Poincari: group

G: = (X;)b
in the sense that for all


I ZET,, :

n r ~ n r i n lvrrsinn of

This point requires further justification. An argument for it was presented in an appendix to the this paper (ICTP/77-78/22), but cannot be included here for reasons of space

In particular we may define the second derivative E, of the energy function E = Po. We shall need one further important assumption, relating to the positivity of energy. (Remember that the zero point is fixed at the energy of the vacuum.) We


T. W.B. Kibble

shall assume specifically that the vacuum is a nondegenerate local minimum of the is a positive definite bilinear function: energy function, i.e. that

Presumably the vacuum should also be a global minimum of E, but we shall not need that condition here.

Consequently, the generator Kk is antisymmetric. We shall use it to introduce a complex structure on T,, in terms of a linear operator J satisfying J2= - 1. It is convenient (though not essential) to consider the complexified space T,,=T,,@iT,,and to define first a corresponding operator J on it. We lnay extcnd K: in ail obvious way to a complex-linear operator K: on 'r,. Then iK:, is essentially self-adjoint, and possesses a purely real spectrum. It may be regarded as an energy operator. We seek to define an operator J which takes the value + i on the subspace wherc iK:, is positive and - i on that whctc it is ncgativc, i.c.

5. Complex Structure
As noted earlier, one of the most important questions to ask about any proposed generalization of quantum mechanics is how the ordinary linear theory can emerge as an approximation. The suggestion made here is that states that are, in a sense to be defined, near the vacuum can be represented by vectors in the tangent space T,, and that on T, one has all the usual structure of linear quantum mechanics, expressed of course in a particular local coordinate system like the one used in defining w. To validate this suggestion, we have to do two things - to show how the structures of linear quantum mechanics emerge on T,, and to specify what is meant by "nearness" to the vacuum. As far as the second question is concerned, no general answer is possible; it must depend on the specific model. In the generalized field theories discusscd carlier 1131 the answer is reasonably clear. A state u is near the vacuum if the expectation value (:42(~):),, everywhere small compared to a is scale constant 1 1 1 ~ . For such states the nonlinear time-evolution equation can be well approximated by its linear counterpart. Note however that there is no guarantcc that a state initially ncar the vacuum will remain so during its subsequent history. 111 general, it will not. In order for states near the vacuum to be represented by vectors in the tangent space, one also needs of course a well defined map from some neighbourhood of v in C into q,. Here again, it is only in the context of a specific model that one can expect to specify it. In our generalized models, it poses no problem because the nonlinear and linear theories share the same space C and the same T,. In the linear theory there is of course a natural map from C to T, defined as in the definition of w in Sect. 2. Let us now turn to the other outstanding problem, that of recovering the structures of linear quantum theory on IT;,. Much of the structure of course already exists. In particular we havc a lincar rcprescntation of the Poincart group. What is lacking however is the complex structure of ordinary quantum mechanics, and the existence of a hermitean inner product. A very important role in recovering this structure is played by the positive definite symmetric bilinear function E':. It is convenient for the moment to think of it as defining a real inner product on T,. (However, it is important to realise that this is not directly related to the hermitean inner product we shall eventually construct.) Thus 7;, becomes a pre-Hilbert space. Now any one-parameter group of time-independent symmetries 4, leaves invariant the energy function E, and so is represented on T, by orthogonal *.--r.,"T -..... "+:-.," ,C I,, ~ , , , . + ; p , , l , , , . t h ; c + ,.,,on I-tI?" t; , . +.."..,.l-r+:-,. iC ,r.?.,t ." .. .., n. ..

J = i sign (iK:)


There is a slight technical problenl here. This construction is certainly possible if the spectrum of i K does not include the point 0. If it does include it, J could still be defined on the Hilbert space obtained by completion of T, but there seems to be no reason to suppose that it would necessarily leave T,, invariant. Physically, the condition that 0 is cxcludcd fro~n spcctruln corresponds to thc rccluirerncn[ the that there be a nonzero ~niilimummass in the theory. For the moment we shall assume that either 0 is excluded or, if it is not, that J can still be defined. It is clear from its definition (28) that J commutes with the co~ljugation operation X iYwX- iY on T,, and so defines a corresponding operator J on q.. In fact J could be defined directly on T, by setting

where the positive square root of the operator is implied. It follows incidentally that J commutes with all Poincare-group generators. Moreover

It may be helpful to describe one way of thinking about this operation in the casc of ordinary quantum mechanics. A vector X E T, may be represented by a density operator which in Dirac notation takes the form

where (v) is a representative of the vacuum, and (vlx) =O. Then J is defined by J X b i l x ) (vl- ilv)


Now that we have dcfined an opcrator J satisfying J 2= - 1 we may introduce a complex structure on T,. We define a complex vector space Tf whose vectors are in one-to-one correspondence XctXc with those of T,. Tf is given a complex structure by defining
iXc (JX)' =


Note that T,' has half as inany complex dimensions as T,. In terms of the representation above, we may regard the passage to projecting out the "positive-energy" part, i.e.
v c 3 .I\ .




T. W. B. Kibble


20 1

On T: we can define a hermitean inner product by setting

2 ( X C , Y ' ) = w(X,J Y )+ iw(X, Y ) .


It is easy to verify that it is indeed hermitean, linear in the second factor and antilinear in the first. 0 1 1 the tangent space we have then all the usual structures of linear quantum theory, in particular the complex structure and hermitean inner product. Moreover the symmetry generators are antihermitean, and multiplying by i makes them essentially self-adjoint. It should be re~narked that the positivity of cncrgy has played a vital role here. Without it K:, would not have left invariant a positive symmetric form and so its complexification would not have had a pure imaginary spectrum.

One aim of this work was to rewrite quantum mechanics in a form better suited to unification with gcncral relativity, but so far wc ha\/c considcrcd only flat space-time. The next step may be to examiile what happens when both structures, space-time and the space of quantum states, are simultaneously freed fro111 the restrictions of linearity. I hope to return to this question at a later date.
Ackt~owledyement.I am indebted to Dr. C. J. Isham and Mr. S. Randjbar-Daemi for criticisms of all earlier versio~iof this papcr.


1. Abraham, R., Marsden, J. E.: Foundations of niccha~iics.New York. Amsterdam: Benj,~ 1967 min 2. Maclanc, S.: Gcomctrical mcclintiics (d~~plic:~tcd Iccturc nolcs). Iltiivcrsi[y of Cl~ic:~go (196s) 3. Souriau, J.M.: Commun. math. I'liys. 1, 374 (1966) 4. Souriau, J.M.: Ann. Inst. Henri Poincari: A6, 31 1 (1967) 5. Kostiuit, B.: Quantizalion and unitary rcprcscntations. Lccturc notcs in mathcmatics, Vol. 170. Berlin, Heidelberg, New York: Springer 1970 6. Chernoff, P.R., Marsden, J.E.: Properties of infinite-dimensional Hamiltonian systems. Lecturc notes in mathematics, Vol. 425. Berlin, Heidelberg, Ncw York: Springcr 1974 7. Haag, R., Kastler, D.: J. Math. Phys. 5, 848 (1964) 8. Edwards, C.M.: Commun. math. Phys. 20, 26 (1971) 9. Davis, E. B., Lewis, J. L. : Commun. math. Pliys. 17, 239 (1970) 10. Birkhoff, G., Ncumann, J.von: Ann. Math. 37, 823 (1936) 11. Varadarajan, V.S.: Geometry of quantum theory, Vol. 1. Princcton, N J : van Nostrand 1968 12. Mielnik, B.: Commun. math. I'liys. 37, 221 (1974) 13. Kibble, T.W.B.: Cornmun. matli. Pliys. 64, 73-82 (1978) 14. Haag, R., Bannier, U.: Corntnun. math. Pliys. 60, 1 (1978) 15. Penrose, R.: Gen. Rel. Grav. 7. 31 (1976); 7, 171 (1976) 16. Jauch, J . M. : Foundations of quantum mcchanics. Reading, Mass. : Addison-Wesley 1968 17. Piron, C.: Foundations of quantum physics. Ncw York, Amstcrdam: Bc~ljamin1976 18. Bogolubov, N.N., Logunov, A.A., Todorov, I.T. : Introduction to axiomatic quantum ficld thcory. Reading, Mass. : Belljamin 1975 19. Lang, S.: Introduction to diffcrcntiablc manifolds. Ncw York, London: l~\tcrscicncc1962 20. Yamamuro, S.: Differential calculus in topological linear spaces. Lecture notcs in mathematics, Vol. 374. Berlin, Heidelbcrg, New York: Springer 1974 21. Kijowski, J., Szczyrba, W.: Studia Math. 30, 247 (1968) 22. Penot, J.-P.: Studia Math. 47, 1 (1973) 23. Lloyd-Smith, J.: Trans. Am. Math. Soc. 187, 249 (1974) 24. Choquet-Bruhat, Y., de Witt-Morctte, C., Willard-Bleick, M. : Analysis, manifolds and physics. Amsterdam : North-Holland 1977

6. Discussion
The principal results of this paper are two. First, I showed in Sect. 2 that quantum dynamics can be expressed in a simple and elegant Hamiltonian form in terms of a symplectic structure on the manifold Cof instantaneous pure states. Thus classical and quantum mechanics, and generalizations thereof, can all be formulated in very similar ways. Second, I have shown that if one makes some rather natural physical assumptions, such as existence of a unique vacuum and local positivity of the energy, then in any of the proposed generalizations of quantum mechanics, the conventional quantum theory will re-emerge as a linear approximation for states near the vacuum. In particular, the con~plex structure appears naturally. Many problems remain. An important feature of the method adopted here is its use of the Schrodinger picturc, but this has the obvious disadvantage of lacking manifest relativistic covariance. It would be useful to find a way of rewriting the formalism in an explicitly covariant manner. points that need to be clarified. The mathematical There are several tecl~nical characterization of Frtcliet-Schwartz manifolds is not at present very simple. It is not even clear that they constitute the correct class to choose. In particular the cohomology of such manifolds is not well known; and an existence theorem for histories is needed. Also the introduction of complex structure in Sect. 5 required a specific assumption about the spectrum of K: which needs to be eliminated or better understood. The only generalizations of quantum mechanics we have considered have used the same quantum phase space C, but a different evolution law. However, one might also examine alternative choices for C. In particular, it would be intriguing to consider spaces with nontrivial first cohomology group. Although the for~nalism was developed to permit generalizations of quantum mechanics, it also provides an interesting starting point for axiomatisation of the convcntional theory. It would be useful to know what conditions imposed on the symplectic structure and Hamiltonian function would allow one to identify the
t J i ~ n r \ \a~illic t : i n r l ' ~ r r l n i i ~ r i t i i m n ~ r h : a n i r s / t

C o m r n ~ ~ n ~ c by Rc d ~ ~ t Haag Rece~vedAugust 2. 1978. In rcv~scdform Dccc~nber 1978 8.