Philosophy of New Music
On Fact and Fraud: Cautionary Tales from the Front Lines of Science
by David Goodstein
Princeton University Press, Princeton, Oxford, 2010
168 pp. Trade, $21.75
Reviewed by Jonathan Zilberg
University of Illinois at Urbana-Champaign
On Fact and Fraud is a classic example of an exercise in clarity and brevity. It is the first book on the subject of the legal definition of fraud in science and codes of conduct for researchers. Richly contextualized with fascinating case studies and written by an intimate insider, a physicist of repute and senior university administrator, it is a study in caution and good measure. In addition, it is of historical import for science writing considering both its elegance and the high profile cases of fraud and alleged fraud covered therein. Indeed, the way in which Goodstein makes complex issues in advanced science accessible and the way in which he deals sensitively with such compromising and tragic circumstances, is a testament to the author's skill as a writer and his level headed compassion. All in all, for undergraduates about to embark on a career in science it makes for indispensable reading.
What will be surprising to the lay reader is how it could be possible to read so effortlessly about what would, otherwise, be overly complicated topics to even consider engaging. Take for instance, the elegant way in which Goodstein revisits the famous posthumous case against Robert Andrew Millikan (the author of The Electron (1917) and first Nobel laureate) for data manipulation in determining the charge of the ion - and for anti-Semitism and male chauvinism in addition. Goodstein settles these accusations deftly for the record. In a masterfully illustrated chapter reproducing the original pages from Milikan's laboratory notes testing Stokes' Law, he concludes that a careful contextual reading of the data and publication at issue "greatly diminishes their apparent significance as evidence of misconduct" (p. 47). From allegations of fraud in the discovery of the charge of the ion to the same issue as it concerned the discovery of the AIDS virus, one finds oneself so quickly at the end of the book that one wonders how one could have read about such high science so easily and with so much pleasure. Surely the answer lies in its presentation. There is not an unnecessary word or comma. In essence, then, for science writers above all, this book presents the ultimate Strunk and White experience . And again, no less important is the ethical value of the text as an exercise in compassion and judicious good measure.
The first chapter sets the stage through describing the nature of the problem of fraud as it drew in Goodstein in practical terms. He begins by deftly outlining the progression of a scientist's career to introduce the reader to the stresses and strains and expectation that have to be met in order to enter into and succeed in the highly competitive world of research and publication. Then come the case studies, and there the reader will learn more about particle physics, molecular biology, immunology, virology and the likes than one could have imagined possible. And while that in itself is a fascinating and unexpected experience adding enormous value to the book, this is, above all, a first hand account of how the Caltech policy on research misconduct came about. On the latter issue alone, the book constitutes an indispensable classic on creating policy and on scientific fraud or "ffp" with "'fabrication' being defined as making up results, 'falsification' as changing and omitting data or results, and 'plagiarism' as the appropriation of ideas without credit" (p. 67).
In the opening discussion on the nature of science, Goodstein revisits Bacon and Popper, falsifiability, and theory building. He describes the reality versus ideal nature of scientific research in generating new hypotheses and the authority and reward structure within which scientific advances and careers are made. More specifically, in the course of the subsequent discussion in each chapter, he underscores the importance of protocols and guidelines for pursuing charges of fraud, especially in universities, and of the importance of not rushing to judgment. As noted above, the study itself is an outcome of the creation of the Caltech campus policy governing the process of investigating and handling fraud. Written by the author in 1988, the policy was put to the test soon after he had completed the final draft . Appointed as a member of the ethics committee dealing with the cases against Caltech immunologists Vipin Kumar and James L. Urban, Goodstein provides us with an inside view of the problem and how it was dealt with on campus. These cases were of particular importance to Caltech's reputation having taken place in the famed laboratory of Leroy E. Hood, the university's most prestigious biologist.
In Kumar's case, a figure had been faked in the Journal of Experimental Medicine. In Urban's case, fake data had been submitted in a manuscript to the journal Cell and then replaced with the actual data by the time of publication. As Goodstein relates, procedure was duly followed and censure taken . While Kumar went on to re-establish his career and become the head of the Laboratory of Autoimmunity at the Torrey Pines Institute of Molecular Studies, Urban's case was different. Disgraced, he disappeared.
A similar tragedy of a ruined career happened in the astounding case of John Hendrick (Jan Hendrik) Schön, a brilliant young experimental physicist working in carbon based semi-conductor research who published a paper every eight days between 1998 and 2001. Shortly after being offered the post of the Director of the Max Plank Institute, his paper, "Field Effect Modulation of the Conductance of Single Molecules," for Science and a similar article to Nature were found to be fraudulent . The ensuing investigation showed that there was no raw data and there were no original samples. The problem had only been noticed by Lydia Sohn and Paul McEuen as the data was "too perfect," the same curves were given for different experiments and the background noise on the curves was identical. These two instances turned out to be only the tip of Schön's iceberg. As Goodstein duly notes, the "ethical fulcrum" of peer review works well at "separating real science from nonsense" but not as well for detecting fraud (p. 17). Schön, the golden boy and almost God-like figure of condensed matter physics, was fired from Bell Labs, and his doctorate from the University of Konstanz and a number of articles revoked .
The case studies serve to illustrate a wide range of fraud, real and alleged. Goodstein presents each in the most precisely measured and yet interesting way. Despite the detail, scientific, procedural and ethical, and presented within the larger social context of each scientific community, he is able to consistently hold one's attention. In doing so, he reveals fascinating aspects of how experimental science differs depending on the field and how this impinges on fraud in terms of reproducibility as noted above. In another case he explores, in the field of super-conductivity and field-effect doping, it is difficult to reproduce results. In fact, this instance of fraud had only been noticed because of duplicated results. In contrast, take the case of Victor Nimov then of the Lawrence Berkeley National Laboratory. Nimov fabricated the evidence for the existence of element 118. Because he was working in a field in which precise reproducibility is expected, his case was easier to detect.
Similarly, in the case of the "too good to be true" discovery of high temperature superconductivity, reproducibility was expected and was not forthcoming. In addition, it was mercilessly and quickly publicly demolished by Lewis, Barnes and Koonin of Caltech as theoretically impossible and, thus, pronounced "effectively dead" early in the scandal. Yet, perhaps out of deference to his eminent and good friend Franco Scaramuzzi, Goodstein goes to great length to show how a scientist of such repute could have found himself in a deeply compromised situation and, furthermore, concludes that in his opinion, the verdict is not yet final on cold fusion. He also shows how Scaramuzzi's laboratory continues to struggle to maintain the research agenda and, in so doing, contributes to technical advances in the field driven on by the occasional inexplicable bursts of energy. It does not, however, come to the same thing as the all important vindication of the virologists Mikulas Popovic and Thereza Imanishi-Kiri respectively in David Gallo's and David Baltimore's laboratories, the principals having in the process also experienced the full, if partial and temporary effects of the professional consequences of allegation of fraud by the mere fact of hierarchical responsibility.
Thereza Imanishi-Kiri was accused by a post-doctoral student of falsifying published immunological data based on the student's inability to reproduce the results. And even though David Baltimore, the head of the laboratory and then President of Rockefeller University, was never charged or even directly involved, he lost his job there. Subsequently, after the Department of Health and Human Services overturned the previous government verdicts against Gallo (and Popovic) as "meaningless," he was appointed President of Caltech. Similarly, while Imanishi-Kiri's tenure was upheld and she was barred from Federal funding for a decade, she was duly tenured at Tufts. Rather differently to the Baltimore-Imanishi-Kiri case, in the case of the Gallo controversy, the issue devolved on whether the laboratory had claimed credit for discovering the AIDS virus and whether Popovic had misappropriated cell cultures from the Pasteur Institute in Paris - in addition to allegations of having misrepresented and falsified both the data and methodology. Though Gallo was not included in the Nobel Prize, the careers of each of these scientists were, in the end, not otherwise adversely affected.
There is a history of tragedy in these cases, considering the arguably unnecessary trauma involved. In the cases of Schön and Nimov, they are exquisitely farcical but exceedingly rare. In Urban's case, claiming that one is sure what the results for an experiment would have been, and despite in this case these turning out to be correct, is hardly an excuse for submitting a fraudulent scientific paper to a prestigious journal. Interestingly enough, Urban would have gotten away with it had the material not surfaced during the Kumar investigation. While that case had minor consequences for the field at large, the consequences of the cold fusion saga were more long lasting. Stanley Pons and Martin Fleishmann who announced the discovery of cold fusion became "scientific outcasts" in a "pariah field" of "pathological science" and most memorable of all of Goodstein's pithy choice of words – a "sincere delusion" – though he is at pains to illustrate that the issue here was not a matter of fraud but bad science. In thus addressing such entirely different allegations and cases of real and alleged fraud and their consequences, Goodstein provides us with an unexpectedly rich insight into a wide range of experimental science that could probably not have been so effectively addressed had it not been for the unifying theme. Hence, we learn as much about science here as fraud and the proven lack or mitigation thereof.
To end, this text should be required reading for all young scientists. The case studies will be of particular value to historians of science, science writers, and journalists in terms of the need to be circumspect in making judgments on areas outside of their expertise, particularly for instance in the best known case of Robert Milliken. It is in all this fascinating how Goodstein is able to write about complicated and highly distasteful matters with such simplicity and grace. The consequence is that one comes away with an enormous respect for the integrity and common sense of the author and of what goes into creating such policies and applying them. The lessons provided herein for future scientists themselves are these: no matter the professional pressure one is under, no matter the expectations of reproducibility in one's field, even the suspicion of "fraud" will grievously damage you, your superiors and your institution - unless of course you are a social scientist who doesn't believe in the distinction between fact and fiction.
[1 ] See Strunk, Jr., William; E.B. White The Elements of Style (4th ed.). Boston: Allyn & Bacon, 1999 . Also see http://en.wikipedia.org/wiki/The_Elements_of_Style where Geoffrey K Pullum is quoted as referring to it as "the book that ate America's brain."
 The Caltech policy on the procedures for dealing with potential ethics violations, and what constitutes fraud is included at the end of the book as an appendix. It will be as useful for administrators as it will be for practicing and future scientists. Also see, Eleanor G. Shore, "Effectiveness of research guidelines in prevention of scientific misconduct", Science and Engineering Ethics Volume 1, Number 4, see http://www.springerlink.com/content/4v2110n7m47mjw72/.
 In Caroline Whitbeck's all important article "Truth and Trustworthiness in Research" published in the Online Ethics Center for Engineering and Research, October 1995, which covers much of the same material in Goodstein's book, she questions the decision that Urban was judged not to have committed fraud but "serious misconduct". The moral basis for this was that Urban did not intend to deceive the editors in that he sincerely believed he would be able to reproduce the invented results but as she explores, as I read it, the decision was legalistically debatable because the editor was in fact deceived and that in any event, the deception certainly did constitute a form of reckless endangerment. See: http://www.onlineethics.org/Topics/RespResearch/ResEssays/cw2.aspx.
 For a scintillatingly detailed account, see the BBC transcript, "The Dark Secret of Hendrik Schön" narrated by Jack Fortune at http://www.bbc.co.uk/science/horizon/2004/hendrikshontrans.shtml. Also see, E. S. Reich, Plastic Fantastic: How the Biggest Fraud in Physics Shook the Scientific World, New York: Macmillan 2009. For similar media scandals as regards physics and post-modernism respectively, see the Bogdanoff and Sokal affairs, particularly Stephen Hilgartner's "The Sokal Affair in Context". Science, Technology, & Human Values 22 (4): 506-522, Autumn 1997.
 For the Bell report and the list of journal retractions, see http://en.wikipedia.org/wiki/Sch%C3%B6n_scandal.