Revolution in Science

To the Editors:

In Ian Hacking’s review [NYR, February 27] of my Revolution in Science (Belknap Press of Harvard University Press, 1985), two examples supposedly show the inadequacy of my criteria for the occurrence of a revolution in science. In each example Hacking finds that the events would not constitute a revolution according to my criteria and he observes that historians do not generally consider them to be revolutions; he himself, however, “instinctively” feels that these were revolutions.

The first is the reform of medicine associated with Xavier Bichat at the end of the eighteenth and beginning of the nineteenth centuries. Citing Michel Foucault (who dealt with this topic in The Birth of the Clinic), Hacking states that Bichat and company produced a significant change in medicine, including the introduction of the practice of pathological anatomy, leading to the creation of new institutions (the clinic). Hence it would be wrong to deny these events a revolutionary status. But according to my standards, Hacking argues, this would be no revolution because “Xavier Bichat—or whomever else we choose as a prime medical revolutionary—seems not to have called this transformation in medical practice a revolution.”

But there is a contemporaneous witness testifying to the revolutionary change, one who is quoted on the opening page of Foucault’s chapter on Bichat in The Birth of the Clinic. This witness wrote of the “astonishing revolution that is characterized by progress in pathological anatomy.” I should add that in my criteria the contemporaneous witness need not be the principal investigator, especially since in many years of research I found less than twenty scientists who announced that their work was revolution-producing.

Hacking’s second example is Humphry Davy’s discovery that chlorine is a chemical element. The significance of this event is explained by Hacking in relation to Lavoisier’s theory of chemistry which postulated that oxygen is the fundamental acid-forming principle (hence the name oxygen). Since common muriatic acid was known to be composed of hydrogen and chlorine, Davy’s discovery that chlorine is an element “proved that oxygen is not the essence of acidity.” In its day, according to Hacking, “this was hailed as revolutionary.” Hacking provides an alleged witness: “In 1817 the new professor of chemistry at Harvard College devoted much of his inaugural lecture to what he called this second revolution in chemistry.”

The witness in this case, Professor John Gorham of Harvard College, in an inaugural lecture of December 11, 1816, published in Boston in 1817, did discuss many aspects of Davy’s discoveries and opinions. But ‘(1) nowhere in this 23-page pamphlet does Gorham ever use the expression “second chemical revolution.” Nor, (2) does he describe any of Davy’s discoveries as a revolution; (3) he does not ever specifically mention Davy’s discovery that chlorine is an element, although he does discuss Davy’s discovery or identification of chemical elements (which are not specified by name). Finally, (4) in the two brief references to the overturning of Lavoisier’s theory of acids, Gorham does not introduce the example of muriatic acid, but rather indicates that oxygen had been found to be a constituent of some alkaline substances.

May I conclude by correcting a factual error about myself. Attributing my supposed “distrust of generalization” to the influence of Herbert Butterfield, Hacking concludes: “It is no accident that when in England Cohen held a Visiting Fellowship at Peterhouse, the Cambridge college of which Butterfield was Master.” I have never been a Visiting Fellow (nor any other kind of Fellow) of Peterhouse although, in my visits to Cambridge, I have been a founding Visiting Fellow of Clare Hall and a Visiting Overseas Fellow of Churchill.

I. Bernard Cohen

Harvard University

Cambridge, Massachusetts

Ian Hacking replies:

Professor Cohen queries two consecutive examples of mine. Between the two I wrote that “the argument [whether or not an event is a revolution in science] will not rest on any lack of use of a word, ‘revolution.”‘ Thus his observations do not bear on our disagreement, namely the importance of his four tests of whether an event is a revolution, namely (i) Do contemporaneous witnesses call it one (ii) Do subsequent textbooks do so? (iii) Do competent historians do so? (iv) Does our present scientific lore do so? Cohen thinks (i) is very important. I suggested two examples, one looking like a major revolution in science, but not passing (i), another not commonly regarded as revolutionary, but passing (i). In both cases I was urging that although we might disagree whether one or the other event is revolutionary, it would be in terms of the event, not in terms of what people said about it. My examples are not crucial to this position, but on present evidence, they still look better than Professor Cohen suggests.

Bichat. Cohen quotes Foucault quoting P.F.O. Rayer as evidence of a contemporaneous witness for test (i). But Foucault’s paragraph begins with the remark that historians were quick to notice the change in pathological anatomy. Rayer’s book which he then quotes is called a history, Sommaire d’une histoire abregée de l’anatomie pathologique. Although a dissertation for the faculty of medicine in Paris, published in 1818, it is a history, doing the subject century by century. Rayer was just nine years old when Bichat died in 1802. I think that his words might be used for tests (ii) or (iii), but not for (i). Alternatively one could argue that Rayer was contemporaneous with an ongoing revolution, begun by Bichat. But it will be the events, not the use of the word by a minor figure such as Rayer, that convince us of that.

Rayer spoke of revolution in imitation of Cabanis, to whom he dedicated his book, and who in 1804 published a Coup d’oeil sur les révolutions et sur la réforme de la médicine. Born before and dying after Bichat, he is contemporaneous. It is of no moment that he does not mention Bichat in this book, for it was written almost a decade earlier, and dated “germinal l’an III.” But Cabanis did express much admiration for his colleague Bichat, even inserting a dolorous footnote in a book he was proofreading on the day of Bichat’s death. Yet Cabanis, the classic contemporary author of the time to write of revolution in medicine, never did call Bichat revolutionary, nor do we yet have a text in which he spoke of the revolution in pathological anatomy of his day.

Gorham. I regret having written as if Gorham used the very words, “second chemical revolution.” Instead he spoke of the “successive revolutions” in chemistry. After presenting that of Lavoisier (now commonly called the chemical revolution) he turns to its overthrow. “Oxygen is not the only supporter of combustion, it is not the principle of acidity,” etc. The “system of Lavoisier”

is already fallen, and the future chemist will regard it with the same feeling with which the scholar contemplates the ruins of a Greek temple over which time has passed his relentless hand, and has withered its beauty, and levelled the harmonious structure with the dust.

To account for the rapid downfall of the French philosophy it will be necessary to take a cursory view of the present state of chemical knowledge.

In my brief paragraph I was even more cursory than Gorham. I used as a relevant example of Gorham’s knowledge the item most accessible to the modern reader. Oxygen cannot be the essence of acidity because one of our most common acids is HCI. The new professor of chemistry at Harvard College mentions other items at length, as Professor Cohen notes. Some of these show what he learned from Davy, and some show how ill he had digested Lavoisier’s achievements. But his words seem to me to establish that Gorham included the work of Davy’s school among his “successive revolutions.”