Monday, 14 December 2009

Realism vs. Anti-Realism and Models of The Atom


Did physicists believe in the reality of their atomic models? To answer this question, one has to look first at what it means to say that one ‘believes in the reality’ of something, and then also understand what is meant by a scientific ‘model’.  (From this point forward, I will refer scientific realism and scientific anti-realism as “realism” and “anti-realism”, respectively.)  “Discerning the aims of physical theory has been an important goal since the Greeks, with realist rather than positivist or instrumentalist views dominating at one time or another.”[i] “For nearly all practicing scientists—not all, to be sure—realism is an unequivocal commitment, rarely reflected upon very deeply.  Science, according to this view, is not merely another cultural activity, not simply fashion or metaphor, not simply an alternative way of viewing the world.  The success of science, its efficacy, its law-giving character—indeed the “progress” of science—clearly distinguishes it from other, no less important, areas of human inquiry.”[ii] The realist makes two claims:

  1. “Scientists ought to seek to formulate true theories that depict the structure of the universe...[and oppose] instrumentalists...who sought to restrict science to the “saving of appearances”.[iii]


  1. “The record of progress indicates that the universe has a structure (largely) independent of human theorizing and that our theories have provided an increasingly more accurate picture of that structure.”[iv]

There are other forms of realism, such as Entity Realism as propounded by Ian Hacking, and Structural Realism by John Worral, that are weaker versions of the realist position.  They don’t require that all of scientific practice aims for and attains truth and knowledge of reality in itself, and that the development is science is progressive.  They just pick out parts that could be so – such as putative entities, and structures.  Anti-realism can take many forms, but at the very least an anti-realist “seek(s) to uncouple the notions of predictive success and truth.”[v] Instrumentalism is a form of anti-realism that says that “scientific theories are calculating devices that facilitate the organization and prediction of statements about observations. [...] Theories are merely “useful” or “not useful”.”[vi] Bas Van Fraassen is a Constructive Empiricist which is a form of Instrumentalism, and he maintains that the goal of science is to “formulate empirically adequate theories... [not to] ... establish the truth of claims about theoretical entities.”[vii] Models can be both realist and anti-realist.  They “have two main functions in physics: they may be proposed either as putatively true representations of the physical characteristics of the objects treated by some theory, or as purely imaginary devices, heuristic fictions (a formal model).”[viii] In either case, whether proposed as putatively true or as a heuristic device, models are suggestive.

In the case of the atom, there was a full range of ontologies that were adopted by practicing physicists.  Developments in the nineteenth century culminated in the development of electromagnetic field theory with Maxwell, and statistical mechanics with Boltzmann.  These two physicists were realists – they believed that their models of the atoms mapped onto reality.  They did have opposition among their contemporaries, such as anti-realist Ernst Mach.  Mach was an anti-realist about unobservable entities.  He had a positivistic approach to the science, and since physical theory during their time did not require the existence of the atom, he did not adopt belief in it.  As regards twentieth century physics, following the Quantum Revolution, there were two main interpretations of Quantum Mechanics that provided models for interpretation of the mathematics, and consequently the atom.  However, following the Solvay Congress in October 1927, the Copenhagen Interpretation as given by Bohr and Heisenberg came to dominate as the most accepted one.[ix] In my paper, I will argue that the question of the ontological status of the atom changed from “does it exist at all (Maxwell/Boltzmann)?”  to a question of “given the atom, what is its nature?” (Bohr/Heisenberg).  Particularly, that there was a shift after 1905 from realist to antirealist attitudes towards the ontological status of the atom.

Nineteenth Century Atomism

Positivist thought “began to be felt at the end of the nineteenth century, promoted by Comte, the Vienna Circle, and the scientist-philosopher-historians such as Pierre Duhem and Ernst Mach.”[x] And this made sense because “before atomic theory became firmly established, and when physics could study only macroscopic phenomena, mechanical models and speculative hypotheses about underlying structure could be counterproductive.  A theory might fail because of such a model, while a macroscopic model only had to describe or reproduce the phenomena.”[xi] “In a loose sense, the distinction between dynamists and mechanists was one between positivists and realists, even though the ideas are not equivalent.  A positivist essentially sees the aim of physical theory as economically summarizing empirical results: as the Greeks saw it, “saving the appearances.”  No mechanical hypotheses are introduced that are not justified by what is observable.  The realist, of course sees the entities introduced to explain the experimental results as objectively real.  The mechanist, in trying to explain the properties of matter on the basis of the nature of its smallest parts, often has recourse to entities that are not accessible to observation.  In some cases entities may ultimately be observed and become part of the empirical world; in others they may disappear from the literature or survive only as heuristic elements.  In some cases, of course, the entities are not intended to be real and serve only as analogy, as an aid in reasoning.  This is sometimes the case in Maxwell’s use of models, which included elements that he never claims to exist fully.  Yet this is no doubt of Maxwell’s commitment to the reality of the molecular vortices on which much of theory of electromagnetism was based.”[xii]

James Clerk Maxwell and Ludwig Boltzmann were realists.  They “did believe in the [realty of their atomic] model(s), particularly in [...] the molecular vortical model.”[xiii] Their collaboration led to the development of a model of the atom where “the particularly simple properties of a molecular model, according to which the molecules are point masses (thus not hard sphere) which interact with a repulsive force inversely proportional to the fifth power of their distance.”[xiv] Their model was visualizable, and was explainable within the current paradigm.  With a visualizable model, they were able to use of analogy to guide their investigations.  “The role of analogy in nineteenth-century physics [...]was used deliberately and self-consciously by some of the most important scientific figures of the time, especially [...] Maxwell.  Indeed, Maxwell not only used a method of physical analogy with great success but also speculated extensively about it, especially the question of whether analogies in the natural world or the human mind. [...] Maxwell employed mechanical models to whose reality he was committed in differing degrees at different times.”[xv] As for Boltzmann’s commitment to his atomic model, in a letter, he wrote: “The realist compares the assertion that he could never imagine how the mental could be represented by the material let alone by the interaction of atoms with the opinion of an uneducated person who says that the Sun could not be 93 million miles from the Earth, since he cannot imagine it.  Just as the ideology is a world picture only for some but not for humanity as a whole, so I think that if we include animals and even the Universe the realist mode of expression is more appropriate.”[xvi]

Twentieth Century

Philosophical positivism was evident in the late nineteenth century practice of physics in the opinions of such physicist/philosophers such as Ernst Mach, and his rejection of atomism.  However, it was the operationalist character of quantum ontology of Neils Bohr’s in the twentieth century that also reflects philosophical positivism.[xvii] An operationalist says that “it is the operations by which values are assigned that give empirical significance to a scientific concept.”[xviii] “Though they were verbally opposed to several theses of the positivism of the philosophers [in the Vienna Circle], the physicists of the Copenhagen School, for their part, built up a quantum mechanics in which certain lines of reasoning when followed closely suggested ... rather similar views.”[xix] And although it was not the only interpretation of Quantum Mechanics, following the Solvay Congress of 1927, the Copenhagen Interpretation of Quantum Mechanics was the dominant one.

As stated, Bohr was a positivist.  He took an instrumentalist’s viewed toward his atomic model.  He was “extremely cautious.  He believe(d) that the models of atomic structure have some realistic significance, but he is acutely conscious of the negative analogy of the models; indeed he doubts whether a complete, realistic model of atomic processes is obtainable. [...] His own anti-realism was inspired by his commitment to Machian positivism. [...] Models help us to construct theories which enable us to explain and predict the course of our sensory experience.  Highly successful models may owe their success to the fact that they faithfully represent at least some aspects of the real entites which lie beneath the appearances.  ... We ought not to put too much  faith in the realistic performance of models.”[xx] In a letter to Hoffding, he very tellingly wrote:

“The question of the role of analogy in scientific investigations which you stressed is undoubtedly an essential feature of every study in the natural sciences, even if it does not always stand out.  It is often quite possible to make use of a picture of a geometrical or arithmetical sort which covers the problem in question in such a clear way that the considerations almost acquire a purely logical character.  In general, and particularly in some new fields of research, one must however constantly keep in mind the obvious or possible inadequacy of the picture, and , so long as the analogies make a strong showing, be content if the usefulness or rather fruitfulness in the area they are used is beyond doubt.  Such a state of affairs holds not least from the standpoint of the present atomic theory.  Here we are in the peculiar situation that we have gained some information about the structure of the atom which may safely be considered just as certain as any of the facts of natural science.  On the other hand we meet with difficulties of such a profound nature that we cannot see any way of solving them; in my personal opinion these difficulties are of such a kind that they scarcely allow us any hope of carrying through in the atomic realm a description in time and space of the kind that matches our ordinary sense impressions.  In these circumstances one must naturally bear in mind that one is operating with analogies, and the point, that the areas of use of these analogies in the individual case are restricted, is of decisive importance for progress.”[xxi]

He wrote this during the decline of his original atomic model, which was already being found to be flawed.

Bohr and Heisenberg’s model Copenhagen Interpretation of Quantum Mechanics depicted the atom as no longer visualizable.  In this interpretation, there is no quantum world that exists independently of our observation.  The observer and observed are inseparable, and that to make a measurement is to define the operation performed in making that measurement.  All that is knowable is what you observe, and what you observe is affected by your action of making the observation.  In adopting instrumentalist views toward their atomic model, they acknowledged the failures of using a model for visualization.  This sort of limitation of a model is analogous to the failures of using a tesseract or hypercube as a 3-dimensional representation of a 4-dimensional object.  Where some useful inferences can be drawn, there may also be ones that fail simply because the model failed.

Concluding Remarks

Nineteenth century and twentieth century physics had entirely different climates.    We’ve looked at nineteenth century realists, and twentieth century instrumentalists, but “there is no single scientific method that [was] applicable in all fields and at all times or to both theorists and experimentalists”[xxii] There were nineteenth century anti-atomists such as Pierre Duhem and Ernst Mach and twentieth century realists such as Einstein.  But there was definite change in attitude with the Quantum Revolution.  Older texts wrote of the way that “independent reality refuses to tell us what it is – or what it is like – it at least condescends to let us know, to some extent, what it is not.  It does not conform to the classical schemes of mechanics, of atomistic materialism, or of objectivist realism – in short, to any variant of ‘near realism’.[xxiii] And textbooks on twentieth century physics stress that “most [scientists] assign a more modest goal to physics, and to knowledge in general.  Science, they say, (and ordinary knowledge as well) is indissolubly linked with human experience.  Once and for all it must therefore give up the unattainable goal of describing whatever some thinkers may mean when they speak of ‘reality in itself’ or ‘reality as it really is’.  The task of science can only be a description of the phenomena, that is, of things, events and so on, as they are organised by human collective experience.”[xxiv]

It was the kinetic theory of gases that changed the ontological status of the physical atom from speculation to reality, and “the understanding of atomic and molecular spectra achieved by the 1860’s and 1870’s, which made possible the use of spectroscopy in chemical analysis, went far toward bridging the gap between the two manifestations of the microscopic structure of matter.  The final resolution came only with a detailed theory of atomic structure, which had to wait for...the quantum revolution.”[xxv]

“But for the most part, the nineteenth century ended the [...] debates about the microscopic structure of matter and provided convincing proof of the reality of the atoms. [...] Models of the internal structure of the atom were being seriously proposed. [...] Some would say that the final blow to the opponents of atomism was Einstein’s 1905 paper on Brownian motion, which showed that it was due to the motion of molecules.”[xxvi] “It is not too much to say that the great revolution in twentieth-century physics—the quantum theory—owes its birth to atomism, not merely in the strict historical sense but because the nineteenth-century success of the corpuscular theory prepared the way for the discontinuities and quantization that lie at the heart of quantum theory.”[xxvii] In this way, we can say that the realist attitudes of Maxwell and Boltzmann paved the way for the instrumentalism of Bohr and Heisenberg that were to come.


Bowler, Peter J. 2005. Making modern science : A historical survey, ed. Iwan Rhys Morus. Chicago: University of Chicago Press.

Buchwald, J. Z. A Brief History of Electric and Magnetic Science (unpublished)

Cassidy, David C. 2008. Beyond uncertainty : Heisenberg, quantum physics, and the bomb. New York: Bellevue Literary Press.

Cercignani, Carlo. 1998. Ludwig boltzmann : The man who trusted atoms. New York: Oxford University Press.

Espagnat, Bernard d. 1989. Reality and the physicist : Knowledge, duration, and the quantum world. New York: Cambridge University Press.

Great experiments in physics : Firsthand accounts from galileo to einstein(1987). In Shamos M. H. (Ed.), . New York: Dover Publications.

Kevles, D. J. (1978). The physicists : The history of a scientific community in modern america. New York: Knopf.

Kragh, H. (1999). Quantum generations : A history of physics in the twentieth century. Princeton, N.J.: Princeton University Press.

Losee, John. 1980. A historical introduction to the philosophy of science. 2d ed. -- ed. London: Oxford University Press.

Murdoch, Dugald. 1987. Niels bohr's philosophy of physics. New York: Cambridge University Press.

Purrington, R. D. (1997). Physics in the nineteenth century. New Brunswick, N.J.: Rutgers University Press.

Wilson, David. 1983. Rutherford, simple genius. London: Hodder and Stoughton.

[i] Purrington, Pg. 19

[ii] Purrington, Pg xi

[iii] Losee. Pg. 253

[iv] Losee. Pg. 253

[v] Losee, Pg. 254

[vi] Losee, Pg. 257

[vii] Losee, Pg. 257

[viii] Murdock, Pg. 74

[ix] Kragh, pg.212-215

[x]Purrington, g. 7

[xi] Purrington, Pg. 21

[xii] Purrington, Pg. 22

[xiii] Purrington, Pg. 67

[xiv] Cercignani, Pg. 199

[xv] Purrington, Pg. 28

[xvi] Cercignani, pg 174

[xvii]Purrington, g. 7

[xviii] Losee, Pg. 160

[xix] D’Espagnat, Pg. 200

[xx] Murdoch, pg 76-77

[xxi] Murdoch, pg 76

[xxii] Purrington, Pg xii

[xxiii] D’Espagnat, Pg. 208

[xxiv] D’Espagnat, Pg. 232

[xxv] Purrington, pg. 131

[xxvi] Purrington, pg. 131

[xxvii] Purrington, Pg. 131
There was an error in this gadget