We clarify several issues involving the concepts of time-reversal invariance and time asymmetry in classical electrodynamics. Specifically, we consider three questions: (I) If electrodynamics is time-reversal invariant, why are the radiative processes that occur in nature time asymmetric? (II) Why doesn't the time-reversal invariance of electrodynamics contradict the fact that a charged particle in motion feels a radiation damping force? (III) Why don't the advanced solutions to Maxwell's equations occur in nature—is there some principle that forbids them? We argue that these questions are not specific to electrodynamics, but also arise in many other systems in which waves are radiated, and answer them in the context of a simple model describing a coupled particle-field system in (1+1) dimensions.

1.
The Abraham-Lorentz formula is derived and discussed in
D. J.
Griffiths
,
Introduction to Electrodynamics
, 2nd ed. (
Prentice-Hall
,
New Jersey
,
1989
), Sec. 9.3.
2.
The claim that the time asymmetry of radiative processes is due to initial conditions has been defended by many authors; see, for example,
A.
Einstein
, A.,
The Collected Papers of Albert Einstein, Volume 2, The Swiss Years: Writings 1900–1909
, translated by
A.
Beck
(
Princeton University Press
,
Princeton
,
1989
), p.
376
;
K.
Popper
, “
The arrow of time
,”
Nature
177
,
538
(
1956
);
and
H. D.
Zeh
,
The Physical Basis of the Direction of Time
(
Springer
,
Berlin
,
2010
), Sec. 2.
3.
Although it is generally accepted that the equation of motion for a charged point particle including radiation damping is not time-reversal invariant, this claim has been disputed by F. Rohrlich; see, for example,
F.
Rohrlich
, “
The arrow of time in the equations of motion
,”
Found. Phys.
28
(
7
),
1045
1056
(
1998
);
F.
Rohrlich
, “
Causality and the arrow of classical time
,”
Stud. Hist. Phil. Mod. Phys.
31
(
1
),
1
13
, (
2000
);
F.
Rohrlich
, “
Time reversal invariance and the arrow of time in classical electrodynamics
,”
Phys. Rev. E
72
,
057601
(
2005
);
F.
Rohrlich
, “
Time in classical electrodynamics
,”
Am. J. Phys.
74
(
4
),
313
315
(
2006
);
and
F.
Rohrlich
,
Classical Charged Particles
, 3rd ed. (
World Scientific
,
New Jersey
,
2007
),
Sec. 9.2. Rohrlich's views are criticized in
H. D.
Zeh
, “
Note on the time reversal asymmetry of equations of motion
,”
Found. Phys. Lett.
12
(
2
),
193
196
(
1999
);
and
C.
Rovelli
, “
Comment on: ‘Causality and the arrow of classical time,’ by Fritz Rohrlich
,”
Stud. Hist. Phil. Mod. Phys.
35
(
3
),
397
405
(
2004
).
4.
This example is motivated by the discussion of the time asymmetry of water waves presented in
K.
Popper
, Ref. 2.
5.
For a detailed discussion of the water wave-buoy system, see
Y. H.
Zhenga
,
Y. G.
Youa
, and
Y. M.
Shenb
, “
On the radiation and diffraction of water waves by a rectangular buoy
,”
Ocean Eng.
31
,
1063
1082
(
2004
) and the references within.
6.
Mass renormalization is discussed in
A. O.
Barut
,
Electrodynamics and Classical Theory of Fields and Particles
(
Dover
,
New York
,
1980
), pp.
190
195
.
7.
The runaway and preaccelerated solutions are discussed in
J. D.
Jackson
,
Classical Electrodynamics
, 2nd ed. (
Wiley
,
New York
,
1975
), Sec. 17.6, and Ref. 1, pp.
435
436
.
8.

The model was originally proposed in Ref. 9, which describes its use as a toy model of classical electrodynamics and provides detailed derivations of many of its properties; topics discussed include the general solution to the initial value problem for the model, derivations of the retarded and advanced fields, conservation laws, and the similarities and differences between the model and ordinary electrodynamics. In Sec. IV we describe the retarded and advanced formulations of the model; further discussion of these formulations is given in Ref. 10. The model can be easily simulated on a computer; a simulation algorithm and the results of simulations for several example radiation processes can be found in Ref. 11.

9.
A. D.
Boozer
, “
A toy model of electrodynamics in (1+1) dimensions
,”
Eur. J. Phys.
28
,
447
464
(
2007
).
10.
A. D.
Boozer
, “
Retarded potentials and the radiative arrow of time
,”
Eur. J. Phys.
28
,
1131
1143
(
2007
).
11.
A. D.
Boozer
, “
Simulating a toy model of electrodynamics in (1+1) dimensions
,”
Am. J. Phys.
77
(
3
),
262
269
(
2009
).
12.

The tensor ϵμν is the Levi-Civita tensor in (1+1) dimensions, defined such that ϵ10=ϵ01=1 and ϵ00=ϵ11=0.

13.

In what follows we will often use the notation Fμ=(B,E) as a convenient way of grouping the E and B fields into a single quantity.

14.

This lack of spacetime symmetry is a generic feature of systems in which nonrelativistic radiators are coupled to wave fields. Such systems are not Lorentz invariant, because the radiators obey nonrelativistic dynamics, and are not Galilean invariant, because the wave equation is not invariant under Galilean transformations.

15.

The calculation is described in detail in Ref. 9; it proceeds analogously to the corresponding calculation for electrodynamics that is presented in J. D. Jackson, Ref. 7, Sec. 14.1.

16.

The damping force can be viewed as a self-force arising from the coupling of the particle to its own retarded field, as given by Eq. (13). Since the retarded field at time t is determined by the particle velocity at time t, the damping force also has this property.

17.

From the form of the damping force we see explicitly that there is a preferred reference frame in which the equations of motion for the model are valid: A particle in motion feels a drag force that ultimately brings it to rest relative to this preferred frame.

18.
Note that the time-reversal transformation of the E and B fields is the same as the time-reversal transformation of the electric and magnetic fields of electrodynamics; see, for example,
J. D.
Jackson
, Ref. 7, p. 249.
19.

We assume that the amplitude A lies in the range (1/2)(e/γ)<A<(3/2)(e/γ), so |v0|<1.

20.

In general, one can show that the in and out fields for processes A and B are related by EB,in(x, t) = EA,out(x, −t) and BB,in(x, t) = −BA,out(x, −t), so if the in fields are simple for process A then the out fields are simple for process B (and vice-versa).

AAPT members receive access to the American Journal of Physics and The Physics Teacher as a member benefit. To learn more about this member benefit and becoming an AAPT member, visit the Joining AAPT page.