Estimating hc/k from sunlight

Peter Pesic

Citation: American Journal of Physics 73, 457 (2005); doi: 10.1119/1.1806479

View online: http://dx.doi.org/10.1119/1.1806479
View Table of Contents: http://scitation.aip.org/content/aapt/journal/ajp/73/5?ver=pdfcov
Published by the American Association of Physics Teachers

Articles you may be interested in

The effect of broad-band Alfvén-cyclotron waves spectra on the preferential heating and differential acceleration
of He ++ ions in the solar wind
AIP Conf. Proc. 1539, 34 (2013); 10.1063/1.4810983

Study on the use of TiO 2 passivation layer to reduce recombination losses in dye sensitized solar cells
AIP Conf. Proc. 1482, 112 (2012); 10.1063/1.4757448

Using eclipse observations to search large-scale velocity field in the range 2–4 R .
AIP Conf. Proc. 471, 753 (1999); 10.1063/1.58802

Commercial GaSb cell and circuit development for the Midnight Sun ® TPV stove
AIP Conf. Proc. 460, 480 (1999); 10.1063/1.57830

A new radiative heater for high T c thin film growth

Rev. Sci. Instrum. 69, 3326 (1998); 10.1063/1.1149097

This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
 NOTES AND DISCUSSIONS

Conventions and inertial reference frames

Alberto A. Martı́neza)
California Institute of Technology, Pasadena, California 91125
共Received 10 May 2004; accepted 21 December 2004兲
This article discusses the role of conventions in defining the concept of inertial reference frame, and
it specifies key historical evidence, up to now widely ignored, connecting Poincaré, Einstein, and
Reichenbach’s analyses of simultaneity. © 2005 American Association of Physics Teachers.
关DOI: 10.1119/1.1858446兴

In special relativity, conventions have been a subject of
debate among physicists and philosophers. We may better
understand such disputes on synchrony, velocity, and inertial
systems by setting them in relation to specific historical tra-
ditions.
Recently, Ohanian argued that dynamical considerations,
applied to inertial systems, necessarily entail the standard
synchronization rule employed in special relativity.1 He de-
fines ‘‘inertial reference frame’’ as a frame in which New-
ton’s laws of motion are valid to a first approximation. He
cites only one, seemingly authoritative, formulation of this
definition. He claims that Einstein, in his famous 1905 paper,
required ‘‘a system of co-ordinates in which the equations of
Newtonian mechanics hold good, i.e., to the first
approximation.’’ 2 That expression, however, is not in the
1905 paper. It actually reads, ‘‘Consider a coordinate system
in which the equations of Newtonian mechanics hold.’’ 3
Ohanian relied on a translation that includes the clause ‘‘i.e.,
to the first approximation’’ as a footnote.4 There is no evi-
dence that Einstein added it. Such footnotes originated in an
edition of 1913 by Otto Blumenthal, which includes notes by
Sommerfeld.5 Presumably, this specific footnote was added
to harmonize the pertinent sentence with special relativity,
which revises many equations of Newton’s mechanics.
Hence, Torretti described Einstein’s original expression as ‘‘a
condition blatantly at odds with the subsequent development
of the paper.’’ 6
At the outset of his argument, Einstein assumed the valid-
ity of Newton’s mechanics before showing its inadequacy.
共This point has been ignored in analyses of the 1905 paper.兲
In particular, in discussing the lack of any absolute meaning
of simultaneity, Einstein assumed that moving clocks keep
the time of the ‘‘stationary’’ system.7 Accordingly, the con-
sideration of a system in which Newton’s equations hold
served to indicate a deficiency in Newtonian mechanics.
Newton’s equations, of course, are valid as a first approxi-
mation in Einstein’s theory. But, does that mean that we must
necessarily define inertial systems in terms of Newton’s
equations?
Other definitions abound. For example, to define an iner-
tial system in a way that is not predetermined by the adop-
tion of particular laws of motion, we may use an approach
formulated by Ludwig Lange. In 1885, he proposed an ideal
construct for identifying any ‘‘inertial system’’ of reference.8
It seems that he actually coined the expression. Lange argued
that any three material points simultaneously projected from
a single point, and moving freely in noncoplanar directions,
constitute an inertial system. He assumed the points’ uniform
rectilinear motion as a matter of ‘‘mere convention.’’ It
seemed impossible to ascertain their motion univocally, be-
cause they were meant to function as the standard by which
to measure other motions.9 Such a system, though estab-
lished by convention, would serve to verify whether other
objects move rectilinearly at constant velocity.
Lange’s definition has served to specify inertial systems in
Newton’s mechanics, as well as in Einstein’s special theory.
For example, in his treatise on relativity, Max Laue used
Lange’s definition.10 It applies also in schemes developed
upon the analysis of special relativity advanced by Hans
Reichenbach.11
In contradistinction, another tradition defines inertial sys-
tems in terms of Newton’s dynamics.12 Arguably, one advan-
tage of that approach is that it need not presuppose the exis-
tence of empty space and hence of any ‘‘free’’ particles. Still,
Newton’s first law of motion refers to free bodies. One of its
virtues is that it continued to be exactly valid even in Ein-
stein’s special relativity.
The law of inertia by itself does not suffice to necessitate
the standard synchrony used in special relativity. Hence,
Ohanian suggests that this law be abandoned or construed as
a corollary of Newton’s second law. Thus, he opposes a tra-
dition 共cultivated by Mach, Kirchhoff, Hertz, Einstein,
among many others兲 that begins the analysis of motion by
mathematically describing varying distances between bodies
rather than by postulating notions of force. Most physicists,
following Newton, prefer to present the law of inertia as
independent and prior to the force law.
In contrast, Ohanian considers dynamical relations as
more real or fundamental than any kinematics that we may
devise by various arbitrary modes of representing empirical
relations. For him, F⫽ma, when applied to multiple bodies,
is ‘‘a true law of nature, that is, a prediction that can be
confirmed or contradicted by experiment.’’ 13 Notwithstand-
ing a century of apparent confirmations of relativistic devia-
tions from Newton’s force law, Ohanian requires that this
very ‘‘law’’ is essential in defining an inertial system. He
shows that different synchrony conventions would entail a
change in the mathematical form of dynamical equations.
The resulting equations involve what he calls ‘‘pseudo-
forces.’’ Yet, such equations do not predict any differences
whatsoever in the actual material behavior of physical sys-
tems. They constitute an unusual but entirely consistent way
of representing and accounting for empirical relations.
Ohanian rejects such equations because they don’t have the

452 Am. J. Phys. 73 共5兲, May 2005 http://aapt.org/ajp © 2005 American Association of Physics Teachers 452
This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
 mathematical form of Newton’s law. He concludes that ‘‘in
an inertial reference frame, there is no freedom of choice in
the synchronization, and in such a reference frame, a con-
vention for synchronization is not needed or permitted.’’ 14
Yet, his argument depends on adopting a particular defini-
tion of inertial frame.
Furthermore, he claims that Einstein was ‘‘sleepwalking’’
in positing the standard synchrony procedure. He construes it
as ‘‘redundant,’’ being involved 共allegedly兲 in the definition
of an ‘‘inertial reference frame,’’ which he claimed was in
the 1905 paper. But, since Einstein did not overtly formulate
a definition of inertial frame consonant with his new kine-
matics, there is no redundancy involved. 共By the way, no-
where in the 1905 paper did Einstein use the phrase ‘‘inertial
reference frame,’’ nor any of these words; he used the ex-
pression ‘‘system of coordinates.’’兲 If Ohanian is correct in
connecting standard synchrony inextricably to inertial
frames, then Einstein’s original formulation is not redundant,
precisely because he only specifies the synchrony procedure,
and that would tacitly fix the definition of an inertial system.
Regardless, Einstein’s kinematics admits other common defi-
nitions of inertial frame.
Another point to clarify is the rise of the notion that the
simultaneity of distant events and the unidirectional speed of
any signal involve an element of convention. Such argu-
ments are often traced to the works of Reichenbach in the
1920s.
However, it is well known that since the 1890s the math-
ematician and physicist Henri Poincaré had highlighted the
importance of conventions in mathematics and physics. His
concerns grew from questions about the validity of non-
Euclidean geometries, as had been the case with Helmholtz.
In 1898 Poincaré characterized the metric notion that light
has the same speed in opposite directions as a ‘‘postulate,’’ a
presupposition ‘‘which could never be verified directly by
experiment.’’ 15 He argued that the simultaneity of distant
events and the equality of two durations ‘‘can acquire mean-
ing only by convention.’’ 16 For Poincaré, conventions were
principles chosen conveniently by all scientists because of
their compatibility with empirical knowledge. In that way,
the fundamental knowledge secured in mathematics and
physics could be understood as not necessarily being fixed
by the structure of the world 共nor by transcendental catego-
ries of the mind, contrary to the philosophy of Kant兲.
Poincaré argued: we voluntarily ascribe certainty to such
conventions because we posit their validity as a matter of
definition. For example, he explained, F⫽ma becomes a
convention when we decide that ‘‘force’’ is just the name
given to the product of mass times acceleration. The propo-
sition cannot then be contradicted by experiment. Further-
more, he noted that if we adopt an unusual way of measuring
time, then ‘‘the experiments on which Newton’s second law
is founded would none the less have the same meaning. Only
the enunciation of the law would be different, because it
would be translated into another language; it would evidently
be much less simple.’’ 17
Poincaré argued that, to understand the principles of me-
chanics, it is useful to compare them with alternative formu-
lations and hypotheses. He complained that analysis of the
origins and validity of principles is often difficult because
textbooks ‘‘do not clearly distinguish between what is expe-
rience, what is mathematical reasoning, what is convention,
and what are hypotheses.’’ 16
It is also well known that Einstein studied and was influ-
453 Am. J. Phys., Vol. 73, No. 5, May 2005
This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
enced by Poincaré’s works.18 At a younger age, Einstein had
been immersed in the philosophy of Kant. But, by 1905,
owing to his readings of Mach, Poincaré, and especially
Hume, Einstein abandoned the idea that notions of space and
time are a priori knowledge. In his 1905 paper, Einstein
acknowledged that different definitions of force are possible.
And, he described also as a ‘‘definition’’ the requirement that
light takes the same time to traverse equal paths in opposite
directions. It is also well known that in his popular book on
relativity he emphasized that this very requirement ‘‘is in
reality neither a supposition nor a hypothesis about the
physical nature of light, but a stipulation which I can make
of my own free will in order to arrive at a definition of
simultaneity.’’ 19
Among French speakers, the notion that light isotropy is
conventional gained prominence because, in the French
translation of Einstein’s book, the German word ‘‘festset-
zung’’ 共‘‘stipulation’’ in the authorized English translation兲
was rendered ‘‘convention.’’ 20 Among English speakers, the
notion that isotropy is conventional gained prominence in the
writings of Eddington, who early on was reputed to be one of
the few physicists who really understood relativity theory.
For example, he argued that, ‘‘Strictly speaking the
Michelson–Morley experiment did not prove directly that
the speed of light was constant in all directions. The experi-
ment compared the times of a journey ‘there-and-back’.’’ Ed-
dington explained that all such experiments compare only
the round trips of signals: ‘‘The measured velocity of light is
the average to-and-fro velocity... there is a deadlock... which
can only be removed by an arbitrary assumption or conven-
tion. The convention actually adopted is that 共relative to the
observer兲 the velocities of light in the two opposite directions
are equal.’’ 21
Such discussions and many that followed on the conven-
tionality of simultaneity have not mentioned whether Ein-
stein himself actually called simultaneity a ‘‘convention.’’ In
fact, he did. In a letter of 1924 to André Metz, a philosopher
of physics, Einstein noted that relativity theory involved
‘‘conventions’’ and ‘‘physical hypotheses.’’ He specified one
such convention: ‘‘simultaneity.’’ 22 Moreover, in 1918 he
had commented to Max Born: ‘‘The ‘a priori’ I must pare
down to ‘conventional’... .’’ 23 Even decades later, Einstein
defended Poincaré’s view that particular geometries are con-
strued as true essentially as a matter of convention.24
Reichenbach, another physicist turned philosopher, main-
tained a friendly correspondence with Einstein. He attributed
to Einstein the merit of having properly extended, to the
concept of time, the role of conventions that had been eluci-
dated by Helmholtz and Poincaré in physics.25 In 1928 Re-
ichenbach published his third book analyzing the concepts of
space and time in Einstein’s theories.26 He elaborated the
thesis that the standard procedure for synchronizing clocks
involves an element of convention, in the stipulation that any
light rays take equal times to travel equal paths in opposite
directions relative to any inertial reference frame.27
Promptly, Einstein wrote a book review, now virtually un-
known. He praised Reichenbach’s efforts to clarify how defi-
nitions connect concepts and experience: ‘‘The clear elucida-
tion of the role of coordinative definitions, especially in the
area of relativity theory, is one of the main objectives to
which the author has aspired and reached.’’ 28
Einstein noted that Reichenbach’s detailed analysis of time
in special relativity was easily intelligible, and he highlighted
that ‘‘value is located in clearly distinguishing what in the
Notes and Discussions 453

relativistic definition of simultaneity is a logically arbitrary
stipulation, and what in it is a hypothesis, that is, a presup-
position about the structure of nature.’’ Unfortunately, nowa-
days, the tendency to not distinguish between conventional
and factual aspects of physical theories continues to be
strong.29
ACKNOWLEDGMENTS
I thank John Stachel and Ronald Anderson for discussions
and helpful comments. I also thank Hans Ohanian and
Michel Janssen for their useful comments.
a兲
Electronic mail: martinez@hss.caltech.edu
1
H. C. Ohanian, ‘‘The role of dynamics in the synchronization problem,’’
Am. J. Phys. 72, 141–148 共2004兲.
2
See Ref. 1, especially pages 146, 148.
3
A. Einstein, ‘‘Zur Elektrodynamik bewegter Körper,’’ Ann. Phys. 共Leipzig兲
17, 891–921 共1905兲.
4
A. Einstein, H. A. Lorentz, H. Minkowski, and H. Weyl, The Principle of
Relativity, translated by W. Perrett and G. B. Jeffery, notes by A. Sommer-
feld 共1923; Dover, New York, 1952兲, p. 38.
5 U. P., Cambridge, 1957兲, p. 19; The Nature of the Physical World 共1927;
Fortschritte der Mathematischen Wissenschaften in Monographien, No. 2:
H. A. Lorentz, A. Einstein, H. Minkowski, Das Relativitätsprinzip, edited
by O. Blumenthal, notes by A. Sommerfeld, foreword by Blumenthal
共Teubner, Stuttgart, 1913兲, p. 28.
6
R. Torretti, Creative Understanding 共University of Chicago Press, Chi-
cago, 1990兲, p. 286. He characterized Einstein as having been ‘‘careless’’
on this matter. Still, Torretti 共1983兲 rejected the later footnote, because it
only avoids the inconsistency ‘‘at a rather high price, for the concept of a
stationary frame thereby becomes irretrievably ambiguous.’’ Torretti, Rela-
tivity and Geometry 共Pergamon, Oxford, 1983兲, p. 294.
7 24
Einstein, Ref. 3, p. 896.
8
L. Lange, ‘‘Ueber das Beharrungsgesetz,’’ Ber. der König. Sachsischen
Gesellschaft der Wissenschaften zu Leipzig, Math.-Phys. Klasse 37, 331–
351 共1885兲.
9
L. Lange, Die Geschichtliche Entwickelung des Bewegungsbegriffes und
Ihr Voraussichtliches Endergebniss 共W. Engelmann, Leipzig, 1886兲, p.
140.
10
M. Laue, Die Relativitätstheorie, Vol. 1, in Die Wissenschaft Sammlung
von Einzeldarstellungen aus den Gebieten der Naturwissenschaft und
Technik, edited by E. Wiedemann, 3rd ed. 共Vieweg, Braunschweig, 1919兲,
Vol. 38. Laue’s discussion of Lange’s definition of inertial system did not
appear in the first 共1911兲 and second 共1913兲 editions.
11
For example, J. A. Winnie, ‘‘Special relativity without one-way velocity
assumptions. Part I,’’ Philos. Sci. 37, 81–99 共1970兲, ‘‘Part II,’’ 37, 223–
238 共1970兲.
12
For example, J. Thomson, ‘‘On the law of inertia, the principle of chro-
nometry and the principle of absolute clinural rest, and of absolute rota-
tion,’’ R. Soc. Edinburgh Proc. 12, 568 –578 共1884兲.
Comment on ‘‘The role of dynamics in the synchronization problem,’’ by
Hans C. Ohanian †Am. J. Phys. 72 „2…, 141–148 „2004…‡
Alan Macdonalda)
Department of Mathematics, Luther College, Decorah, Iowa, 52101
共Received 8 July 2004; accepted 21 December 2004兲
关DOI: 10.1119/1.1858448兴
Hans C. Ohanian1 claims to ‘‘defeat’’ the conventionalist
thesis of clock synchronization2,3 by using an argument
based on dynamics. My aim here is to show that his argu-
ment does not succeed.
454 Am. J. Phys. 73 共5兲, May 2005 http://aapt.org/ajp
This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
13
Ohanian, personal communication, July 2004.
14
Ohanian, Ref. 1, p. 143.
15
共a兲 H. Poincaré, ‘‘La mesure du temps,’’ Revue de Métaphysique et de
Morale 6, 1–13 共1898兲; reissued in 共b兲 Poincaré, Foundations of Science,
authorized translation by G. B. Halsted 共Science, New York, 1913兲, p. 232.
16
H. Poincaré, La Science et l’Hypothèse 共Flammarion, Paris, 1902兲, p. 111;
translated in Foundations of Science, Ref. 15b, p. 92.
17
Poincaré, Ref. 15b, p. 227.
18
For example, Einstein intensely analyzed La Science et l’Hypothèse, re-
called M. Solovine; A. Einstein, Letters to Solovine 共Philosophical Library,
New York, 1986兲, p. 9. Also, in a lecture delivered in 1921, Einstein
argued: ‘‘It is essential here also to pay strict attention to the relation of
experience to our concepts. It seems to me that Poincaré clearly recog-
nized the truth in the account he gave in his book La Science et
l’Hypothèse.’’ A. Einstein, The Meaning of Relativity 共Princeton U.P., Prin-
ceton, 1946兲, p. 3.
19
A. Einstein, Über die Spezielle und die Allgemeine Relativitätstheorie
共Vieweg, Braunschweig, 1917兲; Einstein, Relativity, translated by R. W.
Lawson 共P. Smith/H. Holt and Co., New York, 1931兲, p. 23.
20
A. Einstein, La Théorie de la Rélativité Restreinte et Généralisé, translated
by J. Rouvière 共Gauthier-Villars, Paris, 1921兲, p. 19.
21
A. S. Eddington, The Mathematical Theory of Relativity 共1923; Cambridge
Macmillan, New York, 1929兲, p. 46.
22
A. Einstein to André Metz, 27 Nov. 1924, Einstein Archive, item control
number 18-255. Metz had argued against Einstein’s claim that isotropy of
light is a convention. Instead, he argued that isotropy is a logically neces-
sary consequence of experiments such as that of Michelson and Morley;
see A. Metz, La Relativité, 16th ed. 共Chiron, Paris, 1923兲, pp. 21–22.
23
A. Einstein to M. Born, after 28 June 1918, in Collected Papers of Albert
Einstein, Vol. 8B: Correspondence 1918 共Princeton, Princeton U.P., 1998兲,
edited by R. Schulmann, A. J. Kox, M. Janssen, and J. Illy, p. 818.
A. Einstein, ‘‘Reply to Criticisms,’’ in Albert Einstein: Philosopher-
Scientist, edited by P. A. Schilpp 共G. Banta, Menasha, WI, 1949兲, pp.
676 – 679.
25
H. Reichenbach, Axiomatization of the Theory of Relativity, translated and
edited by M. Reichenbach 共1924; University of California Press, Berkeley
and Los Angeles, 1969兲, p. 13.
26
H. Reichenbach, Philosophie der Raum-Zeit-Lehre 共de Gruyter, Berlin,
1928兲.
27
H. Reichenbach, The Philosophy of Space and Time, translated by M.
Reichenbach and J. Freund 共Dover, New York, 1958兲, pp. 127–129, 146,
168.
28
A. Einstein, ‘‘Philosophie und Pädagogik,’’ Deutsche Literaturzeitung 20,
No. 1 共1928兲, p. 19.
29
Students interested in learning how physicists, especially in the U. S., have
sometimes misconstrued postulates and conventions in relativity as state-
ments of experimental facts may consult S. Goldberg, Understanding
Relativity 共Birkhäuser, Boston, 1984兲.
Ohanian writes, ‘‘关The conventionalist兴 thesis rests on the
belief that the adoption of the nonstandard synchronization
leads to a self-consistent description of physical phenomena,
without any demonstrably erroneous experimental conse-
© 2005 American Association of Physics Teachers 454
 quences.’’ But he does not claim to defeat the thesis on this
basis. Indeed, he writes, ‘‘Reichenbach’s nonstandard syn-
chronization permits a consistent description of physical
phenomena.’’
Why, then, does Ohanian reject the thesis? He shows that
with a nonstandard synchronization Newton’s second law
contains pseudoforces and thus is not in its ‘‘standard form,’’
F⫽ma. He writes, ‘‘The fundamental error of Reichenbach
and his followers . . . 关is that兴 they failed to appreciate that
the time variable must be chosen in such a way that the laws
of dynamics 关i.e., Newton’s laws兴 keep their standard form.’’
共My emphasis.兲
But why must Newton’s laws take their standard form?
Ohanian writes that Einstein’s definition of an inertial frame
‘‘demands’’ that Newton’s laws ‘‘must be valid.’’ Ohanian’s
citation of Einstein’s ‘‘demand’’ reads, ‘‘Let us take a system
of coordinates in which the equations of Newtonian mechan-
ics hold good.’’ There is no indication that Einstein would
have rejected nonstandard synchronizations and their non-
standard forms of Newton’s laws. And if he would have, we
need not follow his definition.
Thus, Ohanian has only shown that if Newton’s laws must
take their standard form, then nonstandard synchronizations
are ruled out. But he has given no reason that the laws must
take their standard form.
Ohanian’s articulation of the conventionalist thesis above,
‘‘nonstandard synchronization leads to a self-consistent de-
scription of physical phenomena,’’ is a triviality. It also
misses the point of the thesis, as I now discuss.
In Newtonian mechanics there is a universal time t that
provides a unique synchronization: two events are simulta-
neous if they occur at the same t. Synchronization methods,
for example clock transport, are merely operational
procedures for actualizing the already existing synchrony
given by t.
In special relativity there is no universal time. Every stu-
dent of the theory learns that synchronization is frame de-
pendent. The conventionalist thesis goes further, recognizing
that even in a single nonaccelerating frame, the theory does
not supply a unique synchronization.4 Thus, synchronization
in the frame must be defined.
This is why Einstein calls the synchronization method in
his 1905 paper a definition 共his emphasis兲, not an operational
procedure. He thought this was so important that, even in a
popularization of relativity, he wrote that his definition ‘‘is in
reality neither a supposition nor a hypothesis about the
physical nature of light, but a stipulation which I can make
of my own freewill in order to arrive at a definition of
simultaneity.’’ 5 Reichenbach also emphasizes that synchro-
nization is a matter of definition.6
Several conditions have been proposed to rule out non-
standard synchronizations.2,3 Malament’s condition, that the
simultaneity relation be defined from the causal connectibil-
ity relation, has elicited the most interest. Ohanian’s condi-
tion, that Newton’s laws take their standard form, is the lat-
est. There is no reason that we must accept any of them.
455 Am. J. Phys., Vol. 73, No. 5, May 2005
This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
Ohanian has not defeated the conventionalist thesis.
There is no need to use Newton’s first or second laws in
the definition of an inertial frame, despite the near universal
use of the first law, and Ohanian’s use of the second. I have
defined an inertial frame as one in which accelerometers read
zero and clocks are Einstein synchronized.7 The accelerom-
eters can be as simple as a cube with identical springs at-
tached from its corners to a weight at its center. If acceler-
ometers read zero, then we can attempt to Einstein
synchronize clocks. My paper gives testable necessary and
sufficient conditions that this is possible. 共Perhaps surpris-
ingly, the conditions do not imply that the speed of light is
isotropic.7兲 If the conditions are satisfied, then we can Ein-
stein synchronize all clocks in the frame with some ‘‘central’’
clock. Then every pair of clocks will be synchronized and
remain so. Because the definition is independent of Newton’s
laws,8 they can then be tested empirically. In our universe
they are valid 共in their standard form兲.
There is a difficulty in using Newton’s laws in the defini-
tion of an inertial frame: how can the laws be laws if they are
part of a definition? I know of no nonsubtle way around this
difficulty. The difficulty does not arise if Newton’s laws are
not part of the definition of an inertial frame.
Note added in proof. Ohanian devotes the entire third
paragraph of his Reply to an argument scarcely mentioned in
his paper: simplicity. I agree that if the laws of physics must
take their simplest form, then nonstandard synchronizations
are ruled out. But he has given no reason that the laws must
take their simplest form. If Ohanian wants a criterion to rule
out nonstandard synchronizations, why not use the simplest
and most direct of all: that the one way speed of light must
be isotropic?
ACKNOWLEDGMENT
I thank Professor Ohanian for a cordial and helpful corre-
spondence.
a兲
Electronic mail: macdonal@luther.edu
1
H. Ohanian, ‘‘The role of dynamics in the synchronization problem,’’ Am.
J. Phys. 72, 141–148 共2004兲.
2
A. Janis, Conventionality of Simultaneity, in The Stanford Encyclopedia of
Philosophy, edited by Edward N. Zalta, http://plato.stanford.edu/archives/
fall2002/entries/spacetime-convensimul/
3
R. Anderson et al., ‘‘Conventionality of synchronisation, gauge depen-
dence and test theories of relativity,’’ Phys. Rep. 295, 93–180 共1998兲.
4
A. Grünbaum, Philosophical Problems of Space and Time 共Knopf, New
York, 1963兲, p. 343ff.
5
A. Einstein, Relativity: The Special and General Theory, 15th ed. 共Crown,
New York, 1952兲.
6
H. Reichenbach, The philosophy of space & time 共Dover, New York,
7
1958兲, p. 127.
A. Macdonald, ‘‘Clock synchronization, a universal light speed, and the
terrestrial redshift experiment,’’ Am. J. Phys. 51, 795–797 共1983兲.
8
Except for a very special case of the first law: to say that accelerometers
read zero is equivalent to saying that objects placed at rest remain at rest.
This involves neither dynamics nor clocks.
Notes and Discussions 455
 Reply to ‘‘Comment„s… on ‘The role of dynamics in the synchronization
problem,’ ’’ by A. Macdonald †Am. J. Phys. 73, 454 „2005…‡
and A. A. Martı́nez †Am. J. Phys. 73, 452 „2005…‡
Hans C. Ohaniana)
Department of Physics, University of Vermont, Burlington, Vermont 05405
共Received 30 November 2004; accepted 21 December 2004兲
关DOI: 10.1119/1.1858449兴
Macdonald1 quotes my paper selectively, giving the im-
pression that the sentence he quotes in his second paragraph
is my description of the conventionalist thesis. Actually, my
description of the conventionalist thesis precedes the quoted
sentence and is clearly labeled: ‘‘Reichenbach claimed that
in the absence of any physical criterion for choosing any
particular value of ␧, this choice is purely a matter of con-
vention. Thus synchronization is a matter of convention, and
so is the value of the one-way speed of light. This is the
conventionalist thesis.’’ After the sentence that Macdonald
quoted, I go on to say ‘‘Reichenbach regarded the freedom in
the choice of ␧ as analogous to the freedom in the choice of
a unit of length, and Anderson et al. regard it as analogous to
the freedom of choice in the additive constant in the electro-
static potential.’’ Of these sentences, the sentence quoted by
Macdonald is the one and only sentence that I do not chal-
lenge and refute. My paper shows that the R synchronization
introduces pseudoforces in the equation of motion, and the
existence of such pseudoforces provides a clear physical cri-
terion for choosing a particular value of ␧. Furthermore, the
presence of such pseudoforces obviously contradicts
Reichenbach’s and Anderson’s claim that the adoption of the
R synchronization is analogous to a choice of a length unit or
a choice of gauge—a change of length unit or of gauge does
not result in modifications of the force. This is why I claim to
have defeated the conventionalist thesis, at least the version
of this thesis that I comprehended from the literature.
I do not claim to defeat all possible ‘‘new, improved’’
versions of this thesis that might hereafter descend upon us,
such as the version proposed by Macdonald, when he says,
‘‘The conventionalist thesis goes further, recognizing that
even in a single nonaccelerating frame, nature does not sup-
ply a unique synchronization.’’ This version is not falsifiable,
because it is merely an instance of the well-known covari-
ance of all the laws of physics under general coordinate
transformations. It is indisputably true that we can adopt any
general time or space coordinates whatsoever to express the
laws of physics, and changes from one set of general coor-
dinates to another will merely alter the form of the laws, not
their content. This applies not only to Newton’s second law,
but also to the first. For example, in an accelerated reference
frame, Newton’s first law is still valid, but it takes a non-
standard form: any particle not acted upon by forces main-
tains a constant acceleration opposite to that of the reference
frame 共more generally, in arbitrary curvilinear spacetime co-
ordinates, the first law takes the form of an equation for
geodesic motion兲. Macdonald vehemently defends non-
standard forms of the second law, but he shies away from
nonstandard forms of the first law. I believe that, in physics,
consistency is a virtue, and if we are going to adopt general
456 Am. J. Phys. 73 共5兲, May 2005 http://aapt.org/ajp
This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
coordinates and accept nonstandard forms for the second
law, then we should be equally willing to accept nonstandard
forms for the first law.
However, although we can adopt any coordinates whatso-
ever, some choices of coordinates and of synchronization
play a preferential role, because they permit us to express the
laws of physics in their simplest form. This is what motivates
Einstein’s definition of inertial reference frames in his book
The Meaning of Relativity,2 where he adopts ‘‘... Cartesian
systems of co-ordinates, the so-called inertial systems, with
reference to which the laws of mechanics 共more generally
the laws of physics兲 are expressed in their simplest form.’’
The adoption of a preferential inertial reference frame in
which all the laws of physics take their simplest form com-
pels the E synchronization and forbids the R synchroniza-
tion. This focus on simplicity gives us insight into a crucial
difference between E synchronization and R synchroniza-
tion: with the former, all the laws of physics take their sim-
plest form; with the latter, only Newton’s first law takes its
simplest form. All other laws containing time dependences
are ‘‘desimplified.’’ Conventionalists believe that nonacceler-
ating reference frames and the simple form of the first law
play a preferential role, but they fail to consider that with R
synchronization the simple form of the first law does not lead
to simple forms of the second and third laws, and that be-
cause of this missing link the practical significance of the
first law is reduced to next to nothing.
Martinez3 is correct in objecting to the footnote I copied
from the Perrett and Jeffery translation of Einstein’s 1905
paper.4 This footnote is a gloss, probably by Sommerfeld. I
was not aware of that, and I am grateful to Martinez for
bringing this to my attention. However, it is evidently a valid
gloss, as can be seen from Article 10 of the paper, where
Einstein states Newton’s equations of motion for a charged
particle and says that these are valid ‘‘insofar as it is moving
slowly’’ 共this is my own, absolutely accurate, translation
from the German original兲. In a later manuscript, written in
1912 and recently published in facsimile format,5 Einstein
deals with the definition of an inertial system 共there called a
‘‘berechtigtes,’’ or authorized, system兲 in the same way: In
Article 6, he selects the coordinate system so as to make
Newton’s equations of motion ‘‘as simple as possible,’’ and
he states these equations in their standard form; and in Ar-
ticle 13 he continues with ‘‘For infinitely slow motions,
Newton’s laws of motion are undoubtedly valid...’’ There is
nothing in either of these papers to support Martinez’s specu-
lation that Einstein intended the statement of Newton’s equa-
tions as a devious polemical device rather than as a straight-
forward definition of inertial reference frame. To this I might
© 2005 American Association of Physics Teachers 456
 add that my usage of ‘‘inertial reference frame’’ is in accord I won’t comment on other tangential arguments made by
with modern practice 共see, e.g., H. Goldstein, Classical Macdonald and Martinez. My silence does not imply con-
Mechanics,6 where the validity of the standard form of New- sent.
ton’s second law is taken as the defining characteristic of an a兲
Electronic mail: HansOhanian@cs.com
inertial system兲. 1
A. Macdonald, Am. J. Phys. 73, 454 – 455 共2005兲.
But the argument in my paper does not really hinge on this 2
A. Einstein, The Meaning of Relativity 共Princeton University Press,
question of terminology. If Macdonald and Martinez dislike Princeton, NJ, 1955兲, p. 24.
my usage, let them replace every instance of ‘‘inertial refer-
3
A. A. Martı́nez, Am. J. Phys. 73, 452– 454 共2005兲.
4
A. Einstein, H. A. Lorentz, H. Minkowski, and H. Weyl, The Principle of
ence frame’’ in my paper by ‘‘preferential reference frame in Relativity, translated by W. Perrett and G. B. Jeffery 共Dover, New York,
which Newton’s laws are valid in their standard form, at low 1952兲, p. 38.
5
speed.’’ The essential point of my paper is that R synchroni- A. Einstein, Einstein’s 1912 Manuscript on the Special Theory of Relativ-
ity 共George Braziller and E. J. Safra Foundation, Jerusalem, 1996兲. Trans-
zation gives rise to extra terms in the equation of motion, and lations of quoted sentences are my own.
conventionalists cannot make this problem go away by dis- 6
H. Goldstein, Classical Mechanics, 2nd ed. 共Addison-Wesley, Reading,
puting matters of terminology. MA, 1980兲, p. 2.

Estimating hc Õ k from sunlight

Peter Pesica)
St. John’s College, 1160 Camino de la Cruz Blanca, Santa Fe, New Mexico 87505-4599
共Received 1 March 2004; accepted for publication 20 August 2004兲
The blackbody approximation is used for the visible solar spectrum to find an order of magnitude
estimate for hc/k. © 2005 American Association of Physics Teachers.
关DOI: 10.1119/1.1806479兴

This paper presents a simple estimate of hc/k using a mann et al. have pointed out that atmospheric extinction will
‘‘back of the envelope’’ argument. If we take c and k as change these temperature results by 11%, which I will dis-
given, the argument gives an estimate of Planck’s constant cuss in the following.3
from basic considerations about the color and temperature of The Planck distribution is given by
the sun. 2hc 2 /␭ 5
The sun’s spectrum is close to that of a blackbody with a B ␭共 T 兲 ⫽ , 共1兲
e hc/kT␭ ⫺1
temperature of about 5800 K, which we can determine
purely from earth-based observations. We first measure what in terms of the wavelength ␭. To find the peak of B ␭ , we set
atmospheric physicists call the ‘‘solar irradiance’’ 共and as- dB/d␭⫽0. If we define x⫽hc/kT␭, this condition leads to
tronomers the ‘‘solar constant’’兲, the energy flux arriving per the equation e ⫺x ⫹x/5⫺1⫽0, whose solution is readily seen
square meter at a sphere whose radius is the mean sun–earth
distance 共without considering the earth’s atmosphere兲,
namely 1370 W/m2 . 1 This value can be estimated by using a
light bulb and a ruler: find the distance from your body at
which a light bulb feels about as warm as the sun does on
your bare skin. The solar irradiance can be estimated as the
wattage of the bulb divided by the surface area of the sphere
whose radius is the bulb–body distance. For instance, using a
55-W light bulb, I found the distance to be 6 cm, leading to
a value of ⬇1200 W/m2 for the solar irradiance.
If we multiply the solar irradiance by the total surface area
of the sphere whose radius is the mean sun–earth distance,
we obtain the sun’s total luminosity. Approximating the sun
by a blackbody of radius R gives its luminosity by the
Stefan–Boltzmann law, 4 ␲ R 2 ␴ T 4 , where T is the sun’s ef-
fective temperature and ␴ ⫽5.67⫻10⫺8 W/m2 K4 is Stefan’s
constant. Equating these two expressions for the luminosity
gives T⬇5800 K if the solar irradiance is 1370 共or T
⬇5600 K, using my light bulb estimate兲. Note that, although
Stefan’s constant can be expressed in quantum theory as ␴ Fig. 1. NASA data for the solar spectral irradiance measured outside the
⫽2 ␲ 5 k 4 /15c 2 h 3 , it can be measured in the laboratory with- atmosphere 共closed dots兲 and at the earth’s surface when the sun is 60° from
out any knowledge of Planck’s constant.2 However, Bier- the zenith 共open circles兲—Ref. 2.

457 Am. J. Phys. 73 共5兲, May 2005 http://aapt.org/ajp © 2005 American Association of Physics Teachers 457
This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
 to be x⬇5, and hence hc/k⬇5T␭. If we measure ␭ in nm,
we have
hc/k⫽ 共 2.9⫻10⫺5 兲 ␭m K.
Note that, though we have used the wavelength representa-
tion, our result is independent of the choice of representation
because hc/k is a universal constant.4
What shall we take for ␭? The sun appears to the naked
eye to be approximately white, and the spectrum of solar
irradiance at the surface of the earth is reasonably flat from
475 to 700 nm 共see Fig. 1兲, although the spectrum above the
atmosphere has a broad peak around 475 nm.2
As a rough estimate, we take the range of visible wave-
lengths, 400–700 nm, and use Eq. 共2兲 to obtain hc/k
⫽1.2– 2.0⫻10⫺2 m K; the mean value is 1.6⫻10⫺2 m K
共corresponding to ␭⫽550 nm, the mean visible wavelength兲
and is within 15% of the known value for hc/k, 1.44
⫻10⫺2 m K. More to the point, the order of magnitude is
correct regardless of which visible wavelength 共or average兲
we choose.
I do not know of any simple way to estimate c, but to find
k⫽R/N we can take the gas constant R from chemical de-
terminations and can estimate Avogadro’s number N from
airlight, the scattered sunlight that makes the sky blue.5 If we
take the known values of c and k over the visible wave-
lengths, we find a range of values for h equal to 5.5– 9.2
⫻10⫺34 J s, which is the right order of magnitude.6 Note that
if we were to include the values T⫽3150 K and ␭
⫽562 nm determined in Ref. 3 using a spectrometer, we
would obtain hc/k⫽1.45⫻10⫺2 m K and h⫽6.67
⫻10⫺34 J s. But my point concerns crude naked-eye esti-
mates. Although the quantum realm often is considered to be
restricted to subatomic phenomena, we can estimate Planck’s
constant by gazing at the sky.
458 Am. J. Phys., Vol. 73, No. 5, May 2005
This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
82.181.36.37 On: Thu, 10 Apr 2014 12:56:54
ACKNOWLEDGMENTS
I thank Craig Bohren, Steve Boughn, Joseph Snider,
共2兲 Michael Vollmer, Ssu Weng, and Andrew Young for their
helpful criticisms and suggestions.
a兲
Electronic mail: ppesic@sjcsf.edu
1
See Craig F. Bohren, What Light Through Yonder Window Breaks?
More Experiments in Atmospheric Physics 共Wiley, New York, 1991兲, pp.
113–120.
2
William R. Kuhn and Susan E. Postawko, ‘‘Solar radiation and the Earth’s
atmosphere,’’ Phys. Teach. 26, 266 –273 共1998兲 present detailed graphs
and discuss the measured spectrum of the sun. See also Günter Wyszecki
and W. S. Stiles, Color Science: Concepts and Methods, Quantitative Data
and Formulae 共Wiley, New York, 1982兲, 2nd ed., p. 6, from which Fig. 1
is taken. I thank Rod Nave for his help in locating and providing this
figure. The air mass m is defined as the ratio between the actual path
length taken by the solar beam when incident at polar angle ␪ from the
zenith to the analogous path length when the sun is directly overhead, so
that m⫽sec ␪; in Fig. 1, ␪ ⫽60° and m⫽2.
3
Mark L. Biermann, Debora M. Katz, Robert Aho, James Diaz-Barriaga,
and Jerome Petron, ‘‘Wien’s law and the temperature of the Sun,’’ Phys.
Teach. 40, 398 – 400 共2002兲.
4
For a discussion of the pitfalls of the different representations, see Bernard
H. Soffer and David K. Lynch, ‘‘Some paradoxes, errors, and resolutions
concerning the spectral optimization of human vision,’’ Am. J. Phys. 67,
946 –953 共1999兲; James M. Overduin, ‘‘Eyesight and the solar Wien
peak,’’ ibid. 71, 216 –219 共2003兲; Geoff Nunes, ‘‘Comment,’’ ibid. 71, 519
共2003兲; and Mark A. Heald, ‘‘Where is the Wien peak?,’’ ibid. 71, 1322–
1323 共2003兲. I thank Craig Bohren for pointing out to me the
representation-independence of hc/k⬇x␭.
5
See Peter Pesic, ‘‘Estimating Avogadro’s number from skylight and air-
light,’’ Eur. J. Phys. 共submitted兲. For a historical survey of views about sky
blue and its physical implications, see Peter Pesic, Sky in a Bottle 共MIT
Press, Cambridge, MA, 2005兲.
6
For more accurate determinations using simple equipment, see R. E. Cran-
dall and J. F. Delord, ‘‘Minimal apparatus for determination of Planck’s
constant,’’ Am. J. Phys. 51, 90–91 共1982兲; J. Dryzek and K. Ruebenbauer,
‘‘Planck’s constant determination from black-body radiation,’’ ibid. 60,
251–253 共1992兲; and Graciela Brizuela and Alfredo Juan, ‘‘Planck’s con-
stant determination using a light bulb,’’ ibid. 64, 819– 821 共1996兲.
Notes and Discussions 458