Einstein and the Development of Twentieth-Century Philosophy of Science. - by Don Howard 2014 - Article review

This document contains article review "Einstein and the Development of Twentieth-Century Philosophy of Science." by Don Howard written in 2014
To order to read the article select: https://www.academia.edu/1294457/_Einstein_and_the_Development_of_Twentieth_Century_Philosophy_of_Science_In_the_Cambridge_Companion_to_Einstein_Michel_Janssen_and_Christoph_Lehner_eds?

Contents

Reflection

Einstein and the Development of Twentieth-CenturyPhilosophy of Science Don HowardUniversity of Notre Dame

I. Introduction

What is Albert Einstein’s place in the history of twentieth-century philosophy of science? Were one to consult the histories produced at mid-century from within the Vienna Circle and allied movements (e.g., von Mises 1938, 1939, Kraft 1950, Reichenbach 1951), then one would find, for the most part, two points of emphasis. First, Einstein was rightly remembered as the developer of the special and general theories of relativity, theories which, through their challenge to both scientific and philosophical orthodoxy made vivid the need for a new kind of empiricism (Schlick 1921) whereby one could defend the empirical integrity of the theory of relativity against challenges coming mainly from the defenders of Kant. Second, the special and general theories of relativity were wrongly cited as straight forwardly validating central tenets of the logical empiricist program,such as verificationism, and Einstein was wrongly represented as having, himself, explicitly endorsed those same philosophical principles. As we now know, logical empiricism was not the monolithic philosophical movement it was once taken to have been. Those associated with the movement disagreed deeply about fundamental issues concerning the structure and interpretation of scientific theories, as in the protocol sentence debate, and about the overall aims of the movement, as in the debate between the left and right wings of the Vienna Circle over the role of politics in science and philosophy. Along with such differences went subtle differences in the assessment of Einstein’s legacy to logical empiricism. Philipp Frank–himself a dissenter from central points of right-wing Vienna Circle doctrine–deserves

Page 2

particular mention for his more accurate reading of Einstein’s position on such issues as the place of convention in scientific theory (Frank 1949b, 1949c). Still, on the whole, the Einstein legacy is rendered with an unhappy mixture of hagiography and persuasive misreading in the service of appropriating the mantle of his authority to legitimate logical empiricism as it sought at mid-century to establish itself as the authoritative philosophical voice on the nature and status of scientific knowledge. A more accurate picture of Einstein’s relationship to logical empiricism began to emerge at about the time in the mid-1960s when challenges to logical empiricist orthodoxy from the likes of Norwood Russell Hanson, Stephen Toulmin, Thomas Kuhn, and Paul Feyerabend began to create a space in which dissent was possible. The single most important early contribution to this new picture of Einstein’s connection with logical empiricism was Gerald Holton’s 1968 study, “Mach,Einstein, and the Search for Reality” (Holton 1968). Though one can argue with Holton’s uninflected interpretation of Einstein’s post-1915 view as a version of what later came to be termed “scientific realism” and with his missing the important theme of Einstein’s sympathy for Pierre Duhem’s brand of conventionalism (see Howard 1993), we nevertheless owe Holton a major debt for making clear the extent to which, already by the late 1920s, Einstein had explicitly forsworn any sympathy for verificationism and the anti-metaphysical agenda of logical empiricism, this at the very time when Vienna Circle pamphleteers were trumpeting his support for the movement (see, for example, Neurath, Hahn, and Carnap 1929). Lacking, however, in both the original logical empiricist hagiography and in Holton’s revisionist picture is a clear sense of what might well be Einstein’s major legacy to twentieth-century

Page 3

philosophy of science, which is the fact that Einstein was, himself, one of the most important, active, constructive contributors to the development of a new empiricism in the 1910s and 1920s. Einstein did not merely provide theoretical grist for the philosophical mill of logical empiricism, that and possibly a benediction. On the contrary, virtually all of the major figures involved in the debates outof which logical empiricism grew worked out their ideas through an engagement not just with general relativity but with the author of the theory, himself. This was true of central founding figures in the logical empiricist camp, such as Moritz Schlick, Hans Reichenbach, Philipp Frank, and Rudolf Carnap. It was true, as well, of figures associated with other traditions, such as the neo-Kantian Ernst Cassirer and Hermann Weyl, whose chief philosophical debt was to the phenomenology of Edmund Husserl. In conversation, in correspondence, and in a wealth of published essays and reviews,Einstein tutored, criticized, and learned, his interventions concerning large questions of doctrine andfine points of detail.Moreover, while Einstein produced no philosophical magnum opus to rival Schlick’s Allgemeine Erkenntnislehre (1918), Weyl’s Philosophie der Mathematik und der Naturwissenchaft (1927), or Reichenbach’s Philosophie der Raum-Zeit-Lehre (1928), the position that he defendedfrom early to late–as mentioned, a variant of Duhemian conventionalism–has much to recommendit as, in its own right, a cogent, comprehensive philosophy of science, from which there is still much to be learned that is of relevance to contemporary philosophical controversies (see, for example,Howard 2004d). A full history of twentieth-century philosophy of science should, therefore, devote a whole chapter to Einstein as a figure at least as important and influential as those already canonized.

Page 4

2. Background and Early Influences

Einstein was like many scientists of his generation in his early and regular engagement with philosophy. He is reported to have read Kant’s first Critique already at the age of thirteen (Beller 2000). As a physics student at the Eidgenössische Technische Hochschule (ETH) in Zurich in thelate 1890s, Einstein was required to attend lectures on “Theorie des wissenschaftlichen Denkens” given by August Stadler, a neo-Kantian philosopher who was Hermann Cohen’s first doctoral student at Marburg, and Einstein elected to enroll also in Stadler’s lectures on Kant (Beller 2000).
That means he should have been happy to be surrounded by so many excellent teachers.
On his own while a student and, a few years later with his friends in the discussion group they constituted in Bern around 1903, the so-called Akademie Olympia, Einstein read carefully many of the most important works of figures like Ernst Mach, Henri Poincaré, John Stuart Mill, RichardAvenarius, Karl Pearson, Richard Dedekind, and David Hume.
He read Friedrich Albert Lange’s Geschichte des Materialismus (Lange 1873-1875), Eugen Dühring’s Kritische Geschichte der Principien der Mechanik (Dühring 1887), and Ferdinand Rosenberger’s Isaac Newton und seine physikalischen Prinzipien (Rosenberger 1895).
During his university years and repeatedly thereafter, Einstein read widely in the works of Arthur Schopenhauer (Howard 1997). From the beginning, Einstein understood such reading in the history and philosophy of science as making an important difference to the way in which one does physics. Thus, in a 1916 memorial notice for Mach, Einstein wrote:
How does it happen that a properly endowed natural scientist comes to concern himself with epistemology? Is there no more valuable work in his specialty? I hear many of my colleagues saying, and I sense it from many more, that they feel this way. I cannot share this sentiment. When I think about the ablest students whom I have encountered in my teaching, that is,those who distinguish themselves by their independence of judgment and not merely their

Page 5

quick-wittedness, I can affirm that they had a vigorous interest in epistemology. They happily began discussions about the goals and methods of science, and they showed unequivocally, through their tenacity in defending their views, that the subject seemed important to them. Indeed, one should not be surprised at this. (Einstein 1916, 101)
Again Einstein should have felt himself lucky
And in a 28 November 1944 letter to Robert Thornton he echoed those words of nearly thirty years earlier:
I fully agree with you about the significance and educational value of methodology as well as history and philosophy of science. So many people today—and even professional scientists—seem to me like somebody who has seen thousands of trees but has never seen a forest. A knowledge of the historic and philosophical background gives that kind of independence from prejudices of his generation from which most scientists are suffering. This independence created by philosophical insight is—in my opinion—the mark of distinction between a mere artisan or specialist and a real seeker after truth. (Einstein to Thornton, 7 December 1944, EA 61-574)
Many of the mentioned authors had an important influence on the development of Einstein’s thinking, foremost among them Kant, Poincaré, Mach, and Hume (Howard 2004a, Beller 2000, Norton 2004). Much has been made of the influence of Mach, in particular (see, for example, Frank 1949c). But the nature of that influence is complicated. Einstein, himself, acknowledged an important debt to Mach in the genesis of both special relativity (Norton 2004 provides helpful detail; see also Holton 1968) and general relativity (seeHoward 1999, Norton 1995, and Hoefer 1995). But on various later occasions Einstein stressed his dissent from Mach’s doctrine of the elements of sensation as the basis of all scientific knowledge, as in a letter to Besso of 6 January 1948:
I see his great service as residing in the fact that he dispelled the dogmatism that reigned inthe foundations of physics in the 18 and 19 centuries. Especially in the Mechanik and the Wärmelehre, he sought to show how concepts grow up out of experience. He convincingly defended the view that these concepts, even the most fundamental ones, obtain their justification only from experience. . . . I see his weakness as residing in the fact that he more

Page 6

or less believed that science consists in the mere “ordering” of empirical materials; i.e., he misunderstood the free, constructive element in the formation of concepts. In a sense, he believed that scientific theories arise through discovery and not through invention. (Speziali1972, 390-391)
More than anything else, the positive lesson that Einstein drew from Mach concerned thehistoricity of scientific concepts and theories. Both the Mechanik (Mach 1883) and the Wärmelehre (Mach 1896) advertise in their subtitles that they are essays in an “historical-critical” method, and,as in the contemporary Biblical hermeneutics literature that employed the same characterization of its critical method (see, for example, Strauss 1835-1836), the point was to exploit the historicity of texts, concepts, or theories–their having been authored in specific historical circumstances byspecific individuals working in specific problematic settings–to deprive those texts, concepts, and theories of their authority as received wisdom. Precisely this aspect of Mach’s legacy was stressed by Einstein in the just-quoted letter to Besso but also, over thirty years earlier, in the above-cited obituary notice for Mach:
The fact is that Mach exercised a great influence upon our generation through his historical-critical writings. . . . Concepts that have proven useful in ordering things easily achieve such an authority over us that we forget their earthly origins and accept them as unalterable givens. Thus they come to be stamped as “necessities of thought,” “a priori givens,” etc. The path of scientific advance is often made impassable for a long time through such errors. For that reason, it is by no means an idle game if we become practiced in analyzing the long commonplace concepts and exhibiting those circumstances upon which their justification and usefulness depend, how they have grown up, individually, out of the givens of experience. By this means, their all-too-great authority will be broken. They will be removed if they cannot be properly legitimated, corrected if their correlation with given things be far too superfluous, replaced by others if a new system can be established that we prefer for what-ever reason. (Einstein 1916, 102)
Mach’s emphasis on the historicity of concepts and theories is continuous with his preferred description of his point of view as a “biological-economical” one, whereby he meant that the

Page 7

capacity for scientific knowledge, with its “economical” representations of experience through concepts and theories, should be viewed as an adaptive trait of the human species.
As great as was the influence of Mach on Einstein, there is another author whom Einstein seems also to have encountered at an early date, an author whose influence was, arguably, still more important, namely, Pierre Duhem. The story of Einstein’s first acquaintance with Duhem is instructive not only as regards the content of Einstein’s philosophy of science but also from the point of view of appreciating the manner in which Einstein and his contemporaries regarded the intermingling of physics and philosophy.
When, in the fall of 1909, Einstein left his job at the patent office in Bern and returned to Zurich to take up his first formal academic position at the University of Zurich, he happened to rent an apartment at Moussonstraße 12, directly upstairs from his old friend and fellow Zurich physicsstudent, Friedrich Adler. As chance would have it, Adler had been the other finalist for the very job Einstein was now assuming and–coincidence added to coincidence–just one year earlier Adler published his German translation of Duhem’s "La Théorie physique: son objet et sa structure" (Duhem 1906, 1908). We know from Adler’s letters to his father, Viktor Adler, co-founder of the Austrian Social-Democratic Party (see Ardelt 1984), that he and Einstein set up for themselves in the attic of the apartment building a shared study where they could work and converse, away from the noise of the children playing in the courtyard below. How did Adler recall their interactions?
We stand on very good terms with Einstein, who lives above us, and indeed as it happens,among all of the academics, we are on the most intimate terms precisely with him. They havea bohemian household similar to ours, one boy of Assinka's age, who is very often at our place. . . . The more I speak with Einstein—and that happens fairly often—the more I see that my favorable opinion of him was justified. Among contemporary physicists he is not only

Page 8

one of the clearest, but also one of the most independent minds, and we are of one mind about questions whose place is generally not understood by the majority of other physicists.(Ardelt 1984, 166)
Adler later gained notariety for assassinating the Austrian Prime Minister, Graf Stürgkh, in 1916 asa protest against Austrian policies in World War I.
A special detail
In 1909, however, he was merely a Privatdozentat the University of Zurich, where he held the venia legendi for “experimental and theoretical physics, as well as their history and epistemological foundations” (Ardelt 1984, 157-166) and was well known in both philosophical and political circles for his writings on the philosophy of science, most of them written from a broadly Machian perspective, as with his intervention on Mach’s sidein the just then raging debate between Mach and Max Planck (see Planck 1909, 1910, Mach 1910,and Adler 1900). Adler’s praise of Einstein’s “independent” mind anticipates Einstein’s own later stress on the independence of judgment that he thought the most valuable consequence of studyingthe history and philosophy of science.The young Adler and the young Einstein were not at all unusual for the time in thus viewingthe history and philosophy of science as an integral, critical part of the larger project of physics.Other examples are everywhere to be found, from Einstein’s good friends Michele Besso (Speziali1972), Maurice Solovine (Solovine 1956), and the Habicht brothers, Conrad and Paul (the latter three part of the Akademie Olympia), to future colleagues, like Philipp Frank (Frank 1949a, Howard2004c), Erwin Schrödinger (Moore 1994), Wolfgang Pauli (Enz 2002), Paul Ehrenfest (Klein 1970),and Max Born (Born 1969 [CPAE 8, Doc. 575]). Many of this generation’s teachers likewise evincedthe same synthetic understanding of a philosophical physics, whether it be Mach (Banks 2003) or Planck (Heilbron 1986), Hermann von Helmholtz (Krüger 1994) or Ludwig Boltzmann (Broda

Page 9

1983). Major journals promoted the integration of philosophy and physics, from Wilhelm Ostwald’s Annalen der Naturphilosophie and Zeitschrift für den physikalischen und chemischen Unterricht to Arnold Berliner’s Die Naturwissenschaften. The philosopher-physicist, if perhaps not quite thenorm, was also far from being the exception.
The main philosophical lesson that Einstein, Adler, and took from their reading of Duhemconcerned the empirical content and warrant for physical theories. Duhem argued that it is notindividual propositions but only whole theories that possess empirical content and that, inconsequence of this holism, theory choice is underdetermined by empirical evidence, since the fit between theory and evidence can be maintained by making adjustments in various different placeswithin the total body of theory. For Duhem, the choice among these options is a conventional one.Duhem’s position stands opposed to a stern empiricism, sometimes wrongly associated withMach, but rightly associated with the later verificationism of the Vienna Circle, according to which each proposition or even each scientific concept must have its own determinate empirical content.Duhem’s position stands equally opposed to Poincaré’s view that some scientific laws are not at risk of empirical refutation as a result of their playing a uniquely definitional role in theory, as with thelaw of free fall (alleged to define the notion of an inertial trajectory), for Duhem argues that thereis no principled basis upon which to parse a theory into propositions that function as definitions and propositions that function as synthetic, empirical assertions.That Einstein learned this lesson from Duhem is evident from the way in which he explainedto his students in the winter semester of 1910-1911 how one can make good empirical sense of the

Page 10

notion of electrical charge within a solid charged body even though one cannot introduce a test particle whereby to gauge the electrostatic field in the interior of that solid:
We have seen how experience led to the introduction of the concept of electrical charge. It was defined with the help of forces that electrified bodies exert on each other. But now we extend the application of the concept to cases in which the definition finds no direct application as soon as we conceive electrical forces as forces that are exerted not on material particles but on electricity. We establish a conceptual system whose individual parts do not correspond immediately to experiential facts. Only a certain totality of theoretical materials corresponds again to a certain totality of experimental facts. We find that such an el[ectrical] continuum is always applicable only for representing relations inside ponderable bodies. Here again we define the vector o[f] el[ectrical] fieldstrength as the vector of the mech[anical] force that is exerted on a unit of pos[istive]electr[ical] charge inside a ponderable body. But the force thus defined is no longer immediately accessible to exp[eriment]. It is a part of a theoretical construction that is true or false, i.e., corresponding or not corresponding to experience, only as a whole. (Einstein 1910-1911,325)
One could not ask for a more succinct statement of the central epistemological lesson of Duhem’s La Théorie Physique.

3. A New Empiricism As an Answer to the Neo-Kantians

Logical empiricism was the philosophical movement with which Einstein later came to bemost closely associated, this in no small measure because of Einstein’s personal connections to several of its chief architects. For example, as early as 1907, Einstein began a correspondence with Philipp Frank, then a central member of what is now known as the “first Vienna Circle” (Haller 1985, Frank 1949a) and to the end of his life prominently associated with the Vienna Circle and itsdescendent institutions (Stadler 1997, 681-687). Einstein had read Frank’s 1907 paper on causality(Frank 1907) and wrote to offer some suggestions about the manner in which causality might be seen

Page 11

as having a conventional and an empirical character (Frank 1949a, 22). In 1912 Frank succeededEinstein to the chair in theoretical physics at the German university in Prague, this upon arecommendation from Einstein that included mention of Frank’s epistemological writings, and inlater years he became Einstein’s biographer (Frank 1947) and an important commentator onEinstein’s philosophy of science (Frank 1949b, 1949c; see also Howard 2004c). Moritz Schlick became, after 1922, the guiding spirit of the Vienna Circle. Schlick andEinstein were in correspondence by late 1915, when Schlick sent Einstein a copy of his paper on the philosophical significance of the theory of relativity (Schlick 1915), a paper that Einsteincommended in part because of Schlick’s having appreciated the importance of Mach and Hume for the development of special relativity. Theirs was an especially close relationship through the early 1920s, Einstein working to promote Schlick’s career and the dissemination of his writings, as withhis helping to arrange an English translation of Schlick’s 1917 monograph, Raum und Zeit in der gegenwärtigen Physik (Schlick 1917), the main aim of which, according to Schlick, was to explainthe chief philosophical implications of the general theory of relativity. Einstein first met Hans Reichenbach in the late 1910s, when, as a student, Reichenbachaudited Einstein’s lectures on relativity at the University of Berlin. Einstein did not wholeheartedlycommend Reichenbach’s early efforts to develop an interpretation of the theory of relativity asdemonstrating a central role in physics for a contingent, constitutive a priori (Reichenbach 1920),a view that Reichenbach himself quickly modified in the direction of the metric conventionalism for which he later was famous. Nevertheless, in the mid-1920s, Einstein’s esteem for Reichenbach wassuch that he collaborated with Planck to create for Reichenbach a chair in the philosophy of science

Page 12

in the physics department at Berlin (Hecht and Hoffmann 1982), from which position Reichenbachwent on to build the Berlin outpost of the Vienna Circle.Schlick and Reichenbach played the leading role in the early- to mid-1920s in shaping the philosophy of science that we remember as logical empiricism, much of the fine structure of whichreflects their efforts, together with Einstein, to craft a form of empiricism adequate to the task of defending the empirical integrity of general relativity against various challenges, including most prominently that issuing from neo-Kantians such as Cassirer (Cassirer 1921). The Kantian challengeto general relativity came in many forms, ranging from those who simply rejected as impossiblegeneral relativity’s assertion of a non-Euclidean metrical structure for space-time, and those whotook refuge in an arguably quiet un-Kantian distinction between physical space and psychologicalspace, to those technically more able philosophers who, like Cassirer, sought to save Kant byretreating from ascribing a priori status to a specific metrical geometry of space-time, locating someweaker a priori structure in, say, local topological structure. As mentioned, in his first book on relativity Reichenbach sought to defend Kant by teasing apart the apodictic and constitutive aspectsof the a priori, rejecting the former and retaining the latter—now regarded as a contingent a priori—as that which effects the needed coordination between theory and world and thereby endowsa theory with empirical content (Reichenbach 1920).Schlick was not enthusiastic about any of the Kantian responses to general relativity,Reichenbach’s included. Schlick’s own earlier exposition of the central philosophical implicationsof general relativity in his Raum und Zeit in der gegenwärtigen Physik (1917) emphasized thetheory’s positing a real space-time event ontology and, thereby, its incompatibility with a philosophy

Page 13

of science that would accord reality only to observables, such as Mach’s elements of sensation. ButSchlick also stressed the ineluctable moment of convention in scientific theorizing, inspired both byhis reading of Poincaré and, even more so, by his own still earlier and then quite influential theoryof truth as involving only a univocal, many-to-one coordination of theory to world (Schlick 1910),which entailed an underdetermination of theory choice by evidence not unlike that central toDuhem’s philosophy of science. Einstein, who first read the book in manuscript, wholeheartedlycommended both the critique of Mach and Schlick’s views on the empirical underdetermination of theory choice. In the first edition of his Allgemeine Erkenntnislehre (1918), Schlick again gave center stage to the role of convention in scientific cognition, basing the argument once more on his viewof truth as univocal coordination but employing also David Hilbert’s idea of a theory’s implicitlydefining its primitive terms via the systematic role they play in the theory’s axioms–as opposed toeach term’s being given an explicit empirical definition–for the purpose of stressing that thecoordination relates theory to world only as a whole and not one concept or one proposition at a time.When, by the early 1920s, the gathering neo-Kantian reaction to relativity finally elicited afocused and thoughtful reply from Schlick, it was the notion of convention that provided Schlick analternative to the a priori. In private correspondence with Reichenbach and in a published review of Cassirer’s book on Einstein’s theory of relativity (Cassirer 1921, Schlick 1921), Schlick suggestedthat the “constitutive principles” whereby experience is ordered and interpreted are more helpfullycharacterized as conventions than as elements of either a contingent or an apodictic a prioricomponent of scientific cognition. Einstein–with whom Schlick was then in close and regular

Page 14

contact–had been suggesting much the same thing for several years. Thus, in a letter to Max Bornof July 1918, Einstein wrote:
I am reading Kant’s Prolegomena here, among other things, and am beginning to comprehend the enormous suggestive power that emanated from the fellow and still does.Once you concede to him merely the existence of synthetic a priori judgments, you aretrapped. I have to water down the “a priori” to “conventional,” so as not to have to contradicthim, but even then the details do not fit. Anyway it is very nice to read, even if it is not asgood as his predecessor Hume’s work. Hume also had a far sounder instinct. (Born 1969, 25-26)
Einstein repeated the same point–that what Kant regards as a priori is more properly seen asconventional–in virtually every one of the many comments he penned on the subject through themid-1920s (see Howard 1994).As the discussion involving Einstein, Schlick, Reichenbach, and Cassirer progressed in theearly 1920s, the effort to craft an empiricist analysis of general relativity that would force the pointwith the neo-Kantians led Schlick and Reichenbach to an historically important refinement in their understanding of where the conventional moment in science was to be located, and it led to a partingof the ways, ironically, with Einstein. Schlick and Reichenbach wanted to be able to say to the neo-Kantian that the attribution of a specific metrical structure to space-time was, in the end, an empiricalmatter, this in spite of there being a conventional aspect to ascriptions of metrical structure. Theytook their cue from Poincaré’s well known view of conventions as disguised definitions (Poincaré1902).As first clearly presented in Reichenbach’s 1924 book on the axiomatization of space-timetheory (Reichenbach 1924) and then more famously in Reichenbach’s 1928 Philosophie der Raum- Zeit Lehre (Reichenbach 1928) or Schlick’s 1935 “Sind die Naturgesetze Konventionen?” (Schlick

Page 15

1935), the idea was to restrict the moment of convention to the coordinating definitions that areassumed to effect links between a theory’s primitive empirical concepts and relevant aspects of theworld that the theory seeks to describe. Paradigmatic coordinating definitions would be theassociations of basic notions like “spatial interval” and “temporal interval” with, respectively, aspecific kind of practically rigid measuring rod and a specific kind of regular clock. Fix the empiricalmeaning of a theory’s primitive terms by such conventional coordinating definitions and each of thesynthetic empirical propositions in a theory thereby acquires a determinate empirical content suchthat its truth or falsity is, in principle, univocally determined by the corresponding experience.Differences do arise through our choosing different coordinating definitions. But such differencesare inconsequential, for they depend, at root, only on arbitrary, conventional definitions, and so can be no more significant than the difference between choosing English or metric units. Moreover, for such empiricists empirical content is the only content, which implies that, since empiricallyequivalent theories have the same empirical content, they can only be alternative ways of expressingthat same content, just as “il pleut” and “es regnet” are merely alternative ways of expressing thesame meteorological proposition, “it is raining.”If this were the right way to think about how theories acquire empirical content, it would haverepresented a persuasive reply to the neo-Kantian critics of general relativity, for it would have madethe choice of a metrical space-time geometry a straightforwardly empirical matter. But the maneuver rests upon one crucial assumption, namely, that one can in a principled manner distinguish withina theory among those propositions that, by their very nature, function as mere conventional

Page 16

definitions and those that function as synthetic, empirical claims. In another guise, this is theanalytic-synthetic distinction.Schlick and Reichenbach might have thought that they were simply elaborating a view aboutempirical content already put forward by Einstein in a widely-read lecture from 1921, Geometrie und Erfahrung (Einstein 1921). But when that lecture’s discussion of the empirical interpretation of general relativity is read against the background of Einstein’s then at least decade-old sympathy for Duhemian holism, it becomes clear that Einstein had in mind a very different picture of the place of convention in scientific theories.In Geometrie und Erfahrung, Einstein distinguished “pure” and “practical” geometry. Puregeometry is an uninterpreted mathematical formalism the primitive terms of which–e.g., “point” and“line”–are defined implicitly, à la Hilbert, via the systematic role they play in the formalism’saxioms. Practical geometry differs by associating or coordinating those primitive terms with someempirical or physical objects whereby the geometry becomes, in effect, a physical theory, a theoryabout the behavior of those objects. The crucial question concerns the manner in which the primitiveterms are thus associated with empirical or physical objects.Consider the primitive term, “segment of a straight line.” Our intuition tells us to associatethis with a physical structure–say the path of a ray of light or the edge of a rigid measuring rod–of such kind that information gleaned from the behavior of those structures concerns exclusively thegeometrical properties of the space or space-time in which the objects live and not physical factsabout the objects themselves. We all know, however, that real measuring rods are not perfectly rigid,suffering thermal deformations and mechanical stresses. Should measurements conducted with such

Page 17

rods appear to show deviations from Euclidean geometry, how much of that “deviant” behavior should be attributed to the geometry of space-time and how much to the physics of the rod itself?In “L’expérience et la géométrie” (1902) Poincaré had, in effect, argued that the defender of Euclidean geometry can always exploit this ambiguity by ascribing the deviance to the physics andthat one would want to do so, in order to save Euclidean geometry, because Euclidean geometry issimpler than its non-Euclidean rivals. Of course, Einstein disagreed, arguing that we should choosenot what yields the simpler geometry, but what maximizes the simplicity of geometry plus physics,as does, by his lights, general relativity.There is, however, a serious obstacle to our maximizing the simplicity of geometry plus physics, for while we know, at one level, that various aspects of the behavior of the rod are physicaleffects, as with thermal deformations, we lack at present a complete, fundamental, physical theoryof structures like measuring rods and clocks and so cannot, in fact, actually pose the simplicityquestion on the physical side. Under such circumstances, the best that we can do is, by stipulationand with a modicum of scientific good sense, to pick some practically rigid body as the physical or empirical interpretation of “segment of a straight line,” likewise for some practically regular clock and the primitive term “temporal interval,” and then to let measurements carried out with these rodsand clocks settle the question of the metrical structure of space-time.The resulting picture looks much like Reichenbach’s. Geometrical primitives are assignedempirical interpretations via stipulative, hence conventional, coordinating definitions, whereafter thetruth or falsity of synthetic, empirical propositions such as that concerning the space-time metric isunivocally determined on the basis of the relevant experience. But what for Reichenbach and Schlick

Page 18

was a first principles story about how all theories get their empirical content was, for Einstein, buta provisional, stop-gap procedure forced upon us by the lack of a sufficiently complete fundamental theory:
The idea of the measuring rod and the idea of the clock coordinated with it in the theory of relativity do not find their exact correspondence in the real world. It is also clear that thesolid body and the clock do not in the conceptual edifice of physics play the part of irreducible elements, but that of composite structures, which must not play any independent part in theoretical physics. But it is my conviction that in the present stage of developmentof theoretical physics these concepts must still be employed as independent concepts; for weare still far from possessing such certain knowledge of the theoretical principles of atomicstructure as to be able to construct solid bodies and clocks theoretically from elementary concepts. (Einstein 1921, 8)
For Einstein, the correct first principles story was very different. It was essentially Duhem’s holism.This was not the first time that Einstein had contrasted an in-principles story akin to Duhem’swith an in-practice story portraying a tighter link between theory and experience. In a 1918 addresscelebrating Planck’s sixtieth birthday, “Motive des Forschens,” Einstein compared the underdeter-mination of theory choice in principle with the well nigh universal impression of physicists that, in practice, a single theory always stands out as superior:
The supreme task of the physicist is . . . the search for those most general, elementary lawsfrom which the world picture is to be obtained through pure deduction. No logical path leadsto these elementary laws; it is instead just the intuition that rests on an empathic under-standing of experience. In this state of methodological uncertainty one can think that arbitrar-ily many, in themselves equally justified systems of theoretical principles were possible; and this opinion is, in principle, certainly correct. But the development of physics has shown that of all the conceivable theoretical constructions a single one has, at any given time, proved itself unconditionally superior to all others. No one who has really gone deeply into the subject will deny that, in practice, the world of perceptions determines the theoretical system unambiguously, even though no logical path leads from the perceptions to the basic principles of the theory. (Einstein 1918a, p. 31)

Page 19

But between 1918 and 1921, the question of the extent to which and manner in which straightforward empirical warrant could be claimed for a space-time theory had become, for Einstein, a most acute one, thanks mainly to the challenge posed by Weyl’s attempt at a unified field theory (Weyl 1918a, 1918b). Weyl’s attempt to unify gravitation with electricity and magnetism employed a geometrical framework that was more strictly “local” than that of Einstein’s general relativity in the sense that it did not assume the cogency of direct distant comparisons of length. In Weyl’s geometry, lengthwas path dependent: two bodies congruent at one space-time point need not be congruent at another if they followed different paths (world-lines) to the latter. Einstein objected that this feature of Weyl’s space-time geometry left it without determinate empirical content, since in a Weyl space thegeometrical concept of the space-time interval could, therefore, not be given a univocal empirical interpretation (Einstein 1918b, 1918c; see also Einstein to Walter Dällenbach, after 15 June 1918[CPAE 8, Doc. 565]). In effect, Einstein was arguing that, in a Weyl space, there could be nounivocal definition of the interval via a practically rigid rod. At the time, Einstein seemed to regard the impossibility of our giving such a direct empirical interpretation to the interval to be a nearly fatal methodological flaw, though he recognized that Weyl would not grant the point, since Weyl would insist on viewing rods and clocks as structures derived from within one’s fundamental theory, asopposed to their being specified from without by a coordinating definition. By 1921, however, Einstein seems to have come to regard Weyl’s point of view as, in principle, the correct one. Had Einstein really changed his mind?

Page 20

The clearest statement of Einstein’s view on the empirical interpretation of general relativity is found in a 1924 review of a book on Kant and relativity by Alfred Elsbach (Elsbach 1924). Elsbach had commended several arguments originally adduced in Paul Natorp’s Die logischen Grundlagen der exakten Wissenschaften (1910), among them the claims (i) that if deviations from Euclidean geometry were discovered they could be accounted for by making changes in physical laws and (ii) that the metrical structure of space cannot be determined experimentally because space is not real but ideal. Einstein comments:
The position that one takes on these claims depends on whether one grants reality to the practically-rigid body. If yes, then the concept of the interval corresponds to something experiential. Geometry then contains assertions about possible experiments; it is a physical science that is directly underpinned by experimental testing (standpoint A).
Experiments (and observations) are the starting points of understanding the physical reality.
If the practically-rigid measuring body is accorded no reality, then geometry alone contains no assertions about experiences (experiments), but instead only geometry with physical sciences taken together (standpoint B). Until now physics has always availed itself of the simpler standpoint A and, for the most part, is indebted to it for its fruitfulness; physics employs it in all of its measurements. Viewed from this standpoint, all of Natorp’s assertions are incorrect. . . . But if one adopts standpoint B, which seems overly cautious at the present stage of the development of physics, then geometry alone is not experimentally testable.
The concept of geometry has to be clearly defined
There are then no geometrical measurements whatsoever. But one must not, for that reason, speak of the “ideality of space.”
Space is a very difficult concept to understand the physical reality
“Ideality” pertains to all concepts, those referring to spaceand time no less and no more than all others. Only a complete scientific conceptual systemcomes to be univocally coordinated with sensory experience. On my view, Kant has influenced the development of our thinking in an unfavorable way, in that he has ascribed a special status to spatio-temporal concepts and their relations in contrast to other concepts.
Viewed from standpoint B, the choice of geometrical concepts and relations is,indeed, determined only on the grounds of simplicity and instrumental utility. . . . Concerning the metrical determination of space, nothing can then be made out empirically, but not“because is not real,” but because, on this choice of a standpoint, geometry is not a complete physical conceptual system, but only a part of one such. (Einstein 1924, 1690-1691)
When Einstein speaks here of granting “reality” to the practically rigid rod, he means simply our taking it as the putative referent of the geometrical primitive, “segment of a straight line.” According

Page 21

no “reality” to the practically rigid rod means simply that there is none such that can stand on theother side of a conventional coordinating definition of “segment of a straight line,” for the reasonearlier mentioned, namely, that real rods (and clocks) are not perfectly rigid (or regular). If there areno such rigid rods and regular clocks, then the alternative is to regard such structures as being defined implicitly within the theory (or some future, complete, fundamental theory), not stipulated from without, which, in turn, implies, as Einstein notes, that it is only the whole theory, “the complete scientific conceptual system” that has determinate empirical content.
Reichenbach and Schlick argue that empirical content attaches to theories one proposition at a time. Einstein, following Duhem, argues that, in principle, it attaches only to whole theories, even if, for practical reasons, we proceed as if the Reichenbach-Schlick picture were the right one. What, then, of the empiricist reply to the neo-Kantians that Reichenbach and Schlick were seeking? Einstein, no less than they, wanted a reply to the neo-Kantians, but he found it precisely in Duhemian holism. Earlier in his review of Elsbach, after asserting that, if one thought relativity a reasonable theory, then one must reject the Kantian doctrine of the a priori character of space and time, Einstein added:
This does not, at first, preclude one's holding at least to the Kantian problematic, as, e.g.,Cassirer has done. I am even of the opinion that this standpoint can be rigorously refuted by no development of natural science. For one will always be able to say that critical philosophers have until now erred in the establishment of the a priori elements, and one will always be able to establish a system of a priori elements that does not contradict a given physical system. Let me briefly indicate why I do not find this standpoint natural. A physical theory consists of the parts (elements) A, B, C, D, that together constitute a logical whole which correctly connects the pertinent experiments (sense experiences).
More detail is required to understand.
Then it tends to be the case that the aggregate of fewer than all four elements, e.g., A, B, D, without C, no longer says anything about these experiences, and just as well A, B, C without D. One is then free to regard the aggregate of three of these elements, e.g., A, B, C as a priori, and only D as

Page 22

empirically conditioned. But what remains unsatisfactory in this is always the arbitrariness in the choice of those elements that one designates as a priori, entirely apart from the fact that the theory could one day be replaced by another that replaces certain of these elements (or all four) by others. (Einstein 1924, 1688-1689)
Einstein’s named target here is Kant. But Einstein’s argument bears just as well against the kind of conventionalism defended by Reichenbach and Schlick, which, recall, arose out of Schlick’s telling the Reichenbach of 1921 that his a priori principles of coordination were more helpfully regarded as conventional coordinating definitions. Call them a priori principles or conventional definitions. The fact remains that one requires a principled basis upon which to distinguish propositions of that kind from the synthetic propositions thereby allegedly endowed with empirical meaning. Einstein’s point is that there is no principled basis for thus distinguishing the components of a theory, any such parsing of theory being, from a principled point of view, purely arbitrary. We see here in germ an argument better known to most philosophers in the form famously given it by Quine in “Two Dogmas of Empiricism” (Quine 1951). The point of view concerning empirical content being developed by Schlick and Reichenbach, with its assumption of some kin to the analytic-synthetic distinction as the basis upon which to distinguish the non-empirical from the empirical components of a theory is the progenitor of the verificationism mainly targeted by Quine(“reductionism” in Quine’s vocabulary), and Einstein’s deploying Duhemian holism against Schlick and Reichenbach anticipates Quine’s principal critical arguments against both verificationism and the analytic-synthetic distinction. Just how close Einstein comes to anticipating Quine’s more famous argument is evident from Einstein’s “Reply” to Reichenbach’s contribution to the Einstein volume in the Library of Living

Page 23

Philosophers. In his essay, Reichenbach had defended a first-principles view of the empiricalcharacter of geometry much like the provisional, stop-gap view that Einstein presented in Geometrieund Erfahrung , with the exception that Reichenbach, invoked explicitly his distinction betweencoordinating definitions and empirical hypotheses, portraying Einstein’s identification of thegeometer’s “rigid body” with the physicist’s “practically rigid rod” as an instance of the former (thisis what is meant by a definition of “congruence”):
The choice of a geometry is arbitrary only so long as no definition of congruence is specified.Once this definition is set up, it becomes an empirical question which geometry holds for physical space. . . . The conventionalist overlooks the fact that only the incomplete statementof a geometry, in which a reference to the definition of congruence is omitted, is arbitrary.(Reichenbach 1949, 297)
In his reply, Einstein constructs an imaginary dialogue bewteen “Reichenbach” and “Poincaré,”about which Ernest Nagel remarked in his review, “the spokesman for Poincaré does not obviouslyhave the worst of the argument” (Nagel 1950, 293-294). The dialogue begins with “Poincaré” asserting that geometrical theorems are not, bythemselves, testable because there are, in fact, no rigid bodies by which to interpret them.“Reichenbach” replies, à la the Einstein of 1921, that we can do well enough with the almost rigid bodies of our experience, as long as we make obvious corrections for such factors as changingtemperature. Einstein’s “Poincaré” notes that in making these corrections we must use physical lawsthat presuppose Euclidean geometry, and concludes that what is at stake in an experiment is, thus,the entire body of law consisting of physics and geometry. Here Einstein interrupts with the following parenthetical remark: “(The conversation cannot be continued in this fashion because the

Page 24

respect of the writer for Poincaré’s superiority as thinker and author does not permit it; in whatfollows therefore, an anonymous non-positivist is substituted for Poincaré—)” (Einstein 1949, 677).
As the dialogue resumes, “Reichenbach” grants a certain attractiveness to “Poincaré’s” pointof view, while insisting that in fact, if not in theory, it would have been impossible for Einstein tohave developed general relativity “if he had not adhered to the objective meaning of length,” that is,to the concept of the practically rigid body regarded as an unanalyzed primitive. “Reichenbach” goeson to argue that, if physics and geometry are tested together, then our aim should he the developmentof the simplest possible total system, not the simplest geometry alone.
Thus far the dialogue has proceeded in a predictable fashion. But now it takes an unexpectedturn. Having wrung from “Reichenbach” the grudging admission that “Poincaré” might be correct,in theory, about physics and geometry being tested together, though with the stipulation that it is thenthe simplicity of this whole–physics plus geometry–that must be judged, the “Non-Positivist” pointsout that “Reichenbach” has thereby contravened one of his own fundamental postulates–the equationof meaning with verification:
Non-Positivist : If, under the stated circumstances, you hold distance to be a legitimateconcept, how then is it with your basic principle (meaning = verifiability)? Do you not haveto reach the point where you must deny the meaning of geometrical concepts and theorems and to acknowledge meaning only within the completely developed theory of relativity(which, however, does not yet exist at all as a finished product)? Do you not have to admit that, in your sense of the word, no “meaning” can be attributed to the individual concepts and assertions of a physical theory at all, and to the entire system only insofar as it makes whatis given in experience “intelligible?” Why do the individual concepts that occur in a theory require any specific justification anyway, if they are only indispensable within the framework of the logical structure of the theory, and the theory only in its entirety validates itself.(Einstein 1949, 678)

Page 25

Notice that Einstein here asserts not just the epistemological holism of Duhem but also the semanticholism fundamental to Quine’s critique of logical empiricism.

4. Not a “Machian,” But Not a “Realist” Either

By the late 1920s Einstein, once Schlick and Reichenbach’s close philosophical friend, had become an open critic of logical empiricist orthodoxy, notwithstanding repeated efforts by membersof the Vienna Circle to claim him as an ally (see, for example, Neurath, Hahn, and Carnap et al.1929). There were many points of disagreement, logical empiricism’s antimetaphysical bias prominent among them. Thus, in a letter to Schlick from November 1930, commenting on amanuscript in which Schlick opposed his own neo-Humean view of causality to more robustly metaphysical notions of causality, Einstein wrote:
From a general point of view, your presentation does not correspond to my way of viewingthings, inasmuch as I find your whole conception, so to speak, too positivistic. Indeed, physics supplies relations between sense experiences, but only indirectly. For me its essence is by no means exhaustively characterized by this assertion. I put it to you bluntly: Physicsis an attempt to construct conceptually a model of the real world as well as of its law-governed structure. To be sure, it must represent exactly the empirical relations between those sense experiences accessible to us; but only thus it it chained to the latter. . . . You will be surprised at the “metaphysician” Einstein. But every four- and two-legged animal is defacto in this sense a metaphysician. (EA 21-603; as quoted in Howard 1994)
Einstein might well have been puzzled over how his old friend, Schlick, could have strayed so far from the realism central to the argument of Raum und Zeit in der gegenwärtigen Physik (Schlick 1917). Einstein had, himself, repeatedly stressed his preference for realism, as when he wrote in theabove-cited review of Elsbach:

Page 26

Is there not an experiential reality [Erlebnis-Realität] that one encounters directly and thatis also, indirectly, the source of that which science designates as real [wirklich]? Moreover,are the realists and, with them, all natural scientists (those who do not philosophize from thevery start) not right if they allow themselves to be led by the startling possibility of orderingall experience in a (spatio-temporal-causal) conceptual system to postulate something realthat exists independently of their own thought and being? (Einstein 1924, 1685)
A disagreement over metaphysics or realism was not, however, the main issue, or at least surely not the only issue, that led to Einstein’s estrangement from logical empiricism. Morefundamental, in my view, was the disagreement with Schlick and Reichenbach over the role of convention in scientific theory. They held that the moment of convention is confined to conventionalchoices of coordinating definitions, the stipulation of which thereby endows each empirical proposition with its own, determinate, empirical content. Einstein, the student of Duhem, held thatonly whole theories have empirical content and so regarded every component of a scientific theoryto be equally conventional and empirical. But can a disagreement over such a recondite questionabout the semantics of scientific theories really have been the main difference separating Einsteinfrom mainstream logical empiricism?Recall Einstein’s major complaint about Mach. It was not his denying the reality of unobservables, however much that had been an issue in Schlick and Einstein’s discussion of the philosophical implications of general relativity when Schlick was writing Raum und Zeit in der gegenwärtigen Physik (Schlick 1917). As Einstein, himself, recalled in his “Autobiographical Notes”(Einstein 1946, 48), Mach turned out not to be dogmatic on this point, readily granting the reality of atoms after experimental confirmation ca. 1909 of Einstein’s own earlier theoretical work on Brownian motion.

Page 27

No, what chiefly bothered Einstein was Mach’s neglect of the moment of invention or free creation in the development of scientific theories.That theories are the “free creations of the human intellect” (Einstein 1921, 5), was arecurring theme in Einstein’s writings (see also, for example, Einstein 1936, 314). It was alwaysopposed to overly strict empiricist accounts of the relation between concepts or theories andexperience: “In general, the weakness of the positivists appears to be that on their view, the logicalindependence of concepts with respect to sense experiences does not stand out clearly” (Einstein1924, 1687). Einstein was not alone in seeing this tension. Philipp Frank saw it as well, contrastingMach’s emphasis upon the empirical legitimation of theories with what he took to be Poincaré’sstress on their character as free creations, and credited both Duhem (Frank 1949a, 26) and Einstein(1949b) with having shown us how to reconcile the two by asserting that, since it is only whole theories that possess empirical content, not individual concepts or propositions, the theorist is,therefore, free to make changes anywhere in the total body of theory, as long as the whole theory conforms with empirical fact. For neither Duhem nor Einstein was the freedom such as to be an obstacle to progress inscience. Duhem emphasized that all creative scientific activity takes place in an historical context that constrains, because of contingent historical circumstance, the freedom that the theorist enjoysas a matter of logic alone (Duhem 1906, ch. 7). Einstein contrasted the fact that, in principle, thereis no univocal, logically determined path from experience to theory with the nearly universal recognition among scientists, themselves, that, in practice, theory choice seems univocally determined (see Einstein 1918b, 1933). Yet both Einstein and Duhem saw in the fact that the theorist

Page 28

is, thus, in principle, a free epistemic agent as that which secures for theory, as such, its central rolein science. Were theoretical concepts eliminable via definitions in terms of phenomenalist protocolterms, or were there, in some other form, an empirical theory choice algorithm, then science could be left to the experimentalist.That realism was not the main issue in Einstein’s dispute with mainstream logical empiricismis evident also from the fact that Einstein was, in fact, rather skeptical about blanket claims of themetaphysical realist kind. Surely the most amusing example of this skepticism is found in a letter of 25 September 1918 from Einstein to the Bonn mathematician, Eduard Study, commenting onStudy’s book Die realistische Weltansicht und die Lehre vom Raume (1914). About Study’s defenseof realism against both positivists and conventionalists, Einstein wrote:
“The physical world is real.” That is supposed to be the fundamental hypothesis.What does “hypothesis” mean here? For me, a hypothesis is a statement, whose truth must be assumed for the moment, but whose meaning must be raised above all ambiguity. Theabove statement appears to me, however, to be, in itself, meaningless, as if one said: “The physical world is cock-a-doodle-doo.” It appears to me that the “real” is an intrinsically empty, meaningless category (pigeon hole), whose monstrous importance lies only in the fact that I can do certain things in it and not certain others. This division is, to be sure, not an arbitrary one, but instead . . . . . .I concede that the natural sciences concern the “real,” but I am still not a realist.(CPAE 8, Doc. 624)
In what sense might science, nonetheless, concern the real on Einstein’s view? Arthur Finehas made the insightful suggestion that it was typical of Einstein to “entheorize” a methodologicalnotion like realism, turning it into a physical thesis, more specifically, the thesis of determinism(Fine 1986). But whereas Fine sees Einstein entheorizing realism as the thesis of determinism, I see

Page 29

him entheorizing it as the thesis of spatial or spatio-temporal separability. Consider what he wroteto Max Born on 18 March 1948:
I just want to explain what I mean when I say that we should try to hold on to physical reality. We are, to be sure, all of us aware of the situation regarding what will turnout to be the basic foundational concepts in physics: the point-mass or the particle is surely not among them; the field, in the Faraday-Maxwell sense, might be, but not with certainty. But that which we conceive as existing (“real”) should somehow be localized in time and space. That is, the real in one part of space, A, should (in theory) somehow “exist” independently of that which is thought of as real in another part of space, B. If a physical system stretches over the parts of space A and B, then what is present in B should some-how have an existence independent of what is present in A. What is actually present in B should thus not depend upon the type of measurement carried out in the part of space, A; it should also be independent of whether or not, after all, a measurement is made in A.
If one adheres to this program, then one can hardly view the quantum-theoreticaldescription as a complete representation of the physically real. If one attempts, nevertheless,so to view it, then one must assume that the physically real in B undergoes a sudden change because of a measurement in A. My physical instincts bristle at that suggestion.However, if one renounces the assumption that what is present in different parts of space has an independent, real existence, then I do not at all see what physics is supposed to describe. For what is thought to be a “system” is, after all, just conventional, and I do not seehow one is supposed to divide up the world objectively so that one can make statementsabout the parts. (Born 1969, 223-224)
That separability was central to Einstein’s conception of physical reality makes sense when it is understood that it was Einstein’s commitment to separability that underlay his decades long opposition to quantum mechanics as a preferred framework for fundamental physics, this because quantum entanglement seems hard to reconcile with the assumption that spatial separation is a sufficient condition for the individuation of physical systems.

5. Conclusion

Page 30

The development of Einstein’s philosophy and the development of logical empiricism were both driven in crucial ways by the quest for an empiricism that could defend the empirical integrity of general relativity in the face of neo-Kantian critiques. But logical empiricism was more than a philosophy of relativity theory, and Einstein’s philosophy of science was more than an answer to Kant.A fuller account of Einstein’s philosophy of science would have to include discussion of his belief in simplicity as a guide to truth, especially in areas of physics comparatively far removed fromextensive and direct contact with experiment, as in his own long search for a unified field theory (see Howard 1998, Norton 2000, Janssen and Renn 2005, and van Dongen 2002). A fuller account wouldalso investigate Einstein’s largely original and, I think, quite profound distinction between “principletheories” and “constructive theories,” the former constituted of mid-level, empirically well-groundedgeneralizations like the light principle and the relativity principle, which, by constraining the searchfor constructive models, often facilitate progress in science, as Einstein thought was the case in hisdiscovery of special relativity (Einstein 1919; see also Howard 2004a). And a fuller account would examine Einstein’s appropriation of what Joseph Petzoldt dubbed “the law of univocalness” (Petzoldt 1895), in effect the requirement that theories determine for themselves unique models of the phenomena they aim to describe, for this idea was central to Einstein’s thinking about a permissible space-time event ontology, his solution of the “hole argument” via the “point-coincidence argument” in the genesis of general relativity, and his more general attitude toward physical reality and objectivity (see Howard 1992 and 1999). And partly through its influence on

Page 31

1. For a helpful survey of philosophical reactions to relativity, see Hentschel 1990.
2. Stadler 1997 is a compendious source of information on all aspects of the history of logicalempiricism. On the protocol sentence debate, see Zhai 1900. Howard 2003 provides background onthe role of politics in the development of logical empiricism.
3. For a complete list of the readings of the Akademie Olympia, see Einstein 1989, xxiv-xxv.
4. Lange 1873-1875, Dühring 1873, and Rosenberger 1895 are among the many titles preserved inEinstein’s personal library, now housed with the Einstein Archive at the Hebrew University andEinstein, this idea of Petzoldt’s also played a significant role in the history of logical empiricism,especially in the development of Carnap’s thinking (see Howard 1996).
Still, the struggle to craft a compelling response to neo-Kantian critiques of general relativitywas, in my view, the single most important factor shaping the development of the story aboutempirical content that defined mainstream logical empiricism, and Einstein’s central role in this development through his personal and intellectual relationships with logical empiricism’s chief architects has yet to receive the attention that it is due in the historical literature. Moreover,understanding Einstein’s role in this history helps us to put into context Einstein’s late and oftenquoted characterization of himself as an “epistemological opportunist” (Einstein 1949, 684). Yes, Einstein’s philosophy of science borrowed from realism, positivism, idealism, and even Platonism.It might appear to the “systematic epistemologist” to be mere opportunism. But when viewed in its proper historical setting, it emerges as an original synthesis of a profound and coherent philosophyof science that is of continuing relevance today, the unifying thread of which is, from early to late,the assimilation of Duhem’s holistic version of conventionalism.

Page 32

NOTES National Library in Jerusalem.
5. For a fuller discussion of Einstein and Duhem, see Howard 1990a.
6. For helpful discussions of the late-nineteenth and early-twentieth century phenomenon of the“philosopher-physicist,” see Stöltzner 2003 and Howard 2004b.
7. That such could not have been Mach’s view is evident from his praise for Duhem both in his preface to Adler’s translation of Duhem 1906 and in the laudatory comments he added to the 1906second edition of his Erkenntnis und Irrtum (Mach 1906).
8. See Einstein to Schlick, 14 December 1915, EA 21-611, CPAE 8, Doc. 165.
9. For more on Einstein’s relationship with Schlick, see Howard 1984 and 1994.
10. Hentschel 1990 surveys the variety of Kantian reactions to relativity.
11. See Schlick to Einstein, 9 October 1920, EA 21-580, and Schlick to Reichenbach, 26 November 1920, HR 015-63-22 (Hans Reichenbach Papers, Archive of Scientific Philosophy, University of Pittsburgh).
12. See Ryckman 2005, especially ch. 3, for further helpful discussion of Einstein’s objection to Weyl’s unified field theory.
13. More or less the same argument, indeed, using almost exactly the same formulations, is repeated in Einstein 1925.
14. Michael Friedman gives a very different reading of the history of verificationism, locating its birth in Einstein’s point-coincidence argument (Friedman 1983, 24). For a critical evaluation of Friedman’s argument, see Howard 1999.
15. An enduring puzzle is why Einstein so frequently attaches the name of Poincaré to the holistic version of conventionalism that was famously defended by Duhem in explicit opposition to the views of Poincaré, another well-known example of his doing so being Einstein 1921. For further discussion, see Howard 1990a.
16. Quine reports that he had not known of Einstein’s quoted remarks from Schilpp 1949 when he wrote “Two Dogmas of Empiricism” (private communication). But given Quine’s close connections with Philipp Frank at Harvard in the late 1940s and the latter’s deep immersion in Einstein’s philosophy of science at that very time (see, e.g., Frank 1947, 1949b, 1949c), one suspects that there is more to the story than Quine, himself, remembered.

Page 33

17. While Einstein disagreed, as indicated, with came to be regarded as mainstream logicalempiricism, his philosophy of science actually had much in common with the views prevalent among the members of what is now dubbed the “left wing” of the Vienna Circle, including Neurath, Hahn,and Frank, all of whom shared with Einstein the experience of being deeply influenced by Duhem.For more on the Vienna Circle’s left wing, see Howard 2003.
18. For more on the role of separability in Einstein’s critique of quantum mechanics,

Index

Duhem page 2, page 7, page 9, page 13, page 18 page 27
Entanglement page 29
Philosopy of Science page 1, page 3, page 5, page 7, page 11, page 13, page 30, page 31

Reflection 1 - Philosophy of Science

The concept 'Philosophy of Science' is important because it does not discuss the act of doing science, but all what is involved to perform science in a scientific way, in a professional way.

If you want to give a comment you can use the following form Comment form
Created: 1 June 2022

Go Back to Article Review
Back to my home page Index