Applications of Reimannian Geometry in Physics

(General Theory of Relativity)

Amirhossein Hassanabadi1
1
Department of Energy Egeineering and Physics, Amirkabir University of
Technology, Tehran, Iran, P.O. Box: 15875-4413; [email protected]

Abstract
I study the underlying geometry of general relativity, through a somewhat basic
discussion of the theory, on which we can construct a rigorous mathematical formalism
that describes gravitation. Einstein’s field equations are derived as an instance of a
classical field theory with an action depending on the metric of a psedo-Riemannian
space that describe the geometry of spacetime. And the article is concluded by the
generalization of the ideas developed here, mainly those of theory of gravitation, and
asking a fundamental question regarding the Equivalence Principle.
Keywords: Differential Geometry, General Relativity, Manifold Theory, Gravitation

1 Introduction
It all starts with Euclid’s fifth axiom which states that if a line segment intersects two
straight lines forming two interior angles on the same side that sum to less than two right
angles, then the two lines, if extended indefinitely, meet on that side on which the angles
sum to less than two right angles.
For about 2000 years, geometers tried to purify Euclidean system by proving that the
fifth axiom can be deduced from the the other four, or substitute it with an equivalent
statement which is more trivial. But as it turns out, no such procedure exists and Euclid
was right to state it as an axiom and not a theorem. This search led to the foundation of a
new geometry, namely the non-Euclidean geometry.
Gauss seems to be the first to accept the non-Euclidean geometry as a logical possibility.
János Bolyai also established a non-Euclidean system between 1820 and 1823. Lobachevsky’s
work also resulted in similar ideas in 1926.
The Interesting point for all these geometers was that they had introduced a geometry
that explained an infinite two dimensional space in which all Euclid’s axioms hold except
for the fifth. In this new geometry the alternatives of the fifth axiom as stated by Proclus,
Wallis, and Legendre didn’t hold either; i.e. from a point exterior to a line, infinite lines
could pass, all parallel to the former line, no two shapes with different sizes were similar, or
the interior angles of a triangle didn’t sum to two right angles.
Gauss also noted that an important property of any surface was the metric function
associated to it which measures the distance of any two points x and y along the shortest
path between them. He assumed in any sufficiently small region of space, it is possible to

1
choose a Euclidean coordinate (ξ1 , ξ2 ) so that the distance between the two points (ξ1 , ξ2 )
and (ξ1 + dξ1 , ξ2 + dξ2 ) obeys the Pythagorean theorem. For instance on a smooth curved
surface, at each point, we can use the tangent surface on that point as the locally Euclidean
coordinates.
In the year 1854 Riemann generalized Gauss’ metric space idea to higher dimensions, and
later Ricci, Levi-Civita, Christoffel, and others developed Riemann’s idea to its complete
mathematical form using tensor analysis.
Theory of gravitation on the other hand was developed completely independent of ge-
ometry by newton and he first published it at the end of his renowned book Philosophiæ
Naturalis Principia Mathematica. Before that Galileo Galilei had observed in his experi-
ments that objects fall with the same rate independent of their masses.
Newton was aware that his results might have been only approximations to reality, for
the inertial mass that appears in his second law wasn’t exactly the same as the gravitational
mass that appears in his law of gravitation.
Newton did not observe any difference between the two kinds of masses in his experiments
with pendulums of same length but different bobs; and Bessel confirmed the same results,
with more precision, in 1830. And lastly Eötvös using another method, confirmed this
equivalence of inertial and gravitational mass with a precision of 1 in 109 [3].
These results led Einstein to his principle of equivalence.

2 Riemannian Geometry
2.1 Basic Topology
A topology on a set M is a collection T of subsets of M having the following properties:

1. ∅ and M are in T.
2. The union of the elements of any subcollection of T is in T .
3. The intersection of the elements of any finite subcollection of T is in T.

A set M for which a topology T has been specified is called a topological space.

Properly speaking, a topological space is an ordered pair (M, T) consisting of a set M

and a topology T on M , but we often omit specific mention of T if no confusion will arise.
Let X and Y be topological spaces; let ϕ : X → Y be a bijection. If both the function ϕ
and the inverse function ϕ−1 : Y → X are continuous, then ϕ is called a homeomorphism.
Geometrically speaking, a homeomorphism is a bijection that can bend, twist, stretch,
and wrinkle the space X to make it coincide with Y , but it cannot rip, puncture, shred, or
pulverize X in the process.
One of the most important and frequently used ways of imposing a topology on a set
is to define the topology in terms of a metric on the set. A metric space is a set M , the
elements of which are referred to as points of M , together with a metric g having the three
properties that distance has in Euclidean space.

A metric on a set M is a function g : M × M → R having the following properties:

1. (Positive definiteness) g(x, y) ≥ 0 for all x, y ∈ M ; equality holds if and only if x = y

2
 2. (Symmetry) g(x, y) = g(y, x) for all x, y ∈ M .
3. (Triangle inequality) g(x, y) + g(y, z) ≥ g(x, z), for all x, y, z ∈ M .
Points x and y in a topological space M can be separated by neighborhoods U of x and
V of y, such that U and V are disjoint (U ∩ V = ∅). M is said to be a Hausdorff space if
all distinct points in X are pairwise neighborhood-separable.

2.2 Manifolds
In the simplest terms, Manifolds are spaces that locally look like some Euclidean space
Rn , and on which we can do calculus. The most familiar examples, aside from Euclidean
spaces themselves, are smooth plane curves such as circles and parabolas, and smooth
surfaces such as spheres, tori, paraboloids, ellipsoids, and hyperboloids. Higher dimensional
examples include the set of unit vectors in Rn+1 (the n-sphere) and graphs of smooth maps
between Euclidean spaces. Another example which is more physical and also a good one to
demonstrate the idea is as follows: if we call a rotation followed by a translation an affine
motion, then the set of all affine motions in R3 is a six-dimensional manifold. Moreover, this
six-dimensional manifold is not R6 . The simplest examples of manifolds are the topological
manifolds, which are topological spaces with certain properties that encode what we mean
when we say that they ”locally look like” Rn .

2.2.1 Topological Manifolds[1]

A topological space M is locally Euclidean of dimension n if every point p in M has a
neighborhood U such that there is a homeomorphism ϕ from U onto an open subset of Rn .
We call the pair (U, ϕ : U → R) a chart, U a coordinate neighborhood or a coordinate open
set, and ϕ a coordinate map or a coordinate system on U . We say that a chart (U, ϕ) is
centered at p ∈ U if ϕ(p) = 0. A chart (U, ϕ) about p simply means that (U, ϕ) is a chart
and p ∈ U .
A topological manifold of dimension n is a Hausdorff, second countable, locally Euclidean
space of dimension n.
A C ∞ atlas or simply an atlas on a locally Euclidean space M is a collection {(Uα , ϕα )}
of C ∞ -compatible charts that cover M , i.e., such that M = ∪α Uα .
An atlas A on a locally Euclidean space is said to be maximal if it is not contained in a
larger atlas; in other words, if M is any other atlas containing A, then M = A.

2.2.2 Smooth Manifolds[1]

A smooth or C ∞ manifold is a topological manifold M together with a maximal atlas.
The maximal atlas is also called a differentiable structure on M . A manifold is said to have
dimension n if all of its connected components have dimension n. A manifold of dimension
n is also called an n-manifold.
Any atlas A = {(Uα , ϕα )} on a locally Euclidean space is contained in a unique maximal
atlas.
The above statement simply means for a topological manifold to be a smooth one, it is
not necessary to exhibit a maximal atlas. The existence of any atlas on M will do.

3
2.3 The Tangent and Cotangent Spaces[1]
2.3.1 The Tangent Space
Just as for Rn , a germ of a C ∞ function at p in M can be defined to be an equivalence class
of C ∞ functions defined in a neighborhood of p in M , two such functions being equivalent if
they agree on some, possibly smaller, neighborhood of p. The set of germs of C ∞ functions
at p in M is denoted Cp∞ (M ). The addition and multiplication of functions make Cp∞ (M )
into a ring; with scalar multiplication, Cp∞ (M ) becomes an algebra over R.
Generalizing a derivation at a point in Rn , we define a derivation at a point in a manifold
M , or a point-derivation of Cp∞ (M ), to be a linear map D : Cp∞ (M ) → R such that
D(f g) = (Df )g(p) + f (p)Dg.
A tangent vector at a point p in a manifold M is a derivation at p.
As for Rn , the tangent vectors at p form a vector space Tp (M ), called the tangent space
of M at p. Tp M is written instead of Tp (M ).
If U is an open set containing p in M , then the algebra Cp∞ (U ) of germs of C ∞ functions
in U at p is the same as Cp∞ (M ). Hence, Tp U = Tp M .
∂
It is easily checked that ∂x i |p satisfies the derivation property and so is a tangent vector
∂ ∂
at p. To simplify the notation, we will often write ∂x i instead of ∂xi |p if it is understood

at which point the tangent vector is located.

Let F : N → M be a C ∞ map between two manifolds. At each point p ∈ N , the map
F induces a linear map of tangent spaces, called its differential at p,
F∗ : Tp N → TF (p) M
as follows. If Xp ∈ Tp N , then F∗ (Xp ) is the tangent vector in TF (p) M defined by

(F∗ (Xp ))f = Xp (f ◦ F ) ∈ R, ∀f ∈ CF∞(p) (M ).

To make the dependence on p explicit we sometimes write F∗,p instead of F∗ .
Let F : N → M and G : M → P be smooth maps of manifolds, and p ∈ N . The
differentials of F at p and G at F (p) are linear maps:
F∗,p G∗,p
Tp N −−−→ TF (p) M −−−→ TG(F (p)) P.
If F : N → M and G : M → P are smooth maps of manifolds and p ∈ N , then
(G ◦ F )∗,p = G∗,F (p) ◦ F∗,p .
The differential of the identity map 1M : M → M at any point p in M is the identity
map 1Tp M : Tp M → Tp M because
((1M )∗ Xp )f = Xp (f ◦ 1M ) = Xp M
for any Xp ∈ Tp M and f ∈ Cp∞ (M ).
If (U, ϕ) = (U, x1 , ..., xn ) is a chart containing p, then the tangent space Tp M has basis
{ ∂x1 |p , . . . , ∂x∂n |p }.
∂

Given a smooth map F : N → M of manifolds and a point p ∈ N , let (U, x1 , . . . , xn ) and

(V, y 1 , . . . , y m ) be coordinate charts about p in N and F (p) in M , respectively. Relative to
∂ ∂
the bases { ∂x j |p } for Tp (N ) and { ∂y i |F (p) } for TF (p) (M ), the differential F∗,p : Tp (N ) →
∂ i
TF (p) (M ) is represented by the matrix [ ∂x j (p)], where F = y i ◦ F is the ith component of
F.

4
2.3.2 The Cotangent Space
Let M be a smooth manifold and p a point in M . The cotangent space of M at p,
denoted by Tp∗ (M ) or Tp∗ M , is the dual space of the tangent space Tp M . An element of the
cotangent space Tp∗ M is called a covector at p. Thus, a covector ωp at p is a linear function

ωp : Tp M → R

A covector field, a differential 1-form, or simply a 1-form on M , is a function that assigns

to each point p in M a covector at p. In this sense it is dual to a vector field on M , which
assigns to each point in M a tangent vector at p. The great utility of differential forms in
manifold theory arises from the fact that they can be pulled back under a map. This is in
contrast to vector fields, which in general cannot be pushed forward under a map.
If f is a C ∞ function on a manifold M , its differential is defined to be the 1-form df on
M such that for any p ∈ M and Xp ∈ Tp M ,

(df )p (Xp ) = Xp f.

The two notions f∗ and df are the same thing and it is justified to call both the differential
of f .
Let (U, ϕ) = (U, x1 , . . . , xn ) be a coordinate chart on a manifold M . Then the differen-
tials dx1 , . . . , dxn are 1-forms on U .
At each point p ∈ U , the covectors (dx1 )p , . . . , (dxn )p form a basis for the cotangent
space Tp∗ M dual to the basis ( ∂x ∂ ∂
1 )p , . . . , ( ∂xn )p for the tangent space Tp M

 
∂
(dxα )p |p = δ αβ .
∂xβ
Thus, every 1-form ω on U can be written as a linear combination
X
ω= ai dxi ,

where the coefficients ai are functions on U .

We can define a (r, s)-tensor at a point p of a smooth manifold as a linear map of r
1-forms and s vectors into a field K using the vector spaces Tp M and its dual Tp∗ M i.e., the
tangent space and cotangent space at the point p

T : Tp∗ M × · · · × Tp∗ M × Tp M × · · · × Tp M → K
| {z } | {z }
r s

The integer r ≥ 0 is called the contravariant degree and s ≥ 0 the covariant degree of T .
Such Multilinear maps form a vector space denoted

V (r,s) = Tp M ⊗ · · · ⊗ Tp M ⊗ Tp∗ M ⊗ · · · ⊗ Tp∗ M .

| {z } | {z }
r s

Let T be an (r, s)-tensor over a finite dimensional manifold, and let {ej } be a basis for
Tp M and {i } be the dual basis, then we can define the (r + s)dim(M ) many real numbers

T i1 ...irj1 ...js ≡ T (i1 , . . . , ir , ej1 , . . . , ejs )

which are called the components of the tensor T.(i1 , . . . , ir , j1 , . . . , js ∈ {1, . . . , dim(M )})

5
2.4 Riemannian and Pseudo-Riemannian Manifolds [4]
A Riemannian manifold is in fact a smooth manifold M , together with an inner product
metric over the tangent space at each point p of the manifold gp : Tp M × Tp M → K which
varies smoothly from one point to another; we call the family of gp ’s the Riemannian metric,
or the Riemannian metric tensor. Riemannian metric tensor is positive definite.
In fact Rimmanian metric allows us to define many of geometric concepts on the man-
ifolds, e.g., angle of intersection, length of curves, area of surfaces, intrinsic and extrinsic
curvature of a manifold, etc. As can be noted the Riemannian spaces are metric spaces.
If the metric is non-singular but not positive definite we get to the notion of pseudo-
Riemannian manifolds. We associate a pair (p, q) to a pseudo-Riemannian metric and call it
its signature, if the metric is diagonalized, p and q count the number the diagonal elements
with + signs and − signs.
g = diag(+, . . . , +, −, . . . , −)
| {z } | {z }
p q

2.5 Analysis on Manifolds[5]

2.5.1 Elements of the Metric Tensor
As was stated in the definition of Riemannian space, the metric is an inner product
metric and according to the definition its elements can be obtained by ei .ej = gij where ei
is the some coordinates of the tangent space. This will be called the covariant metric. Its
inverse g ij , called the contravariant metric, is defined so that g ij gij = δji .

2.5.2 Christoffel Connection [2]

In affine coordinates covariant bases are constant at all points, but in curvilinear coor-
dinates, the bases varies from one point to another. This variation can be measured via
the partial derivatives of the bases with respect the coordinate xi , in an N dimensional
space, this procedure results in N 2 vectors, and all these N 2 vectors can be decomposed
with respect to the covariant bases resulting in N 3 coefficients, these coefficients are called
Christoffel coefficients, also known as Christoffel symbol or Christoffel connection. Christof-
fel connection can be written in a more important way, with respect to derivatives of metric.
This definition is more important because as was stated in the notion of Riemmanian man-
ifolds the metric is an intrinsic property of space, and Christoffel connection being defined
as derivative of the metric would also be an intrinsic property of the space. We denote the
Christoffel connection by Γλµν .

1 λσ
∂µ eν = Γλµν eλ , Γλµν = g (∂ν gσµ + ∂µ gσν − ∂σ gµν )
2

2.5.3 Covariant Derivative ∇µ

An Important application of the Christoffel connection is its use in the definition of
the covariant derivative. The motivation for defining this new differentiation is that in
general the partial derivatives of a tensor of rank greater than 0 is not a tensor, and to
get a tensor out of another tensor by differentiation, Ricci and Levi-Civita introduced this
new differentiation called the covariant derivative, its action on covariant and contravariant

6
vectors is defined by

∇µ V λ = ∂µ V λ + Γλµν V ν
∇µ Vλ = ∂µ Vλ − Γνµλ Vν .

Covariant derivative is linear and obeys the Leibniz product rule. After covariant differ-
entiation of a (r, s)-tensor the resulting tensor is of type (r, s + 1). In cartesian coordinates
the covariant derivative reduces to partial derivetive.

2.5.4 Parallel Transport

The concept of moving a vector along a path, keeping constant all the while, is known as
parallel transport. Parallel transport requires a connection to be welldefined; the intuitive
manipulation of vectors in flat space makes implicit use of the Christoffel connection on this
space. The crucial difference between flat and curved spaces is that, in a curved space, the
result of parallel transporting a vector from one point to another will depend on the path
taken between the points. As a result two vectors can only be compared in a natural way if
they are elements of the same tangent space.
Covariant derivative like other kinds of derivative expresses some rate of change; in
fact covariant derivative expresses the instantaneous rate of change of a tensor field in
comparison with what the tensor would be if it were parallel transported. We then define
parallel transport of the tensor T along the path xµ (λ) to be the requirement that the
covariant derivative of T along the path vanishes.

2.5.5 Riemann Tensor, Ricci Tensor, and Ricci Scalar

If we explore what happens for a vector under the operation of parallel transport we find
a relation for a tensor describing curvature of the manifold on which this vector is parallel
transported.
To obtain this Tensor we investigate the commutation of two covariant derivative. The
relationship between this and parallel transport around a loop should be evident; the co-
variant derivative of a tensor in a certain direction measures how much the tensor changes
relative to what it would have been if it had been parallel transported, since the covari-
ant derivative of a tensor in a direction along which it is parallel transported is zero. The
commutator of two covariant derivatives, then, measures the difference between parallel
transporting the tensor first one way and then the other, versus the opposite ordering.

[∇µ , ∇ν ]V ρ = Rρσµν V σ − T λµν ∇λ V ρ

where T λµν defined by T λµν = Γνµλ − Γλνµ is called the torsion, and Rρσµν defined by
Rρσµν = ∂µ Γρνσ − ∂ν Γρµσ + Γρµλ Γλνσ − Γρνλ Γλµσ is called the Riemann or curvature tensor.
contracting two of the Riemann tensor indices we get the Ricci tensor defined by C31 Rρσµν =
Rλσλν = Rµν , and the trace the Ricci tensor is the Ricci scaler also known as curvature scalar
defined by R = g µν Rµν = Rµµ .
The Ricci tensor and scalar contain all of the information about traces of the Riemann
tensor.
If a coordinate system exists in which the components of the metric are constant, the
Riemann tensor will vanish, and vice versa.

7
3 Elements of Classical Field Theory
In Lagrangian formalism we derive equations of motion by using the principle of least
action, more explicitly, we search for extrema (as a function of trajectory of the particle
being investigated) of an action S, written as
Z
S = dt L(q, q̇)

where the function L(q, q̇) is the Lagrangian. The Lagrangian in point-particle mechanics
is typically of the form
L=T −U
where T is the kinetic energy and U the potential energy. Following the procedure of calculus
of variations, it can be shown that extrema of the action (trajectories q(t) for which S remains
stationary under small variations) are those that satisfy the Euler-Lagrange equations,
∂L d ∂L
− = 0.
∂q dt ∂ q̇
Field theory is a similar story, except that we replace the single coordinate q(t) by a set
of spacetime-dependent fields, Φi (xµ ), and the action S becomes a functional of these fields.
In field theory, the Lagrangian can be expressed as an integral over space of a Lagrange
density L, which is a function of the fields Φi and their spacetime derivatives ∂µ Φi :
Z
L = d3 x L(Φi , ∂µ Φi).

So the action is Z Z
S= dt L = d4 x L(Φi , ∂µ Φi ).

It can be shown that the extrema for this action are those satisfying following equation
∂L ∂L
i
− ∂µ = 0.
∂Φ ∂(∂µ Φi )

These are known as the Euler-Lagrange equations for a field theory in flat spacetime.
Action in flat spacetime can be generalized to curved spacetime by substitution of partial
spacetime derivative to covariant derivative; and using general Stokes’ theorem it can be
shown that the extrema satisfy the Euler-Lagrange equation.
A very important theorem concerning the Lagrangians is Noether’s Theorem, it states
that every symmetry of a Lagrangian implies the existence of a conservation law.

4 General Theory of Relativity

4.1 Principle of Equivalence[6]
Equivalence principle can be stated in more than one way. Weak equivalence principle
(WEP) formally as expressed by Newton states that there is no difference between inertial
mass and gravitational mass. The universality of gravitation, as implied by the WEP, can
be stated in another, more popular, form. Imagine that we consider a physicist in a tightly

8
sealed box, unable to observe the outside world, who is doing experiments involving the
motion of test particles, for example to measure the local gravitational field. Of course she
would obtain different answers if the box were sitting on the moon or on Jupiter than she
would on Earth. But the answers would also be different if the box were accelerating at
a constant velocity; this would change the acceleration of the freely-falling particles with
respect to the box. The WEP implies that there is no way to disentangle the effects of a
gravitational field from those of being in a uniformly accelerating frame, simply by observ-
ing the behavior of freely-falling particles. This follows from the universality of gravitation;
in electrodynamics, in contrast, it would be possible to distinguish between uniform accel-
eration and an electromagnetic field, by observing the behavior of particles with different
charges. But with gravity it is impossible, since the ”charge” is necessarily proportional to
the (inertial) mass.
To be careful, we should limit our claims about the impossibility of distinguishing gravity
from uniform acceleration by restricting our attention to ”small enough regions of space-
time”. If the sealed box were sufficiently big, the gravitational field would change from place
to place in an observable way, while the effect of acceleration would always be in the same
direction.
The WEP can therefore be stated as follows: The motion of freely-falling particles are
the same in a gravitational field and a uniformly accelerated frame, in small enough regions
of spacetime.
Finally Einstein states the principle of equivalence (EEP) as follows: In small enough
regions of spacetime, the laws of physics reduce to those of special relativity; it is impossible
to detect the existence of a gravitational field by means of local experiments.
The EEP implies (or at least suggests) that we should attribute the action of gravity to
the curvature of spacetime. Remember that in special relativity a prominent role is played
by inertial frames– while it is not possible to single out some frame of reference as uniquely
”at rest”, it is possible to single out a family of frames that are ”unaccelerated” (inertial).
The acceleration of a charged particle in an electromagnetic field is therefore uniquely de-
fined with respect to these frames. The EEP, on the other hand, implies that gravity is
inescapable– there is no such thing as a ”gravitationally neutral object” with respect to
which we can measure the acceleration due to gravity. It follows that the acceleration due
to gravity is not something that can be reliably defined, and therefore is of little use.
Instead, it makes more sense to define ”unaccelerated” as ”freely falling”. From here
we are led to the idea that gravity is not a ”force”– a force is something that leads to
acceleration, and our definition of zero acceleration is ”moving freely in the presence of
whatever gravitational field happens to be around”.
In Special Relativity, we have a procedure for starting at some point and constructing
an inertial frame that stretches throughout spacetime, by joining together rigid rods and
attaching clocks to them. But, again due to inhomogeneities in the gravitational field, this
is no longer possible. If we start in some freely-falling state and build a large structure out
of rigid rods, at some distance away freely-falling objects will look like they are accelerating
with respect to this reference frame.
The solution is to retain the notion of inertial frames, but to discard the hope that they
can be uniquely extended throughout space and time. Instead we can define locally inertial
frames, those that follow the motion of individual freely-falling particles in small enough
regions of spacetime.
We therefore would like to describe spacetime as a kind of mathematical structure that
looks locally like Minkowski space, but may possess nontrivial curvature over extended

9
regions.
We can readily choose pseudo-Riemannian manifolds with Minkowkian signature, (1, 3),
for the mathematical structure of spacetime as they have nontrivial curvature over extended
regions and locally look like pseudo-Euclidean space, in the case of (1, 3) signature it will
be the Minkowskian spacetime, as was required.
The EEP arises from the idea that gravity is universal; it affects all particles (and indeed
all forms of energy-momentum) in the same way. This feature of universality led Einstein to
propose that what we experience as gravity is a manifestation of the curvature of spacetime.
The idea is simply that something so universal as gravitation could be most easily described
as a fundamental feature of the background on which matter fields propagate, as opposed
to as a conventional force. At the same time, the identification of spacetime as a curved
manifold is supported by the similarity between the undetectability of gravity in local regions
and our ability to find locally inertial coordinates on a manifold.
Therefore we conclude this section by asserting that wherever there is a gravitational
field present in spacetime we have curvature on our pseudo-Riemannian manifold and, as
was seen in previous sections, curvature depends on the metric, which defines the geometry
of our manifold; more precisely the measure of the curvature of a (pseudo-) Riemannian
manifold is the Riemann tensor which is defined in terms of derivatives of the metric of the
manifold, for this if we want to explore the gravitational effects on a region of spacetime we
should investigate the variations of the metric and its derivatives.

4.2 Einstein Field Equations As A Classical Field[7]

As was stated in the previous section the dynamical variable of gravitation is the metric.
The question is, what scalars can we make out of the metric to serve as a Lagrangian?
Since the metric can be set equal to its canonical form and its first derivatives set to
zero at any one point, any nontrivial scalar must involve at least second derivatives of the
metric. The Riemann tensor is of course made from second derivatives of the metric, and it
can be shown that the only independent scalar that can be constructed from the Riemann
tensor is the Ricci scalar R. It is true that, any nontrivial tensor made from products of
the metric and its first and second derivatives can be expressed in terms of the metric and
the Riemann tensor. Therefore, the only independent scalar constructed from the metric,
which is no higher than second order in its derivatives, is the Ricci scalar. Hilbert figured
that this was therefore the simplest possible choice for a Lagrangian, and proposed
√
Z
SH = −g R dn x, (1)

known as the Hilbert action (or sometimes the Einstein-Hilbert action).

The equation of motion should come from varying the action with respect to the metric.
As the action isn’t quite in the form suggested by section 3, since it can’t be written in terms
of covariant derivatives of gµν (which would simply vanish). Therefore, instead of simply
plugging into the Euler-Lagrange equations, we have to consider directly the behavior of SH
under small variations of the metric. After doing so we get to the following relation
1 δSH 1
√ µν
= Rµν − R gµν = 0. (2)
−g δg 2

Which is the Einstein’s equation in vacuum.

10
 We derived Einstein’s equation ”in vacuum” because we only included the gravitational
part of the action, not additional terms for matter fields. What we would really like, however,
is to get the non-vacuum field equation as well. That means we consider an action of the
form
1
S= SH + SM (3)
16πG
where SM is the action for matter, and we have presciently normalized the gravitational
action so that we get the right answer. Following through the same procedure leads to
1 δS 1  1  1 δSM
√ µν
= R µν − R gµν +√ = 0. (4)
−g δg 16πG 2 −g δg µν

We now boldly define the energy-momentum tensor to be

1 δSM
Tµν = −2 √ (5)
−g δg µν

This allows us to recover the complete Einstein’s equation

1
Rµν − Rgµν = 8πGTµν (6)
2
Why should we think that (5) is really the energy-momentum tensor? In some sense it
is only because it is a symmetric, conserved, (0, 2)-tensor with dimensions of energy density,
it could be called anything though.
In this context energy-momentum conservation arises as a consequence of symmetry of
the Lagrangian under spacetime translations and according to Noether’s Theorem.

5 Prologue
5.1 Beyond Riemannian Spaces
Finsler spaces, Lagrange spaces, and Hamilton spaces, are all geometries beyond that of
Riemannian; in particular Finslerian Geometry is just the next step from the Riemannian
geometry and generalizes Riemannian manifolds in the sense that the metric defined on the
smooth manifold need not be an inner product or otherwise known as 2-norm, nevertheless
the metric of Finsler spaces are necessarily positive definite, so although they generalize the
Riemannian spaces they do not generalize the pseudo-Riemannian spaces.

5.2 Alternative Theories of Gravitation

General relativity has passed a wide variety of experimental tests. Nevertheless, it is al-
ways possible that the next experiment we do will reveal a deviation from Einstein’s original
formulation; nonetheless the theory suffers from singularities, which is quite unacceptable
in complete theory. Let us therefore briefly consider ways in which general relativity could
be modified. There are an uncountable number of such ways, but I will state four different
possibilities and explain the basic ideas of two of these (with minimal depth or details):
• gravitational scalar fields
• extra spatial dimensions

11
 • higher-order terms in the action
• non-Christoffel connections
A popular set of alternative models are known as scaler-tensor theories of gravity since
they involve both the metric tensor, gµν and a scalar field, λ. In particular, the scalar
field couples directly to the curvature scalar, not simply to the metric (as the Equivalence
Principle would seem to imply). The action can be written as a sum of a gravitational piece,
a pure-scalar piece, and a matter piece:

S = Sf R + Sλ + SM

where
√
Z
Sf R = d4 x −g f (λ) R,

√
Z h 1 i
Sλ = d4 x −g − h(λ)g µν (∂µ λ)(∂ν λ) − U (λ) ,
2
and
√
Z
SM = d4 x −g LM (gµν , ψi ).

Here, f (λ), h(λ) and U (λ) are functions that define the theory, and the matter Lagrangian
LM depends on the metric and a set of matter fields ψi , but not on λ. For the Hilbert
action, f is a constant.
One of the earliest scalar-tensor models is known as Brans-Dicke theory, and corresponds
in our notation to the choices
λ ω
f (λ) = , h(λ) = , U (λ) = 0
16π 8πλ
where ω is a coupling constant.
In the Brans-Dicke theory, the scalar field is massless, but in the ω → ∞ limit the field
becomes non dynamical and ordinary GR is recovered. Current bounds from Solar System
tests imply ω > 500, so if there is such a scalar field it must couple only weakly to the Ricci
scalar.
A popular approach to dealing with scalar-tensor theories is to perform a conformal
transformation to bring the theory in to a form that looks like conventional GR. In the
conformal frame, therefore, the curvature scalar appears by itself, not multiplied by any
function of λ.. This frame is sometimes called the Einstein frame, since Einstein’s equations
for the conformal metric g̃µν take on their conventional form. The original frame with
metric gµν is called the Jordan frame, or sometimes the string frame. (String theory typically
predicts a scalar-tensor theory rather than ordinary GR, and the string worldsheet responds
to the metric gµν .)
Another way to modify general relativity is to allow for the existence of extra spatial di-
mensions; in fact the physical consequences of extra dimensions turn out to be closely related
to those of scalar-tensor theories. By extra dimensions we don’t simply mean considering
GR in higher-dimensional spaces, but rather considering models in which the spacetime
appears four-dimensional on large scales even though there are really 4 + d total dimensions.
The simplest way for this to happen is if the extra d dimensions are ”compactified” on some
manifold; Models of this kind are known as Kaluza-Klein theories.

12
 Yet another view of modification focuses on the fact that quantum mechanics and general
theory of relativity don’t quite fit together, and these views tries to develop a quantum theory
of gravitation and main aspect of developing such theories lies in getting rid of singularities
of GR. In fact string theory is such a theory. Also Twistor theory was proposed by Roger
Penrose in 1967 as a possible path to quantum gravity and has evolved into a branch of
theoretical and mathematical physics. Penrose proposed that twistor space should be the
basic arena for physics from which spacetime itself should emerge. It leads to a powerful set
of mathematical tools that have applications to differential and integral geometry, nonlinear
differential equations, representation theory, and in physics to relativity and quantum field
theory, in particular to scattering amplitudes.

5.3 The Main Mystery

We concluded that gravitation means curvature in spacetime and that led us to a specific
mathematical structure for describing the spacetime, but all that happened on account of
Einstein’s Equivalence Principle. The question is : why should gravitation obey the Principle
of Equivalence? The answer may not be found in realm of classical physics, but in constraints
imposed by the quantum theory of gravitation. It seems to be impossible to construct
any Lorentz-invariant quantum theory of particles of mass zero and spin two(that is like
gravitons, which are quanta of gravitation in modern theories), unless the corresponding
classical field theory obeys the Principle of Equivalence.. Thus the Principle of Equivalence
appears as the best bridge between the theories of gravitation and elementary particles.

13
References
[1] L. Auslander and R. E. MacKenzie. Introduction to Differentiable Manifolds. New York,
McGraw-Hill, 1963.
[2] R. W. R. Darling. Differential Forms and Connections. New York, Cambridge University
Press, 1994.

[3] R. V. Eötvös. Eötvös experiment. Math. Nat. Ber. Ungarn, 1890.

[4] T. Frankel. The Geometry of Physics. New York, Cambridge University Press, 1997.
[5] S. Kobayashi and K. Nomizu. Foundations of Differential Geometry. New York, Inter-
science Publishers, 1963.

[6] H. Stephani. General Relativity. Cambridge, Cambridge University Press, 1982.

[7] R. M. Wald. General Relativity. Chicago, The University of Chicago Press, 1984.

14