Hilbert's ‘Foundations of Physics’: Gravitation and electromagnetism within the axiomatic method
Introduction
In November and December 1915, Hilbert gave two presentations to the Royal Göttingen Academy of Sciences under the common title ‘The Foundations of Physics’. Distinguished as ‘First Communication’ (Hilbert, 1915b) and ‘Second Communication’ (Hilbert, 1917), two papers (or ‘notes’, as they are widely known) eventually appeared in the Nachrichten of the Academy.1 The First Communication, which quickly entered the canon of classical general relativity, has recently become the object of renewed scholarly scrutiny since the discovery of a set of printer's proofs (Hilbert, 1915a) dated 6 December 1915 (henceforth the ‘Proofs’).2 With the exception of Renn and Stachel (1999), the Second Communication has not been given the same detailed reconsideration. However, the analysis of the Second Communication by Renn and Stachel seriously misrepresents its aims, content, and significance, and also its links to the First Communication. Our aim in this paper is to show that Hilbert's Second Communication is a natural continuation of his First Communication, that it contains important and interesting further developments of that project, and above all that it sheds needed illumination on Hilbert's assessment of the epistemological novelty posed by a generally covariant physics.
Hilbert's notes on ‘Foundations of Physics’ traditionally have been assessed solely in terms of the contributions they made to general relativity, as that theory is known in its completed form.3 From this vantage point, they present a mixed record of achievement, ranging from genuine insight (the Riemann scalar as the suitable invariant for the gravitational action) through incomprehension (Hilbert's interpretation of electromagnetism as a consequence of gravitation) to abject failure (attachment to the untenable electromagnetic theory of matter of Gustav Mie). The usual implication is that Hilbert's principal intent in November 1915 was to arrive at a theory of gravitation based on the principle of general covariance in one blinding flash, masterfully wielding an arsenal of axiomatized advanced mathematics. Thus arose ‘the legend of a royal road to general relativity’ (Renn & Stachel, 1999, p. 1) through the axiomatic method, whilst Hilbert's reputed remark that ‘physics is much too difficult for physicists’ has been widely understood to epitomize a haughty mathematical arrogance (Reid, 1970, p. 127). Correspondingly, some historians of general relativity have concluded that Hilbert ‘attached a kind of metaphysical significance to variational methods’ (Rowe, 1999, p. 201), while others have regarded Hilbert's approach as evincing an optimistic demonstration of the Göttingen-based ideology of a ‘pre-established harmony’ between mathematics and physics (Pyenson, 1985).4 Finally, some have relished pointing out where Hilbert's elaborate mathematical constructions were either inadequate to the complexities of the initial value problem in general relativity5 or simply led to hopelessly failed physics (Stachel, 1992; Renn & Stachel, 1999, pp. 77, 81–83).
Our contention is that viewing Hilbert's notes solely in terms of contributions made to general relativity as that theory is canonically understood, radically occludes internal motivations, which are largely logical and epistemological, and so casts them in a misleading light. In so doing, the explicitly stated epistemological intent of the ‘axiomatic method’ is willfully ignored, as are Hilbert's own express assertions regarding his construction as a triumph of that method. Although understandable in terms of the intellectual small change of ‘textbook’ histories, such accounts overlook or downplay fundamental philosophical and methodological differences with Einstein, in emphasis as well as in detail, concerning the significance of general covariance, arguably impeding clarification on that vexed issue for decades. But set within the logical and epistemological context of the ‘axiomatic method’, Hilbert's two notes may be seen to have the common goal of pinpointing, and then charting a path toward resolution of, the tension between causality and general covariance that, in the infamous ‘hole argument’, had stymied Einstein from 1913 to the autumn of 1915.6 Unlike Einstein's largely informal and heuristic extraction from the clutches of the ‘hole argument’, Hilbert stated the difficulty in a mathematically precise manner as an ill-posed Cauchy problem in the theory of partial differential equations, and then indicated how it can be resolved. As we will show, material cut from the proofs establishes this essential thematic linkage between the two notes and redeems Hilbert's claim that tension between causality and general covariance, precisely formulated in Theorem I of the First Communication, was the ‘point of departure’ for his axiomatic investigation.
Einstein and Hilbert were engaged in qualitatively different enterprises that only partially overlapped. In contrast to Einstein, Hilbert's goals were at least as much logical and epistemological, according to the character of the axiomatic method, as they were physical. We concur with the judgment of Felix Klein, who wrote, in 1921, that ‘there can be no talk of a question of priority, since both authors pursued entirely different trains of thought (and to be sure, to such an extent that the compatibility of the results did not at once seem assured)’.7 For Hilbert, the principal outcomes arrived at by the axiomatic method concern his revisiting the principle of causality, and his revisions of Kantian epistemology, in the light of generally covariant physics.
The structure of our paper is as follows.
In Section 2 we present what we call ‘the essential context’: Hilbert's axiomatic method and its presupposition of central tenets of Kantian epistemology.
In Section 3, we briefly review the published version of Hilbert's First Communication, proceeding in Section 4 to emphasize differences in content between the December Proofs and the published version. Since this subject has been extensively treated in Sauer (1999), in Renn and Stachel (1999), and more recently in Corry (2004), our treatment will highlight only the central features, omitting many details that can be found in these sources. We shall see that a passage cut from the proofs elucidates the problem pinpointed by Theorem I, and this, we claim, provides essential thematic linkage to Hilbert's Second Communication (see Section 6). Section 5 returns to the topic of the axiomatic method, and examines the aims and achievements of this method as it appears in the First Communication.
Section 6 concerns Hilbert's Second Communication. Following a brief introduction, we review the secondary literature, and then turn our attention to an exposition of the content of the Second Communication. We see how Hilbert now sought to resolve the challenge posed by Theorem I—the tension between general covariance and causality. We show that Hilbert's employment of the axiomatic method identified an epistemological novelty emerging in generally covariant physics regarding the constitution of physical objectivity as this is understood in a broadly transcendental idealist sense, most prominently displayed in his subsequent remarks regarding the axiom of general invariance. In further addressing the related matter of the vexing problem of causality in the new physics of general covariance, deemed a ‘pseudo-geometry’ and not, as before, as field physics set within a ‘background geometry’, Hilbert sought to remove all global ‘pseudo-Euclidean’ presuppositions, equivalent to ‘action-at-a-distance’.
Armed with our new understanding of Hilbert's ‘problem of causality’, Section 7 explains why this is not the same problem as Einstein faced in his ‘hole argument’.
In Section 8, we look beyond Hilbert's 1915 and 1917 papers to consider his further reflections on the epistemological significance of the respective principles of general covariance and causality and the ground of Hilbert's subordination of the latter to the former. We explain in detail the revisions of Kant that Hilbert believed were required in the face of the new generally covariant physics.
Section snippets
The essential context: Hilbert's axiomatic method and Kantian epistemology
There are two pieces of context that we believe are crucial to correctly understanding Hilbert's treatment of generally covariant physics: his axiomatic method, and his appeal to Kantian epistemology. These themes infuse our discussion throughout.
Hilbert's First Communication opens with a declaration that his investigation of the foundations of physics is undertaken ‘in the sense of the axiomatic method’ (‘im Sinne der axiomatischen Methode’), and it concludes with the striking claim that the
Hilbert's First Communication on ‘The Foundations of Physics’ (published version)
According to the annotation on the published version of the paper, Hilbert's First Communication was presented at the 20 November 1915 session of the Royal Göttingen Academy of Sciences. Traditionally, the date of submission was the only date appearing on publications in the Nachrichten of the Academy (Rowe, 2001, p. 418). However, with the discovery of the December Proofs in 1993 it was learned that the version submitted on 20 November differs considerably from that appearing in the published
Comparison of the Proofs with the published version of the First Communication
Felix Klein is reported to have commented on the ‘completely disordered’ character of Hilbert's First Communication, remarking that it was ‘evidently a product of great exertion and excitement’).34 We do not know whether Klein was referring to the Proofs or the published version, or perhaps to both, but there is evidence that Klein and Einstein each encountered
The First Communication and the axiomatic method
We recall that the task of the axiomatization of physics was the sixth in the famous list of 23 mathematical problems Hilbert posed at the International Congress of Mathematicians in Paris in 1900.
The investigations on the foundations of geometry suggest the problem: To treat in the same manner, by means of axioms, those physical sciences in which mathematics plays an important part …. If geometry is to serve as a model for the treatment of physical axioms, we shall try first by a small number
Hilbert's Second Communication on ‘The Foundation of Physics’
On 4 December 1915, ‘The Foundation of Physics (Second Communication)’ was presented to the Göttingen Academy. A further presentation under that rubric was delivered on 26 February 1916. However, both of these were withdrawn before publication, and no version of either has apparently survived (Sauer, 1999, p. 557 and n. 120, p. 560 and n. 129; 2001, p. 3). From Hilbert's correspondence, it may be conjectured that one principal topic concerned a much anticipated derivation of ‘the electron’ from
Hilbert and the ‘hole argument’
The tension between general covariance and causality is just the problem encountered by Einstein in the ‘hole argument’. But in Hilbert's hands, is there any reason to think that, as with Einstein, the difficulty arose from wrongly attributing physical meaning to the spacetime coordinates? That Hilbert's four non-covariant energy equations do implicate such a misunderstanding is affirmed by Renn and Stachel (1999, pp. 77, 83):
… we will discuss Hilbert's treatment of the problem of causality in
Hilbert's revision of Kant in the light of general invariance
In the context of a discussion of Hilbert's views on foundations of mathematics, Peckhaus (1994, p. 91) has recently remarked that
However, this statement should be qualified, for its truth largely depends on what is meant by ‘Kant's philosophy’. To many of the neo-Kantians active in the first quarter of the twentieth century, Kant's philosophy was largely a work in progress, not a finished edifice, and nearlyHilbert's preference for Kantianism was largely incompatible with Kant's philosophy.
Conclusions
Theorem I of Hilbert's First Communication, based on the axiom of general invariance, provided a purely mathematical diagnostic of the tension between causality and general covariance appearing as a problem of causal underdetermination. The material cut from the Proofs is his first attempt at resolving this tension.
The Second Communication (1917) presents Hilbert's later, and different, attempt at a resolution. For Hilbert, general covariance is a refinement, first identified as such by
Acknowledgments
We would like to thank Carlos Casanova, Sol Feferman, Brandon Fogel, Michael Friedman, Ulrich Majer, Brian Pitts, Tilman Sauer, and Joseph Zepeda, for their comments and suggestions. We are especially grateful to two anonymous referees for their detailed comments and advice. Material based on this research was presented at the Philosophy of Physics Research Seminar, Oxford University (2002); HOPOS (2004); the Logical Methods Workshop, Stanford University (2005); the Belgian Society for Logic
References (102)
- et al.
Einstein and singularities
Studies in History and Philosophy of Modern Physics
(1999) Electromagnetism as an aspect of geometry? Already unified field theory—The null field case
Annals of Physics
(1966)Energy conservation in GTR
Studies in History and Philosophy of Modern Physics
(2000)- et al.
Introduction to general relativity
(1975) Der Impuls-Energie-Satz in der Elektrodynamik von Gustav Mie, Nachrichten von der Königliche Gesellschaft der Wissenschaften zu Göttingen
Mathematisch-Physikalisch Klasse
(1914)Hilbert und die Physik
Die Naturwissenschaften
(1922)A note on general relativity, energy conservation, and Noether's theorems
- et al.
General covariance from the perspective of Noether's theorems
Diálogos
(2002) Metaphysics and the philosophy of science: The classical origins from Descartes to Kant
(1969)- Corry, L. (1999). Hilbert and physics (1900–1915). In Gray (1999, pp....
David Hilbert and the axiomatization of physics (1898–1918): From Grundlagen der Geometrie to Grundlagen der Physik
Belated decision in the Hilbert–Einstein priority dispute
Science
Bangs, crunches, whimpers, and shrieks: Singularities and acausalities in relativistic spacetimes
Einstein and Hilbert: Two months in the history of general relativity
Archive for History of Exact Sciences
Entwurf einer verallgemeinerten Relativitätstheorie und einer Theorie der Gravitation
The early interpretation of the Schwarzschild solution
The symbolic universe: Geometry and physics 1890–1930
The Hilbert challenge
Contribution to the history of Einstein's geometry as a branch of physics
Hilbert's axiomatic method and the laws of thought
David Hilbert's lectures on the foundations of geometry 1891–1902
Die Grundlagen der Physik. Typescript of summer semester lecture notes
Neubegründung der Mathematik. Erste Mitteilung
Abhandlungen aus dem mathematischen Seminar der Hamburgischen Universität
Die Grundlagen der Physik
Mathematische Annalen
Über das Unendliche
Mathematische Annalen
Gesammelte Abhandlungen. Bd. III
Out of the labyrinth? Einstein, Hertz, and the Göttingen answer to the hole argument
Untying the knot: How Einstein found his way back to field equations discarded in the Zurich notebook
Cited by (20)
Hilbert-style axiomatic completion: On von Neumann and hidden variables in quantum mechanics
2022, Studies in History and Philosophy of ScienceWhy history matters to philosophy of physics
2015, Studies in History and Philosophy of Science Part AErich Kretschmann as a proto-logical-empiricist: Adventures and misadventures of the point-coincidence argument
2013, Studies in History and Philosophy of Science Part B - Studies in History and Philosophy of Modern PhysicsCitation Excerpt :in Howard & Norton, 1993, 46). Hertz's suggestion probably reflects a familiarity in the use of arbitrary coordinates, which might have been widespread in the Göttingen community formed around David Hilbert (Brading & Ryckman, 2008; Howard & Norton, 1993), who, at about that time had started to work intensively on Einstein's theory (Einstein had just visited Göttingen between June and July of 1915; Corry, 2003). It is hard to establish precisely when Einstein came to realize that two sets of coefficients which are “developable onto another”—that is, that differ only by a coordinate transformation—represent the same physical situation.
Tuning up mind's pattern to nature's own idea: Eddington's early twenties case for variational derivatives
2010, Studies in History and Philosophy of Science Part B - Studies in History and Philosophy of Modern PhysicsThe relativity of inertia and reality of nothing
2010, Studies in History and Philosophy of Science Part B - Studies in History and Philosophy of Modern PhysicsEinstein Studies, volume 11: A retrospective review
2008, Studies in History and Philosophy of Science Part B - Studies in History and Philosophy of Modern Physics