This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:
This content was downloaded by: aarathikrishnag
IP Address: 14.139.174.34
This content was downloaded on 21/01/2019 at 17:04

Please note that terms and conditions apply.

You may also be interested in:

Of Clocks and Time: Space and time forever entwined

L Hüwel

From Newton to Einstein: Special relativity

F T Baker

A Concise Introduction to Quantum Mechanics: Wave mechanics

M S Swanson

An Introduction to the Physics of Nuclear Medicine: Radiation interactions with matter

L Harkness-Brennan

Fourier Transform and Its Applications Using Microsoft EXCEL®

: The
S ChoFourier transform

Stacking the nines: relativistic steps to the stars

James R Claycomb

The relativistic wave vector

Jens Madsen Houlrik

Tippe top paradox in relativity

Aniket Basu, R S Saraswat, Kedar B Khare et al.

Elementary analysis of the special relativistic combination

Kane O’Donnell and Matt Visser
 IOP Concise Physics

General Relativity: An Introduction to Black Holes, Gravitational

Waves, and Cosmology
Michael J W Hall

Chapter 1
Concepts in special relativity

Special relativity is a fascinating and important subject. Its main concepts are
recalled in this book only to the extent needed to formulate and understand the basic
elements of general relativity. However, there are many textbooks which explore
further details of special relativity, of which one by Rindler is particularly
recommended for further reading [1].

1.1 Galilean relativity

In 1632 a brilliant Italian physicist, Galileo Galilei, published a book that was to end up
getting him into a lot of trouble. It was called Dialogue Concerning the Two Chief World
Systems [2], and gave strong arguments for the consistency of the Copernican (Sun-
centred) model of the solar system. The arguments also supported the superiority of this
model over the Ptolemaic (Earth-centred) model, which is what caused the later trouble.
At that time, one of the arguments against the Copernican model was that if the
Earth is rotating to the east, then objects that are thrown or dropped will tend to fall
behind, to the west. For example, a stone dropped from the mast of a stationary ship
should land westward of the mast. Galileo had already contented himself by
experiment that this was not so—but also that, contrary to the very basis of the
argument, that even if the ship was sailing, no matter in which direction, the stone
always landed at the foot of the mast. The motion of a (smoothly) moving ship could
not be detected in such a way! In his book he further developed this idea [2]:

Shut yourself up with some friend in the main cabin below decks on some large
ship and have with you there some flies, butterflies, and other small flying
animals. Have a large bowl of water with some fish in it; hang up a bottle that
empties drop by drop into a narrow vessel beneath it. With the ship standing
still, observe carefully how the little animals fly with equal speed to all sides of

doi:10.1088/978-1-6817-4885-6ch1 1-1 ª Morgan & Claypool Publishers 2018

 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

the cabin. The fish swim indifferently in all directions; the drops fall into the
vessel beneath; and, in throwing something to your friend, you need throw it
no more strongly in one direction than another, the distances being equal;
jumping with your feet together, you pass equal spaces in every direction.
When you have observed all these things carefully (though there is no doubt
that when the ship is standing still everything must happen in this way), have
the ship proceed with any speed you like, so long as the motion is uniform and
not fluctuating this way and that. You will discover not the least change in all
the effects named, nor could you tell from any of them whether the ship was
moving or standing still [emphasis added].
In jumping, you will pass on the floor the same spaces as before, nor will
you make larger jumps toward the stern than toward the prow even though the
ship is moving quite rapidly, despite the fact that during the time that you are
in the air the floor under you will be going in a direction opposite to your
jump. In throwing something to your companion, you will need no more force
to get it to him whether he is in the direction of the bow or the stern, with
yourself situated opposite. The droplets will fall as before into the vessel
beneath without dropping toward the stern, although while the drops are in the
air the ship runs many spans. The fish in their water will swim toward the front
of their bowl with no more effort than toward the back, and will go with equal
ease to bait placed anywhere around the edges of the bowl. Finally the
butterflies and flies will continue their flights indifferently toward every side,
nor will it ever happen that they are concentrated toward the stern, as if tired
out from keeping up with the course of the ship, from which they will have
been separated during long intervals by keeping themselves in the air.

The idea in the middle paragraph above (in italics), illustrated in figure 1.1, is that
the laws of physics are the same in frames of reference moving at constant velocities

Figure 1.1. Galileo’s thought experiment: an observer below deck cannot determine, by local experiments,
whether the ship is motionless on the water or sailing smoothly in some direction.

1-2
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

with respect to one another. This is now called Galilean relativity. It may be
sharpened in Newtonian mechanics to invariance of the laws of motion under any
coordinate transformation of the form
x′ = Rx − vt + a ,
(1.1)
t′ = t + b
for any rotation R, velocity boost v, and translations a and b. For example,
Newton’s universal law of gravitation,
GmM(x − X )
mx¨ = − , (1.2)
x−X3
takes precisely the same form in the transformed coordinates:
GmM(x′ − X ′)
mx¨′ = − , (1.3)
x′ − X ′ 3
as may easily be checked.
The above ‘Galilean’ transformations form a ten-parameter group. Note that
they separately preserve spatial and temporal distances, i.e., writing x ≡ (x , y, z )
and X ≡ (X , Y , Z ),
(x′ − X ′)2 + (y′ − Y ′)2 + (z′ − Z′)2 = (x − X )2 + (y − Y )2 + (z − Z )2 ,
(1.4)
t′ − T ′ = t − T .

1.2 Inertial frames

For Galileo’s principle of relativity to apply to Newtonian mechanics, one has to
begin with a coordinate system in which Newtonian mechanics is itself valid. In
particular, Newton’s first law of motion, that free particles remain at rest or move in
a straight line, must hold. That is, a free particle trajectory must satisfy
d 2x
= 0.
dt 2
This would not be the case, for example, if the coordinate system was fixed relative to
a spinning roundabout (try rolling a ball to a friend while on one of these).
Newton’s first law of motion is often called the law of inertia, and hence
coordinate systems which respect it are called ‘inertial frames of reference’, or,
more simply, inertial frames. Thus, in any inertial frame free particles move in
straight lines. Galilean relativity may be broadly stated in the form that the laws of
physics have the same form in all inertial frames.
Note that Newton did not regard inertial frames as fundamental. He believed that
space and time are ‘absolute’, with each point in space and time having meaningful
fixed coordinate values. He considered the use of relative coordinates, defined with
respect to some arbitrary spatial and temporal origin, to be a mere convenience.

1-3
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

Galilean relativity is not incompatible with this belief, but it does imply that no
experiment can distinguish such an ‘absolute’ frame of reference from any other
inertial frame.
For this reason, Galileo’s principle of relativity was put in doubt by early
interpretations of James Clerk Maxwell’s equations for the electromagnetic field,
which postulated an ‘ether’ through which electromagnetic waves propagated,
analogously to the propagation of sound waves through air. The ether was thought
to be at rest with respect to absolute space. Surprisingly, however, all experiments
made to try and detect the motion of the Earth through this ether—most notably the
Michelson–Morley experiment—failed, consistent with Galilean relativity.
Nevertheless, Maxwell’s equations still raised a difficulty for Galilean relativity,
at least when embedded into Newtonian mechanics. For example, the equations
predicted (consistently with the Michelson–Morley experiment) that the speed of
light in free space is a fixed constant in all inertial frames, despite the light source
(e.g., a lamp) having different speeds with respect to different frames. This cannot be
reconciled with the form of the Galilean transformations in equation (1.1), under
which Newtonian mechanics is invariant. It was this incompatibility that led Albert
Einstein to his ‘Special Theory of Relativity’, published in 1905 [3, 4].

1.3 Special relativity

Postulates
The basic elements of Einstein’s special relativity can be formulated as follows.
First, inertial frames are defined via the motion of free particles:

Postulate 1: Free particles move in straight lines in inertial frames.

Thus, an inertial frame is a spacetime coordinate system in which Newton’s first law
of motion is valid (the other laws need to be slightly modified).
Second, motivated by the prediction of Maxwell’s equations in the previous
section, it is further assumed that:

Postulate 2: The speed of light in a vacuum is a constant, c, in all inertial frames,

when measured using a standard set of clocks and rulers at rest in each frame.

Standard time and length standards are required because otherwise the speed of light
could be numerically different in different directions, even in the same inertial frame—
e.g., if coordinates were measured in the x-direction and the y-direction using rulers
having different units. Postulate 2 implies that only agreement on a standard clock is
actually needed, as the spatial distance between two points at rest in a given inertial
frame can then be defined via the time for light to propagate between them.
The first and second postulates still allow the possibility that standard length and
time scales can only be defined up to a common scale factor, since straight-line
motion and the speed of light are invariant under such a rescaling. If this were so,
then it would not be physically meaningful, for example, to compare the rates of

1-4
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

clocks at rest in different inertial frames, as there would be no ‘absolute’ rate even for
clocks at rest. There are various choices for removing this possibility, a suitable one
being:

Postulate 3: If a standard clock at rest in a first inertial frame moves at velocity v

with respect to a second inertial frame, then a standard clock at rest in the second
frame moves at velocity −v with respect to the first frame, and the moving clock
in each frame ticks at the same rate.

It could alternatively be required that the volume of a given spacetime region is the
same for all observers, or that the laws of physics are not scale-invariant (implying
the existence of a standard time scale for each inertial frame), as discussed in
appendix A.
Finally, we cannot in fact fully adopt the broad notion of Galilean relativity in
section 1.2, that the laws of physics have the same form in all inertial frames. A
postulate is needed to rule out invariance of physical laws between inertial frames
related by transformations such as spatial reflections and reversing the direction of
time (under which the laws of physics are known not to be invariant, from particle
decay experiments), or related by swapping the time coordinate with a spatial
coordinate (which converts subluminal motion to superluminal motion and vice
versa):

Postulate 4: The laws of physics are invariant in inertial frames connected by a

continuous sequence of intermediate coordinate transformations.

In the language of group theory, this implies that the physically relevant group of
transformations between inertial frames is continuously connected to the identity. It
makes plausible the idea that systems and observers can, by continuous operations,
move from being at rest in a first inertial frame to being at rest in a second inertial
frame, with no change in the laws of physics.

The invariant spacetime interval

The above four postulates are sufficient to derive the group of spacetime coordinate
transformations that link inertial frames (see appendix A). These are called (proper
orthochronous) Lorentz transformations, after their first discoverer, Hendrik
Lorentz [4], and have some significant differences from the Galilean transformations
in (1.1).
The derivation in appendix A shows, in particular, that transformations between
inertial frames preserve the quantity
c 2(Δτ )2 ≔ c 2(Δt )2 − (Δx )2 − (Δy )2 − (Δz )2 (1.5)
linking any two points (t , x , y , z ) and (t + Δt , x + Δx , y + Δy , z + Δz ) in space-
time, where c is the invariant speed of light in vacuum. This may be directly
compared with (1.4) for Galilean transformations, under which the spatial and

1-5
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

temporal distances are separately preserved. In special relativity only the above
combination of the spatial and temporal distances is invariant, and is called the
invariant spacetime interval. The two points are said to be time-like separated, light-
like separated, and space-like separated for values of (Δτ )2 that are, respectively,
positive, zero, and negative (see figure 1.2). Note that Δτ has units of time.
Any particle travelling at the speed of light in some inertial frame has Δτ = 0
from (1.5), and hence must travel at the speed of light in all inertial frames. Thus the
lightcones in figure 1.2 are invariant.
Further, if (t , x , y , z ) and (t + Δt , x + Δx , y + Δy , z + Δz ) refer to successive
spacetime locations of a free particle, then in the rest frame of the particle one has
Δx = Δy = Δz = 0, and so
(Δτ )2 = (Δt )2 > 0.
Thus, a free particle follows a time-like trajectory, and the duration Δτ is equal to the
elapsed time experienced by the particle in moving between the two spacetime
locations, relative to its own rest frame. Since Δτ has the same value in all inertial
frames, it has the same physical meaning in all inertial frames, and is called the

light signal (lightlike) particle trajectory (timelike)

spacelike curve

Figure 1.2. Spacetime diagrams and causal structure. Each point in spacetime has an invariant past and future
lightcone, defined by the regions from which light signals can be directly received and transmitted. The
interiors of the lightcones define the absolute past and absolute future of each point. Particle trajectories are
time-like, and hence move from the absolute past to the absolute future of any given point. All events outside
the lightcones of a point are space-like separated from the point, and so no (subluminal) signal or particle can
be sent or received from them. In the Galilean limit c → ∞ the past and future lightcones of a given point (t ,x,
y, z) flatten and merge into a single hyperplane, corresponding to the set of spatial positions at a fixed time t.

1-6
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

invariant or proper time. Note, it also follows that any particle that has a rest frame
cannot move faster than the speed of light, as this would require (Δτ )2 to change,
from a positive value to a negative value.
Finally, if (t , x , y , z ) and (t + Δt , x + Δx , y + Δy , z + Δz ) refer to the two
endpoints of a body such as a ruler, then in the rest frame of the body one has Δt = 0
at any given time. Hence (Δτ )2 < 0 from (1.5) (implying Δτ is imaginary). Hence the
endpoints of a body at rest are space-like separated, and the invariant interval
c 2(Δτ )2 = −(Δx )2 − (Δy )2 − (Δz )2 ≕ − D 2
defines the distance D between these endpoints in its rest frame, called the proper
distance or rest length.

Lorentz transformations
General Lorentz transformations are those coordinate transformations between
inertial frames that preserve the invariant interval c 2(Δτ )2 in (1.5). They have the
linear form (see appendix A)
⎛ct′⎞ ⎛ct ⎞
⎜ x′ ⎟ ⎜x⎟
⎜ ⎟ = L⎜ ⎟ + a , LT GL = G , (1.6)
⎜ y′ ⎟ y
⎜ ⎟
⎝ z′ ⎠ ⎝z⎠

for some 4 × 4 matrix L and spacetime translation vector a, where G denotes the 4 ×
4 matrix G ≔ diag[1, −1, −1, −1]. The second equation ensures that c 2(Δτ )2 has the
same value in all inertial frames. Postulate 4 above further requires that det L = 1
and L 00 ⩾ 1, corresponding to the subgroup of proper orthochronous Lorentz
transformations (see appendix A).
The proper orthochronous Lorentz transformations form a ten-parameter group,
similarly to the Galilean transformations in equation (1.1). Six of these parameters
correspond to spatial rotations and translations, and one to time translations, just as
for the Galilean group. However, the transformations between two frames having a
nonzero relative velocity are different in the two cases (as they must be, for the speed
of light to be invariant). In particular, a velocity boost v in the x-direction
corresponds to the Lorentz transformation
⎛ γ − γv c 0 0 ⎞
⎜ ⎟
L = ⎜− γv c γ 0 0⎟, a = 0, (1.7)
⎜⎜ 0 0 1 0 ⎟⎟
⎝ 0 0 0 1⎠
where
1
γ≔ ⩾ 1. (1.8)
1 − v2 c2

1-7
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

Thus, the x and t coordinates transform as

x′ = γ (x − vt ) , t′ = γ (t − vx c 2 ) , (1.9)
which reduces to the corresponding Galilean boost x′ = x − vt , t′ = t in the limit
v /c → 0. Hence special relativity is well approximated by Galilean relativity for
frames having small relative speeds. It is easy to check that LT GL = G , as required
by (1.6). The generalisation to an arbitrary velocity boost is considered in question
1.3 at the end of this chapter.

1.4 Velocity addition, length contraction, and time dilation

Velocity addition
To check that the boost in (1.9) does indeed correspond to two frames with relative
velocity v in the x-direction, consider an infinitesimal portion of a spacetime trajectory
between ξ and ξ + dξ , where ξ denotes the spacetime vector (ct , x , y, z ). Now, taking
the inverse of the transformation in (1.9) (equivalent to replacing v by −v ), one has
dx = γ (dx′ + vdt′) , dt = γ (dt′ + vdx′ c 2 ).
Hence, if the trajectory velocity in the two frames is denoted by u = dx /dt and
u′ = dx′/dt′, respectively, it follows that
dx γ (dx′ + vdt′) dx′ dt′ + v
u= = 2
=
dt γ (dt′ + vdx′ c ) 1 + v(dx′ dt′) c 2
u′ + v (1.10)
= u′v
.
1+ 2
c

Hence, a particle at rest in the primed coordinates, i.e., with u′ = 0, has velocity u =
v in the unprimed coordinates. That is, the primed frame moves at velocity v with
respect to the unprimed frame. Note also that a particle at rest in the unprimed
coordinates, i.e., with u = 0, has velocity u′ = −v in the primed coordinates, as
required by postulate 3 in section 1.3.
The velocity addition formula in (1.10) may further be used to check the
invariance of the speed of light, as required by postulate 2. In particular, taking
u′ = c immediately gives u = c also, independently of the relative velocity v.

Length contraction
If x1′ and x2′ denote the endpoints of a ruler in its rest frame, then the distance
between them, is given by D0 = ∣x1′ − x2′∣. As noted previously, this is called the
‘proper distance’ or ‘rest length’. Further, in a frame where the ruler is moving at
speed v in the x-direction, let x1 and x2 be the positions of the endpoints at equal
times t1 = t2 = t . Equation (1.9) then yields
D0 = x1′ − x2′ = γ (x1 − vt1) − (x2 − vt2 ) = γ x1 − x2 .
Hence, defining the ‘length’ D of the ruler in a given frame to be the distance
between its endpoints at equal times, one has

1-8
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

D = γ −1D0 = D0 1 − v 2 c 2 . (1.11)

Since D < D0 for v ≠ 0, this is referred to as length contraction.

Note that length contraction should not be confused with the apparent length of a
moving rod. This is because one does not actually ‘see’ the endpoints simultaneously
in general, as light will take different times to reach the eye from each endpoint if
they are located at different distances. It may be shown, for example, that the outline
of a moving sphere is circular for any inertial observer, rather than contracted in the
direction of relative motion [1]. In relativity, what you get is not necessarily what
you see! Relativistic optics will not be reviewed further here.

Time dilation
If t1′ and t2′ denote the times of successive ticks of a clock in its rest frame, then its rest
period is given by T0 = t2′ − t1′. In a frame where the clock is moving at speed v in the
x-direction, the ticks will be separated by T = t2 − t1, and by a spatial distance
x2 − x1 = v(t2 − t1). Hence, using (1.9), one has
T0 = t2′ − t1′ = γ [(t2 − t1) − v(x2 − x1) c 2 ] = γ (t2 − t1) (1 − v 2 c 2 ) = γ −1T .
The period relative to the moving frame is therefore given by

T = γT0 = T0 1 − v2 c2 . (1.12)

Since T > T0 for v ≠ 0, this is referred to as time dilation. That is, moving clocks tick
slowly.
Like length contraction, time dilation should not be confused with the apparent
period of a moving clock. In particular, since the clock will be at different distances
between successive ticks, these ticks will take different lengths of time to arrive to an
observer at rest. This leads to the relativistic Doppler effect, discussed in chapter 2.

1-9
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

1.5 Questions
Readers are encouraged to attempt questions marked with an asterisk (*), both to
develop understanding and because they are typically relevant to later chapters.
Other questions are aimed at those wishing to explore more deeply.

Question 1.1* Bert and Ernie started out at the spacetime point (0, 0, 0, 0) in some
inertial frame. Bert moved in a straight line to (5, 3, 0, 0), while Ernie first moved to
(3, 2, 0, 0) and then on to (5, 3, 0, 0). Who had aged the most when they met up
again? (units are chosen such that c = 1).
Hint: calculate the elapsed proper times via (1.5).

Question 1.2* Show that the defining property of Lorentz transformations in (1.6),
i.e., LT GL = G for G = diag[1, −1, −1, −1], ensures that the interval (Δτ )2 in (1.5)
has the same value in all inertial frames.

Question 1.3 (General form of Lorentz boost)

Check that the Lorentz transformation in (1.7), corresponding to a velocity boost
v = (v, 0, 0) in the x-direction, can be generalised to the block-matrix form
⎛ γ ⎞
⎜ γ − vT ⎟
c
L (v ) = ⎜ ⎟, (1.13)
⎜ γ γ − 1 T⎟
⎜− v I + vv ⎟
⎝ c v2 ⎠

where I is the 3 × 3 identity matrix. Show for a general 3-vector v, with ∣v∣ = v < c ,
that a particle at rest in the unprimed frame (i.e., with Δx = 0) has velocity −v in the
primed frame. Check that L−1(v ) = L( −v ). How are these properties related to
postulate 3 in section 1.3?

Question 1.4 (Topological twin paradox)

Note that there is no paradox arising from question 1.1 above: Bert behaves as a free
particle, moving in a single straight line (relative to any inertial frame), whereas
Ernie changes direction and speed (relative to any inertial frame). Thus there is a
fundamental physical asymmetry in their motions, which underlies the different
amounts by which they age.
Consider, however, the case of a cylindrical spacetime, with one spatial dimension
that is periodic in some inertial frame, so that x is identified with x + L. Suppose Bert
and Ernie again start from spacetime point (0, 0, 0, 0), but with Bert remaining at
rest in this frame and Ernie moving in the x-direction at speed v. Thus both move in
a straight line, and the physical asymmetry of question 1.1 above is no longer
present. Indeed, from Ernie’s point of view, he is at rest and Bert is moving in the

1-10
 General Relativity: An Introduction to Black Holes, Gravitational Waves, and Cosmology

x-direction with speed −v . Check that, nevertheless, Bert will have aged the most
when they meet again at time t = L /v . Is this a true twin paradox?
[The resolution within special relativity is discussed by Dray [5]].

Question 1.5 An arrow having rest length A0 is shot horizontally, and flies at
constant speed into the open end of a cylinder having rest length C0 < A0.
(i) When the tip of the arrow reaches the other end of the cylinder, is the arrow
wholly within the cylinder relative to:
(a) the rest frame of the arrow?
(b) the rest frame of the cylinder?
Is this paradoxical?
(ii) If the far end of the cylinder is closed by an immovable solid wall, that
prevents the arrow from travelling further, at what minimum speed must the
arrow travel at to guarantee that it is enclosed by the cylinder relative to
both rest frames?
Hint: the feathered end of the arrow cannot be affected by the presence of
the wall prior to some physical signal or shockwave, generated by the tip
hitting the wall, being transmitted at light speed or less along the length of
the arrow.

References
[1] Rindler W 1991 Introduction to Special Relativity (Oxford: Clarendon Press)
[2] Galilei G 1967 Dialogue Concerning the Two Chief World Systems, Ptolemaic And Copernican
(Berkeley, CA: University of California Press)
[3] Einstein A 1905 Zur Elektrodynamik bewegter Körper Ann. Phys. 322 891–921
[4] Lorentz H A, Einstein A, Minkowski H, Weyl H and Sommerfeld A 1952 The Principle of
Relativity: A Collection of Original Memoirs on the Special and General Theory of Relativity
(New York: Dover) This book contains translations of a number of important papers in
relativity theory originally published in German
[5] Dray T 1990 The twin paradox revisited Am. J. Phys. 58 822–5

1-11