The Whig History of Science: An Exchange

To the Editors:

Steven Weinberg suggests, in “Eye on the Present—The Whig History of Science” [NYR, December 17, 2015], that historians of science may be permitted to view the past in terms of modern knowledge (called “whig history”), whereas others may not. This is because in science, we know who in the past was “right,” and who was “wrong,” whereas in religion, politics, and other sociological areas, ultimate truths based upon scientific methodology and mathematical applications can never be decided.

But is this really true of science? Before the understanding of atomic energy and radioactivity, could a historian of science in the early 1900s declare that Lord Kelvin was wrong in his mathematical calculation of the age of the earth based upon thermodynamic laws as 100 million years, as many recent historians of science have done? Weinberg implies that we now know these ultimate truths, and are thus in a position to objectively evaluate who was right and who was wrong in the past. But has this not been generally true throughout science’s past, and may not today’s scientific truths give way to further progress? Permit me an example from my own field.

When Edward Jenner developed vaccination for smallpox and Louis Pasteur generalized this approach for other diseases, they did not know what was responsible. Then, in 1890, Emil von Behring and Shibasaburo Kitasato showed that specific antibodies are produced by immunization (or infection) to neutralize dangerous organisms and toxins. Paul Ehrlich then proposed that these antibodies are spontaneously formed by cells and affixed to their surface; interaction with their specific pathogen or toxin will stimulate further production into the blood and thus protect the host from disease. Ehrlich shared the 1908 Nobel Prize for these studies, and his famous theory held sway for several decades.

But then it became increasingly evident that such specific antibodies could be stimulated against a wide variety of even nontoxic substances, including synthetic chemical structures newly produced in the test tubes. Soon, the number of different antibodies found became so large as to defy any hint of a possible Darwinian evolutionary basis. Ehrlich’s theory was cast aside as impossible. Over the following decades, various theories were advanced by such prominent figures as Felix Haurowitz, Linus Pauling, and Macfarlane Burnet, involving the instruction by antigen of how specific antibody structures are formed.

Surely any historian of immunology in the period from the 1920s to the 1960s would have applied his whiggish approach to a declaration that Paul Ehrlich had been wrong in his theory of antibody formation, for everyone knew that the genome could not have produced so large a repertoire of specific proteins. But with advances in molecular genetics, that same Macfarlane Burnet shifted gears, and advanced his clonal selection theory of antibody formation, substantially identical to Ehrlich’s original but long-disavowed concept. This is now the guiding paradigm in immunology, and the whig historian may once more suggest that Ehrlich was right all along!

Scientific understanding is a continuously advancing wave of knowledge, even in the physical sciences, and one cannot be certain that similar progress will not continue beyond the present day. Thus, today’s whiggish condemnations or validations of the historical past, based upon what we now believe, may change with tomorrow’s progress. Most history is still best understood in its own contemporary setting. Let us not be too harsh in dealing with Herbert Butterfield’s critique of whig history. Perhaps a better quote from Butterfield might be: “History is not the study of origins; rather it is the analysis of all the mediations by which the past was turned into our present.” The accent is on the “all”!

Arthur M. Silverstein
Emeritus Professor
Johns Hopkins Medical School
Falmouth, Massachusetts

Steven Weinberg replies:

Arthur Silverstein makes an interesting point about whig history of science: that historians may have trouble judging who in the past was correct or incorrect because our present scientific knowledge may be wrong. Of course, no one thinks we know everything, but we do know some things. To go back to the example in my article, it has been known for centuries that, about the solar system, Copernicus was right against the adherents of Ptolemy, and Newton was right against the followers of Descartes. Knowing this, the historian can appreciate the power of aesthetic judgment in the work of Copernicus, and of mathematical creativity and philosophical open-mindedness in the work of Newton.

Silverstein is right that progress will continue in science (at least, I hope so). But some judgments won’t change. Indeed, since the downfall of the phlogiston theory of fire and the caloric theory of heat in the nineteenth century, there has been no generally accepted theory in the exact sciences that has turned out to be simply wrong. The calculation of the age of the earth and sun by Lord Kelvin, mentioned by Silverstein, is not a good counterexample, even though Kelvin’s ages are now known to be much too short. His results were never generally accepted, but were vigorously contested by geologists, and by biologists who needed more time for the evolution of life. Kelvin himself realized that the earth and sun could be older than he had calculated if “some new sources [of heat], now unknown to us, were prepared in the great storehouse of Creation.” He did not then know of the nuclear reactions that replenish the heat of the earth and sun, but he did take the trouble to estimate the heat contributed to the sun through its collisions with meteors.

This is not to say that, once a successful theory is well established, our understanding of it will then never change. With the advent of special and general relativity, it became clear that Newton’s theories of motion and gravitation are approximations, valid for bodies moving much slower than light in gravitational fields that are not too strong. But this did not make Newton’s work a mistake, or revive Descartes’s. Just the opposite—relativity theory explained why Newton’s theory works, when it does work. In future, relativity theory will doubtless in turn experience a similar reinterpretation, possibly in terms of string theory, but this will not change our whiggish appreciation of Einstein’s success.

Silverstein also attributes to me the view that while historians of science may be whiggish, others may not. I didn’t intend in my article to deny the opportunity of whiggery to anyone, but it is true that I don’t see how it can work in the history of art. Some years ago Richard Rorty attacked me for this, as follows: “Does he really want to disagree with those who think that poets and artists stand on the shoulders of their predecessors, and accumulate knowledge about how to write poems and paint pictures?”

Yes, I do disagree. I don’t think that poets and artists knew more about their craft in the twentieth century than in the nineteenth. Wonderful as were some twentieth-century poems and pictures, I don’t know of any that were superior to the work, say, of Keats or Manet. Of course, this is a matter of taste, an issue that can’t be settled by the sort of investigation that settled the age of the earth. But that is precisely why a whig interpretation makes sense in the history of science, in ways that it does not in the history of art.