1: \input ictp.tex
2: \hfuzz=50pt
3:
4: \TITLE{EINSTEIN AND HILBERT:}
5: \TITLE{THE CREATION OF GENERAL RELATIVITY \FOOTNOTE{$^*$}
6: {Expanded version of a Colloquium lecture held at the International Centre for
7: Theoretical Physics, Trieste, 9 December 1992 and (updated) at the
8: International University Bremen, 15 March 2005.}}
9: \vskip1.5truecm
10:
11: \AUTHOR{Ivan T. Todorov}
12: \centerline{Institut f\"ur Theoretische Physik, Universit\"at
13: G\"ottingen, Friedrich-Hund-Platz 1}
14: \centerline{D-37077 G\"ottingen, Germany;
15: e-mail: itodorov@theorie.physik.uni-goe.de}
16: \centerline{and}
17: \centerline{Institute for Nuclear Research and Nuclear Energy,
18: Bulgarian Academy of Sciences}
19: \centerline{Tsarigradsko Chaussee 72, BG-1784
20: Sofia, Bulgaria;\FOOTNOTE{$^{**}$}{Permanent address.} e-mail:
21: todorov@inrne.bas.bg}
22: \vskip1truecm
23: \ABSTRACT
24:
25: It took eight years after Einstein announced the basic physical ideas
26: behind the relativistic gravity theory before the proper mathematical
27: formulation of general relativity was mastered. The efforts of the greatest
28: physicist and of the greatest mathematician of the time were involved and
29: reached a breathtaking concentration during the last month of the work.
30:
31: Recent controversy, raised by a much publicized 1997 reading of
32: Hilbert's proof-sheets of his article of November 1915, is also discussed.
33:
34: \vfill\eject
35:
36: \bigskip
37:
38:
39:
40:
41: {\bf Introduction}
42:
43: Since the supergravity fashion and especially since the birth of superstrings
44: a new science emerged which may be called ``high energy mathematical
45: physics''. One fad changes the other each going further away from accessible
46: experiments and into mathematical models, ending up, at best, with the
47: solution of an interesting problem in pure mathematics. The
48: realization of the grand original design seems to be, decades later,
49: nowhere in sight. For quite some time, though,
50: the temptation for mathematical physicists (including leading
51: mathematicians) was hard to resist. Yuri Manin characterized the
52: situation as ``an extreme romanticism of the theoretical high energy
53: physics of the last quarter of our century''.
54:
55: There does exist, on the other hand, a true example of a happy competition
56: between mathematics and physics which has led to the most accomplished
57: among the three claimed revolutions in our science in the first
58: quarter of the twentieth century: the creation of the general
59: theory of relativity. It illustrates how difficult it has been
60: - even for the founding fathers of the theory - to fully understand
61: and adopt such basic notions as reparametrization invariance, Bianchi
62: identities, the concept of energy, which
63: nowadays enter a student curriculum. The presence of a controversy - if
64: not so much among the participants in the events, at least among the
65: historians of science nearly a century later - could serve one good
66: purpose: to attract a wider audience to this remarkable story.
67:
68: \bigskip
69:
70: \noindent{\bf 1.\ \ Prologue: Einstein (and Grossmann): 1907--1915}
71:
72: Einstein seemed never happy with what he had achieved. He was not
73: satisfied by the special principle of relativity because it did not
74: incorporate accelerated motion. Since his student years he had
75: absorbed with sympathy Ernst Mach criticism of the ``monstrous
76: [Newtonian] notion of absolute space''. (It was his lifelong friend
77: Michele Besso who induced the 18 year old Einstein -- back in 1897 -- to
78: read Mach's ``History of Mechanics''.)
79:
80: In his recently reprinted address to the 1904 International Congress
81: of Arts and Science St. Louis, \FOOTNOTE{$^{*)}$}{H. Poincar\'e, L'
82: \'etat actuel et l'avenir de la physique math\'ematique, {\it
83: Bulletin des Sciences Math\'ematiques} {\bf 28} (1904) 302-324;
84: reprinted as: The
85: present and future of mathematical physics, {\it Bull. Amer. Math.
86: Soc.} {\bf 37} (2000) 25-38.}\ as well as in his fundamental paper
87: ``Sur la dynamique de l'\'electron'' which appeared in 1906, Poincar\'e
88: already states the problem of modifying Newton's gravity theory in
89: order to make it consistent with relativity. The problems of gravity
90: and of relativity of accelerated motion are combined in what Einstein
91: will call 13 years later ``{\it the happiest thought in my life}''
92: (Pais 82, Chap.9, pp.178--179). Then, in 1907, while working on the
93: review article ``\"Uber das Relativit\"atsprinzip und die aus
94: demselben gezogenen Folgerungen'' (``On the principle of relativity and
95: its consequences''), Jahrbuch der Radioaktivit\"at und Elektronik {\bf
96: 4} (1907) 411--461 (written, ironically, on the request of Johannes
97: Stark who appears as a bitter enemy of relativity during the Nazi
98: period), he has the idea that
99: ``{\it for an observer falling freely from the roof of a house there
100: exists} -- at least in his immediate surroundings -- {\it no
101: gravitational field}''. In his Kyoto lecture (cited by Pais) Einstein
102: recalls: ``I was sitting in a chair in the patent office at Bern when
103: all of a sudden a thought occurred to me: `If a person falls freely he
104: will not feel his own weight'. I was startled . . . ''
105:
106: Thus the celebrated equivalence principle first appears just two
107: years after the formulation of special relativity (although it was
108: only so baptized by Einstein 5 years later). The 1907 paper does not
109: stop at that. It contains a derivation of the gravitational red
110: shift. Einstein also deduces the formula
111: %
112: $c(\phi )=c\left( 1+{\phi\over c^2}\right)$
113: %
114: for the velocity of light along the direction $\xi$ of a constant
115: gravitational field (the mass in the gravitational potential $\phi$
116: being identified with the unit of mass
117: so that $\phi /c^2$ appears to be dimensionless). He infers
118: that \FOOTNOTE{$^{*)}$} {In predicting the bending of light Einstein
119: has a prominent predecessor. In the first ``Query'' to his
120: ``Opticks'' Isaac Newton writes: ``Do not Bodies act upon Light at a
121: distance, and by their action bend its Rays; and is not this action
122: strongest at the least distance?"}\ ``the light rays which do not run
123: in the $\xi$ direction are bent by the gravitational field''. As if
124: all this was not enough for a first probe into relativistic gravity, Einstein
125: wrote to his friend Konrad Habicht on Christmas 1907 (just 3 weeks after
126: submitting the paper): ``I hope to clear up the so--far unexplained secular
127: change of the perihelion length of Mercury . . .'' (Pais 82, p.182).
128:
129: All three observational implications of general relativity were in the
130: mental view of its creator already at this preliminary stage. Einstein's
131: genius is here manifest with all its flare as well as with its limitations.
132: The limitations are most honestly described by Einstein himself. In his
133: ``Autobiographische Skizze'' completed in March 1955, a month before his
134: death, he deplors his attitudes towards advanced mathematics during his
135: student years. (Maurice Solovin, Einstein's close friend during the period
136: just after he graduated from the ETH -- the Z\"urich Polytechnic,
137: remembers that ``Einstein . . . often spoke against abusive use of
138: mathematics in physics. Physics, he would say, is essentially a concrete
139: and intuitive science.'' ``I do not believe in mathematics'', Einstein is
140: reported to have affirmed before 1910 -- see Pyenson 85, p.21, and
141: references cited there.) Already in 1907 Einstein is striving for a
142: generally covariant theory, but he is not aware that such a theory, the
143: Riemannian geometry, has been created in the 19th century. The (local)
144: equivalence principle purporting to generalize ``the happiest thought'' of
145: 1907 to inhomogeneous fields is instrumental in Einstein's tracing the
146: road to identifying gravity with space time geometry. Yet, nowadays, a
147: mathematically minded student of relativity feels embarrassed if he has to
148: explain what such an ``equivalence principle'' does actually mean. The
149: presence of a non--zero gravitational field strength is manifested by a
150: non--zero curvature tensor and cannot be eliminated -- not even locally --
151: by a coordinate transformation -- no matter what acceleration one chooses.
152: (In order to make our point in a few words we are oversimplifying matters.
153: A detailed treatment of Einstein's principle of equivalence is contained in
154: (Norton 86).)
155:
156: To summarize: by Christmas 1907 Einstein had all physical consequences of
157: the future gravity theory in his hands, yet, he had another eight
158: years to go and to appeal for mathematicians' help before arriving at
159: the proper mathematical formulation of general relativity.
160:
161: At first, though, Einstein behaved much like Michael Atiyah's physicist who
162: ``not being able to solve a problem moves on to the next more difficult one''.
163: For three and a half years, from 1908 to mid 1911, Einstein's main
164: preoccupation was quantum theory: light quanta, blackbody radiation. (In
165: 1910 he also completed a paper on critical opalescence, his last major
166: work on classical statistical physics.) Characteristically this was before
167: the early (1913) work of Niels Bohr after which quantum theory started
168: becoming popular. (Even later, in 1915, Robert Millikan, who spent 10 years
169: to test Einstein's prediction of the photoeffect will write: ``Einstein's
170: photoelectric equation . . . appears in every case to predict exactly the
171: observed results . . . Yet the semicorpuscular theory by which Einstein
172: arrived at his equation seems at present wholly untenable.'' -- see Pais 82,
173: p.357.) Einstein's own appraisal of his 1909--1910 assault on the light
174: quantum problem was not complimentary either. (In today's perspective, though,
175: he appears as a true pioneer in the quantum theory of those days -
176: see, e.g., Sec.2 {\it Einstein and the early quantum theory} in M.J. Klein's
177: lectures ``The Beginning of Quantum Theory'' in {\it History of
178: Twentieth Century Physics}, Proceedings of the International School
179: of Physics ``Enrico Fermi'', Course LVII, ed. by C. Weiner, Academic
180: Press, New York and London 1977, pp.19-39.)
181:
182:
183: By June 1911, after a four month stay in Prague, Einstein is again on the
184: general relativity track and on the bending of light. The next important
185: breakthrough comes a year later, in August 1912 when Einstein is back in
186: Z\"urich and is literally crying for help to his friend and fellow student
187: from ETH: ``Grossmann, you must help me or else I'll go crazy!'' (Since 1907
188: Marcel Grossmann was a professor of geometry in ETH.) As witnessed in
189: Einstein's correspondence (letter to L. Hopf of 16 August, 1912) at this time
190: the two of them understood that gravity should be described not by a single
191: scalar field (which Einstein related in previous publications with the
192: variation of the velocity of light), but by the symmetric tensor metric
193: field $g_{\mu\nu}(x)$ which has 10 independent components. Grossmann
194: quickly realized that the generally covariant formalism Einstein was
195: looking for (equivalence of arbitrarily moving frames) was provided by
196: Riemannian geometry. Yet, this was only the beginning of the hard work.
197: Einstein was absorbed to a point at which he refused to talk about quantum
198: theory anymore. On 29 October 1912 he wrote to Sommerfeld: ``I assure you
199: that with respect to quantum I have nothing new to say . . . I am now
200: exclusively occupied with the problem of gravitation and I hope to master
201: all difficulties with the help of a friendly mathematician here. But one
202: thing is certain: in all my life I have labored not nearly as hard, and I have
203: become imbued with great respect for mathematics, the subtler part of
204: which I had in my simple--mindedness regarded as pure luxury until now.
205: Compared with this problem the original relativity is a child's play.'' (Mehra
206: 73, p.93 and Pais 82, p.216).
207:
208: It appears that in the subsequent months Einstein still trusts better his
209: (uncommon!) physical intuition, rather than the mathematical wisdom. In
210: their first joint paper with the glorious title: ``Entwurf einer
211: verallgemeinerten Relativit\"atstheorie und einer Theorie der Gravitation. I.
212: Physikalischer Teil von Albert Einstein. II. Mathematischer Teil von Marcel
213: Grossmann'' (Leipzig und Berlin, B.G. Teubner 1913; reprinted with added
214: ``Bemerkungen'' in Zeitschrift f\"ur Mathematik und Physik {\bf 62} (1914)
215: 225--261). Grossmann notes in the mathematical part that the Ricci tensor
216: $R_{\mu\nu}$ may be used for the formulation of a generally covariant
217: gravity theory - an important step towards the ultimate formulation of
218: the basic equation of general relativity. (As stressed in (Win 04)
219: Grossmann deserves more credit for this achievement than usually
220: given.) But the authors reject this possibility, since it allegedly
221: violates ``the physical requirements''. The crucial mistake comes from
222: Einstein's ``causality requirement'': the metric tensor $g_{\mu\nu}$
223: should be completely determined from the stress energy tensor. This is
224: certainly not correct for the true equations of general relativity,
225: %
226: $$G_{\mu\nu}=\kappa\ T_{\mu\nu}\quad {\rm where}\quad
227: G_{\mu\nu}:=R_{\mu\nu}-{1\over 2}\ R\ g_{\mu\nu}, \eqno(1)$$
228: %
229: first reported by Hilbert in his paper submitted (to
230: Nach. Ges. Wiss. G\"ottingen), 20 November 1915.
231: Indeed, $G_{\mu\nu}$ satisfies the Bianchi identities
232: $(G^{\mu\nu})_{;\nu}=0$ in accord with the (covariant) energy momentum
233: conservation law. Hence, only six of its ten components are independent, so
234: that the $g_{\mu\nu}$, far from being uniquely determined, depend on four
235: arbitrary functions. We now, sure, understand what it means. General
236: covariance says that the choice of coordinates is a matter of convention
237: which should not affect physics. The metric tensor much like the
238: electromagnetic potential is not an observable. To determine it one needs
239: (on top of $T_{\mu\nu}$) four ``coordinate conditions'' corresponding to the
240: gauge fixing in electrodynamics. It would be too easy to criticize Einstein
241: on the ground of knowledge acquired by physics decades later. (The notion of
242: gauge invariance (Eichinvarianz) first appears six years later in
243: Hermann Weyl's ``Gravitation und Elektrizit\"at''
244: (Sitzungsber. d. Preuss. Akad. d. Wiss. (1918) 465-478)\FOOTNOTE
245: {$^*)$} {After over half a century Paul Dirac (Proc. Roy. Soc. {\bf
246: A333} (1973) 403-418) still views this paper of Weyl as ``unrivalled
247: by its simplicity and beauty''.}\ describing a hypothetical dilation
248: symmetry - in an early attempt to construct a unified field theory. It made
249: its way to where it really belongs -- Maxwell--Dirac electrodynamics --
250: again thanks to Weyl after another 10 years. In 1912--1915 Einstein was
251: well ahead of his time exploring, in the words of Pais, a ``no man's land''.)
252:
253: In short, in the ``Entwurf'' (``Outline'') Einstein and Grossmann back
254: down from general covariance and settle for a set of not quite
255: geometric equations only invariant under linear coordinate
256: transformations. Einstein is not happy with it. In August 1913 he
257: writes to Lorentz: ``The gravitational equations unfortunately do not
258: have the property of general covariance . . . However, the whole faith
259: in the theory rests on the conviction that acceleration of the
260: reference system is equivalent to a gravitational field. Thus, if not all
261: equations of the theory . . . admit transformations other than linear ones,
262: then the theory contradicts its own starting point . . . all is then
263: up in the air'' (Pais 82, p.228).
264:
265: In early 1914 Einstein and Adriaan Focker (who had just received his Ph.D
266: under Lorentz) restored general covariance but at a high price. They derived
267: the scalar equation $R=-\kappa\ T(R=R^\nu\ _\nu ,T=T^\nu\ _\nu )$
268: assuming that the metric is conformally flat,
269: $g_{\mu\nu}=\psi^2\eta_{\mu\nu}$, i.e., returning, essentially, to the
270: scalar theory of gravity. In October 1914 Einstein completes a 56-page long
271: paper ``Die formale Grundlage der allgemeinen Relativit\"atstheorie'' which
272: goes back to the Einstein--Grossmann theory. (Einstein's strength is
273: not in the mathematical formalism: his 1914 treatment of the covariance
274: properties of the field equations, criticized by Levi-Civita, would
275: hardly have impressed educated geometers - see Sect. 4 and footnote 124 of
276: (Sau 99); it does not please its author either.) In the beginning
277: of 1915 Einstein appears fed up (if not fully satisfied) with general
278: relativity, and he goes ahead to do some experimental work with the Dutch
279: physicist Wander de Haas (they discover a new effect: the torque of a
280: suspended iron cylinder as a consequence of an abrupt magnetization).
281: \bigskip
282:
283: \noindent{\bf 2.\ \ Berlin -- G\"ottingen (1915)}\FOOTNOTE {$^*)$} {A
284: well researched and lively account of Einstein's Berlin years is
285: provided in (Goen 05); (Reid 96) is a standard source for Hilbert's
286: life (1862-1943), most of which (since 1895) is spent in G\"ottingen.}\
287:
288: On 29 November 1971 Eugene Wigner writes to Jagdish Mehra asking him one
289: of those questions, people, who have come to know Wigner, can easily
290: imagine: ``. . . I was under the impression that, simultaneously with
291: Einstein, Hilbert also found the now accepted equations of general
292: relativity. Is this correct? If so, is there a reason no one seems to mention
293: this now? I realize that the basic idea was due to Einstein but it is
294: interesting that, even after the promulgation of the basic idea, it took a
295: rather long time to find the correct equations incorporating that idea --
296: even though both Einstein and Hilbert seem to have worked on it.'' Mehra
297: replies to Wigner within two weeks by a long letter and later publishes
298: an 87-page paper on the subject (Mehra 73). But the real answer to
299: Wigner's question comes another five years later in an article by Earman and
300: Glymour who have digged into the Einstein Papers at Princeton University.
301:
302: David Hilbert, whose 23 ``Honors Class'' problems (Yan 02) occupy
303: mathematicians throughout the 20th century, is, in his fifties (after
304: Poincar\'e, 58, dies in 1912), the uncontested leader of the
305: world of mathematics. Having published a (by now, classic) book, ``{\it
306: Foundations of Geometry}'' Hilbert states his sixth problem (Paris,
307: 1900): ``To treat in the same manner, by means of axioms, those
308: physical sciences in which already today mathematics plays an
309: important part...''. His lifelong belief that every scientific
310: problem can - and will - be solved is reflected in the words engraved
311: on his tombstone in G\"ottingen: ``Wir mussen wissen, wir werden
312: wissen.'' (``We must know, we shall know.''). Starting with 1912,
313: after completing his book
314: on linear integral equations, Hilbert's main preocupation becomes
315: mathematical physics: the realization of the program encoded in his
316: sixth problem. He thinks of unifying within the axiomatic approach the
317: new electromagnetic theory of the electron, put forward in 1912 by
318: Gustav Mie (1869-1957), with the Einstein-Grossmann theory
319: (reported to the G\"ottingen Mathematical
320: Society in 1913, shortly after its publication - see (Sau 99),
321: Sect. 2.1). Hilbert tries to have Einstein visiting G\"ottingen
322: (he invites him more than once - first in 1912) and this time
323: he succeeds.
324:
325: In late June - early July 1915 Einstein spends a week in G\"ottingen where (as
326: he witnesses in a letter to Zangger of 7 July) he ``gave six two--hour lectures
327: there''. By all accounts he seems happy with the outcome: ``To my great
328: joy, I succeeded in convincing Hilbert and Klein completely'' (E. to de Haas)
329: ``I am enthusiastic about Hilbert'' (E. to Sommerfeld). The feelings appear
330: to be mutual. Hilbert recommends Einstein for the third Bolyai Prize in
331: 1915 for ``the high mathematical spirit of his achievements'' (the first and
332: the second recipients of the Bolyai prize have been Poincar\'e and Hilbert
333: -- see Mehra 73).
334: Nevertheless, the G\"ottingen discussions seem to have reinforced Einstein's
335: uneasiness about the lack of general covariance of his (and Grossmann's)
336: equations. He is reluctant (he writes to Sommerfeld in July 1915) to
337: include his papers on general relativity in a new edition of ``{\it The
338: Principle of Relativity}'', ``because none of the presentations to date is
339: complete''. After the November race Einstein will state more
340: precisely (in letters to friends) the grounds for his discontent with the old
341: theory: (1) its restricted covariance did not include uniform rotations; (2)
342: the precession of the perihelion of Mercury came out 18$''$ instead of the
343: observed 45$''$ per century; (3) his proof of October 1914 of the uniqueness
344: of the gravitational Hamiltonian is not correct.
345:
346: In the meantime Einstein receives a letter by Sommerfeld (perhaps in late
347: October 1915 -- the letter is lost) from which he learns that he is not the
348: only one dissatisfied with his 1914 theory. Hilbert also has objections to it
349: and is working on his own on ``Die Grundlagen der Physik'' originally
350: conceived as ``Die Grundgleichungen /basic equations/ der Physik'' - see
351: (Sau 99) footnotes 73 and 90). Will Einstein let someone else, be it
352: Hilbert himself, share with him the fruit of years of hard work
353: and great inspiration? Not he! At 36, he can still fight. The Einstein
354: papers reveal an unprecedented activity in November 1915.
355:
356: Einstein submits four communications to the ``Preussische Akademie der
357: Wissenschaften'': on 4, 11, 18 and 25 November -- no Thursday is skipped!
358: These are not different parts of a larger work. The first, ``Zur allgemeine
359: Relativit\"atstheorie'' rejects his formulation of 1914 and proposes a new
360: fundamental equation. The second, with the same title, rejects the first
361: and starts anew. The fourth, ``Die Feldgleichungen der Gravitation'' rejects
362: the first two and finally contains the right equations. It is like in a movie
363: when the film is turned on a high speed. Nothing similar has happened
364: either before or after in Einstein's life.
365:
366: But this is not all. Einstein only answers (the lost) Sommerfeld's
367: letter on 28 November (three days after his last talk at the Academy).
368: ``Don't be angry with me'' he writes ``for only today answering your
369: friendly and interesting letter. But last month I had one of the most
370: exciting, most strenuous times of my life,
371: also one of the most rewarding. I could not concentrate on writing''. Indeed,
372: from late October to late November Einstein stops writing to any of his
373: habitual addressees: Besso, Ehrenfest, Lorentz, . . . But he does write
374: letters (or, rather, postcards). He only replaces all his regular
375: correspondents by a single new one -- Hilbert. Four postcards are preserved
376: from Einstein to Hilbert dated 7, 12, 15, 18 November and two of the four
377: Hilbert answers.
378:
379: On 7 November Einstein sends to Hilbert the proofs of his
380: November-four paper and in the accompanying card writes
381: ``I recognized four weeks ago that my
382: earlier methods of proof were deceptive''. He alludes to the above mentioned
383: letter of Sommerfeld which reports on Hilbert's objections to the October
384: 1914 paper; and closes by saying: ``I am curious whether you will be well
385: disposed towards this solution''.
386:
387: Hilbert would have hardly been well disposed towards the new equation,
388: since it assumes that the determinant of the metric tensor is a constant
389: (-1) and is hence still not generally covariant. Probably, after having
390: Hilbert's criticism (which has been lost) Einstein opted on 11 November for
391: the generally covariant equation
392: %
393: $$R_{\mu\nu}=\kappa\ T_{\mu\nu} \eqno(2)$$
394: %
395: which Grossmann and he have rejected two years earlier. It only coincides,
396: however, with the correct equation (1) if $T_{\mu\nu}$ (and hence also
397: $R_{\mu\nu}$) is traceless. This is the case of Maxwell electrodynamics
398: and Einstein speculates that it may be more general.
399:
400: The next day, 12 November, Einstein sends a second postcard to Hilbert
401: announcing that he had finally achieved generally covariant field equations.
402: He also thanks Hilbert for his ``kind letter'' (which is lost). Hilbert replies
403: on 14 November a long message on two postcards. He is excited about his
404: own ``axiomatic solution of your grand problem''. In a postscript Hilbert adds
405: that his theory is ``wholly distinct'' from Einstein's and invites Einstein to
406: come to G\"ottingen and hear his lecture on the subject. The tone is cordial:
407: Hilbert urges Einstein to come to G\"ottingen the day before the lecture and
408: pass the night at Hilbert's home. The next day, Monday, 15 November,
409: Einstein already answers Hilbert's cards. (One cannot fail to notice how
410: accurately the mail service is working in Germany in the midst of the
411: European war.) ``The indications on your postcards lead to the greatest
412: expectations''. He apologizes for his inability to attend the lecture,
413: since he is overtired and bothered by stomach pains.
414: Asks for a copy of the proofs of
415: Hilbert's paper. Apparently, he does receive the requested copy within
416: three days, because on 18 November, the day of his third talk at the
417: Academy, Einstein writes his fourth postcard: ``The system [of equations]
418: given by you agrees -- as far as I can see -- exactly with what I found in
419: recent weeks and submitted to the Academy''. Then Einstein remarks that he
420: has known about Eq.(2) ``for three years'' but that he and Grossmann have
421: rejected it on the grounds that in the Newtonian limit they are not
422: compatible with ``Newton's law'' (meaning Poisson's field equation). Finally,
423: Einstein informs Hilbert that he is finally explaining the advance of
424: the perihelion of Mercury from general relativity alone without the aid of
425: any subsidiary hypotheses.
426:
427: Two remarks are in order.
428:
429: First, it is not true that Hilbert's Eq.(1) is equivalent to Einstein's
430: Eq.(2) of the paper submitted to the Academy on 11 November. (It will be
431: equivalent to the equation Einstein is going to write a week later. It seems,
432: however, that Einstein does have in mind his Eq.(2) in this postcard since he
433: is adding the priority claim that he knew it for three years.) The two
434: equations are only consistent with one another for $T(=T^\nu_\nu )=0$,
435: the case Einstein has been mostly interested in at the time.
436:
437: Second, Einstein does derive the correct value for the advance of the
438: perihelion of Mercury in his third communication ``Erkl\"arung der
439: Perihelbewegung des Merkur aus der allgemeinen Relativit\"atstheorie''
440: from his not exactly correct equation. This is possible since he is
441: actually solving the homogeneous equation (with $T_{\mu\nu}=0$) in the
442: post Newtonian approximation (allowing for point singularities). - In
443: seeing the physical implications of the theory Einstein has no competitor.
444:
445: The next day, Friday the 19th, Hilbert congratulates Einstein for having
446: mastered the perihelion problem and adds cheerfully: ``If I could calculate
447: as quickly as you, then the electron would have to capitulate in the face of
448: my equations and at the same time the hydrogen atom would have to offer
449: its excuses for the fact that it does not radiate'' (Pais 82, p.260).
450:
451: On 20 November Hilbert presents to the Gesellschaft der Wissenschaften in
452: G\"ottingen his work. He {\it derives} the correct equations from the
453: variational principle assuming general covariance (we would say today
454: reparametrization invariance) and a second order equation for
455: $g_{\mu\nu}$. He gives full credit to Einstein's ideas. On the first page of
456: his article he writes: ``Einstein . . . has brought forth profound thoughts and
457: unique conceptions, and has invented ingeneous methods for dealing with
458: them . . . Following the axiomatic method, in fact from two simple axioms, I
459: would like to propose a new system of the basic equations of physics. They
460: are of ideal beauty and I believe they solve the problems of Einstein and Mie
461: at the same time''. In the published version Hilbert refers to all Einstein
462: November papers. About the one of 25 November, submitted after his talk,
463: he says: ``It seems to me that [our] differential equations of gravitation are
464: in agreement with the noble theory of general relativity proposed by
465: Einstein in his later memoire''.
466:
467: On 25 November Einstein proposes {\it without derivation} the equation
468: %
469: $$R_{\mu\nu}=\kappa \left( T_{\mu\nu}-{1\over 2}\ T\ g_{\mu\nu}
470: \right) \eqno(3)$$
471: %
472: which is exactly equivalent to Hilbert's Eq.(1), since they both imply
473: $R+\kappa\ T=0$. He chooses not to mention Hilbert's name in the published
474: paper. Later commentators have a hard time to understand what was
475: Einstein's argument at the time to include the trace term. Only Norton
476: makes a well documented (59 pages long) case
477: (including the study of a Z\"urich notebook
478: of Einstein) for an independent Einstein's road to the correct equations.
479: \bigskip
480:
481: \noindent{\bf Aside:}\quad
482: Today's student will easily find the $-{1\over 2}\ R$ term (or equivalently,
483: the $-{1\over 2}\ T$ term) using the Bianchi identity. The trouble was,
484: Einstein did not know them. We should not be too hard on him on that
485: account. Hilbert, too, does not know them: he derives four identities
486: among the fields in his theory - anticipating, three years in advance,
487: a special case of Noether's theorem (Viz 94) (complemented with a
488: stronger statement - see (Sau 99), Sect. 3.3 and footnote 120) but
489: he conjectures
490: erroneously that they will enable him to express the electromagnetic
491: potential in terms of the gravitational field. He corrects his error in a
492: later version of the paper.
493:
494: Felix Klein who (in 1918) reduces the vanishing of the covariant
495: divergence,
496:
497: %
498: $$\left( R^{\mu\nu}-{1\over 2}\ g^{\mu\nu}\ R\right)_{;\nu}=0,$$
499: %
500: to Noether's theorem as well, does not realize that it is a
501: consequence of
502: the Bianchi identities for the Riemann curvature tensor either.
503: ``Bianchi identities'' are known before Bianchi to
504: Aurel Voss (1880) and to Ricci (1889); Luigi Bianchi (a pupil of Klein's!)
505: rediscovers them in 1902.
506: \bigskip
507:
508: \noindent{\bf 3.\ \ Aftermath. Controversy among historians of science}
509:
510: In his speech on the occasion of Planck's 60$^{th}$ birthdate (in 1918)
511: Einstein talks about different categories of people that have devoted
512: themselves to science. For some science is a sport which allows them to
513: satisfy their pride or vanity. If the angel chases all such people from the
514: temple of science, he continues, then the temple would remain almost
515: empty, but Planck will be among the precious few who will remain.
516:
517: The chronicle of the last month of the creation of general relativity
518: demonstrates that the spirit of competition has not been alien to Einstein
519: himself (as it was not to Leibniz and Newton). It is to the credit of both
520: Hilbert and Einstein that their November 1915 rivalry did not grow into a
521: public argument. Yet the outcome of the November events resulted in some
522: uneasy feelings between the two men. On 20 December 1915, Einstein writes
523: to Hilbert: ``I want to take the opportunity to say something to you which is
524: important to me. There has been a certain spell of coolness between
525: us, the cause of which do not want to analyze. I have, to be sure,
526: struggled against any resentment, and with complete success. I
527: think of you once again with untroubled friendliness, and ask you to
528: try to think of me in the same way. It is really
529: a shame when two such real fellows [zwei wirkliche Kerle], whose work
530: has taken them above the shaby world, give one another no pleasure''. (EG
531: 78, p.306; a slightly different reading of the German original the reader
532: will find in Pais 82, p.260.)
533:
534: In his expository paper ``Die Grundlage der allgemeinen
535: Relativit\"atstheorie'', Annalen der Physik {\bf 49}, 769-822 (received 20
536: March, 1916) Einstein already refers (albeit superficially) to Hilbert's work.
537: In May 1916 he gives a colloquium in Berlin on Hilbert's paper. On that
538: occasion he again writes to Hilbert asking him to
539: explain his work (and complaining about its obscurity).
540:
541: Hilbert's appreciation of Einstein appears unequivocal. His biographer,
542: (Reid 96), attributes to him the words: ``Every boy in
543: the streets of G\"ottingen understands more about four--dimensional
544: geometry than Einstein. Yet, . . . Einstein did the work and not the
545: mathematicians''.
546:
547: The story does not end here, however: it is continued by the next
548: generation of Einstein biographers and students of science history.
549:
550: In 1997 a noteworthy addition to existing Einstein's biographies, (FL 97),
551: appeared in English, providing a nice complement to (Pais 82).
552:
553: Summing up the decisive phase of his work on general relativity
554: (Fl 97) quotes Einstein's letter to Heinrich Zangger (see also an
555: earlier discussion of this letter in (Med 84)) which says:
556: ``{\it Only one colleague truely understood it, and he
557: now tries skillfully to 'nostrify' it}'' [i.e. appropriate ('make it ours')].
558: We already know that the colleague in question was none other than
559: David Hilbert. F\"olsing justly refutes the accusation on the basis
560: of available evidence.
561:
562: Later the same year an article in the 14 November issue of {\it
563: Science}, (CRS 97) made the news. This paper has a direct bearing on
564: our topic. It points out that a lately discovered proof-sheet of
565: Hilbert's paper, with a publisher's stamp of 6 December 1915, i.e. after the
566: publication of the fourth of Einstein's communications, involves substantial
567: changes in the manuscript. The fact that Hilbert modified his paper
568: after its submission has been known before: as we noted he had cited
569: all four Einstein's November papers and had commented on
570: the last one (submitted after his) in the published version of his
571: November 20 article. The authors strive to attribute a
572: great significance to the fact that the original text only involves
573: the Hilbert action, while the field equations, which are derived from
574: it, appear to be first inserted at the stage of the proofreading. Their
575: attempt to support on this ground Einstein's accusation of
576: ``nostrification'' goes much too far. A calm, non-confrontational
577: reaction was soon provided by a thorough study (Sau 99) of Hilbert's
578: route to the ``Foundations of Physics'' (see also the relatively even
579: handed survey (Viz 01)). A direct critical comment on the unfounded
580: accusations in (CRS 97), (Win 04), originally rejected by the editors
581: of {\it Science}, \FOOTNOTE {$^*)$}
582: {There seems to be a concerted effort to present the view of (CRS 97)
583: as a final generally accepted ``Decision''. In a little more than a
584: 3-page long article on Hilbert (by J.J. O'Connor and E.F. Robertson),
585: available at
586: http://www-groups.dcs.st-and.ac.uk/~history/Mathematicians/Hilbert.html,
587: the authors have found it necessary to devote a
588: paragraph citing the (CRS 97) accusation of inappropriate
589: behaviour against Hilbert, preceded by the words ``the authors /of
590: (CRS 97)/ show convincingly ...''. It is this type of overtly prejudiced
591: attitude that provokes uncommonly angry reactions as (LMP 04) and
592: gives credibility to extremist publications as (Bje 03).}\
593: finally appears in a more specialized journal (Win 04).
594: \FOOTNOTE {$^**)$} {In their
595: ``Response'' (CRS 04) to this comment the authors of (CRS 97)
596: continue to assert that taking the variational
597: derivative of the Hilbert action (a routine 3-line exercise for an
598: average graduate student) is something Hilbert was not able to do by
599: himself in 1915, and even compare it with the calculation of ``the
600: one-billionth digit of {$\pi$}'' (that would require a supercomputer
601: and the dedication of someone - like the Chudnovsky brothers (see
602: R. Preston, The Mountains of PI, {\it New Yorker}, March 2, 1992) -
603: to program it).}\
604:
605: The polemics is getting rough. A new book, (Wuensch 05), is advertised with
606: a question mark: ``Ein Kriminalfall in der
607: Wissenschaftsgeschichte?'' (``A criminal case in the history of
608: science?''). The author asserts - already in the abstract to the book
609: - that a missing fragment (also discussed in (Sau 99) and in (Win 04))
610: of the text on
611: pages 7 and 8 of Hilbert's proof-sheets, used in (CRS 97), contained
612: ``in all probability ... the explicit form of the field equations...''
613: She further argues that
614: ``the passage ... was not excised originally but rather ... it must
615: have been deliberately removed in more recent times in order to
616: falsify the historical truth.''
617:
618: It is quite clear from the November correspondence (and from recently
619: discovered letters of Max Born to Hilbert of the fall of 1915 - see
620: (Som 05)) - without appealing to criminal proceedings - that Hilbert's
621: competitive influence was crucial for Einstein's acceptance of general
622: covariance - in spite of his long time reservations and doubts. The
623: analysis of the new evidence, detailed in (Sau 99) and in (Viz 01),
624: indicates, on the other hand, that Hilbert appears to have been misled
625: for a while, during the final race, in the opposite direction.
626: After formulating the generally covariant action principle he appeals,
627: in his original text, to Einstein's long-promoted ``causality
628: principle'' and restricts the general covariance by a (non-covariant
629: formulation of) the energy momentum conservation law. Only at the
630: stage of proofreading does Hilbert suppress all extra conditions and
631: recognize the unqualified physical relevance of the covariant equation (1).
632:
633: Einstein and Hilbert had the moral strength and wisdom - after a
634: month of intense competition, from which, in a final account,
635: everybody (including science itself) profited - to avoid a lifelong
636: priority dispute (something in which Leibniz and Newton failed). It
637: would be a shame to subsequent generations of scientists and
638: historians of science to try to undo their achievement.
639:
640: \ACK
641:
642: An early version of this Colloquium lecture was held at the
643: International Centre for Theoretical Physics (Miramare-Trieste) in
644: December 1992. Its text (which appeared as an ICTP internal report,
645: IC/90/421) was prepared
646: while the author was visiting at the Laboratorio Interdisciplinare
647: per le Scienze Naturali ed Umanistiche of the International School
648: for Advanced Studies (SISSA/ISAS) in Trieste. The present extended
649: version of the March 15, 2005 Colloquium talk, held at the Physics
650: Department of the International University Bremen, was written during
651: the author's visit as an Alexander von Humboldt awardee at the
652: Institut f\"ur Theoretische Physik, Universit\"at G\"ottingen. The
653: author is grateful to all these institutions for their hospitality
654: and support. It is a pleasure to thank Hubert Goenner for providing
655: relevant recent references (including copies of unpublished Born's
656: letters to Hilbert) and Karl-Henning Rehren for a critical reading of
657: the manuscript.
658:
659: \bigskip
660:
661: \vfill\eject
662:
663: \centerline {REFERENCES}
664:
665: \noindent (Bje 03) C.J. {\it Bjerkenes}, {\it Anticipations of Einstein
666: in the General Theory of Relativity} (XTX Inc. Downers Grove, Illinois
667: 2003).
668:
669: \noindent (CRS 97) Leo {\it Corry}, J\"urgen {\it Renn}, John {\it
670: Stachel}, Belated decision in the Einstein-Hilbert priority dispute,
671: {\it Science} {\bf 278} (1997) 1270-1273.
672:
673: \noindent (CRS 04) Leo {\it Corry}, J\"urgen {\it Renn}, John {\it
674: Stachel}, Response to F. Winterberg, ``On 'Belated decision in the
675: Hilbert-Einstein priority dispute' published by L. Corry, J. Renn, and
676: J. Stachel'', anounced in a ``Retraction Notice'', {\it
677: Z. Naturforsch.} {\bf 59a} (2004) 1004 at
678: http:www.mpiwg-berlin.mpg.de/texts/Winterberg-Antwort.html.
679:
680: \noindent (EG 78) John {\it Earman}, Clark {\it Glymour}, Einstein and
681: Hilbert: Two months in the history of general relativity, {\it Arch.
682: Hist. Exact Sci.} {\bf 19} (1978) 291--308.
683:
684: \noindent (Fl 97) A. {\it F\"olsing}, {\it Albert Einstein: A
685: Biography} (Viking, N.Y. 1997).
686:
687: \noindent (Goen 05) Hubert {\it Goenner}, {\it Einstein in Berlin
688: 1914-1933} (C.H. Beck, M\"unchen 2005).
689:
690: \noindent (LMP 04) A.A. {\it Logunov}, M.A. {\it Mestvirishvili},
691: V.A. {\it Petrov}, How were the Hilbert-Einstein equations discovered?
692: {\it Uspekhi Fizicheskikh Nauk} {\bf 174} (2004) 663-621 (English
693: translation: {\it Physics-Uspekhi} {\bf 47} (2004) 607-621); arXiv:
694: physics/0405075.
695:
696: \noindent (Med 84) H.A. {\it Medicus}, A comment on the relations
697: between Einstein and Hilbert, {\it Am. J. Phys.} {\bf 52}:3 (1984)
698: 206-208.
699:
700: \noindent (Mehra 73) Jagdish {\it Mehra}, Einstein, Hilbert and the theory of
701: gravitation in: {\it The Physicist's Conception of Nature}, edited by Jagdish
702: Mehra (D. Reidel Publ. Co., Dordrecht--Holland, Boston, USA 1973)
703: pp. 92--178.
704:
705: \noindent (Norton 86) John {\it Norton}, What was Einstein's principle of
706: equivalence? pp. 5--47;
707: How Einstein found his field equations, 1912--1915, pp. 101--159, in {\it
708: Einstein and the History of General Relativity}, Don Howard, John Stachel,
709: eds., Einstein Studies, Vol.1 (Birkh\"auser, Boston, Basel, Berlin 1989).
710: Based on the Proceedings of the 1986 Osgood Hill Conference, North
711: Andover, Massachusetts.
712:
713: \noindent (Pais 82) Abraham {\it Pais}, `{\it Subtle is the Lord . . .}', The
714: Science and the Life of Albert Einstein (Clarendon Press, Oxford 1982).
715:
716: \noindent (Pyenson 85) Lewis {\it Pyenson}, {\it The Young Einstein}, The
717: advent of relativity (Adam Hilger Ltd., Bristol and Boston 1985).
718:
719: \noindent (Reid 96) Constance {\it Reid}, {\it Hilbert} (Springer,
720: Berlin et al. 1996).
721:
722: \noindent (Sau 99) Tilman {\it Sauer}, The relativity of discovery:
723: Hilbert's first note on the Foundations of Physics, {\it Arch. Hist.
724: Exact Sci.} {\bf 53} (1999) 529-575; arXiv:physics/9811050.
725:
726: \noindent (Som 05) Klaus P. {\it Sommer} ``Nicht das Deutschland von
727: Hindenburg und Ludendorff'', sondern das von Hilbert und Einstein. Der
728: Fund von Briefen von Einstein, Planck, Born, Debye, Nernst,
729: Sommerfeld, Ehrenfest, Weyl, Courant und Althoff an David Hilbert auf
730: einem G\"ottinger Dachboden, {\it Berichte f\"ur
731: Wissenschaftsgeschichte} (to be published).
732:
733: \noindent (Viz 01) V.P. {\it Vizgin}, On the discovery of the
734: gravitational field equations by Einstein and Hilbert: new material,
735: {\it Uspekhi Fizicheskikh Nauk} {\bf 171} (2001) 1347-1363 (English
736: translation: {\it Physics-Uspekhi} {\bf 44} (2001) 1283-1298).
737:
738: \noindent (Viz 94) V.P. {\it Vizgin}, {\it Unified Field Theories in
739: the First Third of the 20th Century} (Birkh\"auser, Basel et al. 1994).
740:
741: \noindent (Win 04) F. {\it Winterberg}, On ``Belated decision in the
742: Hilbert-Einstein priority dispute'' published by L. Corry, J. Renn,
743: and J. Stachel, {\it Z. Naturforsch.} {\bf 59a} (2004) 715-719.
744:
745: \noindent (Wuensch 2005) Daniela {\it Wuensch}, {\it ``zwei wirkliche
746: Kerle''} Neues
747: zur Entdeckung der Gravitationsgleichungen der Allgemeinen
748: Relativit\"atstheorie durch Albert Einstein und David Hilbert
749: (Termessos Verlag, G\"ottingen 2005).
750:
751: \noindent (Yan 02) Benjamin H. {\it Yandell}, {\it The Honors Class:
752: Hilbert's Problems and Their Solvers} (A.K. Peters, Natik, MA 2002).
753:
754:
755:
756: \vfill\eject
757: \bye
758: