Fraud and Trust in Science

FRAUD AND TRUST IN SCIENCE STEPHAN FUCHS and SAUNDRA DAVIS WESTERVELT* It is more ignominious to mistrust than to be deceived.—La Rochefoucauld Two Forms of Scientific Deviance "Deviance in science" should be, and often is, separated from "scientific deviance." The former is much more frequent than the latter [1, p.552]. Deviance in science includes all behaviors and communications that violate established law or other social conventions, such as sexual harassment or the improper handling of research funds. Modern science is a work organization, and as such subject to similar rules and regulations as govern other work organizations. As opposed to this very broad category, we understand "scientific deviance" to refer to misconduct in the act of doing science. We deal only with scientific deviance or misconduct, not with deviance in science. Following Zuckerman [2, p.521], we distinguish here two common forms of scientific deviance, fraud and plagiarism. There are many contested , ambiguous, and changing definitions for both. Various agencies and institutions define and interpret "misconduct" in different ways. However, the main problem seems to be how to demarcate fraud and plagiarism from other "deviant" research practices—such as divergent interpretations, error, sloppiness, selective presentation of data, or innovations . One popular way to distinguish scientific misconduct from other The authors thank the Biology Department at the University of Virginia, and especially Bob Kretzinger, for providing the medium within which this argument was first presented. Special thanks are due Paul Gross, whose most generous support was critical for completing this project. *Department of Sociology, University of Virginia, Charlottesville, Virginia 22903.©1995 by The University of Chicago. All rights reserved. 0031-5982/95/3901-0942$01.00 248 Fuchs and Westervelt ¦ Fraud and Trust in Science deviance, with precedents in law, is by the intention to deceive [3-5]. While not all relevant agencies have adopted this practice, it is nevertheless influential, and colors even definitions that do not explicitly require the intention to deceive before establishing fraud. For example, the federal Office of Research Integrity (ORI) will have to establish intent to deceive in the highly publicized Imanishi-Kari case to overrule her appeal of ORI's guilty verdict [6]. As of September 1993, the ORI, probably the foremost judicial authority on misconduct, stated that misconduct "does not include honest error or honest differences in interpretations or judgments of data" [7, p.2]. This modification was in response to the criticism that the previous definition of misconduct— "fabrication, falsification, plagiarism, or other practices that deviate from those that are commonly accepted within the scientific community "—gave too much room for accusations against innovative science. The importance of intent is also indirectly stressed in the definition of misconduct given by the National Academy of Sciences (NAS): "Misconduct in science does not include errors of judgment; errors in the recording, selection, or analysis of data; differences in opinions involving the interpretation of data; or misconduct unrelated to the research process" [8, p.5]. Even the National Science Foundation's (NSF) definition of misconduct—which still includes the phrase "serious deviations from accepted research practices"—emphasizes intent, insofar as "ordinary errors" and "ordinary differences in interpretations orjudgements of data" are not "considered to be misconduct under this definition" [4, p.290]. The NSF also cautions that its definition of misconduct does not endorse an "interpretation that would make it misconduct in science just to do something novel or unorthodox" [9, p.647]. Intent to deceive, then, appears significant in most understandings and definitions of fraud, even if it is not included expressis verbis. We define "fraud," then, as the deliberate fabrication or falsification of evidence with the intention to deceive or misinform peers and other audiences of scientific knowledge. In addition, some, but not all, definitions now require that falsifications and fabrications have a "significant effect" on the research and its results [10, H]. This shows that what is sometimes considered decisive is not falsification or fabrication per se, but only insofar and inasmuch as it influences what is eventually communicated to others. That is, the emphasis is on communication, not action. "Plagiarism" is the intentional failure to credit known others for some idea, invention, or discovery. Plagiarism is theft of intellectual...