In response to:

The Pioneer Defended from the December 21, 1995 issue

To the Editors:

As a scientist interested in history and philosophy of science, I am dismayed by the gulf that separates a few of my friends and colleagues on some fundamental issues. The review by M.F. Perutz [NYR, December 21, 1995] of The Private Science of Louis Pasteur by Gerald L. Geison reflects this deplorable situation. This review is unfair and inaccurate.

Thomas Kuhn noted that scientists generally think of history in terms of current outcomes of past science, i.e., what pre-saged the present. Yet the historian’s task is to interpret the past in its own terms: how did Pasteur think about his experiments rather than what a modern scientist thinks was “really” happening. Perutz’s review was sensitive to neither of these issues.

Repeatedly, Perutz writes that Geison “accuses,” “castigates,” and “insinuates [things about]” Pasteur. Such pejorative words do not occur in Geison’s test. Consider how Perutz’s points relate to Geison’s text:

  1. Perutz writes: “By a painstaking comparison of Pasteur’s notebooks with his publications, Geison claims to have found him guilty of deception, of stealing other peoples’ ideas, and of unsavory and unethical conduct.” Geison’s analyses, however, yield more muted conclusions. Geison elaborates the context of Pasteur’s experimental path and juxtaposes it to his public pronouncements. “Stealing” and “unsavory” as used by Perutz reflect neither the tone nor tenor of Geison’s book.

  2. Perutz writes: “Geison insinuates that Pasteur cheated because the effects on polarized light of his right- and left-handed salts of tartaric acid, which according to Pasteur’s interpretation should have been exactly equal and opposite, were in fact slightly different.” What Geison did say was that Pasteur believed in his interpretation even though he observed differences in optical rotation, and in order to save his hypothesis, he offered a (presumably reasonable) explanation for the difference. In Geison’s phrase, “Pasteur minimized the difference—in effect, he explained it away ….” In making such a judgment about his limited confidence in his experimental results, Pasteur, indeed, was acting as scientists do today.

  3. Perutz objects to Geison’s suggestion that Pasteur slighted his collaborator, Auguste Laurent, and claims that Geison accuses Pasteur of “opportunism.” From the full text of the Pasteur-Dumas letter (p. 84), it appears that Pasteur wrote an ingratiating letter to Dumas in the hope Dumas would become his new patron. This and Geison’s other evidence on the politics of Laurent and Dumas make it reasonable to suspect that Pasteur’s gratuitous deprecation of Laurent was intended to help insinuate himself into Dumas’s circle.

  4. Perutz writes that Geison’s analysis of the importance of philosophical, religious, and political interests in the Pasteur-Pouchet debate are without “any clear evidence for the claim.” Yet contemporary accounts (e.g., Richard Owen, p. 124) described this controversy as in “curiously close” analogy to that between Cuvier and Geffroy which, Owen observed, involved theological, philosophical, and political issues.

  5. Perutz says that “Geison implies that Pasteur acted dishonestly…” in his discussion of Pasteur’s failure to repeat Pouchet’s experiments exactly as described by Pouchet. Perutz apparently missed Geison’s point on pages 129-132, where he asks us, as an act of historical methodological analysis, to take quite literally Pasteur’s public statement that he approached the problem of spontaneous generation “without preconceived idea.” Geison says: “Let us pretend for a while that we believe that statement and that we also believe in the so-called Scientific Method. If so, we can only be dismayed by some surprising lapses in Pasteur’s modus operandi.” On page 132, Geison concludes this analysis, and notes that “for what sounded like criticisms of Pasteur just paragraphs ago are really criticisms of a simplistic and passé notion of the Scientific Method. It is not Pasteur who has fallen short; it is this Scientific Method.”

  6. Perutz focuses on outcomes when he reviews Geison’s analysis of Pasteur’s famous anthrax vaccine trials. Pasteur’s notebooks and the recollections of his laboratory technician support Geison’s conclusion that Pasteur did not use the vaccine he had publicly announced (based on his theory of air-oxygen attenuation); instead, he used a bichromate-inactivated vaccine prepared by the method of his competitor, Toussaint. As a modern chemist, Perutz notes that both air-oxygen and bichromate are oxidizers, so it didn’t matter which vaccine Pasteur used because they both work in the same way. As a historian, however, Geison is rightly concerned with how Pasteur thought about his work; modern chemistry is off limits. All the evidence suggests that bichromate was conceived as an anti-septic quite different from air; thus, it did matter whether he used air or bichromate. Based on the evidence presented, Pasteur must have known that his public statements about the Pouilly-le-Fort vaccine were at odds with the actual vaccine used.

These trials served “rhetorical” purposes, that is, to convince the public and win support. The pressure to achieve the predicted outcome was enormous. In another episode (not described by Geison) Pasteur carried out trials with a vaccine for rouget of swine. The Pasteur-Thuillier correspondence* reveals that Pasteur prepared explanations in advance for his young colleague, Thuillier, to use, no matter what the outcome. This correspondence, too, shows that the rhetorical strategies in these public trials took precedence over “scientific” purposes.

  1. Perutz, again with clear hindsight, argues that Pasteur should have had no ethical qualms about his rabies vaccination trials, and that his critics were unfair or wrong. Geison carefully describes, as best he can from the notebooks, the state of Pasteur’s knowledge of the safety and efficacy of the rabies vaccine when he decided to proceed with human trials. Geison carefully distinguishes between therapeutic trials on Pasteur’s two “private patients,” already known to have rabies, and experimental trials such as that on Joseph Meister, for whom the diagnosis was unknown. Geison sets this information in the context of the ethics of his time, and in his view, Pasteur comes out wanting. Geison’s picture is of a scientist whose ideas about vaccines and immunity are in great flux, but who is under immense pressure to perform dramatic cures.
  2. Perutz attacks Geison for having an ideology which denies the very existence of correct explanations of phenomena. Perutz, in his examples, adopts an absolutist view of truth. What is at issue here is not that Pasteur’s ideas have been strongly confirmed in many ways, and hence are believed to be “true” today. The issue is how Pasteur went about his work as a scientist. The historian cannot unravel such activities in terms of ultimate outcomes; he or she must analyze the relevant data. Geison’s book is grounded in laboratory research notebooks; this is as close as a historian can get to empirical evidence, yet Perutz attacks him for “unethical and unsavory conduct” and “contrived evidence.”

The real villain here is the philosophical position Perutz labels “relativism.” Professor Geison’s position is by no means a simple relativist one, and he has done much to recognize and contextualize the legitimate work of scientists. Yet he and his book have drawn the misdirected ire and scorn of Dr. Perutz. Thus: “The entire approach emphasizing ‘relative’ truth seems to me a piece of humbug masquerading as an academic discipline; it pretends that its practitioners can set themselves up as judges over scientists whose science they fail to understand.” Unfortunately, in this review Perutz passes judgment on historical scholarship, the nature of which he, himself, apparently fails to understand.

William C. Summers, M.D., Ph.D.
Departments of Molecular Biophysics and Biochemistry and Genetics
Yale University
New Haven, Connecticut

M.F Perutz replies:
  1. Dr. Summers writes that pejorative words such as “unethical” and “unsavory conduct” do not occur in Geison’s text. Here are some quotations from it: “Here we deal not with mere acquiescence in the formulaic genre of scientific papers and the associated ‘inductivist’ image of science, but with discrepancies between Pasteur’s public and private science in cases where the word ‘deception’ does not seem so inappropriate, and even ‘fraud’ does not seem entirely out of line…” (p. 16); “But the effort to analyze Pasteur’s ethically dubious deceptions…” (p. 17). In commenting on the public demonstration of Pasteur’s anthrax vaccine, Geison accuses Pasteur of “actively misrepresenting the nature of the vaccine actually used” and calls this “a significant and undeniable element of deception.” In commenting on Pasteur’s vaccination of Meister, Geison writes: “boldly, even recklessly, Pasteur was willing to apply vaccines in the face of ambiguous experimental evidence about their safety and efficiency.”
    He continues: “In fact, as Dr. Peter suspected and as Dr. Roux knew full well, the decision to treat Meister was ethically dubious by the then prevailing standards, as was some of the rest of Pasteur’s conduct in his headlong and headstrong quest for vaccines.” In my review of Geison’s book and my later exchange of letters with Geison I showed these accusations to have been false.

  2. Dr. Summers denies that Geison’s phrase “Pasteur minimized the difference, in effect he explained it away” insinuates that Pasteur cheated. Surely you explain away only things that are dubious or untrue. Elsewhere Geison writes: “Pasteur displayed an almost magus-like capacity for almost literally creating crystals of a sort that confirm his alleged [my italics] correlation between optical activity and crystalline symmetry.” The word “alleged” implies once more that the correlation was untrue. In fact, the small discrepancy between the optical rotations of the solutions of the two kinds of crystals found by Pasteur was a normal experimental error. There is no evidence that Pasteur had only limited confidence in his results. They were spectacular, and there was nothing to be explained away or to be alleged.

  3. There was no need to attribute Pasteur’s failure to acknowledge the guidance he had received from his teacher Auguste Laurent to political opportunism, because Laurent’s ideas on the relationship between the crystalline forms of tartaric acid and the transmission of polarized light when they are dissolved in water had proved misleading. Laurent had started Pasteur off on the wrong track.

4-5. The controversy over spontaneous generation between Cuvier and Geffroy may have involved philosophical, religious, and political questions, but Pasteur’s disproof of spontaneous generation was based on solid evidence which would have been the same if he had been an atheistic anarchist and not a conservative Catholic. Nor was there any lapse in scientific method by the standards of Pasteur’s day. His experiment could not have been done more carefully and methodically by either this or the last century’s standards.

  1. Toussaint used Lister’s antiseptic, carbolic acid, called phenol today, rather than bichromate. You don’t need to be a modern chemist to know that bichromate is an oxidant, because it is a fact of elementary chemistry that was already well known in the last century. Pasteur’s collaborators would have known this when they used it as an alternative to oxygen. Unlike Toussaint’s carbolic acid, it left the anthrax bacilli alive so that Chamberland could further attenuate them by three passages through mice. Geison’s accusations of “active misrepresentation” are therefore groundless. Pasteur carried out the public trial of the anthrax vaccine not as a rhetorical stratagem, but because he was challenged to it by the skeptical vets.
  2. Concerning Pasteur’s vaccination of Joseph Meister, Summers’s letter gives a misleading impression.

The dog attacked the nine-year-old boy, threw him on the ground, and bit him fourteen times on his hands and legs. A bricklayer who saw the scene beat the dog off with an iron bar and picked up the boy, who was covered in blood and dog saliva. The dog returned to his master, the grocer Théodore Vone, and bit him in the arm. Vone shot the dog, whose stomach was found to be full of hay, straw, and bits of wood. These facts leave no doubt that the dog was rabid.

Joseph, his mother, and Vone travelled to Paris to see Pasteur, who assured Vone that his own bite on the arm was superficial and presented no danger. Pasteur then consulted the neurologist Professor Vulpian and Dr. Grancher, a physician in the same laboratory, about Joseph. On seeing the many bites, some very deep, Vulpian expressed the view that Pasteur’s experiments with dogs were conclusive enough to apply his vaccine to Joseph, and the two physicians went ahead. This account does not support Geison’s damaging accusation that “boldly, even recklessly, Pasteur was willing to apply vaccines in the face of ambiguous experimental evidence about their safety and efficacy.” The dog was clearly rabid, and experience had shown bites of the severity of Joseph Meister’s to be fatal. Therefore Pasteur would have been timid and heartless not to try to save Joseph’s life. Pasteur was not under public pressure to perform dramatic cures; he was pressed by Joseph’s desperate mother to save her son’s life.

  1. According to Geison, “Pasteur shared with many of his peers a rather simple minded and absolutist notion of scientific truth, rarely conceding the possibility of its being multifaceted and relative.” According to Geison, Pasteur’s “scientific beliefs and modus operandi were sometimes profoundly shaped by his personal concerns, including his political, philosophical and religious instincts,” while the real individual scientist “…tries to navigate a safe passage between the constraints of empirical evidence on the one hand and personal or social interests on the other.”

Dr. Summers has not been able to explain to the readers of The New York Review why Pasteur’s discoveries of molecular asymmetry, of fermentation as a manifestation of life, of the impossibility of spontaneous generation, or of vaccination by attenuated micro-organisms were multifaceted and relative. Nor do I think he could name a single Nobel Prize-winning discovery in physics, chemistry, or medicine where the discoverers acted as he charges, by trying to navigate a safe passage between the constraints of empirical evidence on the one hand and their personal or social interests on the other. If neither Dr. Summers, nor social historians of science can support their views by such concrete examples, then the entire approach is clearly humbug.

This Issue

February 6, 1997