Fedosov Quantization of Lagrange-Finsler and Hamilton-Cartan Spaces and Einstein Gravity Lifts On (Co) Tangent Bundles

JOURNAL OF MATHEMATICAL PHYSICS 50, 013510 共2009兲
Fedosov quantization of Lagrange–Finsler

and Hamilton–Cartan spaces and Einstein
gravity lifts on „co… tangent bundles
Mihai Anastasiei1,a兲 and Sergiu I. Vacaru2,b兲
1
Faculty of Mathematics, University “Al. I. Cuza” Iaşi, 11, Carol I Boulevard,
Iaşi 700506, Romania and Mathematical Institute “O. Mayer,” Romanian
Academy Iaşi Branch, 8, Carol I Boulevard, Iaşi 700506, Romania
2
The Fields Institute for Research in Mathematical Science, 222 College Street,
2nd Floor, Toronto M5T 3J1, Canada and Faculty of Mathematics, University
“Al. I. Cuza” Iaşi, 11, Carol I Boulevard, Iaşi 700506, Romania
共Received 27 February 2008; accepted 13 November 2008; published online 8 January 2009兲
We provide a method of converting Lagrange and Finsler spaces and their Legendre
transforms to Hamilton and Cartan spaces into almost Kähler structures on tangent
and cotangent bundles. In particular cases, the Hamilton spaces contain nonholo-
nomic lifts of 共pseudo兲 Riemannian/Einstein metrics on effective phase spaces. This
allows us to define the corresponding Fedosov operators and develop deformation
quantization schemes for nonlinear mechanical and gravity models on Lagrange–
and Hamilton–Fedosov manifolds. © 2009 American Institute of Physics.
关DOI: 10.1063/1.3043786兴
I. INTRODUCTION
To construct a quantum theory for a given classical model is usually understood that it is
necessary to elaborate a quantization procedure adapted to certain fundamental field equations and
corresponding Lagrangians/Hamiltonians and their symmetries, constraints, and locality. In vari-
ous attempts to develop quantum versions of gravity and nonlinear physical theories, one provided
different computation schemes when it is supposed that all constraints can be solved, gauge
symmetries can be represented by shift symmetries, and there are certain canonical forms for the
Poisson brackets. Nevertheless, such quantization usually destroys global symmetries, locality, and
may result into violation of local Lorentz symmetry.
There were proposed different sophisticate constructions with formal and partial solutions for
quantum gravity and field interactions theories. We cite here the BRST quantization methods for
non-Abelian and open gauge algebras,1–4 deformation quantization,5–8 quantization of general
Lagrange structures, and, in general, BRST quantization without Lagrangians and
Hamiltonians,9,10 W-geometry and Moyal deformations of gravity via strings and branes,11–13 and
quantum loops and spin networks.14–16
In the above-mentioned approaches, it is necessary to quantize curved phase spaces in a
manner that is explicitly covariant on phase-space coordinates. Indeed, for instance, the key
ingredients of Fedosov and BRST methods, see Refs. 9, 10, 17, and 18 and references therein, are
the embedding of the system into the cotangent bundle over its phase space. There were also
elaborated such models following standard geometric constructions over Riemannian manifolds
and formal schemes with symplectic groupoids and contravariant connections.17,19
A rigorous geometric approach to deformation quantization of gravity, gauge theories and
geometric mechanics models with constraints and related generalized Lagrange–Finsler theories,
see Refs. 20–24, shows that the quantization schemes have to be developed for nonholonomic
a兲
Electronic mail: anastas@uaic.ro.
b兲
Electronic mail: sergiu.vacaru@gmail.com. URL: http://www.scribd.com/people/view/1455460-sergiu.
0022-2488/2009/50共1兲/013510/23/$25.00 50, 013510-1 © 2009 American Institute of Physics
Author complimentary copy. Redistribution subject to AIP license or copyright, see http://jmp.aip.org/jmp/copyright.jsp
013510-2 M. Anastasiei and S. I. Vacaru J. Math. Phys. 50, 013510 共2009兲
manifolds1 and tangent and cotangent bundles endowed with nonlinear connection 共N-connection兲
structure. The natural step in this direction is to apply the methods of the geometry of Hamilton
and Cartan spaces and generalizations31,32 共such spaces are, respectively, dual to the Lagrange and
Finsler spaces and generalizations33,34兲.
The aim of this work is to show how Karabegov’s approach to Fedosov deformation
quantization19,35,36 can be naturally extended for almost Kähler manifolds endowed with canonical
geometric structures generated by semi-Riemannian and/or Einstein metrics and Lagrange–Finsler
and Hamilton–Cartan fundamental generating functions.
This paper is motivated by the following results: In Refs. 30, 21, 22, 37, and 29, we concluded
that classical and quantum gravity models on 共co兲 tangent bundles positively result in generalized
Finsler-like theories with violation of local Lorentz symmetry. The conclusion was supported also
by a series of works on definition of spinors and field interactions on 共in general, higher order兲
locally anisotropic-space times,38,39 on low energy limits of 共super兲 string theory40–42 and possible
Finsler-like phenomenological implications and symmetry restriction of quantum gravity.43–45
Here, we emphasize that the nonholonomic quantum deformation formalism can be redefined for
nonholonomic 共pseudo兲 Riemannian, or Riemann–Cartan, manifolds with fibered structure. Such
nonholonomic spaces, under well defined conditions and for corresponding Lagrange–Finsler
variables, possess local Lorentz invariance for classical theories and seem to preserve it for
nonholonomic quantum deformations, see results from Refs. 23 and 24.
The work is organized as follows. In Sec. II we recall the basis of the Lagrange–Finsler and
Hamilton–Cartan geometry, nonholonomic lifts of Einstein metrics on cotangent bundles, and
almost Kähler models of such spaces. We also collect there some geometric constructions that we
need for further considerations. Section III is devoted to the formalism of canonical nonlinear
connections and distinguished connections induced by Lagrange and Hamilton fundamental func-
tions. We introduce the concept of Hamilton–Fedosov spaces and define the corresponding almost
symplectic structures. In Sec. IV we consider a generalization of the concept of connection to that
of connection-pair resulting in Fedosov–Hamilton operator pairs which is necessary for definition
of deformation quantization models being invariant under symplectic morphisms and Legendre
transform. We provide Fedosov’s theorems for connection pairs which allows us to develop an
approach to geometric quantization of Hamilton and Einstein 共generalized on cotangent bundles兲
spaces in Sec. V. We speculate on possible quantum generalizations of gravitational field equations
on phase spaces and their deformation quantization. Finally, in Sec. VI we conclude the results.
II. LAGRANGE–FINSLER AND HAMILTON–CARTAN GEOMETRY AND EINSTEIN

SPACES
In this section, we outline some results from the geometry of Lagrange–Finsler33,34 and
Hamilton–Cartan31,32 spaces.
A. Canonical geometric objects on Lagrange and Hamilton spaces

Let us consider a real sufficiently smooth manifold M, dim M = n 艌 2. We label the local
coordinates x = 兵xi其, with “base” indices i , j , . . . = 1 , 2 , . . . , n, and write by TM and T*M, respec-
tively, the total spaces of tangent and cotangent bundles ␲ : TM → M and ␲* : T*M → M, with local
coordinates u = 共x , y兲 = 兵u␣ = 共xi , y a兲其, for “fiber” indices a , b , . . . = n + 1 , . . . , n + n on any local cart
U 傺 TM, and *u = 共x , p兲 = 兵 *u␣ = 共xi , pa兲其 on any local cart *U 傺 T*M, when pa are dual to y a. In
order to apply the Einstein summation rule for contracting base and fiber indices, we shall use
identifications of type y i ⯐ y n+i, i.e., we suppose that indices such as a , b , . . . can split into respec-
tive n + i, n + j.
Definition 2.1: A generalized Hamilton space is defined by a pair GHn = 共M , gij共x , p兲兲, where
g 共x , p兲 is a contravariant symmetric tensor field, nondegenerate, and of constant signature on
ij
1
That is, manifolds endowed with nonholonomic 共equivalent, anholonomic, or nonintegrable distributions兲, see details and
references in Refs. 25–30.
013510-3 Fedosov quantization, Hamilton spaces and gravity J. Math. Phys. 50, 013510 共2009兲

T*M = T*M / 兵0其 for 兵0其 being the null section of T*M.
We note that a contravariant tensor of type
gij共x,p兲 = eii⬘共x,p兲e j j⬘共x,p兲gi⬘ j⬘共x兲共1兲
includes, for some vielbein fields eii⬘, the dual tensor gi⬘ j⬘共x兲 as the inverse of a 共semi兲 Riemann
metric gi⬘ j⬘共x兲 on M. In this paper, we consider that any classical solution of the Einstein equations
defines a space-time manifold M and a corresponding deformation quantization procedure on T*M
defines the values eii and other fundamental quantum geometric objects 共like almost symplectic
⬘
structure and generalized connection兲 with nontrivial dependence on variables pa. In general, we
shall provide our constructions for 共pseudo兲 Riemannian spaces with 共co兲 metrics of types gi⬘ j⬘共x兲
and gij共x , p兲 and discuss if there are any important particular properties for Einstein manifolds
共spaces兲 when gi⬘ j⬘共x兲 is subjected to the condition to solve on M the gravitational field equations
with nonzero cosmological constant ␭,
1Ri⬘ j⬘共x兲 = ␭gi⬘ j⬘共x兲, 共2兲
where 1Ri⬘ j⬘共x兲 is the Ricci tensor for the Levi–Civita connection ⵜ = 兵 1⌫i⬘ j⬘k⬘共x兲其 completely
defined by gi⬘ j⬘.
For simplicity, we shall work with a more particular class of spaces when gij共x , p兲 is defined
by a Hamilton function H共x , p兲.
Definition 2.2: A Hamilton space Hn = 共M , H共x , p兲兲 is defined by a function T*M 苹共x , p兲
→ H共x , p兲苸 R, i.e., by fundamental Hamilton function, which is differentiable on T*M and con-
tinuous on the null section ␲* : T*M → M and such that the (Hessian, equivalently, fundamental)
tensor field
*gab共x,p兲 = ⳵ 2H
共3兲
⳵ p a⳵ p b
is nondegenerate and of constant signature on T* M .

Let L共x , y兲 be a regular differentiable Lagrangian on U 傺 TM, with nondegenerate Hessian
共equivalently, fundamental tensor field兲
⳵ 2L
gab共x,y兲 = . 共4兲
⳵ y a⳵ y b
Definition 2.3: A Lagrange space Ln = 共M , L共x , y兲兲 is defined by a function TM 苹共x , y兲
→ L共x , p兲苸 R, i.e., a fundamental Lagrange function, which is differentiable on
TM and continu-
ous on the null section of ␲ : TM → M and such that the (Hessian) tensor field gab共x , y兲 (4) is
nondegenerate and of constant signature on TM .
We can define the Legendre transform L → H,
H共x,p兲 = pay a − L共x,y兲, 共5兲

where y = 兵y a其 are solutions of the equations pa = ⳵L共x , y兲 / ⳵y a and 共inversely兲 the Legendre trans-
form H → L,
L共x,y兲 = pay a − H共x,p兲, 共6兲

where p = 兵pa其 is the solution of the equations y = ⳵H共x , p兲 / ⳵ pa.
a 2
2
In some monographs 共for instance, see Refs. 34 and 31兲, it is considered the factor of 1 / 2 in the right sides of 共3兲 and 共4兲.
We emphasize that in this paper 共for simplicity兲 the Hamilton and Lagrange functions will be supposed to be regular and
related mutually by Legendre transforms.
Following terminology from Ref. 31, we say that Nai 共14兲 and *Nij 共15兲 are L-dual if L and H
are related by Legendre transform 共5兲 or 共6兲. In the following constructions, we shall consider that
to Legendre transform there are associated the diffeomorphisms,
␸:TM 傻 U → *U 傺 T*M,共xi,y a兲 → xi,pa = 冉 ⳵L共x,y兲

⳵ya
冊
and
␺:T*M 傻 *U → U 傺 TM,共xi,pa兲 → xi,y a = 冉 ⳵H共x,p兲

⳵ pa
, 冊
allowing to define, respectively, pull back and push forward of geometric objects 共functions,
vectors, differential forms, connections, tensors, etc.兲 from *U to U and from U to *U, i.e., we
define L-dual geometric objects. For instance, for a differentiable function 1 f on U, we define a
differentiable function 1 f * ⯐ 1 f ⴰ ␺ = 1 f ⴰ ␸−1 on *U, and 共inversely兲 for a differentiable function 2 f
on *U, we have a differentiable 共 2 f兲0 ⯐ 2 f ⴰ ␸ = 2 f ⴰ ␺−1 on U. Note that H = L ⴰ ␸−1 and L = H ⴰ ␺−1.
Similarly, for any vector field X on U, we get a vector field *X ⯐ T␸ ⴰ X ⴰ ␸−1 = T␺−1 ⴰ X ⴰ ␺ on *U,
and 共inversely兲 for a vector field *X on *U, we get a vector field ⴰX ⯐ T␺ ⴰ X ⴰ ␺−1 = T␸−1 ⴰ *X ⴰ ␸ on
U, where, for example, T␸ is the tangent map to ␸. Dualizing the vector constructions, we obtain
that for any 1-form ␻ on U, there is 1-form *␻ ⯐ 共T␸兲* ⴰ ␻ ⴰ ␸−1 = 共T␺−1兲* ⴰ ␻ ⴰ ␺ on *U and 共in-
versely兲 for any form *␻ on *U, we can consider ⴰ␻ ⯐ 共T␺兲* ⴰ *␻ ⴰ ␺−1 = 共T␸−1兲 ⴰ *␻ ⴰ ␸ on U, where,
for example, 共T␺兲* denotes the cotangent map to 共T␺兲.
Let vTM and vT*M be the vertical distributions on TM and T*M, respectively.
Definition 2.4: Any Whitney sums
TTM = hTM 丣 vTM 共7兲

and
TT*M = hT*M 丣 vT*M 共8兲

define, respectively, nonlinear connection (N-connection) structures parametrized by the local
vector fields,
⳵ ⳵
i − Ni 共x,y兲
a
ei = on TM
⳵x ⳵ya
and
*e = ⳵ + *N 共x,p兲 ⳵ on T*M .
i
⳵xi ia
⳵p a
One says that a N-connection defines on TM, or T*M, a conventional horizontal 共h兲 and
vertical 共v兲 splitting 共decomposition兲.
Let us consider a regular curve c共␶兲 with real parameter ␶, when c : ␶ 苸关0 , 1兴 → xi共␶兲傺 U. It
can be lifted to ␲−1共U兲傺 TM as c̃共␶兲 : ␶ 苸关0 , 1兴 → 共xi共␶兲 , y i共␶兲 = dxi / d␶兲 since the vector field
dx / d␶ does not vanish on
i
TM . Following techniques from variational calculus, one proves the
following.
Theorem 2.1: The Euler–Lagrange equations,
d ⳵L ⳵L
− = 0, 共9兲
d␶ ⳵ y i ⳵xi
are equivalent to the Hamilton–Jacobi equations
dxi ⳵H dpi ⳵H
= and =− i 共10兲
d␶ ⳵ pi d␶ ⳵x
and to the nonlinear geodesic (semispray) equations
d 2x i
+ 2Gi共x,y兲 = 0, 共11兲
d␶2
where
1
Gi = gij
2
冉
⳵ 2L k ⳵ L
⳵ y j⳵xk
y − j
⳵x
冊
for gij being the inverse to gij (4).
Let us consider on T*M the canonical symplectic structure
␪ ⯐ dpi ∧ dxi . 共12兲

The Hamiltonian H defines an unique vector field on T*M,
⳵H ⳵ ⳵H ⳵
XH = − ,
⳵ pi ⳵xi ⳵xi ⳵ pi
by the equation
iXH␪ = − dH,
where iXH denotes the interior product by XH. The same holds for any function on T*M.
By Theorem 2.1 one has the following.
Corollary 2.1: The Hamilton–Jacobi equations (10) are equivalent to
dxi dpa
= 兵H,xi其 and = 兵H,pa其,
d␶ d␶
where the Poisson structure is defined by brackets
兵 1 f, 2 f其 = ␪共X 1 f ,X 2 f 兲共13兲
for any functions 1 f共x , p兲 and 2 f共x , p兲 on T*M.

Proof: It can be obtained by a standard calculus in geometric mechanics. 䊐
The following theorem holds.
Theorem 2.2: The are canonical N-connections defined, respectively, by regular Lagrange
L共x , y兲 and/or Hamilton H共x , p兲 fundamental functions,
⳵Ga
Nai ⯐ 共14兲
⳵yi
and
ij
2
冋
*N ⯐ 1 兵 *g ,H其 −
ij
⳵ 2H *
g
⳵ pk⳵x共i兲共j兲k
, 册共15兲
where *gij is the inverse to *gij (3) and, for instance, a共ij兲 = aij + a ji denotes symmetrization of
indices.
Proof: We can verify, respectively, that on any open sets U 傺 TM and U* 傺 T*M, coefficients
共14兲 and 共15兲 satisfy the conditions of Definition 2.4. For details, see Ch. 9 in Ref. 34 and Ch. 5
in Ref. 31. 䊐
On 共co兲 tangent bundles endowed with N-connection structure, it is convenient to elaborate a

covariant calculus adapted to this structure, i.e., preserving the conventional splitting of tensors
and other geometric objects 共such as connections, differential forms, etc.兲 into horizontal 共h兲 and
vertical 共v兲 components. In brief, such distinguished 共by N-connection兲 components are called,
respectively, d-objects, d-field 共for some physical fields of tensor, spinor nature, etc.兲, d-tensors,
d-vectors, d-forms, d-connections, etc., see details in Refs. 31–34, 38, and 29.
Proposition 2.1: There are canonical frame structures [local N-adapted (co-)bases] defined
by canonical N-connections,
e␣ = ei =冉 ⳵
⳵x
a ⳵
i − Ni
⳵y a ,eb =
⳵
⳵yb
冊 on TM , 共16兲
␣ 冉
*e = *e i = ⳵ + *N ⳵ , *e b = ⳵
i ia
⳵x ⳵ pa ⳵ pb
冊 on T*M , 共17兲
and their dual (coframe) structures
e␣ = 共ei = dxi,eb = dy b + Nbi dxi兲 on 共TM兲* , 共18兲
*e␣ = 共 *ei = dxi, *e = dp − *N dxi兲 on 共T*M兲* , 共19兲

p b ib
when e␣e␤ = ␦␣␤ and *e␣ *e␤ = ␦␣␤, where by we note the interior products and ␦␣␤ being the
Kronecker delta symbol.
Proof: It follows by construction under the condition that such frames should depend linearly
on coefficients of respective N-connections. 䊐
One says that certain geometric objects are defined on TM 共or T*M兲 in N-adapted form
共equivalently, in distinguished form, in brief, d-form兲 if they are given by coefficients defined with
respect to frames e␣ 共16兲 and coframes e␣ 共18兲 and their tensor products 关with respect to frames
*e 共17兲 and coframes *e␣ 共19兲 and their tensor products兴. We shall use “boldface” letters in order
␣
to emphasize that certain spaces 共or geometric objects兲 are in N-adapted form.
Definition 2.5: The N-lifts of the fundamental tensor fields *gab (3) and gab (4) are, respec-
tively,
*g = *g *e␣ 丢 *e␤ = *g 共x,p兲ei 丢 e j + *gab共x,p兲 *e 丢 *e , 共20兲

␣␤ ij a b
on T*M, where *gij is inverse to *gab, and
g = g␣␤e␣ 丢 e␤ = gij共x,y兲ei 丢 e j + gab共x,y兲ea 丢 eb ,

on TM, where gij is stated by gab following gij = gn+i n+j.
The following proposition holds.
Proposition 2.2: The canonical N-connections N (14) and *N (15) define, respectively, the
canonical almost complex structures J, on TM, and *J, on T*M.
Proof: On TM one introduces the linear operator J acting on e␣ = 共ei , eb兲共16兲 as follows:
J共ei兲 = − en+i and J共en+i兲 = ei .

It is clear that J defines globally an almost complex structure 共J ⴰ J = −I for I being the unity
matrix兲 on TM completely determined for Lagrange spaces by a L共x , y兲. Now we provide the proof
for T*M. Let us introduce a linear operator *J acting on *e␣ = 共 *ei , *eb兲共17兲 by following formulas
*J共 *e 兲 = − g* en+i and *J共 *en+i兲 = *ei .

i ia
Then *J defines globally an almost complex structure 共*J ⴰ *J = −I for I being the unity matrix兲 on
T*M completely determined for Hamilton spaces by a H共x , p兲. 䊐
Definition 2.6: The Neijenhuis tensor field for the almost complex structure *J on T*M, or J
on TM, defined by a N-connection (equivalently, the curvature of N-connection) is
*J
⍀共X,Y兲 = − 关X,Y兴 + 关 *JX, *JY兴 − *J关 *JX,Y兴 − *J关X, *JY兴,
J
⍀共X,Y兲 = − 关X,Y兴 + 关JX,JY兴 − J关JX,Y兴 − J关X,JY兴, 共21兲
for any d-vectors X and Y.
Hereafter, for simplicity and if one shall not result in ambiguities, we shall present the
N-adapted component formulas for geometric objects on T*M 共those for TM being similar兲, or
inversely.
With respect to N-adapted bases, the components of the Neijenhuis tensor *J⍀ involve the
coefficients *⍀ija,
⳵ *Nia ⳵ *N ja ⳵ *N ja ⳵ *Nia
*⍀ = − + *Nib − *N jb . 共22兲
ija
⳵x j ⳵xi ⳵ pb ⳵ pb
They define the coefficients of the N-connection curvature. One gets a complex structure, i.e.,
*J = 0 under some quite complicated conditions on gab共x , p兲 and N which will be not written
⍀ ia
here.
It should be noted here that the N-adapted 共co-兲 bases 共16兲–共19兲 are nonholonomic with
nontrivial anholonomy coefficients. For instance,
␥
关e␣,e␤兴 = e␣e␤ − e␤e␣ = W␣␤ e␥ , 共23兲
with 共antisymmetric兲 anholonomy coefficients b
Wia = ⳵aNbi and Waji = ⍀aij, with
⳵Nai ⳵Naj b ⳵N j b ⳵Ni

a a
⍀aij = − + N − N .
⳵x j ⳵xi i
⳵ pb j
⳵ pb
B. Almost Kähler Lagrange–Hamilton structures

We can adapt to N-connections various geometric structures on TM and T*M. For instance,
we can consider the following.
Definition 2.7: One calls an almost symplectic structure on T*M a nondegenerate N-adapted
2-form

␪ = 21 ␪␣␤共u兲 *e␣ ∧ *e␤ .
The following Proposition holds.
Proposition 2.3: For any T␪ on T*M, there is a unique N-connection *N = 兵 *Nia其 satisfying
the conditions

␪ = 共h *X, v *Y兲 = 0 and ␪ ⬟ h ␪ + v ␪ , 共24兲
for any *X = h *X + v *X, *Y = h *Y + v *Y and h T␪共 *X , *Y兲 ⯐ T␪共h *X , h *Y兲,

v ␪共 X , *Y兲 ⯐ ␪共v *X , v *Y兲.
T * T
Proof: For *X = *e␣ = 共 *ei , *ea兲 and *Y = *e␤ = 共 *el , *eb兲, where *e␣ is a N-adapted basis of
type 共17兲, we write the first equation in 共24兲 in the form
T
␪ = T␪ 共 *e i, *e a兲 = T␪ 冉 ⳵ ⳵
i,
⳵x ⳵ pa
冊
− *Nib T␪
⳵ ⳵
,
⳵ pb ⳵ pa
冉
= 0. 冊
Such conditions uniquely define *Nib because T
␪ is nondegenerate, that is,
rank兩 T␪共⳵ / ⳵ pb , ⳵ / ⳵ pa兲兩 = n. Setting locally
T
␪ = 21 T␪ij共u兲ei ∧ e j + 21 T␪ab共u兲 Tea ∧ *eb , 共25兲
where the first term is for h T␪ and the second term is v T␪, we get the second formula in 共24兲.
Finally, we note that in this proposition the constructed N-connection *Nib, in general, is not a
canonical one 共15兲. 䊐
In a similar form, as in Proposition 2.3, we can construct a unique N-connection N = 兵Nai 其 for
any almost symplectic structure ␪ on TM 共from formal point of view, we have to omit in formulas
the symbols “ⴱ” and “T” and use variables y a instead of pa兲.
A N-connection *N 共8兲 defines a unique decomposition of a d-vector *X = *Xh + *Xv on T*M
for *Xh = h *X and *Xv = v *X, where the projectors h and v define, respectively, the distributions
*N and *V. They have the properties
h + v = I, h2 = h, v2 = v, h ⴰ v = v ⴰ h = 0.
This allows us to introduce on T*M the almost product operator
*P ⯐ I − 2v = 2h − I
acting on *e␣ = 共 *ei , *eb兲共17兲 following formulas

*P共 *e 兲 = *e and *P共 *eb兲 = − *eb .
i i
In a similar form, a N-connection N 共7兲 induces an almost product structure P on TM. One uses
also the almost tangent operators
⳵
J共ei兲 = en+i and J共ea兲 = 0 or J= 丢 dxi ,
⳵yi
*J共 *e 兲 = *g *eb *J共 *eb兲 = 0 ⳵

and or J = *gia 丢 dxi .
i ib
⳵ pa
The operators *P, *J, and *J are, respectively, L-dual to P, J, and J if and only if *N and N are
L-dual.
For the above-introduced almost complex and almost product operators, it is straightforward
to prove the following.
Proposition 2.4: Let 共N , *N兲 be a pair of L-dual N-connections. Then, we can construct
canonical d-tensor fields [defined, respectively, by L共x , y兲 and H共x , p兲 related by Legendre trans-
forms (5) and/or (6)],
J = − ␦ai ea 丢 ei + ␦iaei 丢 ea, *J = − *g *ea 丢 *ei + *gia *e 丢 *e

ia i a
corresponding to the L-dual pair of almost complex structures 共J , *J兲,
P = e i 丢 e i − e a 丢 e a, *P = *e 丢 *e i − *e a 丢 *e
i a
corresponding to the L-dual pair of almost product structures 共P , *P兲, and almost symplectic
structures
␪ = gaj共x,y兲ea ∧ ei and *␪ = ␦ a *e ∧ *e i . 共26兲

i a
Let us consider an important example.

A Finsler manifold 共space兲 Fn = 共M , F共x , y兲兲 is a particular case of Lagrange space, when the
regular Lagrangian L = F2 is defined by a fundamental Finsler function F共x , y兲 satisfying the
following conditions: 共1兲 the positive function F is differentiable function on TM continuous on
the null section of projection ␲ : TM → M; 共2兲 F共x , ␭y兲 = 兩␭兩F共x , y兲, i.e., it is 1-homogeneous on the
fibers of TM; and 共3兲 the Hessian 共4兲 defined in this case by F2 is positively defined on TM .
It is used also the notion of Cartan space Cn = 共M , C共x , p兲兲 for H = C2共x , p兲 as a particular
共1-homogeneous on fiber coordinates兲 case of Hamilton space when C satisfies the same condi-
tions as F but with respect to coordinates pa 共in brief, we can say that Cartan spaces are Finsler
spaces on T*M, see details Ref. 31兲. In a similar manner as for Lagrange and Hamilton spaces, we
can introduce the concept of L-dual geometric objects on Finsler and Cartan spaces. For simplic-
ity, in this work we shall emphasize the bulk constructions for Hamilton spaces considering that by
Legendre transform, we can generate similar ones for Lagrange spaces and, in particular, for
respective Finsler and Cartan geometries.
Definition 2.8: An almost Hermitian model of a cotangent bundle T*M (or tangent bundle
TM) equipped with a N-connection structure *N (or N) is defined by a triple *H2n
= 共T*M , *␪ , *J兲, where *␪共 *X , *Y兲 ⯐ *g共 *JX , Y兲 [or H2n = 共TM , ␪ , J兲, where ␪共X , Y兲 ⯐ g共JX , Y兲].
A space *H2n is almost Kähler, denoted *K2n, if d *␪ = 0.
The following theorem holds Theorem 2.3: The Lagrange and Hamilton spaces can be
represented as almost Kähler spaces on, respectively, on TM and T*M endowed with canonical
N-connection structures N 共14兲 and *N 共15兲.
Proof: It follows from the existence on TM and T*M of canonical 1-forms, respectively,
defined by a regular Lagrangian L and Hamiltonian H related by a Legendre transform,
⳵L i
␻= e and *␻ = p dxi ,
i
⳵yi
for which
␪ = d␻ and *␪ = d *␻ ,
see 共12兲. As a result, we get that d␪ = 0 and d *␪ = 0, which correspond to the Definition 2.8. 䊐
In this paper, we shall work with almost Kähler models on 共co兲 tangent bundles defined
canonically by 共pseudo兲 Riemannian metrics on base manifolds, see 共1兲, and/or 共effective, or for
regular mechanics兲 Lagrangians 共Hamiltonians兲. Finally, we emphasize that realistic classical and
quantum models are elaborated in explicit form for some classes of linear connections defined to
satisfy certain physical principles and constructed geometrically to be adapted, or not, to a
N-connection structure. We shall perform such classical and quantum constructions in the follow-
ing sections.
III. NONLINEAR CONNECTIONS AND ALMOST SYMPLECTIC GEOMETRY

In this section, we consider the almost symplectic geometry induced by regular Hamiltonians
and corresponding canonical N-connections defined naturally, for gravitational and/or geometric
mechanics models, on 共co兲 tangent bundles.
A. Canonical N-connections and d-connections for Lagrange and Hamilton spaces

Let D be a linear connection on TM when for a L-duality between the tangent and corre-
sponding cotangent bundles there are defined pull-back and push-forward maps. We can define a
linear connection *D on T*M as follows:
*D *Y ⯐ 共D ⴰ ⴰY兲* ,
*X X
for any vector fields *X and *Y on T*M. Inversely, for any linear connection *D on T*M, we get
a linear connection ⴰD on TM, following the rule
ⴰ
DXY ⯐ 共 *D *X *Y兲ⴰ
for any vector fields X and Y on TM.

Definition 3.1: A linear connection D or (*D兲 on TM (or T*M) is a distinguished connection
(d-connection) if it is compatible with the almost product structure DP = 0 (or *D *P = 0).

For L-dual Lagrange and Hamilton spaces, one follows that DP = 0 induces *D *P = 0, and
inversely. The coefficients of d-connections can be defined with respect to N-adapted frames,
De␤e␥ ⯐ ⌫␣␤␥e␣ and D *e *e␥ ⯐ *⌫␣␤␥ *e␣

␤
with corresponding N-adapted splitting,
Deke j ⯐ Li jkei, Dekeb ⯐ Ĺabkea, Dece j ⯐ Ći jcei, Deceb ⯐ Cabcea ,
and
*D *e ⯐ *L i *e , *D *eb ⯐ − *Ĺ b *ea ,

*e j jk i ek a k
k
*D *e ⯐ *Ći c *e , *D *eb ⯐ − *C bc *ea ,

*ec j j i *ec a
when
⌫␣␤␥ = 兵Li jk,Ĺabk,Ći jc,Cabc其 and *⌫ ␣

␤␥ = 兵 *Li jk, *Ĺabk, *Ći jc, *Cabc其
define corresponding h- and v-splittings of covariant derivatives
D = 共 hD, vD兲 and *D = 共 *D, *D兲,

h v
where hD = 兵L jk , Ĺ bk其, vD = 兵Ć jc , C bc其 and h*D = 兵 *L jk , *Ĺa k其, *v D = 兵 *Ćicj , *Cabc其.
i a i a i b
We shall work with a more special class of d-connections.

n
Definition 3.2: A linear connection nD ( or * D) on TM ( or T *M) is N-linear if it preserves
under parallelism the Whitney sum N (7) [or *N (8)] i.e., it is a d-connection and is compatible
n
with the almost tangent structure J (or *J), i.e., nDJ = 0 (or * D *J = 0).
This is a class of N-linear connections characterized by the property.
Proposition 3.1: The N-adapted coefficients of N-linear connections nD ⯐ 兵 n⌫␣␤␥
= 共L jk , Ći jc兲其 and * D ⯐ 兵⌫␣␤␥ = 共 *Li jk , *Ći jc兲其 of respective L-dual Lagrange and Hamilton spaces
i n
are
n
⌫␣␤␥ = 兵Li jk,Ĺn+in+jk = Li jk,Ći jc,Cn+in+jc = Ći jc其共27兲
and
*n⌫ ␣ 兵 *Li jk, *Ĺn+jn+ik = *Li jk, *Ći jc, *Cn+jn+ic = *Ći jc其. 共28兲
␤␥ =
Proof: By a straightforward computation for coefficients 共27兲 and 共28兲, when, for instance,
*nD = 共 *D , *Da兲 for *D = 兵 *Li 其 and *D = 兵 *Ći c其, we can verify that the conditions considered in
␣ h i v h jk v j
Definition 3.2 are satisfied. 䊐
The connection 1-form of N-linear connection * ⌫␣␤ = * ⌫␣␤␥ *e␥ = 兵 * ⌫i j其 is defined by
n n n
*n⌫i = *Li *ek + *Ći c *e , 共29兲

j jk j c
n n
where the v-components * ⌫ab are identified with the h-components * ⌫i j following the formulas
*n⌫n+i = *n⌫a . The following theorem holds.
n+j b
n
Theorem 3.1: On a Hamilton space, the structure equations for * ⌫i j are
n
d *e k − *e j ∧ * ⌫ k j = − *T k ,
n
d *e c + *e a ∧ * ⌫ ac = − *T c ,
n n n
d * ⌫ i j − * ⌫ ik ∧ * ⌫ k j = − *R i j ,
where the 2-form of torsion T␣ = 兵Tk , Tc其 is computed
*Tk = 1 *Tk *ei ∧ *e j + *Ćk c *e j ∧ *e ,

2 ij j c
*T = 1 *⍀ *ei ∧ *e j + 1 * P *ei ∧ *e + 1 *S bc *e ∧ *ec ,

a 2 ija 2 aic c 2 a b
for the N-connection curvature *⍀ija (22) and
*T k = *L k − *L k , *S bc = *Ćb c − *Ćc b ,
ij ij ji a a a
*P
aic = *gae共 *Leic − *ee共 *Nic兲兲, 共30兲
and the 2-form of curvature *R␣␤ = 共 *Ri j , *Rab兲, with *Ri j = *Rn+in+j, is computed
*R i = 1 *R i 1 * i bc *
jkm e ∧ e + P jk e ∧ ec + 2 S j e b ∧ *e c ,
* k * m * i c* k *
j 2
where
*R i
jkm = *em共 *Li jk兲 − *ek共 *Li jm兲 + *Lo jk *Liom − *Lo jm *Liok + *Ći jo *⍀oka ,
* Pi c = *ec共 *Li 兲 − *D 共 *Ći c兲 + *Ći o * Pc ,

jk jk h k j j ko
*Si bc = *ec共 *Ći b兲 − *eb共 *Ći c兲 + *Ćk b *Ći c − *Ćk c *Ći b 共31兲
j j j j k j k
for “nonboldface” *ec = ⳵ / ⳵ pc.

Proof: It is a straightforward differential computation for 1-form 共29兲. 䊐
There is a “L-dual” Theorem for Lagrange spaces,34 see generalizations of nonholonomic
manifolds and deformation quantization of gravity in Refs. 30 and 24, similarly to Theorem 3.1.
From formal point of view, we have to change H into L and consider the constructions on TM,
omitting the labels “ⴱ” and using coordinates 共xi , y a兲 instead of 共xi , pa兲.
B. Hamilton–Fedosov spaces and almost Kähler structures

There are canonical N-linear connections on TM and T *M completely defined, respectively,
by the fundamental Lagrange 共see Ref. 34 and, for applications to geometric quantization, Refs. 21
and 24兲 and Hamilton functions.
ˆ
Theorem 3.2: There exists a canonical N-linear connection *D̂ ⯐ 兵 *⌫ˆ ␣ = 共 *L̂i , *Ćic兲其 on a
␤␥ jk j
Hamilton space Hn = 共M , H共x , p兲兲 endowed with canonical N-connection *Nij (15) satisfying the
conditions: (1) h*D̂i *gkj = 0 and *v D̂a *gkj = 0, (2) *Tkij = 0 and *Sabc = 0, and (3) *D̂ is completely
defined by H共x , p兲, i.e., by d-metric (20).
Proof: Let us consider the N-adapted coefficients
*L̂i = 1 *gis共 *e 共 *g 兲 + *e 共 *g 兲 − *e 共 *g 兲兲,

jk 2 j sk k js s jk
ˆi c 1*
j = − 2 g js e 共 g 兲共32兲
*Ć * c * si
defined with respect to N-adapted bases *e␣ 共17兲 and coframes *e␣ 共19兲 defined by *Nij 共15兲. By
a direct computation, we can verify that the conditions of this Theorem are satisfied only for such
coefficients and their coordinate/frame transform. 䊐
The coefficients 共32兲 are just the Christoffel symbols on 共co兲 tangent space T *M, defined by
H. Nevertheless, for a Hamilton space endowed with canonical N-connection, d-connection, and
metric structure, there are nontrivial torsion components induced by the nonholonomic distribution
defined by H, see * Paic 共30兲.
From Theorem 3.2, one follows an important property.
Corollary 3.1: The canonical d-connection *D̂ is an almost symplectic d-connection satisfy-
ing the conditions
*D̂ *␪ = 0 and *D̂ *J = 0 共33兲
and being completely defined by a Hamiltonian H共x , p兲 for *␪共 *X , · 兲 ⯐ *g共 *JX , · 兲.
Proof: A Hamilton space can be equivalently transformed into almost Kähler space *K2n, see
Definition 2.8. Considering the d-tensor fields associated with *␪ and *J, see Proposition 2.4, and
a covariant N-adapted calculus defined by the canonical d-connection coefficients 共32兲 we can
verify that the almost symplectic structure compatibility conditions 共33兲 are satisfied. 䊐
In Ref. 20, it was introduced the concept of Lagrange–Fedosov manifold as nonholonomic
manifold with the N-connection and almost symplectic structure defined by a fundamental 共in
general, effective兲 Lagrange function L共x , y兲. On cotangent bundles, we can consider the follow-
ing.
Definition 3.3: A Hamilton–Fedosov space is a cotangent bundle endowed with canonical
N-connection and the almost Kähler structure induced by a fundamental Hamilton function
H共x , p兲.
There are Hamilton–Fedosov spaces defined completely by a lift 共1兲 of a 共pseudo兲 Riemannian
metric on base M.
Theorem 3.3: Any Einstein manifold associated with a solution of (2), for a lift (1) on T *M,
defines canonically a Hamilton–Fedosov space.
Proof: We sketch the idea for such constructions. Let us fix any values eii⬘共x , p兲 in 共1兲 and
associate gab共x , p兲 with a *gab共x , p兲共3兲. This defines correspondingly the values *g 共20兲 and *N
共15兲. As a result, we construct an effective Hamilton space, which can be modeled as a canonical
almost symplectic structure as we described above. For classical configurations, the values eii⬘ can
be ␦ii⬘, but for quantum models they should be defined by a scheme of dequantization or semi-
classical approximation in quantum gravity.
There is a particular case of Cartan–Fedosov spaces with 0-homogeneous on variables p
components eii⬘共x , p兲 resulting in a similar homogeneity for gab共x , p兲 and H = gab pa pb when *N is
determined from 共15兲 with *g from 共21兲.
The problem is more sophisticated in the case of general Hamilton–Fedosov spaces. For
certain physical important four dimensional spaces 共used in general relativity兲 with nonholonomic
splitting of dimensions as 2 + 2 and, for instance, if dim M = 2, we can fix such local coordinate
systems when 共3兲 is integrable for certain solvable partial differential equations for eii⬘共x , p兲, but
this may not hold true for other parametrizations and higher dimensions. A general approach
should include the case of Eisenhart–Hamilton spaces and their Fedosov quantum deformation
analogs, with both symmetric and nonsymmetric components for 共3兲 resulting because of any
general quantum nonholonomic Legendre transform. Such constructions should dub on cotangent
bundles those for Eisenhart–Lagrange/–Finsler spaces 共see Chap. 8 in Ref. 33 and Refs. 46 and
47兲. The length of this paper does not allow us to present a detailed proof because it is connected
with a sophisticated geometric techniques for nonlinear connections and nonsymmetric metrics
arising both in the case of nonholonomic Ricci flows and quantum nonholonomic deformations
and/or symplectic transform in gravity and geometric mechanics, see recent results in Refs.
48–50. 䊐
Finally, we note that similar almost symplectic models can be performed for Cartan spaces
when the Hamiltonian is homogeneous on vertical coordinates.
IV. FEDOSOV OPERATOR PAIRS FOR HAMILTON SPACES

In this section, we shall apply the method of deformation quantization elaborated in Refs. 5,
6, and 35 to define two classes of canonical operators which are necessary to quantize the
Hamilton–Fedosov spaces and related subspaces on cotangent bundles defined by lifts of Einstein
metrics. We shall address precisely the question how the geometry of cotangent bundles and
related deformation quantization change under symplectic transforms and elaborate a formalism
which preserves the form of Hamilton–Jacobi equations both on classical and quantum levels.
A. Canonical Fedosov–Hamilton operators

The formalism of deformation quantization can be developed by using the space
C⬁共 *K2n兲关关v兴兴 of formal series in the variable v with coefficients from C⬁共 *K2n兲 on an almost
Poisson manifold 共 *K2n , 兵· , · 其兲, see the almost symplectic form *␪ and the Poisson brackets 共13兲.
Using the associative algebra structure on C⬁共 *K2n兲关关v兴兴 with a v-linear and v-adically continuous
star product
⬁
1
f ⴱ f = 兺 rC共 1 f, 2 f兲vr ,
2
共34兲
r=0
where rC , r 艌 0, are bilinear operators on C⬁共 *K2n兲 with 0C共 1 f , 2 f兲 = 1 f 2 f and 1C共 1 f , 2 f兲
− 1C共 2 f , 1 f兲 = i兵 1 f , 2 f其; for i being the complex unity, we can construct a formal Wick product
1
冉v ⳵2
冊
a ⴰ 2a共z兲 ⯐ exp i *⌳␣␤ ␣ ␤ 兩 1a共z兲 2a共z关1兴兲兩z=z关1兴
2 ⳵z ⳵z关1兴
共35兲
for two elements a and b defined by series of the type
a共v,z兲 = 兺 ar,兵␣其共u兲z兵␣其vr , 共36兲

r艌0,兩兵␣其兩艌0
where by 兵␣其 we label a multi-index and *⌳␣␤ ⯐ *␪␣␤ − i *g␣␤, where *␪␣␤ is the symplectic form
共26兲, with “up” indices and *g␣␤ is the inverse to d-tensor 共20兲. This defines a formal Wick algebra
*W associated with the tangent space T *K2n for u 苸 *K2n, where the local coordinates on *K2n
u u
are parametrized in the form u = 兵u␣其 and the local coordinates on Tu *K2n are labeled 共u , z兲
= 共u␣ , z␤兲, where z␤ are fiber coordinates.
We trivially extend the fiber product 共35兲 to the space of *W-valued N-adapted differential
forms *W 丢 ⌳, where by ⌳ we note the usual exterior product of the scalar forms and *W is the
sheaf of smooth sections of *W. There is a standard grading on ⌳ noted dega. We also introduce
gradings degv , degs , dega on *W 丢 ⌳ defined on homogeneous elements v , z␣, *e␣, when
degv共v兲 = 1, degs共z␣兲 = 1, dega共 *e␣兲 = 1, and all other gradings of the elements v , z␣, *e␣ are set to
zero. As a result, the product ⴰ from 共35兲 on *W 丢 ⌳ is bigraded, written with respect to the
grading Deg= 2 degv + degs and the grading dega.
ˆ
The canonical d-connection *D̂ ⯐ 兵 *⌫ˆ ␣ = 共 *L̂i , *Ćic兲其 with coefficients 共32兲 can be extended
␤␥ jk j
to an operator
ˆ ␥ z*e 共␣兲兲丢共 *e␣ ∧ ␭兲 + a 丢 d␭

*D̂共a 丢 ␭兲 ⯐ 共 *e 共a兲 − u␤ *⌫ 共37兲
␣ ␣␤ ␣
z*
on *W 丢 ⌳, where the N-adapted basis e␣ is *e␣ redefined in z-variables. This canonical almost
symplectic d-connection *D̂ is a N-adapted dega-graded derivation of the distinguished algebra
共 *W 丢 ⌳ , ⴰ 兲, in brief, called d-algebra: this follows from formulas 共35兲 and 共37兲.
Definition 4.1: The Fedosov-Hamilton operators *␦ and *␦−1 on *W 丢 ⌳ are defined
冦冧
i ␣*
*␦␣共a兲 = *e␣ ∧ z*e 共a兲 *␦−1共a兲 =
z e␣共a兲, if p + q ⬎ 0
␣ and p+q 共38兲
0, if p = q = 0,
where any a 苸 *W 丢 ⌳ is homogeneous with respect to the gradings degs and dega with degs共a兲
= p and dega共a兲 = q.
The d-operators 共38兲 define the formula a = 共 *␦ *␦−1 + *␦−1␦ + ␴兲共a兲, where a 哫 ␴共a兲 is the
projection on the 共degs , dega兲-bihomogeneous part of a of degree zero, degs共a兲 = dega共a兲 = 0; *␦ is
also a dega-graded derivation of the d-algebra 共 *W 丢 ⌳ , ⴰ 兲.
B. Fedosov–Hamilton operator pairs

Having defined d-operators 共38兲, we can perform a Fedosov-type quantization of Hamilton
spaces in a L-dual form to Lagrange spaces.21,24 Nevertheless, such constructions would not
reflect completely the symplectic properties of Hamilton spaces and their quantum deformations.
For a cotangent bundle ␲* : T*M → M, a N-connection *N 共15兲 is a supplementary distribution
hT*M of the vertical distribution vT*M = ker ␲*, where ␲* is the tangent map of ␲*. It is often
more convenient to consider a N-connection as an almost product structure *P, see Proposition
2.4, such that vT*M = ker共I + *P兲. If f 苸 Diff共T*M兲, the push forward of *N by f generally fails to
be a connection. This constrains us to extend the definition of connection, see details in Chap. 8 of
Ref. 31.
Definition 4.2: A connection pair *␾ is an almost product structure *␾ on T*M such that the
horizontal bundle hT*M ⯐ ker共I − *␾兲 is supplementary to vT*M and the oblique bundle
wT*M ⯐ ker共I + *␾兲 defines the nonholonomic splitting (Whitney sum),
TT*M = hT*M 丣 wT*M . 共39兲

We can consider preferred *␾-adapted frame and coframe structures induced by the coeffi-
cients of N-connection.
Proposition 4.1: There are a canonical connection pair *␾ and associated frames and cof-
rames,
␾
e␣ = 冉 ␾
ei =
⳵ *
⳵x i + Nia
⳵ w b
⳵ p␣
, e =
⳵
⳵ pb
冊
− *gbi ␾ei , 共40兲
␾ ␣
e = 共 ␾ei = dxi + *gbi web, web = dpb − *Nibdxi兲, 共41兲
where *gab (3) and *Nia (15) are generated by a Hamilton fundamental function H共x , p兲.
Proof: The proof is similar to that for Proposition 2.1 but adapted both the splitting 共39兲, with
respective h- and v-projections, 2h⬘ = I + *␾ and 2w = I − *␾, and to the splitting 共8兲, with respective
h- and v-projections, 2h = I + *N and 2v = I − *N. 䊐
We note that for Hamilton spaces a connection-pair structure *␾ is symmetric, i.e., *Nia
= *Nai and *gab = *gba. For simplicity, in this work, we shall restrict our considerations only to
*␾-symmetric configurations.
In order to perform geometric constructions adapted both to the N-connection and almost
symplectic structure, it is necessary to work with *␾-adapted bases 共40兲 and 共41兲 instead of,
respectively, 共17兲 and 共19兲. For instance, every vector field X has two components,
X = hX + vX = h⬘X + wX,
where wX = w共X兲. This defines the class of *␾-tensor fields alternatively to that of d-tensor fields
considered in the previous sections.
Definition 4.3: A linear connection ␾D on T*M is a *␾-connection if
␾ ␾
D *␾ = 0 and Dw = 0.
We can characterize ␾D = 共 h␾D = 兵 ␾Lijk其 , w␾D = 兵 ␾Cijk其兲 by the coefficients computed with respect
to *␾-adapted frames 共40兲 and 共41兲,
␾
␾
D ␾e e j = ␾Lijk ␾e j, ␾
D ␾e e = ␾Lck
w b b w c
e ,
k k
␾
D wea we b = ␾C bca we c, ␾
D wea ␾e j = ␾C k j a ␾e k ,
for ␾Ln+jk
n+i
= −Ljki␾ and ␾Cn+j
n+ka
= − ␾Cka
j .
Theorem 4.1: There is a canonical *␾-connection ␾D̂ = 共 ␾L̂ijk , ␾Ĉka
j 兲 on T M completely de-
*
fined by a Hamilton fundamental function H共x , p兲 and satisfying the conditions
␾ ␾
h D̂共 g 兲
* ab =0 and *gab兲 = 0,
v D̂共
␾ k ␾
T̂ ij =0 and Ŝabc = 0.
Proof: The torsion and curvature of ␾D̂ are computed as in Theorem 3.1, see formulas 共30兲,
but with respect to 共40兲 and 共41兲. By straightforward computations, we can verify that for
␾ i
L̂ jk = 21 *gkm共 ␾ei *g
mj + ␾e j *gim − ␾em *gij兲 − 21 *gkm共A jim + Aijm + Amij兲,
␾
C jik = − 21 *gim共 we j *gmk + wek *gim + wem *g jk兲 − 21 *gim共B jmk + Bkmj + Bmkj兲, 共42兲
where
Aijm = *⍀ijm and Bmkj = *gmi共 *gka we j *Nia − *g ja wek *Nia兲,

the conditions of Theorem are satisfied. 䊐
The diffeomorphism symmetry of Hamilton mechanical models and possible lifts of Einstein
spaces on cotangent bundles are important characteristics of classical theories. In deformation
quantization models, this property can be preserved for regular N-connection structures.
Definition 4.4: A diffeomorphism f 苸 Diff共T*M兲 is called *N-regular if the restriction of
tangent map 共␲ f兲* : hT*M → TM to hT*M is a diffeomorphism.
For a connection pair *␾, the concept of *N-regularity imposes the equivalence of statements:
共1兲 the push-forward map of *␾ by f, defined as *␾ ¯ = f 共 *␾兲f −1, is a connection pair and 共2兲 f is
* *
*N-regular. This follows from the mapping ␲ : ker共I − *␾ ¯ 兲 → TM and the equality f 共ker共I
* *
− *␾兲兲 = ker共I − *␾
¯ 兲. The N-connection *N̄ associated with *␾¯ is the push forward of *N by f. One
says that *␾ and *␾ ¯ are f-related.
Following a calculus with local coordinate transforms, one proves the following.
Corollary 4.1:
(1) For ␾e␣ = 共 ␾ei , web兲 in 共x , p兲 and ␾ē␣ = 共 ␾ēi , wēb兲 in 共x̄共x , p兲 , p̄共x , p兲兲 induced, respectively, by
*␾ and *␾¯ , the following formulas hold:
f *共 ␾ei兲 = ␼ki ␾ēk and f *共 we b兲 = ␼

˜ bc wec ,
where
␼ki *Nka = ␾ei共p̄a兲 and ␼

˜ bc = web共p̄c兲 − web共x̄k兲 *N̄kc .
(2) The push forward of a N-connection *N by a *N-regular diffeomorphism is a N-connection

if f is fiber preserving, i.e., locally f共x , p兲 = 共x̄共x兲 , p̄共x , p兲兲.
Symplectic morphisms are diffeomorphisms which transform a symplectic form *␪

= dpi ∧ dxi into a symplectic form *¯␪ = f *共*␪兲 = dp̄i ∧ dx̄i. By coordinate transforms, one proves the
following.
Theorem 4.2: For a f 苸 Diff共T*M兲 being *N-regular and satisfying the condition
f *共 ␪兲共 ␾ei , web兲 = ␦bi , two from the next statements imply the third: (1) f is a symplectic morphism;
*
(2) *␾ is symmetric; (3) *␾ ¯ is symmetric.
The push forward of g 共20兲 results in the local form,
*
*ḡ = *ḡ *ē␣ 丢 *ē␤ = *ḡ 共x,p兲ēi 丢 ē j + *ḡab共x,p兲 *ē 丢 *ē ,

␣␤ ij a b
where *ḡij ⴰ f = ␼ik␼lj *gkl.

If ⵜ is a linear connection on T*M, we define its push forward by f on T*M as
¯ ¯ Ȳ ⯐ f 共ⵜ Y兲
ⵜ X * X
for Ȳ = f *共Y兲 and X̄ = f *共X兲. By coordinate parametrizations of diffeomorphisms, we can prove the
following.
Proposition 4.2:
(1) A connection ⵜ is a *␾-connection if and only if ⵜ ¯ is a *␾

¯ -connection.
(2) * * ¯ is compatible
ⵜ is compatible to metric g (almost symplectic ␪) structure if and only if ⵜ
to metric *ḡ (almost symplectic *¯␪) structure.
Proof: Locally, we can prove that the coefficients of *␾-connection ␾D are related to coeffi-
¯ -connection ␾D̄ by formulas
cients of *␾
␾ k
L̄ij =␼ ˜ jj⬘␼k ␾L j⬘ k + ␼i ␾ē j ␾ēi共xi⬘兲,
˜ ii⬘␼ k i k
⬘ ⬘⬘ ⬘
␾ i j
C̄ k
i
˜ kk⬘ ␾Ci⬘k j⬘ + ␼i wē j ␾ēi共xi⬘兲.
= ␼ i⬘␼ j ⬘␼
j i
⬘ ⬘
Using formulas
␾
D̄共 *␾兲 = 0 ⇔ ␾D共 *␾兲 = 0,
␾
D̄ *ḡ = ␾D̄关f *共 *g兲兴 = f *共 ␾D *g兲,
␾
D̄ *¯␪ = ␾D̄关f *共 *␪兲兴 = f *共 ␾D *␪兲,
where for *␾-connections we use the symbol ␾D instead of ⵜ, and by local computations, we can
verify that there are satisfied conditions,
␾
D̄ *ḡ = 0 ⇔ ␾D̄ *g = 0,
␾
D̄ *¯␪ = 0 ⇔ ␾D *␪ = 0.
䊐
Summarizing the above-presented results we get the proof of the following.
Theorem 4.3: There is a unique canonical *␾-connection ␾D̂ with coefficients (42) generated
by a Hamilton fundamental function H共x , p兲 which is almost symplectic and metric compatible and
adapted to symplectic morphism.
We can use ␾D̂ and frames 共40兲 and 共41兲 to construct instead of 共37兲 an extended on *W
丢 ⌳ operator
␾
D̂共a 丢 ␭兲 ⯐ 共 ␾e␣共a兲 − u␤ ␾⌫ˆ ␣␤
␥ z␾
e␣共a兲兲丢共 ␾e␣ ∧ ␭兲 + a 丢 d␭, 共43兲
where the local basis z␾e␣ is ␾e␣ redefined in z-variables. This allows us to introduce a new class
of operators adapted both to the N-connection structure and symplectic morphisms.
Definition 4.5: The Fedosov–Hamilton operator pairs ␾␦ and ␾␦−1 on *W 丢 ⌳ are defined
冦冧
i ␣␾
␾ ␾ ␣ z␾ ␾ −1 z e␣共a兲, if p+q⬎0
␦共a兲 = e ∧ e␣共a兲 and ␾ 共a兲 = p + q 共44兲
0, if p = q = 0,
where any a 苸 *W 丢 ⌳ is homogeneous with respect to the gradings degs and dega with degs共a兲
= p and dega共a兲 = q.
We note that the formulas 共44兲 are different from 共38兲 because they are defined for different
nonholonomic distributions and related adapted frame structures.
Using differential calculus of forms on *W 丢 ⌳, we prove the following.
Proposition 4.3: The torsion and curvature canonical d-operators of ␾D̂共a 丢 ␭兲 (43) are
computed
z␥ *
z␾
T̂ ⯐ ␪ ␾T̂␶ 共u兲 ␾e␣ ∧ ␾e␤ 共45兲
2 ␥␶ ␣␤
and
z ␥z ␸ ␾
z␾
R̂ ⯐ ␪␥␶ ␾R̂␶␸␣␤共u兲 ␾e␣ ∧ ␾e␤ , 共46兲
4
where the nontrivial coefficients of ␾T̂␣␶␤ and ␾R̂␶␸␣␤ are defined, respectively, by formulas (30)
and (31) for the canonical *␾-connection ␾D̂ = 共 ␾L̂ijk , ␾Ĉk ja兲 (42) with respect to (40) and (41).
By straightforward verifications, one gets the proof of the following.
Theorem 4.4: Any Fedosov–Hamilton operator pairs (44) and related extended operator pair
(43) are defined by torsion (45) and curvature (46) by following the formulas
i ␾ i
关 ␾D̂, ␾␦兴 = adWick共 z␾T̂兲 and D̂2 = − adWick共 z␾R̂兲, 共47兲
v v
where 关·,·兴 is the dega-graded commutator of endomorphisms of *W 丢 ⌳ and adWick is defined via
the dega-graded commutator in 共 *W 丢 ⌳ , ⴰ 兲.
The formulas 共47兲 can be restated for any metric compatible d-connection and *␾-connection
structures on T *M. Finally, we conclude that the operators constructed in this section are invariant
under diffeomorphisms and in particular, under symplectic morphisms.
V. DEFORMATION QUANTIZATION OF HAMILTON AND EINSTEIN SPACES

The aim of this section is to provide the main Fedosov’s theorems for Hamilton spaces and
show how the Einstein manifolds can be encoded into the topological structure of such quantized
nonholonomic spaces.
A. Fedosov’s theorems for connection pairs
The theorems will be formulated for the canonical *␾-connection ␾D̂ 共42兲.
Theorem 5.1: A Hamilton fundamental function H共x , p兲 defines a flat Fedosov *␾-connection
␾ i
D̂ ⯐ − ␾␦ + ␾D̂ − adWick共 ␾r兲
v
satisfying the condition ␾D̂2 = 0, where the unique element ␾r 苸 *W 丢 ⌳, dega共 ␾r兲 = 1, ␾␦−1 ␾r = 0,
is a solution of equation
i␾ ␾
␾
␦ ␾r = ␾T̂ + ␾R̂ + ␾D̂ ␾r − r ⴰ r.
v
The solution for ␾r can be computed recursively with respect to the total degree Deg in the form
␾ 共0兲
r = ␾r共1兲 = 0, ␾ 共2兲
r = ␾␦−1 ␾T̂,
␾ 共3兲
r = ␾␦−1 冉 ␾
R̂ + ␾D̂ ␾r共2兲 −
i ␾ 共2兲 ␾ 共2兲
v
r ⴰ r , 冊
␾ 共k+3兲
r ␾ −1 ␾
= ␦ 冉 ␾ 共k+2兲
D̂ r
i
v l=0
k
− 兺 ␾r共l+2兲 ⴰ ␾r共l+2兲 , 冊 k 艌 1,
where a共k兲 is the Deg-homogeneous component of degree k of an element a 苸 *W 丢 ⌳.

Proof: It follows from a local component calculus with *␾-adapted coefficients of ␾D̂, see
similar considerations in Refs. 5, 6, and 35. 䊐
For Hamilton spaces, we can define a canonical star product. By an explicit construction, we
prove the following.
Theorem 5.2: A *␾-adapted star product for Hamilton spaces is defined on C⬁共T *M兲关关v兴兴 by
formula
1
f ⴱ 2 f ⯐ ␴共␶共 1 f兲兲 ⴰ ␴共␶共 2 f兲兲,
where the projection ␴ : *W → C⬁共T *M兲关关v兴兴 onto the part of degs-degree zero is a bijection and
the inverse map ␶ : C⬁共T *M兲关关v兴兴 → *W can be calculated recursively with respect to the total
degree Deg,
␶共f兲共0兲 = f
and, for k 艌 0,
␶共f兲共k+1兲 = ␦ˇ −1 冉 ␾
D̂␶共f兲共k兲 −
i
k
兺 adWick共 ␾r共l+2兲兲共␶共f兲共k−l兲兲 .
v l=0
冊
Let f X the Hamiltonian vector field corresponding to a function f 苸 C⬁共T *M兲 on space
共T *M , *␪兲 and consider the antisymmetric part
−
C共 1 f, 2 f兲 ⯐ 21 共C共 1 f, 2 f兲 − C共 2 f, 1 f兲兲
of bilinear operator C共 1 f , 2 f兲. A star product 共34兲 is normalized if 1C共 1 f , 2 f兲 = 共i / 2兲兵 1 f , 2 f其, where
兵·, ·其 is the Poisson bracket, see 共13兲. For the normalized ⴱ, the bilinear operator −2 C defines a de
Rham–Chevalley 2-cocycle, when there is a unique closed 2-form ␾⑂ such that
1␾
2 ⑂共 X, X兲共48兲
1 f1 f2
2C共 f, 2 f兲 =
for all 1 f , 2 f 苸 C⬁共T *M兲. This is used to introduce c0共ⴱ兲 ⯐ 关 ␾⑂兴 as the equivalence class.
Computing 2C from 共48兲 and using the Theorem 5.2, we get the proof for the following.
Lemma 5.1: The unique 2-form defined by the unique canonical connection-pair can be
computed as
i i
␾
Ꭽ = − *J␶␣⬘ ␾R̂␶␣⬘ − d共 *J␶␣⬘ ␾T̂␶␣⬘␤ ␾e␤兲.
8 6
Let us define the canonical class ␾␧, for the splitting 共39兲. We can perform a distinguished
complexification of such second order tangent bundles in the form TC共 ␾TT *M兲 = TC共hT *M兲
␾
丣 TC共wT * M兲 and introduce ␧ as the first Chern class of the distributions TC⬘ 共 ␾TT *M兲
= TC⬘ 共hT M兲丣 TC⬘ 共wT M兲 of couples of vectors of type 共1, 0兲 both for the h- and w-parts. We
* *
compute ␾␧ using ␾D̂ and the h- and v-projections h⌸ = 21 共Idh − iJh兲 and v⌸ = 21 共Idv − iJv兲, where
Idh and Idv are respective identity operators and Jh and Jv are almost complex operators, which are
projection operators onto corresponding 共1, 0兲-subspaces. Introducing the matrix 共h⌸ , v⌸兲
R̂共h⌸ , v⌸兲T, where 共¯兲T means transposition, as the curvature matrix of the N-adapted restriction
of ␾D̂ to TC⬘ 共 ␾TT *M兲, we compute the closed Chern–Weyl form,
␾
␥ = − i Tr关共h⌸, v⌸兲 ␾R̂共h⌸, v⌸兲T兴 = − i Tr关共h⌸, v⌸兲 ␾R̂兴 = − 41 *J␶␣⬘ ␾R̂␶␣⬘ . 共49兲
Using the canonical class ␾␧ ⯐ 关 ␾␥兴, we prove the following.

Theorem 5.3: The zero-degree cohomology coefficient c0共ⴱ兲 for the almost Kähler model of a
Hamilton space *K2n, defined by a Hamilton fundamental function H共x , p兲, is computed
␾
c0共ⴱ兲 = −共1 / 2i兲 ␧.
The coefficient c0共ⴱ兲 contains as a particular case the class of zero-degree cohomologies
computed for a metric of type gi⬘ j⬘共x兲 on M, defining a solution of the Einstein equations and lifted
on cotangent bundle by formula 共1兲. In such cases, this zero-degree coefficient defines certain
quantum properties of the gravitational field. A more rich geometric structure should be considered
if we define a value similar to c0共ⴱ兲 encoding the information about Einstein manifolds lifted to
the cotangent bundle.
B. Quantum gravitational field equations

Any solution in classical Einstein gravity can be embedded into a Hamilton space model and
quantized on cotangent bundle following the Fedosov quantization adapted to *␾- and
d-connections. Considering a dequantization formalism,36 we construct certain quantum deforma-
tions of the classical Einstein configurations in the classical limit. Such a model defines a non-
holonomic almost Kähler generalization of the Einstein gravity on cotangent bundle. The solutions
for the “cotangent” gravity are, in general, with violation of Lorentz symmetry induced by quan-
tum corrections. The nature of such quantum gravity corrections is different from those defined by
Finsler–Lagrange models on tangent bundle 共see, for instance, Refs. 33, 34, and 43–45, locally
anisotropic string gravity40–42 with corrections from extradimensions, and nonholonomic spinor
gravity37–39 and noncommutative gravity, see reviews of results in Refs. 29 and 30. The aim of this
section is to analyze how a generalization of Einstein gravity can be performed on cotangent
bundles in terms of canonical *␾-connections, with geometric structures induced by an effective
Hamiltonian fundamental function, when the Fedosov quantization can be naturally performed.
For a canonical *␾-connections ␾D̂ = 兵 ␾⌫其共42兲, we can define the Ricci tensor,
Ric共 ␾D̂兲 = 兵 ␾R̂␤␥ ⯐ ␾R̂␣␤␥␣其,
and the scalar curvature, ␾R ⯐ *g␣␤ ␾R̂␣␤. On cotangent bundle T *M, we postulate the field equa-
tions
␾
R̂␣គ ␤ − 21 共 ␾R + ␭兲 ␾e␣គ ␤ = 8␲G⌼␣គ ␤ , 共50兲
where ⌼␣គ ␤ is the effective energy-momentum tensor, ␭ is the cosmological constant, G is the
Newton constant in the units when the light velocity c = 1, and ␾e␤ = ␾e␣គ ␤⳵ / ⳵u␣គ is the *␾-adapted
base 共40兲.
We consider the effective source 3-form,
ឈ = ⌼␣គ ␧
⌼ ␤គ ␥គ ␦គ
␤ ␤ ␣គ ␤គ ␥គ ␦គ du ∧ du ∧ du ,
where ␧␣␤␥␦ is the absolute antisymmetric tensor, and the action for the cotangent gravity and
matter fields,
S关e,⌫, ␾兴 = grS关e,⌫兴 + matterS关e,⌫, ␾兴.

Theorem 5.4: Equation (50) can be represented as 3-form equations
ឈ
␧␣␤␥␶共 ␾e␣ ∧ ␾R␤␥ + ␭ ␾e␣ ∧ ␾e␤ ∧ ␾e␥兲 = 8␲G⌼ 共51兲
␶
following from action by varying the components of e␤, when
ឈ = m⌼
⌼ ឈ and mឈ
⌼␶ = m⌼␣គ ␶␧␣គ ␤គ ␥គ ␦គ du␤គ ∧ du␥គ ∧ du␦គ ,
␶ ␶
where m⌼␣គ ␤ = ␦matterS / ␦e␣គ ␶.

Proof: It is a usual textbook and/or differential form calculus 共see, for instance, Refs. 51 and
14兲. In our case, we have to use the *␾-adapted bases 共40兲 and 共41兲 for ␾D̂. 䊐
The Chern–Weyl 2-form 共49兲 can be used to define the quantum version3 of Einstein equations
共51兲 in the approaches with deformation quantization.
Corollary 5.1: The quantum field equations on cotangent bundle generalizing the Einstein
gravitations in general relativity are
␾ ␣ ឈ − ␭*
e ∧ ␾␥ = ␧␣␤␥␶2␲G *J␤␥⌼ ␶ J ␾e ␣ ∧ ␾e ␤ ∧ ␾e ␥ . 共52兲
4 ␤␥
Proof: Multiplying ␾e␣∧ with 共49兲, taking into consideration Eq. 共51兲, and introducing the
almost complex operator *J␤␥, we get the almost symplectic form of the Einstein equations 共52兲.䊐
An explicit computation of ␾␥ for nontrivial matter fields has to be performed for a deforma-
tion quantization model with interacting gravitational and matter fields geometrized in terms of an
almost Kähler model defined for spinor and fiber bundles on cotangent bundles.
VI. CONCLUSIONS
In this paper we outlined a method of converting any regular Lagrange and Hamilton dynam-
ics into equivalent almost Kähler geometries with canonical nonlinear connection 共N-connection兲
and adapted almost symplectic structures. The formalism was performed to be invariant under
symplectic morphisms and adapted to Legendre transforms of Lagrangians into Hamiltonians and
3
In the sense of deformation quantization 共i.e., when quantum equations are derived following a deformation procedure兲
but not of perturbative quantum theory.
inversely. The geometry of cotangent bundles endowed with nonholonomic distributions, formu-
lated as certain Hamilton–Cartan spaces being dual to the corresponding Lagrange–Finsler spaces,
described in Secs. II and III, presents a key prerequisite of this approach to deformation quanti-
zation.
Given a regular Lagrange 共in particular, Finsler and related Legendre transforms兲, or Hamilton
共in particular, Cartan兲, generating function, we completely define the fundamental geometric ob-
jects of a Hamilton geometry modeled on cotangent bundle, inducing canonical almost symplectic
and compatible symplectic connection. The connection pair and canonical *␾-connection uniquely
constructed to preserve the invariance under symplectic morphisms are introduced in Sec. IV B.
Such geometric objects are crucial for defining Fedosov–Hamilton operator pairs. This allows us
to generalize the Fedosov’s theorems for deformation quantization to the case of Hamilton spaces
and to show how the Einstein manifolds can be encoded into such a quantization scheme.
Any classical solution of the Einstein equations can be lifted to the cotangent bundle and
embedded into a Hamilton space geometry using frame transform variables depending both on
space-time and phase 共momentumlike兲 coordinates. Such variables can be defined from a dequan-
tization procedure like that considered in Ref. 36 but in our case reformulated for nonholonomic
cotangent bundles. In general, such constructions result in quasiclassical effects of quantum grav-
ity with violations of local Lorentz symmetry. The surprising result advocated in this paper is that
we can model classical and quantum gravitational effects by corresponding effective classical and
quantum Hamilton mechanics systems. Nevertheless, certain additional phenomenological and/or
experimental data for quantum gravity effects have to be assigned to the scheme in order to define
the nonholonomic frame transforms of locally isotropic gravitational fields into quantized and
semiclassical ones on cotangent bundle.
On cotangent bundles and curved phase spaces, there were developed different methods of
quantization of nonlinear field theories and mechanical systems with nonholonomic constraints.
For instance, in Refs. 52 and 53, there were constructed in explicit form examples of cotangent
star products 共in such cases, one derive certain compatible symplectic/Levi Civita–type connec-
tions兲. The formalism was revised and developed by introducing auxiliary variables, with further
restrictions, and/or higher spin systems, in analogy with gauge theories with generalized symme-
tries, in the framework of the BRST approach.54–56,10,1 In general, the methods of quantization of
nonlinear physical systems on contangent bundles are very different from that on tangent bundles.
The source of this is in the fact that Lagrange and Hamilton 共both classical and quantum兲 schemes
may result in very different quantum models for nonlinear classical theories: On phase spaces, we
have to consider Legendre transform and additional symmetries related to symplectic morphisms
共i.e., morphisms preserving a symplectic structure into a symplectic structure兲. In such a case, the
author of Ref. 19 had to work with symplectic groupoids and introduce contravariant connections
which modified substantially the Fedosov scheme of quantization.
There are various ideas and approaches to quantize gravity theories, including deformation
quantization. In our partner works23,24 we elaborated a direction related to effective Lagrange–
Finsler geometries by performing nonholonomic deformations to quantum versions of 共semi兲
Riemannian manifolds preserving the local Lorentz invariance at least for semi-classical approxi-
mations. Following the same geometric methods developed for Finsler spaces, but extended for
nonholonomic manifolds, we proved that there are similar quantization schemes of gravity and
Lagrange–Finsler spaces modeled on tangent bundles.21,22 Such constructions also result in viola-
tions of local Lorentz symmetry but with physical effects which are very different from those for
models on cotangent bundles. The general conclusion of this paper is that a deformation quanti-
zation scheme for Hamilton spaces and related generalizations of the Einstein gravity on cotangent
bundle results in more rich geometric structures and requests more advanced geometric methods
with nonlinear connections and connection pairs. We can elaborate a standard Fedosov formalism
also on cotangent bundle 共not involving groupoid structures兲 by introducing canonical nonlinear
connection structures and generalizing the concept of linear and distinguished connections to that
of connection pairs.
We are planing to compare different approaches to deformation quantization of gravity 共pre-

serving or violating the local Lorentz invariance兲 and other quantization schemes in our further
works.
ACKNOWLEDGMENTS
M.A. was partially supported by Grant No. CNCSIS,1158/2007, Romania. S.V. performed his
work as a visitor at Fields Institute.
1
I. Batalin, E. Fradkin, and T. Fradkina, Nucl. Phys. B 332, 224 共1990兲.
2
B. de Wit and J. W. van Holten, Phys. Lett. 79B, 389 共1978兲.
3
M. Grigoriev and S. Lyakhovich, Commun. Math. Phys. 218, 437 共2001兲.
4
M. Henneaux, Phys. Rep. 126, 1 共1985兲.
5
B. Fedosov, J. Diff. Geom. 40, 213 共1994兲.
6
B. Fedosov, Deformation Quantization and Index Theory, Mathematical Topics Vol. 9 共Akademie-Verlag, Berlin, 1996兲.
7
M. Kontsevich, Lett. Math. Phys. 66, 157 共2003兲.
8
M. Kontsevich, Lett. Math. Phys. 48, 35 共1999兲.
9
P. Kazinski, S. Lyakhovich, and A. Sharapov, J. High Energy Phys. 0507, 076 共2005兲.
10
S. Lyakhovich and A. Sharapov, J. High Energy Phys. 0503, 011 共2005兲.
11
C. Castro, J. Geom. Phys. 33, 173 共2000兲.
12
C. Castro, Gen. Relativ. Gravit. 36, 2605 共2004兲.
13
C. Castro, Phys. Lett. B 668, 442 共2008兲.
14
C. Rovelli, Quantum Gravity 共Cambridge University Press, Cambridge, 2004兲.
15
A. Ashtekar and J. Lewandowski, Class. Quantum Grav. 21, R53 共2004兲.
16
T. Thiemann, Modern Canonical Quantum General Relativity 共Cambrdigde University Press, Cambridge, 2006兲.
17
I. Gorbunov, S. Lyakhovich, and A. Sharapov, J. Geom. Phys. 53, 98 共2005兲.
18
I. Batalin, M. Grigoriev, and S. Lyakhovich, J. Math. Phys. 46, 072301 共2005兲.
19
A. Karabegov, Commun. Math. Phys. 258, 223 共2005兲.
20
F. Etayo, R. Santamaría, and S. Vacaru, J. Math. Phys. 46, 032901 共2005兲.
21
S. Vacaru, J. Math. Phys. 48, 123509 共2007兲.
22
S. Vacaru, e-print arXiv:0707.1526.
23
S. Vacaru, Phys. Lett. A 372, 2949 共2008兲.
24
25
G. Vrǎnceanu, Acad. Sci., Paris, C. R. 103, 852 共1926兲.
26
G. Vrǎnceanu, Bull. Fac. St. Cernauti 5, 177 共1931兲.
27
G. Vrǎnceanu, Leçons de Geometrie Differentielle 共l’Academie de la Republique Populaire de Roumanie, Bucharest,
1957兲, Vol. II.
28
A. Bejancu and H. R. Farran, Foliations and Geometric Structures 共Springer, New York, 2005兲.
29
S. Vacaru, P. Stavrinos, E. Gaburov, and D. Gonţa, Clifford and Riemann-Finsler Structures in Geometric Mechanics and
Gravity, Selected Works, Differential Geometry—Dynamical Systems, Monograph 7 共Geometry Balkan Press, Buchar-
est, 2006兲; www.mathem.pub.ro/dgds/mono-va-t.pdf; e-print arXiv:gr-qc/0508023.
30
31
R. Miron, D. Hrimiuc, H. Shimada, and V. S. Sabau, The Geometry of Hamilton and Lagrange Spaces 共Kluwer
Academic, Dordrecht, 2000兲.
32
R. Miron, The Geometry of Higher-Order Hamilton Spaces 共Kluwer Academic, Dordrecht, 2003兲.
33
R. Miron and M. Anastasiei, Vector Bundles and Lagrange Spaces with Applications to Relativity 共Geometry Balkan
Press, Bucharest, 1997兲关translation from Romanian of Editura Academiei Romane 共1987兲兴.
34
R. Miron and M. Anastasiei, The Geometry of Lagrange Spaces: Theory and Applications 共Kluwer Academic, Dordrecht,
1994兲.
35
A. Karabegov and M. Schlichenmaier, Lett. Math. Phys. 57, 135 共2001兲.
36
A. Karabegov, Lett. Math. Phys. 65, 133 共2003兲.
37
S. Vacaru and P. Stavrinos, Spinors and Space-Time Anisotropy 共Athens University Press, Athens, 2002兲; e-print
arXiv:gr-qc/0112028.
38
S. Vacaru, J. Math. Phys. 37, 508 共1996兲.
39
S. Vacaru, J. High Energy Phys. 09, 011 共1998兲.
40
S. Vacaru, Ann. Phys. 共N.Y.兲 256, 39 共1997兲.
41
S. Vacaru, Nucl. Phys. B 434, 590 共1997兲.
42
N. Mavromatos, e-print arXiv:0708.2250.
43
F. Girelli, S. Liberati, and L. Sindoni, Phys. Rev. D 75, 064015 共2007兲.
44
S. Mignemi, Phys. Rev. D 76, 047702 共2007兲.
45
G. Gibbons, J. Gomis, and C. N. Pope, Phys. Rev. D 76, 081701 共2007兲.
46
R. Miron and Gh. Atanasiu, Kodai Math. J. 6, 228 共1983兲.
47
Gh. Atanasiu, M. Hashiguchi, and R. Miron, Rep. Fac. Sci. Kagoshima Univ. Math. Phys. Chem. No. 18, 19 共1985兲.
48
S. Vacaru, e-print: arXiv:0806.3810.
49
S. Vacaru, Symmetry, Integr. Geom.: Methods Appl. 4, 071 共2008兲.
50
S. Vacaru, Int. J. Theor. Phys. 共to be published兲.
51
C. Misner, K. Thorne, and J. Wheeler, Gravitation 共Freeman, San Francisco, 1973兲.
52
M. Bordemann, N. Neumaier, and S. Waldman, Commun. Math. Phys. 198, 363 共1998兲.
53
M. Bordemann, N. Neumaier, and S. Waldman, J. Geom. Phys. 29, 199 共1999兲.
54
G. Barnich, M. Grigoriev, A. Semikhatov, and I. Tipunin, Commun. Math. Phys. 260, 147 共2005兲.
55
M. A. Vasiliev, Int. J. Geom. Methods Mod. Phys. 3, 37 共2006兲.
56
M. Grigoriev, e-print arXiv:hep-th/0605089.

Fedosov Quantization of Lagrange-Finsler and Hamilton-Cartan Spaces and Einstein Gravity Lifts On (Co) Tangent Bundles

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fedosov Quantization of Lagrange-Finsler and Hamilton-Cartan Spaces and Einstein Gravity Lifts On (Co) Tangent Bundles

Uploaded by

Copyright:

Available Formats

JOURNAL OF MATHEMATICAL PHYSICS 50, 013510 共2009兲

Fedosov quantization of Lagrange–Finsler

0022-2488/2009/50共1兲/013510/23/$25.00 50, 013510-1 © 2009 American Institute of Physics

II. LAGRANGE–FINSLER AND HAMILTON–CARTAN GEOMETRY AND EINSTEIN

A. Canonical geometric objects on Lagrange and Hamilton spaces

gij共x,p兲 = eii⬘共x,p兲e j j⬘共x,p兲gi⬘ j⬘共x兲 共1兲

1Ri⬘ j⬘共x兲 = ␭gi⬘ j⬘共x兲, 共2兲

is nondegenerate and of constant signature on T* M .

H共x,p兲 = pay a − L共x,y兲, 共5兲

L共x,y兲 = pay a − H共x,p兲, 共6兲

␸:TM 傻 U → *U 傺 T*M,共xi,y a兲 → xi,pa = 冉 ⳵L共x,y兲

␺:T*M 傻 *U → U 傺 TM,共xi,pa兲 → xi,y a = 冉 ⳵H共x,p兲

TTM = hTM 丣 vTM 共7兲

TT*M = hT*M 丣 vT*M 共8兲

␪ ⯐ dpi ∧ dxi . 共12兲

for any functions 1 f共x , p兲 and 2 f共x , p兲 on T*M.

On 共co兲 tangent bundles endowed with N-connection structure, it is convenient to elaborate a

and their dual (coframe) structures

e␣ = 共ei = dxi,eb = dy b + Nbi dxi兲 on 共TM兲* , 共18兲

*e␣ = 共 *ei = dxi, *e = dp − *N dxi兲 on 共T*M兲* , 共19兲

*g = *g *e␣ 丢 *e␤ = *g 共x,p兲ei 丢 e j + *gab共x,p兲 *e 丢 *e , 共20兲

on T*M, where *gij is inverse to *gab, and

g = g␣␤e␣ 丢 e␤ = gij共x,y兲ei 丢 e j + gab共x,y兲ea 丢 eb ,

J共ei兲 = − en+i and J共en+i兲 = ei .

*J共 *e 兲 = − g* en+i and *J共 *en+i兲 = *ei .

⳵Nai ⳵Naj b ⳵N j b ⳵Ni

B. Almost Kähler Lagrange–Hamilton structures

for any *X = h *X + v *X, *Y = h *Y + v *Y and h T␪共 *X , *Y兲 ⯐ T␪共h *X , h *Y兲,

acting on *e␣ = 共 *ei , *eb兲 共17兲 following formulas

*J共 *e 兲 = *g *eb *J共 *eb兲 = 0 ⳵

J = − ␦ai ea 丢 ei + ␦iaei 丢 ea, *J = − *g *ea 丢 *ei + *gia *e 丢 *e

corresponding to the L-dual pair of almost complex structures 共J , *J兲,

␪ = gaj共x,y兲ea ∧ ei and *␪ = ␦ a *e ∧ *e i . 共26兲

Let us consider an important example.

III. NONLINEAR CONNECTIONS AND ALMOST SYMPLECTIC GEOMETRY

A. Canonical N-connections and d-connections for Lagrange and Hamilton spaces

for any vector fields X and Y on TM.

(d-connection) if it is compatible with the almost product structure DP = 0 (or *D *P = 0).

De␤e␥ ⯐ ⌫␣␤␥e␣ and D *e *e␥ ⯐ *⌫␣␤␥ *e␣

with corresponding N-adapted splitting,

Deke j ⯐ Li jkei, Dekeb ⯐ Ĺabkea, Dece j ⯐ Ći jcei, Deceb ⯐ Cabcea ,

*D *e ⯐ *L i *e , *D *eb ⯐ − *Ĺ b *ea ,

*D *e ⯐ *Ći c *e , *D *eb ⯐ − *C bc *ea ,

⌫␣␤␥ = 兵Li jk,Ĺabk,Ći jc,Cabc其 and *⌫ ␣

D = 共 hD, vD兲 and *D = 共 *D, *D兲,

We shall work with a more special class of d-connections.

*n⌫i = *Li *ek + *Ći c *e , 共29兲

*Tk = 1 *Tk *ei ∧ *e j + *Ćk c *e j ∧ *e ,

*T = 1 *⍀ *ei ∧ *e j + 1 * P *ei ∧ *e + 1 *S bc *e ∧ *ec ,

for the N-connection curvature *⍀ija (22) and

* Pi c = *ec共 *Li 兲 − *D 共 *Ći c兲 + *Ći o * Pc ,

for “nonboldface” *ec = ⳵ / ⳵ pc.

B. Hamilton–Fedosov spaces and almost Kähler structures

*L̂i = 1 *gis共 *e 共 *g 兲 + *e 共 *g 兲 − *e 共 *g 兲兲,

*D̂ *␪ = 0 and *D̂ *J = 0 共33兲

IV. FEDOSOV OPERATOR PAIRS FOR HAMILTON SPACES

A. Canonical Fedosov–Hamilton operators

for two elements a and b defined by series of the type

a共v,z兲 = 兺 ar,兵␣其共u兲z兵␣其vr , 共36兲

ˆ ␥ z*e 共␣兲兲 丢 共 *e␣ ∧ ␭兲 + a 丢 d␭

B. Fedosov–Hamilton operator pairs

TT*M = hT*M 丣 wT*M . 共39兲

Aijm = *⍀ijm and Bmkj = *gmi共 *gka we j *Nia − *g ja wek *Nia兲,

gij共x,p兲 = eii⬘共x,p兲e j j⬘共x,p兲gi⬘ j⬘共x兲共1兲

␸:TM 傻 U → U 傺 TM,共xi,y a兲 → xi,pa = 冉 ⳵L共x,y兲

␺:TM 傻 U → U 傺 TM,共xi,pa兲 → xi,y a = 冉 ⳵H共x,p兲

TTM = hTM 丣 vT*M 共8兲

e␣ = 共 ei = dxi, e = dp − N dxi兲 on 共TM兲 , 共19兲

g = g e␣ 丢 e␤ = g 共x,p兲ei 丢 e j + gab共x,p兲 e 丢 e , 共20兲

on TM, where gij is inverse to *gab, and

J共 e 兲 = − g* en+i and J共 en+i兲 = *ei .

for any X = h X + v X, Y = h Y + v Y and h T␪共 X , Y兲 ⯐ T␪共h X , h Y兲,

acting on e␣ = 共 ei , *eb兲共17兲 following formulas

J共 e 兲 = g eb J共 eb兲 = 0 ⳵

J = − ␦ai ea 丢 ei + ␦iaei 丢 ea, J = − g ea 丢 ei + gia e 丢 *e

␪ = gaj共x,y兲ea ∧ ei and ␪ = ␦ a e ∧ *e i . 共26兲

(d-connection) if it is compatible with the almost product structure DP = 0 (or D P = 0).

De␤e␥ ⯐ ⌫␣␤␥e␣ and D e e␥ ⯐ ⌫␣␤␥ e␣

D e ⯐ L i e , D eb ⯐ − Ĺ b ea ,

D e ⯐ Ći c e , D eb ⯐ − C bc ea ,

D = 共 hD, vD兲 and D = 共 D, *D兲,

n⌫i = Li ek + Ći c *e , 共29兲

Tk = 1 Tk ei ∧ e j + Ćk c e j ∧ *e ,

T = 1 ⍀ ei ∧ e j + 1 * P ei ∧ e + 1 S bc e ∧ *ec ,

* Pi c = ec共 Li 兲 − D 共 Ći c兲 + Ći o Pc ,

L̂i = 1 gis共 e 共 g 兲 + e 共 g 兲 − e 共 g 兲兲,

D̂ ␪ = 0 and D̂ J = 0 共33兲

ˆ ␥ ze 共␣兲兲丢共 e␣ ∧ ␭兲 + a 丢 d␭

TTM = hTM 丣 wT*M . 共39兲

Aijm = ⍀ijm and Bmkj = gmi共 gka we j Nia − g ja wek Nia兲,

f 共 ␾ei兲 = ␼ki ␾ēk and f 共 we b兲 = ␼

(2) The push forward of a N-connection N by a N-regular diffeomorphism is a N-connection

ḡ = ḡ ē␣ 丢 ē␤ = ḡ 共x,p兲ēi 丢 ē j + ḡab共x,p兲 ē 丢 ē ,

where ḡij ⴰ f = ␼ik␼lj gkl.

(1) A connection ⵜ is a ␾-connection if and only if ⵜ ¯ is a ␾