Chapter 7

The General Theory of

Relativity

The General Theory of Relativity is, as the name indicates, a generalization

of the Special Theory of Relativity. It is certainly one of the most remarkable
achievements of science to date, it was developed by Einstein with little or
no experimental motivation but driven instead by philosophical questions:
Why are inertial frames of reference so special? Why is it we do not feel
gravity’s pull when we are freely falling? Why should absolute velocities be
forbidden but absolute accelerations by accepted?

Figure 7.1: Einstein

1
2

7.1 The happiest thought of my life.

In 1907, only two years after the publication of his Special Theory of Rel-
ativity, Einstein wrote a paper attempting to modify Newton’s theory of
gravitation to fit special relativity. Was this modification necessary? Most
emphatically yes! The reason lies at the heart of the Special Theory of Rel-
ativity: Newton’s expression for the gravitational force between two objects
depends on the masses and on the distance separating the bodies, but makes
no mention of time at all. In this view of the world if one mass is moved,
the other perceives the change (as a decrease or increase of the gravitational
force) instantaneously. If exactly true this would be a physical effect which
travels faster than light (in fact, at infinite speed), and would be inconsis-
tent with the Special Theory of Relativity (see Sect. ??). The only way
out of this problem is by concluding that Newton’s gravitational equations
are not strictly correct. As in previous occasions this does not imply that
they are “wrong”, it only means that they are not accurate under certain
circumstances: situations where large velocities (and, as we will see, large
masses) are involved cannot be described accurately by these equations.
In 1920 Einstein commented that a thought came into his mind when
writing the above-mentioned paper he called it “the happiest thought of my
life”:

The gravitational field has only a relative existence... Because for

an observer freely falling from the roof of a house – at least
in his immediate surroundings – there exists no gravitational
field.

Let’s imagine the unfortunate Wile E. Coyote falling from an immense

height 1 . As he starts falling he lets go of the bomb he was about to drop
on the Road Runner way below. The bomb does not gain on Wile nor does
it lag behind. If he were to push the bomb away he would see it move with
constant speed in a fixed direction. This realization is important because
this is exactly what an astronaut would experience in outer space, far away
from all bodies (we have good evidence for this: the Apollo 10–13 spacecrafts
did travel far from Earth into regions where the gravitational forces are quite
weak).
Mr. Coyote is fated to repeat the experience with many other things:
rocks, magnets, harpoons, anvils, etc. In all cases the same results are
obtained: with respect to him all objects, irrespective of composition, mass,
1
I ignore air resistance
 3

etc. behave as if in free space. So, if he should fall inside a closed box, he
would not be able to tell whether he was plunging to his death (or, at least,
severe discomfort), or whether he was in outer space on his way to Pluto at
constant speed.
This is reminiscent of Galileo’s argument: the observer lets go of some
objects which remain in a state of uniform motion (with respect to him!).
This behavior is independent of their chemical or physical nature (as above,
air resistance is ignored). The observer (Wile), as long as he confines his/her
observations to his/her immediate vicinity (that is, as long as he/she does
not look down) has the right to interpret his state as ‘at rest’. Just as
Galileo argued that experiments in a closed box cannot determine the state
of uniform motion of the box, Einstein argued that experiments in a freely
falling small2 closed box cannot be used to determine whether the box is in
the grip of a gravitational force or not.
Why would this be true? The answer can be traced back to the way in
which gravity affects bodies. Remember (see Sect. ??) that the quantity we
called m (the mass) played two different roles in Newton’s equations. One
is to determine, given a force, what the acceleration of the body would be:
F = ma (the inertial mass). The other is to determine the intensity with
which the said body experiences a gravitational force: F = mM G/r2 (the
gravitational mass). As mentioned before these two quantities need not be
equal: the first “job” of m is to tell a body how much to accelerate given
any force, a kick, an electric force (should the body be charged), etc. The
second “job” tells the body how much of the gravitational force should it
experience and also determines how strong a gravitational force it generates.
But, in fact, both numbers are equal (to a precision of ten parts per billion).
What does this imply? Well, from Newton’s equations we get

mM G MG
= ma so that = a;
r2 r2
this equation determines how a body moves, which trajectory it follows,
how long does it take to move from one position to another, etc. and is
independent of m! Two bodies of different masses, composition, origin and
guise will follow the same trajectory: beans, bats and boulders will move in
the same way.
So the equality of the two m’s was upgraded by Einstein to a postulate:
the Principle of Equivalence; this one statement (that the m in ma and
the m in mM G/r2 are identical) implies an incredible amount of new and
2
The reasons behind the requirement that the box be small will become clear soon.
 4

surprising effects. The m in F = ma is called the inertial mass and the m in

The m in F = ma is mM G/r2 the gravitational mass. Then the Principle of Equivalence states
called the inertial mass that the inertial and gravitational masses are identical.
and the m in mM G/r 2
the gravitational mass
The whole of the General Theory of Relativity rests on this postulate,
The inertial and
gravitational masses are
and will fail if one can find a material for which the inertial and gravita-
identical tional masses have different values. One might think that this represents a
defect of the theory, its Achilles heel. In one sense this is true since a single
experiment has the potential of demolishing the whole of the theory (people
have tried...hard, but all experiments have validated the principle of equiv-
alence). On the other hand one can argue that a theory which is based on a
minimum of postulates is a better theory (since there are less assumptions
involved in its construction); from this point of view the General Theory of
Relativity is a gem 3 .

The completed formulation of the General Theory of Relativity was pub-

lished in 1916 (Fig. 7.2).

Figure 7.2: Einstein’s General Theory of Relativity paper.

3
The Special Theory of Relativity is equally nice, it is based on the one statement that
all inertial frames of reference are equivalent.
 5

7.1.1 Newton vs. Einstein

I have stated that Newton’s mechanics and his theory of gravitation are but
approximations to reality and whose limitations are now known 4 . So it
might be questionable to use F = ma and Fgrav = mM G/r2 as basis to any
argument as was done above. Einstein was careful to use these expressions
only in situations where they are extremely accurate (small speeds com-
pared to c and small gravitational forces). In these cases the inertial and
gravitational masses are identical, as shown by experiment. Then he postu-
lated that the same would be true under all circumstances. This statement,
while consistent with Newton’s equations, cannot, in a strict logical sense,
be derived from them.

7.2 Gravitation vs acceleration

Consider the following experiment: a person is put in a room-size box high
above the moon (chosen because there is no air and hence no air friction)
with a bunch of measuring devices. This box is then taken high above the
lunar surface and then let go: the box is then freely falling. The question
is now, can the observer determine whether he/she is falling or whether
he/she is in empty space unaffected by external forces (of course the answer
is supposed to come before the box hits the surface). The answer to that is
a definite NO! The observer can do experiments by looking at how objects
move when initially at rest and when given a kick, he/she will find that they
appear to move as is there were no gravitational forces at all! Similarly any
experiment in physics, biology, etc. done solely inside the box will be unable
to determine whether the box is freely falling or in empty space.
Why is that? Because of the equality of the gravitational and inertial
masses. All objects are falling together and are assumed to be rather close
to each other (the box is not immense) hence the paths they will follow will
be essentially the same for each of them. So if the observer lets go of an
apple, the apple and the observer follow essentially the same trajectory, and
this implies that the observer will not see the apple move with respect to
him. In fact, if we accept the priniciple of equivalence, nothing can be done
to determine the fact that the observer is falling towards the Moon, for this
can be done only if we could find some object which behaved differently from
all the rest, and this can happen only if its gravitational and inertial masses
4
For all we know our present theories of mechanics and gravitation may also be invalid
under certain conditions.
 6

are different. The principle of equivalence then implies that the observer
will believe that he/she is an inertial frame of reference...until disabused of
the notion by the crash with the surface.
The principle of equivalence is of interest neither because its simplicity,
nor because it leads to philosophically satisfying conclsions. It’s importance
is based on the enormous experimental evidence which confirms it; as with
the Special Theory, the General Theory of Relativity is falsifiable.

Figure 7.3: An observer cannot distinguish between acceleration produced

by a rocket and the acceleration produced by gravity.

The lesson is that for any gravitational force we can always choose a
frame of reference in which an observer will not experience any gravitational
For any gravitational force effects in his/her immediate vicinity (the reason for this last qualification
we can always choose a will become clear below). Such a frame of reference is, as stated above,
frame of reference in which
an observer will not
experience any gravitational
effects in his/her immediate
vicinity
 7

freely falling.
Conversely one can take the box an attach it to a machine that accel-
erates it (Fig. 7.3). If an observer drops an apple in such an accelerated
box he/she will see the apple drop to the floor, the observer will also feel
hi/her-self pressed against the bottom of the box, etc. The observer can-
not distinguish between this situation and the one he/she would experience
in the presence of gravitational forces! As long as we do experiments in a
small region, the effects produced by a gravitational force are indistinguish-
able from those present in an accelerated reference frame. In a small region the effects
produced by a gravitational
Does this mean that the gravitational forces are a chimera, an illusion?
force are indistinguishable
Of course no. Consider for example Fig. 7.4, two apples fall to the Moon from those present in an
inside a box which is also falling. If they are separated by a sufficiently accelerated reference frame
large distance an observer falling with the apples and box will find that the
distance between the apples shortens as time goes on: this cannot be an
inertial frame he argues (or else it is, but there is some force acting on the
apples).
This same set-up can be used to distinguish between a box under the
influence of a gravitational force and one being pulled by a machine; again
we need a very big box (planet-sized). An observer places an two apples at
the top of the box and releases them, he/she carefully measures its initial
separation. The apples fall to the bottom of the box and the observer
measures their separation there. If it is the same as above, and is the same
irrespective of their initial separation, the observer is being pulled by a
machine (box and all). If the separation is different, he/she can conclude
that he/she is experiencing the effects of a gravitational force.

7.3 Light
A very surprising corollary of the above is that light paths are bent by
gravitational forces! I will argue this is true in a slightly round-about way.
Consider an elevator being pulled by a crane so that it moves with con-
stant acceleration (that is its velocity increases uniformly with time). Sup-
pose that a laser beam propagating perpendicular to the elevator’s direction
of motion enters the elevator through a hole on the left wall and strikes the
right wall. The idea is to compare what the crane operator and the elevator
passenger see.
The crane operator, who is in an inertial frame of reference, will see the
sequence of events given in Fig. 7.5. Note, that according to him/her, light
travels in a straight line (as it must be since he/she is in an inertial frame!).
 8

Figure 7.4: Experiment that differentiates between a gravitational effect and

the effects of uniform acceleration: for an observer in the box the apples will
draw closer.

The elevator passenger will see something very different as shown in Fig.
7.5: the light-path is curved! Thus for this simple thought experiment light
paths will be curved for observers inside the elevator.
Now we apply the equivalence principle which implies that we cannot
distinguish between an elevator accelerated by a machine and an elevator
experiencing a constant gravitational force. It follows that the same effect
should be observed if we place the elevator in the presence of a gravitational
Light light paths are curved force: light paths are curved by gravity
by gravity
That gravity affects the paths of planets, satellites, etc. is not something
strange. But we tend to think of light as being different somehow. The above
argument shows that light is not so different from other things and is indeed
affected by gravity in a very mundane manner (the same elevator experiment
could be done by looking at a ball instead of a beam of light and the same
 9

Figure 7.5: Left: sequence of events seen by an crane operator lifting an

elevator at constant acceleration (the speed increases uniformly with time).
The short horizontal line indicates a laser pulse which, at the initial time,
enters through an opening on the left-hand side of the elevator. At the
final time the light beam hits the back wall of the elevator. Right: same
sequence of events seen by a passenger in an elevator being hoisted by a
crane. The line joining the dots indicates the path of a laser which, at the
initial time, enters through an opening on the left-hand side of the elevator.
At the final time the light beam hits the back wall of the elevator.

sort of picture would result).

A natural question is then, why do we not see light fall when we ride
an elevator? The answer is that the effect in ordinary life is very small.
Suppose that the height of the elevator in Fig. 7.5 is 8 ft. and its width
is 5 ft; if the upward acceleration is 25% that of gravity on Earth then the
distance light falls is less than a millionth of the radius of a hydrogen atom
(the smallest of the atoms). For the dramatic effect shown in the figure
the acceleration must be enormous, more than 1016 times the acceleration
of gravity on Earth (this implies that the passenger, who weights 70 kg on
Earth, will weigh more than 1,000 trillion tons in the elevator).
10

This does not mean, however, that this effect is completely unobservable
(it is small for the case of the elevator because elevators are designed for
very small accelerations, but one can imagine other situations). Consider
from example a beam of light coming from a distant star towards Earth
(Fig. 7.6) which along the way comes close to a very massive dark object.
The arguments above require the light beam to bend; and the same thing
will happen for any other beam originating in the distant star. Suppose that
the star and the opaque object are both prefect spheres, then an astronomer
on Earth will see, not the original star, but a ring of stars (often called an
“Einstein ring). If either the star or the massive dark object are not perfect
spheres then an astronomer would see several images instead of a ring (Fig.
7.7). This effect has been christened gravitational lensing since gravity acts
here as a lens making light beams converge.

Figure 7.6: Diagram illustrating the bending of light from a star by a massive
compact object. If both the bright objects and the massive object are prefect
sphere, there will be an apparent image for every point on the “Einstein
ring”.

How do we know that the multiple images which are sometimes seen (Fig.
7.7) are a result of the bending of light? The argument is by contradiction:
 11

Figure 7.7: The Einstein Cross: four images of a quasar GR2237+0305 (a

very distant, very bright object) appear around the central glow. The split-
ting of the central image is due to the gravitational lensing effect produced
by a nearby galaxy. The central image is visible because the galaxy does
not lie on a straight line from the quasar to Earth. The Einstein Cross is
only visible from the southern hemisphere.

suppose they are not, that is suppose, that the images we see correspond to
different stars. Using standard astronomical tools one can estimate the dis-
tance between these stars; it is found that they are separated by thousands
of light years, yet it is observed that if one of the stars change, all the others
exhibit the same change instantaneously! Being so far apart precludes the
possibility of communication between them; the simplest explanation is the
one provided by the bending of light. It is, of course, possible to ascribe
these correlations as results of coincidences, but, since these correlations
are observed in many images, one would have to invoke a “coincidence” for
hundreds of observations in different parts of the universe.

The bending of light was one of the most dramatic predictions of the
General Theory of Relativity, it was one of the first predictions that were
verified as we will discuss below in Sect. 7.12.
12

7.4 Clocks in a gravitational force.

When comparing a clock under the influence of gravitational forces with one
very far from such influences it is found that the first clock is slow compared
to the second. To see this consider the same clock we used in the Special
Theory of Relativity. For this experiment, however, imagine that the clock
is being accelerated upward, being pulled by a crane. The clock gives off a
short light pulse which moves towards the mirror at the top of the box, at
the same time the mirror recedes from the pulse with even increasing speed
(since the box accelerates). Still the pulse eventually gets to the mirror
where it is reflected, now it travels downward where the floor of the box is
moving up also with ever increasing velocity (see Fig. 7.8).

Figure 7.8: An accelerated clock. The circle denotes a pulse of light which
at the initial is sent from a source; after a time it reaches the top of the the
box and is reflected. The time it takes to do the trip is longer than for a
clock at rest.

On the trip up the distance covered by light is larger than the height
 13

of the box at rest, on the trip down the distance is smaller. A calculation
shows that the whole distance covered in the trip by the pulse is larger than
twice the height of the box, which is the distance covered by a light pulse
when the clock is at rest.
Since light always travels at the same speed, it follows that the time it
takes for the pulse to go the round trip is longer when accelerating than
when at rest: clocks slow down whenever gravitational forces are present.
This has an amazing consequence: imagine a laser on the surface of
a very massive and compact planet (so that the gravitational field is very
strong). An experimenter on the planet times the interval between two crests
of the laser light waves and gets, say, a millionth of a second. His clock ,
however, is slow with respect to the clock of an observer far away in deep
space, this observer will find that the time between two crests is larger. This
implies that the frequency of the laser is larger on the planet than in deep
space: light leaving a region where gravity is strong reddens. This is called Light leaving a region where
the gravitational red-shift (see Fig. 7.9). gravity is strong reddens

Figure 7.9: The gravitational redshidft. Since clocks slow down in a strong
gravitational field then light, whose oscillations can be used as clocks, will
be shifter towards the red as it leaves a region where gravity is strong.

As for time dilation, the slowing down of clocks in the presence of gravi-
tational forces affects all clocks, including biological ones. A twin trveling to
14

a region where gravity is very strong will come back a younger than the twin
left in a rocket in empty space. This is an effect on top of the one produced
by time dilation due to the motion of the clocks. The gravitational forces
required for a sizable effect, however, are enormous. So the twin will return
younger...provided she survives.

7.5 Black holes

So gravity pulls on light just as on rocks. We also know that we can put
rocks in orbit, can we put light in orbit? Yes! but we need a very heavy
object whose radius is very small, for example, we need something as heavy
as the sun but squashed to a radius of less than about 3km. Given such an
object, light moving towards it in the right direction will, if it comes close
enough land in an orbit around it. If you place yourself in the path of light
as it orbits the object, you’d be able to see your back.
But we can go farther and imagine an object so massive and compact
that if we turn on a laser beam on its surface gravity’s pull will bend it
back towards the surface. Think what this means: since no light can leave
this object it will appear perfectly black, this is a black hole. An object
which comes sufficiently close to a black hole will also disappear into it
(since nothing moves faster than light if an object traps light it will also
trap everything else).
The effect of a black holes, like all gravitational effects, decreases with
distance. This means that there will be a “boundary” surrounding the black
hole such that anything crossing it will be unable to leave the region near
the black hole; this boundary is called the black-hole horizon see Fig. 7.10
Anything crossing the horizon is permanently trapped. Black holes are
prefect roach motels: once you check in (by crossing the horizon), you never
check out.
The distance from the black hole to the horizon is determined by the
mass of the black hole: the larger the mass the mode distant is the horizon
from the center. For a black hole with the same mass as out sun the horizon
is about 3 km from the center; for black holes with a billion solar masses
(yes there are such things) this is increased to 3×109 km, about the distance
from the sun to Uranus. For very massive black holes the horizon is so far
away from the center that an observer crossing it might not realize what has
just happened, only later, when all efforts to leave the area prove futile, the
dreadful realization of what happened will set in.
Imagine a brave (dumb?) astronaut who decides to through the horizon
 15

Figure 7.10: Illustration of the horizon surrounding the black hole. The
black holes is represented by the small heavy dot, the light rays or particle
trajectories which cross the dotted line cannot cross it again.

and into the nearest black hole and let us follow his observations. The first
effects that becomes noticeable as he approaches the event horizon is that
his clock ticks slower and slower with respect to the clocks on his spaceship
very far from the black hole (see Sect. 7.4) to the point that it will take
infinite spaceship time for him to cross the horizon. In contrast it will take
a finite amount of astronaut time to cross the horizon, an extreme case of
the relativity of time.
As the astronaut approaches the horizon the light he emits will be more
and more shifted towards the red (see Sect. 7.4) eventually reaching the
infrared, then microwaves, then radio, etc. In order to see him the spaceship
will eventually have to detect first infrared light, then radio waves, then
microwaves, etc.
After crossing the horizon the astronaut stays inside. Even though the
crossing of the horizon might not be a traumatic experience the same cannot
be said for his ultimate fate. Suppose he decides to fall feet first, then, when
sufficiently close to the black hole, the gravitational pull on his feet will be
much larger than that on his head and he will be literally ripped to pieces.
So far black holes appear an unfalsifiable conclusion of the General The-
16

ory of Relativity: their properties are such that no radiation comes out of
them so they cannot be detected from a distance, and if you should decide
to go, you cannot come back to tell your pals whether it really was a black
hole or whether you died in a freak accident. Doesn’t this contradict the
basic requirement that a scientific theory be falsifiable (Sect. ??)?

Figure 7.11: Artist’s version of a black hole accreting matter from a compan-
ion star. The Star is on the left of the picture and is significantly deformed
by the gravitational pull of the black hole; the object on the right represents
the matter which surrounds the black hole and which is being sucked into
it. The black hole is too small to be seen on the scale of this picture

Well, no, General Theory of Relativity even in this one of its most ex-
treme predictions is falsifiable. The saving circumstance is provided by
the matter surrounding the black hole. All such stuff is continuously being
dragged into the hole (see Fig. 7.11) and devoured, but in the process it gets
extremely hot and radiates light, ultraviolet radiation and X rays. More-
over, this cosmic Maelstrom is so chaotic that the radiation changes very
rapidly, sometimes very intense, sometimes much weaker, and these changes
come very rapidly (see Fig. 7.12). From this changes one can estimate the
size of the object generating the radiation.
On the other hand astronomers can see the gravitational effects on near-
by stars of whatever is making the radiation. And from these effects they
 17

Figure 7.12: X-ray emission from a black hole candidate (Cygnus X1)

can estimate the mass of the beast. Knowing then the size, the manner in
which matter radiates when it comes near, and the mass one can compare
this to the predictions of General Theory of Relativity and decide whether
this is a black hole or not. The best candidate for a black hole found in this
way is called Cygnus X1 (the first observed X ray source in the constellation
Cygnus, the swan).
All the ways we have of detecting black holes depend on the manner in
which they affect the matter surrounding them. The most striking example
is provided by some observation of very distant X-ray sources which are
known to be relatively compact (galaxy size) and very far away. Then the
very fact that we can see them implies that they are extremely bright objects,
so bright that we know of only one source that can fuel them: the radiation
given off by matter while being swallowed by a black hole 5 . So the picture
we have of these objects, generically called active galactic nuclei, is that of
a supermassive (a billion solar masses or so) black hole assimilating many
stars per second, and in disappearing these stars give off the energy that
announces their demise.
All this from the (apparently) innocent principle of equivalence.

7.6 Gravitation and energy

Consider a beam of sunlight falling on your skin; after a while your skin
warms and, eventually, will burn: light carries energy (which is absorbed
by your skin thus increasing its temperature). Recall also that a body with
5
This is much more efficient than nuclear power which would be incapable of driving
such bright sources.
 18

mass m, by its very existence, carries and energy mc2 (Sec. ??). There is
no way, however, in which we can associate a mass with light; for example,
we can always change the speed of a mass (even if only a little bit), but this
cannot be done with light.
The force of gravity affects both light and all material bodies; since
both carry energy, but only the bodies carry mass, it follows that gravity
Gravity will affect anything will affect anything carrying energy. This conclusion lies at the root of the
carrying energy construction of Einstein’s equations which describe gravity.
Note that this conclusion has some rather strange consequences. Con-
sider for example a satellite in orbit around the Earth, when the Sun shines
on it it will increase its energy (it warms up), and gravity’s pull with it.
When the satellite is in darkness it will radiate heat, lose energy and the
force of gravity on it will decrease 6 .
Again let me emphasize that this argument is not intended to imply that
light carries mass, but that gravity will affect anything that carries energy.

7.7 Space and time.

When considering the Special Theory of Relativity we concluded that the
state of motion of an observer with respect to, say, a laboratory, determines
the rate at which his/her clocks tick with respect to the lab’s clocks (see
Sect.??). Thus, in this sense, time and space mingle: the position of the
observer (with respect to the lab’s measuring devices) determines, as time
evolves, his/her state of motion, and this in turn determines the rate at
which his/her clocks tick with respect to the lab’s.
Now consider what happens to objects moving under the influence of a
gravitational force: if initially the objects set out at the same spot with the
same speed they will follow the same path (as required by the principle of
equivalence). So what!? To see what conclusions can be obtain let me draw
a parallel, using another murder mystery.
Suppose there is a closed room and a line of people waiting to go in. The
first person goes in and precisely two minutes afterward, is expelled through
a back door, dead; it is determined that he died of a blow to the head. The
police concedes that the room is worth investigating, but procrastinates,
alleging that the person was probably careless and his death was accidental.
Soon after, however, a second victim enters the room with precisely the
same results, she also dies of an identical blow to the head; the police claims
an astounding coincidence: two accidental deaths. This goes on for many
6
Needless to say this is a very small effect, of the order of one part in a trillion.
 19

hours, each time the victim dies of the same thing irrespective of his/her
age, occupation, habits, color, political persuasion or taste in Pepsi vs. Coke;
animals suffer the same fate, being insects of whales. If a rock is sent flying
in, it comes out with a dent of the same characteristics as the ones suffered
by the people and animals.
The police finally shrewdly concludes that there is something in the room
that is killing people, they go in and... But the result is not important, what
is important for this course is the following. We have a room containing
something which inflicts a certain kind of blow to everything going through
the room, I can then say that this inflicting of blows is a property of the
room.
Consider now a region of empty space relatively near some stars. Assume
that the only force felt in this region is the gravitational pull of these stars,
hence all objects, people, animals, etc. going into this region will accelerate
in precisely the same way. Then I can state that the region in space has a
property which generates this acceleration 7 .
Remember however that the region considered was in empty space (it
only contains the objects we send into it), yet some property of this re-
gion determines the motion of anything that goes through it; moreover this
property is a result of the gravitational pull of nearby heavy objects. The
conclusion is then that gravity alters the properties of space, we also saw
that the rates of clocks are altered under the influence of a gravitational
force, it follows that gravity alters the properties of space and time. Space Gravity alters the properties
and time is in fact very far from the unchanging arena envisaged by New- of space and time

ton, they are dynamical objects whose properties are affected by matter and
energy. These changes or deformations of space and time in turn determine
the subsequent motion of the bodies in space time: matter tells space-time
how to curve and space-time tells matter how to move (Fig. 7.13). Matter tells space-time how
to curve and space-time
tells matter how to move

Figure 7.13: An illustration of the bending of space produced by a massive

object

7
I assume that the objects coming into this region are not too heavy, so that their
gravitational forces can be ignored and that the start from the same spot with identical
velocities.
20

7.8 Properties of space and time.

Up to here I’ve talked little of the implications of the Special Theory of
Relativity on the General Theory of Relativity, I have only argued that in
special relativity time and space are interconnected. In a separate discussion
I argued that gravity alters space. In this section I will use what we know
about length contraction together with the equivalence principle to deter-
mine how space is altered by gravity and to show that it is this deformation
of space that is responsible for the gravitational force.
Consider two identical disks one of which is made to rotate uniformly
as in Fig. 7.14. In the non-rotating disk we select a small segment of its
circumference of length `o . For the rotating disk this same segment will
be measured to have length ` which is smaller, due to length contraction
(Sec. ??), than `o . Since the little segment we focused on is no different
from any other (small) segment of the circumference, we can conclude that
the circumference of the rotating disk is smaller than the circumference of
the non-rotating disk.

Figure 7.14: A rotating vs a non-rotating disk. The bit labeled ` in the

rotating disk is shorter, due to length contraction to the corresponding bit
`o in the non-rotating disk.

Consider now a radius of the disks. This is a length that is always

perpendicular to the velocity of the disk and it is unaffected by the rotation.
thus both disks will continue to have the same radius (see Sect. ??).
So now we have one non-rotating disk whose circumference is related to
the radius by the usual formula, circumference = 2π× radius, and a rotating
disk whose circumference is smaller than this number!
How can this be? Isn’t it true that the perimeter always equals 2π
 21

radius? The answer to the last question is yes...provided you draw the circle
on a flat sheet of paper. Suppose however that you are constrained to draw
circles on a sphere, and that you are forced to measure distances only on
the sphere. Then you find that the perimeter measured along the sphere is
smaller than 2π×radius (with the radius also measured along the sphere,
see Fig. 7.15).

Figure 7.15: The distance from the equator to the pole on a sphere is larger
than the radius. For being constrained to move on the surface of the sphere
this distance is what they would call the radius of their universe, thus for
them the circumference is smaller than 2π×radius and they can conclude
that they live in curved space.

We conclude that the uniformly rotating disk behaves as a (piece of a)

sphere due to length contraction. So much for the effects of special relativity.
Now let us go back to the principle of equivalence. One of its conse-
quences is that, by doing experiments in a small region one cannot distin-
guish between a gravitational force and an accelerated system. So if we
attach a small laboratory of length `0 (at rest) to the small section of the
perimeter, experiments done there will not be able to tell whether the lab.
is in a rotating disk or experiences a gravitational force (remember that a
rotating object is changing its velocity – in direction – and it is therefore
accelerating!).
Putting together the above two arguments we get
Gravitation curves space and time. Gravitation curves space
and time
22

Conversely curved space and time generate effects which are equivalent
to gravitational effects. In order to visualize this imagine a world where
all things can only move on the surface of a sphere. Consider two beings
labeled A and B as in Fig. 7.16, which are fated live on the surface of this
sphere. On a bright morning they both start from the equator moving in a
direction perpendicular to it (that is, they don’t meander about but follow
a line perpendicular to the equator).

Figure 7.16: Two beings moving on a sphere are bound to come closer just
as they would under the effects of gravity

As time goes on the two beings will come closer and closer. This effect
is similar to the experiment done with two apples falling towards the moon
(Fig. 7.4): an observer falling with them will find their distance decreases
as time progresses; sentient apples would find that they come closer as time
goes on.
So we have two descriptions of the same effect: on the one hand grav-
itational forces make the apples approach each other; on the other hand
the fact that a sphere is curved makes the two beings approach each other;
mathematically both effects are, in fact, identical. In view of this the conclu-
sion that gravity curves space might not be so peculiar after all; moreover,
in this picture the equivalence principle is very natural: bodies move the
way they do due to the way in which space is curved and so the motion is
 23

independent of their characteristics 8 , in particular the mass of the body Bodies move the way they
does not affect its motion. do due to the way in which
space is curved and so the
motion is independent of
their characteristics

Figure 7.17: Just as bugs fated to live on the surface of a sphere might find
it peculiar to learn their world is curved, so we might find it hard to realize
that our space is also curved.

Now the big step is to accept that the same thing that happened to the
above beings is happening to us all the time. So how come we don’t see that
the space around us is really curved? The answer is gotten by going back
to the beings A and B: they cannot “look out” away from the sphere where
they live, they have no perception of the perpendicular dimension to this
sphere, and so they cannot “see it from outside” and realize it is curved.
The same thing happens to us, we are inside space, in order to see it curved
we would have to imagine our space in a larger space of more dimensions
8
I am assuming here that the moving things are not massive enough to noticeably curve
space on their own.
24

and then we could see that space is curved; Fig. 7.17 gives a cartoon version
of this.

7.9 Curvature
When considering the beings living on a sphere it is easy for us to differen-
tiate between the sphere and some plane surface: we actually see the sphere
being curved. But when it comes to us, and our curved space, we cannot see
it since this would entail our standing outside space and looking down on
it. Can we then determine whether space is curved by doing measurements
inside it?
To see that this can be done let’s go back to the beings on the sphere.
Suppose they make a triangle by the following procedure: they go form the
equator to the north pole along a great circle (or meridian) of the sphere, at
the north pole they turn 90o to the right and go down another great circle
until they get to the equator, then they make another 90o turn to the right
until they get to the starting point (see Fig. 7.18). They find that all three
lines make 90o angles with each other, so that the sum of the angles of this
triangle is 270o , knowing that angles in all flat triangles always add up to
180o they conclude that the world they live on is not a flat one. Pythagoras’
theorem only holds on flat surfaces
We can do the same thing: by measuring very carefully angles and dis-
tances we can determine whether a certain region of space is curved or not.
In general the curvature is very slight and so the distances we need to cover
to observe it are quite impractical (several light years), still there are some
special cases where the curvature of space is observed: if space were flat
light would travel in straight lines, but we observe that light does no such
thing in regions where the gravitational forces are large; I will discuss this
further when we get to the tests of the General Theory of Relativity in the
following sections.
The curvature of space is real and is generated by the mass of the bodies
in it. Correspondingly the curvature of space determines the trajectories of
all bodies moving in it. The Einstein equations are the mathematical em-
bodiment of this idea. Their solutions predict, given the initial positions and
velocities of all bodies, their future relative positions and velocities. In the
limit where the energies are not too large and when the velocities are signif-
icantly below c the predictions of Einstein’s equations are indistinguishable
from those obtained using Newton’s theory. At large speeds and/or energies
significant deviations occur, and Einstein’s theory, not Newton’s, describes
 25

Figure 7.18: A path followed by a determined being living on the surface of

a sphere; each turn is at right angles to the previous direction, the sum of
the angles in this triangle is then 270o indicating that the surface in which
the bug lives is not flat.

the observations.

7.10 Waves
A classical way of picturing the manner in which heavy bodies curve space
is to imagine a rubber sheet. When a small metal ball is made to roll on
it it will go in a straight line at constant speed (neglecting friction). Now
imagine that a heavy metal ball is placed in the middle of the sheet; because
of its weight the sheet will be depressed in the middle (Fig 7.13). When a
small ball is set rolling it will no longer follow a straight line, its path will
be curved and, in fact, it will tend to circle the depression made by the
heavy ball. The small ball can even be made to orbit the heavy one (it will
eventually spiral in and hit the heavy ball, but that is due to friction, if
the sheet is well oiled it takes a long time for it to happen). This toy then
realizes what was said above: a heavy mass distorts space (just as the heavy
ball distorts the rubber sheet). Any body moving through space experiences
this distortion and reacts accordingly.
26

Now imagine what happens if we drop a ball in the middle of the sheet.
It will send out ripples which spread out and gradually decrease in strength.
Could something similar happen in real life? The answer is yes! When there
is a rapid change in a system of heavy bodies a large amount of gravitational
waves are produced. These waves are ripples in space which spread out form
their source at the speed of light carrying energy away with them.
A computer simulation of a gravitational wave is given in Fig. 7.19. The
big troughs denote regions where the wave is very intense, the black dot
at the center denotes a black hole, the ring around the hole represents the
black hole’s horizon.

Figure 7.19: A computer simulation of a gravitational wave generated by a

collision of two black holes, which have now merged and are represented by
the heavy black dot in the middle.

Can we see gravitational waves? Not yet directly, but we have very strong
indirect evidence of their effects. Several systems which according to the
General Theory of Relativity ought to lose energy by giving off gravitational
waves have been observed. The observations show that these systems lose
energy, and the rate at which this happens coincides precisely with the
predictions from the theory.
Observing gravitational waves directly requires very precise experiments.
The reason is that, as one gets farther and farther away from the source these
waves decrease in strength very rapidly. Still, if a relatively strong gravita-
tional wave were to go by, say, a metal rod, its shape would be deformed
by being stretched and lengthened periodically for a certain time. By ac-
curately measuring the length of rods we can hope to detect these changes.
 27

The technical problems, however, are enormous: the expected variation is

of a fraction of the size of an atom! Nonetheless experiments are under way.
Gravitational waves are generated appreciably only in the most violent
of cosmic events. During the last stages in the life of a star heavier than 3
solar masses, most of the stellar material collapses violently and inexorably
to form a black hole (n the rubber sheet picture this corresponds to drop-
ping a very small and very heavy object on the sheet). The corresponding
deformation of space travels forth from this site site as a gravitational wave.
High intensity gravitational waves are also produced during the collision of
two black holes or any sufficiently massive compact objects.

7.11 Summary.
The conclusions to be drawn from all these arguments are,

• All frames of reference are equivalent, provided we are willing to in-

clude possible gravitational effects (in non-inertial or accelerated frames
forces will appear which are indistinguishable from gravitational forces).

• Space-time is a dynamic object: matter curves it, and the way in which
it is curved determines the motion of matter in it. Since all bodies are
affected in the same way by the curvature of space and time the effects
of gravity are independent of the nature of the body. Changes in the
distribution of matter change space-time deforming it, and, in some
instances, making it oscillate.

7.12 Tests of general relativity.

After Einstein first published the General Theory of Relativity there was a
very strong drive to test its consequences; Einstein himself used his equa-
tions to explained a tiny discrepancy in the motion of Mercury. Yet he most
dramatic effect was the shifting of the positions of the stars (see below).
Since 1916 there have been many measurements which agree with the Gen-
eral Theory of Relativity to the available accuracy. Here I will concentrate
on the “classical” tests of the thoery.

7.12.1 Precession of the perihelion of Mercury

A long-standing problem in the study of the Solar System was that the orbit
of Mercury did not behave as required by Newton’s equations.
28

To understand what the problem is let me describe the way Mercury’s

orbit looks. As it orbits the Sun, this planet follows an ellipse...but only ap-
proximately: it is found that the point of closest approach of Mercury to the
sun does not always occur at the same place but that it slowly moves around
the sun (see Fig. 7.20). This rotation of the orbit is called a precession.
The precession of the orbit is not peculiar to Mercury, all the plane-
tary orbits precess. In fact, Newton’s theory predicts these effects, as be-
ing produced by the pull of the planets on one another. The question is
whether Newton’s predictions agree with the amount an orbit precesses; it
is not enough to understand qualitatively what is the origin of an effect,
such arguments must be backed by hard numbers to give them credence.
The precession of the orbits of all planets except for Mercury’s can, in fact,
be understood using Newton;s equations. But Mercury seemed to be an
exception.

Figure 7.20: Artist’s version of the precession of Mercury’s orbit. Most of

the effect is due to the pull from the other planets but there is a measurable
effect due to the corrections to Newton’s theory predicted by the General
Theory of Relativity.

As seen from Earth the precession of Mercury’s orbit is measured to

be 5600 seconds of arc per century (one second of arc= 1/3600 degrees).
Newton’s equations, taking into account all the effects from the other planets
(as well as a very slight deformation of the sun due to its rotation) and the
fact that the Earth is not an inertial frame of reference, predicts a precession
of 5557 seconds of arc per century. There is a discrepancy of 43 seconds of
 29

arc per century.

This discrepancy cannot be accounted for using Newton’s formalism.
Many ad-hoc fixes were devised (such as assuming there was a certain
amount of dust between the Sun and Mercury) but none were consistent
with other observations (for example, no evidence of dust was found when
the region between Mercury and the Sun was carefully scrutinized). In con-
trast, Einstein was able to predict, without any adjustments whatsoever,
that the orbit of Mercury should precess by an extra 43 seconds of arc per
century should the General Theory of Relativity be correct.

7.12.2 Gravitational red-shift.

We saw in Sec. 7.4 that light leaving a region where the gravitational force
is large will be shifted towards the red (its wavelength increases; see Figs.
7.21,7.9); similarly, light falling into a region where the gravitational pull
is larger will be shifted towards the blue. This prediction was tested in
Harvard by looking at light as it fell from a tower (an experiment requiring
enormous precision since the changes in the gravitational force from the
top to the bottom of a tower are minute) and the results agree with the
predictions from the General Theory of Relativity.
The gravitational red-shift was also tested by looking at the light from
a type of stars which are very very well-studied. The observations showed
that the light received on Earth was slightly redder than expected and that
the reddening is also in agreement with the predictions from the General
Theory of Relativity.

Figure 7.21: Illustration of the gravitational red-shift predicted by the Gen-

eral Theory of Relativity. A heavy object is denoted by a deformation of
space represented by the funnel. As light leaves the vicinity of this object
it is shifted towards the red: for a sufficiently compact and massive object
a blue laser on the surface will be seen as red in outer space.
30

7.12.3 Light bending

If we imagine observing a beam of light in an accelerated elevator we will see

that the light path is curved. By the equivalence principle the same must
be true for light whenever gravitational forces are present. This was tested
by carefully recording the position of stars near the rim of the sun during
an eclipse (see Fig. 7.23) and then observing the same stars half a year later
when there is no eclipse.
During the eclipse the observed starlight reaches us only after passing
through a region where gravitational effects from the sun are very strong
(that is why only stars near the rim are used), but the observation half a
year later are done at a time where the gravitational effects of the sun on
starlight is negligible.
It is found that the position of the stars are displaced when photographs
of both situations are compared (see Fig. 7.23). The deviations are the same
as the ones predicted by General Relativity. Eddington first observed this
effect in 1919 during a solar eclipse. The early 20th century telegram (see
Fig. ??) announcing this observation for the frist time marks the change in
our views about the structure of space time.

Figure 7.22: Illustration of the effects of the gravitational bending of light:

during an eclipse the observed positions of the stars will be shifted away
from the Sun.
 31

Figure 7.23: Eddington’s telegram to Einstein announcing the observation

of the bengin of light by a gravitational force as predicted by the General
Theory of Relativity.

7.12.4 The double pulsar

There are certain kind of stars which are called pulsars (see Sect. ??).
These are very compact objects (they have a diameter of about 10km but
are several times heavier than the sun) which emit radio pulses at very
regular intervals.
In the early 80’s, Taylor and Hules (recent Nobel prize winners for this
work) discovered a system where one pulsar circles another compact object.
Because the pulsar pulses occur at very regular intervals, they can be used
as a clock. Moreover there are several physical effects which can be used to
determine the shape of the orbits of the pulsar and the compact object. It
was found that these objects are slowly spiraling into each other, indicating
that the system is losing energy in some way.
This system can also be studied using the General Theory of Relativity
which predicts that the system should radiate gravitational waves carrying
energy with them and producing the observed changes. These predictions
are in perfect agreement with the observations. This is the first test of
General Theory of Relativity using objects outside our solar system.