it had few parallels in contemporary American government.rnYet throughout the 1960’s and 70’s, the research system wasrnhealthy and productive, the science agencies monitored performancernthoroughly and conscientiouslv, and no one questionedrnthe overall trustworthiness of grant recipients. Politicalrncontroversy surrounded the use of human subjects and animalsrnin experiments, the safety of genetic manipulation, and the disposalrnof hazardous byproducts of research, but fabrication,rnfalsification, and plagiarism were considered to be matters ofrnethics not policy, moral rather than political issues.rnIn the 1980’s, criticism of university management practices,rnespecially indirect cost-accounting practices and the hiring ofrnpublic relations firms to acquire “pork barrel” grants withoutrnpeer review, helped to create a less favorable political climate forrnacademic science. Soon, congressional committees chargedrnwith overseeing R&D began to investigate why the managementrnand evaluation systems created to monitor research andrnresearch communication had failed to detect or prevent severalrnoutrageous examples of misconduct.rnI’he first significant legislative attention to scientific fraudrntook place in 1981, in hearings of the House Committee onrnScience and Technology’s Subcommittee on Oversight and Investigations,rnchaired by Albert Gore, Jr. (then U.S. Representativernfrom Tennessee). Gore opened the hearing with a statementrnthat encapsulated Congress’s interest: “At the base of ourrninvestment in research lies the trust of the American peoplernand the integrity of the scientific enterprise.” The senior officialsrnat the National Institutes of Health and the other prominentrnscientists who testified seemed unresponsive to such reminders.rnNo “real” scientist could ever commit fraud (andrnhence those who committed fraud were never real scientists)rnand, because so few cases of fraud had been found, no elaboraternmanagement or training programs were needed to prove it.rnWhen congressmen heard testimony that NIH was notrnonly funding scientists who had been accused of wrongdoingrnbut that its policy also did not prohibit funding after admissionrnof guilt, they expressed outrage. These were not, however, thernusual ritual statements of Capitol Hill: they typified instead arngrowing, genuine sense of political betrayal, a violation of arntrust relationship of the utmost seriousness. As delays in thernrule-making process within Nil I continued through the I980’s,rnand few universities moved forward with developing proceduresrnto investigate allegations or promote integrity, members ofrnCongress continually warned that, if scientists and universitiesrndid not soon demonstrate a commitment to change, thenrntough and unpleasant federal regulations would follow.rnIn addition to the Breuning case, several other extensive,rnwell-publicized episodes of scientific fraud helped to shape therncontent and direction of political attention. The activities ofrnJohn Darsee, for example, which had been discussed in thernGore hearings, became front-page news in 1981. This caserndrew special attention because of the research topic and sourcernof funding (Darsee was participating in an important, multiinstitutionalrncardiology study funded by the NIH) and thernprestige of Darsce’s mentors and affiliations (training at EmoryrnUniversitv, post-doctoral appointment at Brigham and Women’srnHospital in Boston, and assistant professorship at HarvardrnMedical School). Steriing credentials did not provide a shieldrnagainst corruption. Coworkers at one point reportedly watchedrnin disbelief as Darsee falsified data from a heart monitor attachedrnto a dog, attempting to make it appear that the informationrnhad been collected over several davs not hours.rnThe Darsee case also drew attention to the growing problemrnof “irresponsible coauthorship.” Some of his colleagues hadrnwillingly accepted credit as “coauthors” on articles they hadrnneither written nor read and then blithely disavowed responsibilityrnwhen the articles were challenged. Many coauthors discoveredrnthat Darsee had done them no favor—their resumesrnwere longer but their names were attached to tainted manuscripts.rnFor some of Breuning’s coauthors, professional problemsrndragged on for months. Over 50 articles remained in disputernwhile Breuning refused to discuss the matter publiclyrnand the journals refused to retract them unless all coauthors cooperated.rnIn a case against cardiological radiologist RobertrnSlutsky, a University of California-San Diego investigation determinedrnthat of 137 manuscripts he had written betweenrn1978 and 1985, 77 were valid, 48 “questionable,” and 12rn”fraudulent.” Slutsky, too, had “favored” junior colleagues withrncoauthorship (in part to disguise his improbable rate of production),rnbut was less cooperative in assisting retraction.rnThe attention that these cases brought to the issue of coauthorrnresponsibility did result in some positive change. Certainly,rnthere is increased awareness among young coauthors andrnsome reassessment of publication policies within laboratories.rnThe greatest change has occurred with the journals, many ofrnwhich now require that submissions be accompanied by a letterrnattesting that all coauthors have read and approved thernmanuscript.rnNot all issues raised by scientific fraud have been settled sorneasily. Retractions, for example, help to preserve a journal’srnreputation for accuracy, but the concept of “retracting”rnerrors also relates to maintenance of a field’s intellectualrnintegrity. Scholars see forgeries as “corrupting” the body ofrnknowledge defining a field’s substance; they can create misunderstandings,rnlend support to false theories, or routerncausative explanations down the wrong path. Unfortunately,rnfake scientific knowledge is not like fake art—we cannot simplyrntear an offending page from a journal like we remove forgedrnpaintings from museum walls. “Retraction” is actually just arnnotice (usually published in a later issue of the same journal)rnthat the previous article should be disregarded. Handling retractionsrnefficiently and fairly continues to be a significantrnproblem for scientific publishing. With electronic data bases,rnretraction notes can be tagged or linked to the original article.rnBut what about subsequent references to the work? Or crossreferences?rnAnd what about data in limbo—questioned butrnnot yet proven false?rnEven a sympathetic observer of the political negotiations surroundingrnscientific fraud might conclude that none of thernparties to the debate—universities, government officials, accusedrnscientists and their lawyers, whistleblowers and theirrnlawyers, congressional oversight committees, and people claimingrnto speak for the scientific community—has much interestrnin resolving it swiftly or simply. At no time has everyonernagreed on a common goal or acceptable outcome to the debate.rnSome of the hearings presided over by John Dingell resemblernMonty Python sketches—silly scientists, overly serious whistleblowers,rnhardhearted legislators—and might even be funnvrnwere the stakes not so high—millions in grants, tenuous careers,rnand Nobel reputations. Some of this hype and hysteria appearsrnto have receded in the past year, but it would be difficult to decidernwhether this is due to journalistic boredom, political hiatus,rna temporary truce, or a genuine shift to reasonableness onrnSEPTEMBER 1993/23rnrnrn