A Genealogy of Nothing: Ether and the Case for Fallibilism

[The following essay is directed toward an imaginary positivist with unflinching faith in the veracity of Einstein’s research programme of relativity, because of relativity’s overwhelming empirical success. Rather than being anti-science (quite the opposite, actually), my humble goal here is simply to show that empiricism does not provide the full picture, and that fallibilism is justified as a default position when considering contemporary science. I would have liked to explicitly dwell upon specific philosophers of science (particularly Kuhn, Lakatos, and Feyerabend), but that will have to wait for another time. Lastly, this is unfortunately not an introductory essay, and is directed toward those who are at least superficially familiar with relativity and the history of physics preceding it.]

The pessimistic meta-induction is the supposition that just as so many theories in the history of science have been superseded, current theories will likewise be found to be unsatisfactory, despite their empirical success. This can be taken in a strong or a weak sense: the strong sense implies that current theories are completely wrong (just as phlogiston, to contemporary scientists, is completely wrong), and the weak sense (fallibilism) acknowledges the empirical success of current theories while insisting that they may be incomplete—epiphenomena, of sorts, of a larger pattern. It is the aim of this essay to make a case for fallibilism, illustrating its case with examples from special relativity, general relativity, and quantum theory; once the latter case is made, the strong pessimistic meta-induction will be left as a possibility, since by definition no positive case (save the explicit falsification of current theories) can be made for its correctness, but only a negative case. Starting with a brief glance into Einstein’s epistemology, the historical development of the concept of ether will be documented, and upon finding that it is not necessarily as “superfluous” as Einstein may have once thought, the implications of this incompleteness will be examined.

“You do not really understand something unless you can explain it to your grandmother,” Einstein is reputed to have said. This strikes the reader as a surprising statement to come from one so notorious for the abstruseness of his theories, but it reveals a striking distinction for philosophies of science: that between how a theory works (in all its mathematical intricacy) and what it means. As Hegel writes in his Shorter Logic,[1] “The chemist places a piece of flesh in his retort, tortures it in many ways, and then informs us that it consists of nitrogen, carbon, hydrogen, etc. True: but these abstract matters have ceased to be flesh.” Here we see that Hegel rejects the mechanical in favor of the conceptual, presumably reacting to the reductionist tendency of scientists to favor the former at the expense of the latter, but we see in Einstein a desire to retain the two in all their incommensurability. Yet, we can also proceed backwards from Einstein’s distinction: if mathematics is a formal delineation of the relations between terms, then insofar as mathematical physics is an empirical science, its terms cannot merely be mathematical variables, but objects, to which correspond concepts. With physics in particular, however, the boundaries separating concepts are of prime importance, and it is these mutable boundaries that pose the primary weak point of scientific research, to the point where fallibilism becomes a rational mindset for scientists regardless of the empirical success of any given theory taken on its own.

The ether is perhaps the most striking example of a concept which was heuristically invented to account for empirical data, but which underwent radical change, arguably culminating in Einstein’s contention that it is “superfluous.” Its origins were the result of a humble analogy: if forces such as electricity require a medium through which to travel, in the lack of conflicting evidence, it was natural for physicists of the late 1800s to suppose that light, related by Faraday with electromagnetism in 1845, likewise required a medium. (The same argument applies to the not-fully-correct analogy with the behaviour of waves.) Also, and just as importantly, the ether as a concept was required to make sense of the Newtonian theory of gravity as ‘action-at-a-distance’[2]: in order to avoid this conclusion, which had no analogue in other physical phenomena, the ether served as a medium to transmit gravitational ‘forces’. The ether was postulated as having the properties of relative immobility[3] (with the exception of “the small movements of deformation which correspond to light waves”[4]), solidity (“because transverse waves are not possible in a fluid, but only in a solid”[5]), and “a definitely assigned velocity throughout the whole of space”[6] (thus causing a ‘drag’ in response to the movement of objects), as well as a contradictory mix of mechanical and electromagnetic qualities. Fresnel’s coefficient was brought in to reconcile these theories with conflicting theoretical evidence in terms of light refraction, deriving the result that light refraction would behave just as if there were no ether. It was Lorentz who brought about the first significant conceptual change in the concept of the ether by divesting it of its mechanical properties, doing likewise to matter of its electromagnetic properties—all of this in response to Michelson and Morley’s famous experiment which attempted (and failed) to record the effect of ‘ether drift’.[7] The ether was now “exclusively the seat of electromagnetic fields”,[8] and deprived of all its qualities except its relative immobility.

The influence of special relativity on the concept of ether was to divest it of the final quality ascribed to it by physicists, namely, its relative immobility, which would have entailed the falsity of the principle of relativity, since the laws of physics would be affected by their position relative to the movement of the ether. This concept was therefore rendered superfluous because, as a medium without qualities, it was no different from empty space. Nevertheless, Einstein himself, in a May 5, 1920 lecture entitled “Ether and the Theory of Relativity,” draws attention to the ways in which the Ether may still be (or become) a helpful concept in light of general relativity. For example, Einstein outlines the fact that Mach’s principle (in which rotation is viewed, in lieu of absolute space, in terms relative to the ‘fixed stars’) presupposes the phenomenon of action-at-a-distance, the only escape from this being by positing an ether as “medium for the effects of inertia”.[9] Still, this postulate “differs essentially from the ether as conceived by Newton, by Fresnel, and by Lorentz” – it “not only conditions the behaviour of inert masses but is also conditioned in its state by them.”[10]

With the introduction of curved space, however, the ether loses its homogeneous and isotropic qualities, “compelling us to describe its state by ten functions”[11] – the Einstein tensor. Still, Einstein insists, insofar as space must be supposed to possess qualities, general relativity requires the ether, though one entirely different from that of Lorentz. After expounding some of this new ether’s properties (most notably its lack of a uniform relation to both gravity and electromagnetism), Einstein propounds that it can only be the discovery of a physical connection between gravitation and electromagnetism that “[t]he contrast between ether and matter would fade away, and through the general theory of relativity, the whole of physics would become a complete system of thought, like geometry, kinematics, and the theory of gravitation”,[12] though he is sure to add the caveat that quantum theory may set limits to this endeavour.

What the above example shows is how concepts heuristically ‘evolve’ when exposed to new mathematical relations, new research programmes (namely, relativity), and new data; that is to say, it would be misleading to portray the validity of the ether as being accepted, denied, and accepted again: it has evolved new functions, and still serves its heuristic use.[13] Though contemporary physicists possess the great advantage of knowing the conditions according to which the concept of ether may be consistently abandoned (viz. field theory), no clear path to this goal has yet been made. In his book Science and Hypothesis, Henri Poincaré, after an incisive analysis of the concept of energy, ultimately concludes that the law of conservation of energy is effectively tautologous—deriving its validity primarily from our definition of what comprises energy rather than any inherent necessity on energy’s part. Poincaré goes on to speculate that the concept of energy as used in his time will be discarded after it ceases to be useful, adding: “some day, no doubt, the ether will be thrown aside as useless”.[14] As is made evident by the unrewarded efforts of generations of physicists, the answers to their most pressing questions are not ‘tautologies’ waiting to be discovered: Einstein’s ideal of physics as a “complete system of thought” has not been realized. Therefore, there is certainly a need for conceptual evolution in physics, hence some skepticism is merited for our current theories—skepticism which new Einsteins in the years to come will devote their lives to dispelling.

Having considered the above, the following situation presents itself for the most likely means of finding a lacuna in relativity and/or quantum theory: the properties of black holes. Typically, relativity and quantum theory do not interact with one another, and indeed they are manifestly incompatible: general relativity treats space as a smooth, continuous sheet, while quantum reality focuses on discontinuous and discrete entities.[15] The two collide, however, when considering singularities, which are both immensely massive (hence amenable to relativity) and immensely small (perhaps, in some cases, even at the Planck scale).[16] It is in examining such extreme phenomena that a hidden lemma is most likely to be found in relativity, though as yet physicists have made little progress in devising a theory of quantum gravity. The above analysis extends to the singularity that preceded the Big Bang, but once again physicists’ equations are not adequate to comprehend such phenomena.

Perhaps the most egregious shortcoming of relativity is its inability to explain how space is curved. This is, of course, overcome by considering ‘intrinsic curvature’ rather than extrinsic curvature, but this is not satisfactory by any stretch of the imagination.[17] In lieu of discovering ‘gravitons’, however, relativity is forced to sweep this question under the rug. Perhaps, as new questions arise, the ether may even come back into vogue. And what of ‘dark matter’ and ‘dark energy’, which physicists surmise makes up nearly 96% of the material in the universe, with matter comprising only the few percentage points left over? Einstein never explicitly dealt with dark matter, and dark energy was not hypothesized until 1992, and it is immensely presumptuous to surmise that his equations have a priori taken its influence into account, the cosmological constant notwithstanding. In short, a theory that does not take into account 96 percent (26% dark matter and 70% dark energy) of the universe does not deserve to be considered complete, despite its flawless empirical accuracy for the experiments it has undergone.

We have examined the heuristic development of the concept of ether and found that it is useful not only in itself, but also—by displaying relations among other concepts—as a means of identifying new theoretical possibilities. We have argued that such a ubiquitous entity must necessarily have a role in the reconciliation of relativity and quantum theory, and we have endeavoured, with our modest means, to show that great conceptual revolutions must still take place in order to realize Einstein’s hope for physics to be a complete system of thought.


Einstein, A.; Jeffrey, G. & Perrett, W. (trans.). (1983 [1922]). Sidelights on Relativity. New York: Dover Publications.

Hegel, G.W.F.; Garaets, T., Harris, H., & Suchting, W. (trans.). (1991). The Shorter Logic. Indianapolis, IN: Hackett Publishing Co.

Larvor, B. (1998). Lakatos: An Introduction. New York: Routledge.

Norton, J. (2007). Einstein for Everyone. Pittsburgh, PA: Nullarbor Press.

Poincaré, H.; Gould, S. (Ed.); Halsted, G. (trans.). (2001). The Value of Science: Essential Writings of Henri Poincaré. New York: The Modern Library.

Sanders, J. (2006). Einstein’s Revolution. Ashland, OR: Blackstone Audiobooks, Inc.

Seife, C. (2000). Zero: The Biography of a Dangerous Idea. New York: Penguin.


[1]: Hegel, § 227.

[2]: Einstein, pp. 3-5.

[3]: By this is meant that no part of the ether moves with respect to its other parts.

[4]: Einstein, 6.

[5]: ibid.

[6]: Einstein, 9.

[7]: Sanders, ch. 6.

[8]: Einstein, 10.

[9]: Einstein, 18.

[10]: ibid.

[11]: ibid.

[12]: Einstein, 22-3.

[13]: For more on the heuristic evolution of scientific concepts, cf. Larvor, pp. 14-5.

[14]: Poincaré, 156.

[15]: Seife, ch. 7.

[16]: ibid.

[17]: For more on this point, cf. Norton, ch. 18.


About Graham Joncas

We are a way for capital to know itself.

Posted on April 13, 2012, in History, Science and tagged , , , , , , . Bookmark the permalink. Leave a comment.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: