Sir — Although biometric technology isbelieved to be a product of the hi-tech era,it is not, in fact, our generation’s invention.People were using biometric technologylong before the word ‘biometric’ wascoined (see Nature 418, 583; 2002; and420, 15; 2002). Not only that, but attemptsto fool it were as common in ancient timesas they are today.

The oldest written testimony of identitytheft we can find dates back to biblicaltimes, when Jacob fraudulently used theidentity of his twin brother Esau to benefitfrom their father’s blessing. Genesisdescribes a combination of hand scan andvoice recognition that Isaac used to attemptto verify his son’s identity, without knowingthat the smooth-skinned Jacob hadwrapped his hands in kidskin: “And Jacob

went near unto Isaac his father; and he felthim, and said, ‘The voice is Jacob’s voice,but the hands are the hands of Esau’. And herecognized him not, because his hands werehairy, as his brother Esau’s hands.” The falseacceptance which resulted from this veryinaccurate biometric test had historicalconsequences of unmatched proportions.

In Greek mythology, too, we are likelyto find surprises. A primitive tactile sensorused by the one-eyed Cyclops afterOdysseus and colleagues had destroyed his monocular face-recognition system —and which they evaded by hiding under his sheep — was actually the firstbiometric lock, operated more than twomillennia before James Bond conqueredthe screen with his hi-tech gadgets.

Turning the dusty pages of the classics,

we discover a wide spectrum of biometrictechnologies, from voice recognition —triumphantly deceived by Dante’s GianniSchicchi, who impersonated a dead man to change a will in his own favour — to the unbeatable feature-matching face-recognition algorithm implemented by thefairy-tale heroine Little Red Riding Hood,who was unconvinced by a wolf disguisedas her grandmother.

Scientists would be envious to find out that many state-of-the-art approachesin biometrics are merely a rediscovery: toquote Ecclesiastes, “nothing is new under the sun”. Michael M. Bronstein, Alexander M. Bronstein Department of Electrical Engineering, Technion -Israel Institute of Technology, Haifa 32000, Israel

correspondence

NATURE | VOL 420 | 19/26 DECEMBER 2002 | www.nature.com/nature 739

Biometrics was no match for hair-raising tricksPeople have been fooling the latest thing in security for a very long time.

Scientific misconduct:the state’s role has limitsSir — The Association of American MedicalColleges (AAMC) has long fostered researchintegrity, developed ethical codes forscientific societies, promulgated guidelinesfor inquiries into allegations of scientificmisconduct, issued policies and recommen-dations for dealing with financial conflicts of interest in clinical research, and, mostrecently, collaborated with the Office ofResearch Integrity (ORI) in educationalefforts and workshops on the responsibleconduct of research. The AAMC has longrecognized that misconduct breaches thesocial contract underpinning academicscience and undermines a scientificestablishment that sets the standard ofinternational excellence. Contrary to yourassertion, the AAMC strongly supportsethics training for graduate students andpostdoctoral research fellows.

In his annual address to academic leadersand the public, the AAMC president, JordanCohen, stated “either we are trustworthy anddeserve the privilege of self-regulation, or weare suspect and warrant the close scrutiny ofgovernment.” We are proud of our record,which speaks for itself.

Your Opinion article (Nature 420, 253;2002) mischaracterizes the oppositionexpressed by the AAMC and the Federationof American Societies for ExperimentalBiology (FASEB) to the ORI’s proposeddata-collection instrument. Our objectionsfocus on two critical elements: first, thelikelihood that the survey would result inunusable, invalid and easily misinterpreted‘data’ based on subjective criteria and

imprecise measures. One example of this is mentioned in your editorial — asking for instances where a researcher observed a colleague “citing an article they had notread firsthand”. Our second objection is to the ORI’s inexplicable defiance of the settled federal definition of scientificmisconduct, which can lead only to confusion.

The crux of the issue, which you fail tocomprehend, is that the federal definitionof ‘scientific misconduct’ is a marker for the proper role of government inoverseeing the conduct of federally fundedacademic research. On this matter, the USscientific community has consistentlyspoken with a single voice in arguing thatthis role be circumscribed and focused on transgressions that are reasonablyunambiguous and are unacceptable acrossall scientific and scholarly disciplines.

This unanimity does not imply,however, that the boundaries of unethicalscientific and professional behaviourshould be so circumscribed — quite thecontrary. It is not the role of the federalgovernment to define or prescribe thoseethical boundaries. Rather, it is theobligation of academia and scientificsocieties, and it is to stimulate and assist ourmember institutions to meet that obligationthat the AAMC has been so actively anddemonstrably engaged for so long.

Perhaps Nature believes that scienceneeds a federally sanctioned ‘high church’ asthe final arbiter of scientific morals, ethicsand integrity. The AAMC, and, we believe,all of US science, profoundly disagrees. David KornSenior Vice President, AAMC, 2450 North Street,Northwest Washington, DC 20037, USA

Scientific misconduct: ORI survey is flawedSir — The Federation of AmericanSocieties for Experimental Biology(FASEB) abhors misconduct in researchand has repeatedly emphasized the needfor clear, unambiguous and consistentdefinitions of misconduct. The accusationsyou make in Opinion (Nature 420, 253;2002) misrepresent our criticism of the US Office of Research Integrity’s (ORI’s)flawed survey questionnaire.

We do not object to data collection onmisconduct. Institutions currently providethis information to the ORI on an annualbasis. Our opinion is that the proposedORI survey has serious deficiencies andwill not produce useful data.

The ORI itself stated that previousattempts to measure misconduct wereunsuccessful because they strayed from the federal misconduct definition. Theissues of fabrication, falsification andplagiarism are too important to beconfused with other questions, many ofwhich involve legitimate differences ofopinion. The survey’s vague questions,such as asking respondents how manytimes they have observed colleagues“failing to cite references that contradicttheir current research” or “refusing to give peers reasonable access to uniqueresearch materials”, will give a misleadingimpression of how research is done.Although it is easy to circle a number, there may be wide variation in the ethicalstatus of the examples being reported byindividuals. Simple summaries of complexissues will lump legitimate actions together

© 2002 Nature Publishing Group

with malevolent cases. Your editorial fails to inform readers

that the ORI is not merely proposing tomeasure “other” misbehaviour, but also“perceived” misbehaviour, which is key toour objection. The survey will not generatea measure of misconduct, but a recordingof hearsay or innuendo. For example, onequestion asks whether the respondentknows of colleagues “citing an article theyhad not read firsthand”. This orwellianapproach, which encourages scientists tospy on each other’s reading habits, will notlead to clarification of the ethical status ofbiomedical research.

You imply that there was a conspiracybehind the creation of the currentdefinition of research misconduct. This isincorrect. The record clearly shows thatthere was an extended debate and manyopportunities for public comment. We have supported efforts to improveeducation in research ethics, as I statedpublicly in remarks at the 10 OctoberInstitute of Medicine town meeting.FASEB’s August 2000 letter commenting on the draft PHS Policy for Instruction in the Responsible Conduct of Researchstates: “Students and trainees must haveinstruction in the responsible conduct of research. But the extension of thisrequirement to ‘all staff ’ including subcon-tractors and consultants will result in anenormous involvement of time andresources.” Our policy statements show ourconsistent commitment to the responsibleconduct of research. There is no basis forimplying that our position condones,supports or protects unethical behaviour.Steven L. TeitelbaumPresident, FASEB, Dept of Pathology andImmunology, Washington University School of Medicine, MS 90-31-649, 216 Kingshighway, St Louis, Missouri 63110, USA

Sounding the alarm onunderwater noiseSir — Your News feature on fisheriesmanagement (Nature 419, 662; 2002) raisesthe important point of considering eachfishery in the context of its larger ecologicalcontext, although you identify overfishingas the root cause of fisheries depletion. Thisperspective puts the onus of fisheries healthexclusively on the fishing industry.Agricultural runoff, the pollution anddisappearance of estuarine and wetlandnurseries and, increasingly, anthropogenic‘noise’ are all compromising ocean health— to such a degree that fisheries populationcrashes might occur even withoutcommercial overharvesting.

Only recently has anthropogenic noisebeen acknowledged as a threat to marine

ecology, as evidenced by whale anddolphin strandings caused by militarysonar (Nature 415, 106; 2002). Noisescaused by shipping, underwater seismicexploration, sonar, underwater telemetryand military exercises have all increaseddramatically over the past decade. Manyfish rely on acoustical perception to hunt,school, evade predators and find mates.Cluttering their acoustical niches withnoise affects their survival prospects.Michael StockerSeaflow.org, PO Box 559, Lagunitas, California 94938, USA

Why astronomy is thestar of the news showSir — Space science news is good news,according to the Pew Research Center forthe People and the Press, which tracks the“most closely followed” news stories in theUnited States (http://people-press.org/reports). From 1986 to 1999, the Pew studyfound 689 such items, 39 of them related toscience, medicine and the weather. To a striking degree, these 39 stories haddisturbing news to tell — earthquakes orother natural calamities, nuclear power,AIDS or controversies over cloning.Virtually every ‘good news’ science storywas about space science, for examplereports of the Hubble Space Telescope(1990) and Mars Pathfinder (1997)missions. The only other scientific subjectreported in a positive light was Viagra.

Perhaps this ‘good news’ feature ofastronomy and space science helps toexplain their broad popularity. A recentNSF survey reports that 74% of adults in1999 were interested in space exploration,and many people (57% in 1999) agreedthat it is worth its costs (see “PublicAttitudes Towards Space Science”, byH.A.S., in Space Science Reviews Vol. 102,3–4, Kluwer, 2002.)

Politicians mindful of public moralemight want to note this phenomenon, asmight budget-conscious managers whoseem to suspect that astronomy and spacescience do not give adequate value-for-dollar (and/or are badly managed). Thecurrent US administration, for example,and its Office of Management and Budget(OMB), in an effort to remedy theseperceived faults, are “measuring theperformance” of government-supportedresearch so that “spending on fundamentalresearch will be judged by formal‘performance criteria’” (see Nature 415,466–467; 2002). In the spirit of “improvingmanagement”, the OMB first proposed tomove the National Science Foundation’sastronomy funding to other agencies —with little clear reason, support or success

— and subsequently proposed theopposite: moving some astronomy (andother research) funding into thefoundation (see Nature 414, 680; 2002).

How does one calculate value-for-dollar in a discipline such as astronomy,whose primary product is knowledge? Assuccinctly (but unhelpfully) put by oneastronomer, Nicholas Copernicus, the aimof a scientist is “to seek the truth in allthings”. But without a sense of the worth ofresearch accomplishments, how can thoseformal performance criteria for measuringand managing be identified, and thenseriously and sensibly applied? Surely onelesson of the collapse of the telecommuni-cations giant Enron is that ‘measuring’ and‘managing’ are techniques that arethemselves subject to error, incompetence,inefficiency and worse.

This is where the good-news aspect ofastronomy comes in. Gerald Holton andGerhart Sonnert (Issues in Science andTechnology, 61–65; Fall 1999) haveproposed a model in which basic researchfalls into three categories whose respectivegoals are knowledge, applied knowledgeand (after Thomas Jefferson) knowledgewith the realization that somethingpractical might ensue. John F. Kennedyexpressed a fourth goal: the public spirit.As he put it when justifying the Apolloprogramme: “We choose to go to theMoon in this decade, and do the otherthings, not because they are easy, butbecause they are hard... No single spaceproject in this period will be moreimpressive to mankind.” In the words ofthe current NASA Administrator, SeanO’Keefe, NASA’s “mandate is to pioneerthe future... NASA’s work inspiresAmericans and unites people.”

Public spirit and public interest, thoughintangible, are reflected in the media. Newplanets, new insights into the creation of the Universe, and other cosmic discoveriesregularly merit upbeat front-page coveragebecause the pictures are inspirational, thestories are relatively easy to understand, theadventures are exciting, the discoveries areoften meaningful, and the successes makeus happy to be alive. Other science researchdisciplines (and their public outreachprogrammes) might also benefit fromperformance criteria that include a media-based appraisal of public attitudes. At leastfor astronomy the news is good —accountants, please add 10 points.Howard A. SmithHarvard–Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, Massachusetts 02138, USA

correspondencecontributions to correspondence may be submitted tocorres@nature.com. They should be no longer than500 words, and ideally shorter.

correspondence

740 NATURE | VOL 420 | 19/26 DECEMBER 2002 | www.nature.com/nature© 2002 Nature Publishing Group