ESSAY: PROFESSIONAL ISSUES

554 Fisheries • VOL 31 NO 11 • NOVEMBER 2006 • WWW.FISHERIES.ORG The scientific community gave a collective sigh of relief just before Christmas 2005 when Judge John E. Jones III ruled that intelligent design is not a scientific theory and cannot be taught alongside evolution as an alternative scientific hypothesis. There is no better way to unite any group of fractious scientists than to bring up creationism and intelligent design as alternative scientific hypotheses—scientists know that these faith-based ideas are not scientific and have no place in a scientific course. The court’s ruling is a triumph for the scientific method of hypotheses being confronted by data and a setback for those with a political agenda masquerading as science. However, before we congratulate ourselves too much for the triumph of the scientific method over belief, I suggest the fisheries community needs to look at itself and question whether there is not a within our own field a strong movement of faithbased acceptance of ideas, and a search for data that support these ideas, rather than critical and skeptical analysis of the evidence. This faith-based fisheries movement has emerged in the last decade, and it threatens the very heart of the scientific process—peer review and publication in the top journals. Two journals with the highest profile, Science and Nature, clearly publish articles on fisheries not for their scientific merit, but for their publicity value. Beginning in at least 1993 with an article I co-authored (Ludwig et al. 1993), Science and Nature have published a long string of papers on the decline and collapse of fisheries that have attracted considerable public attention, and occasionally gaining coverage in the New York Times and the Washington Post. I assert that the peer review process has now totally failed and many of these papers are being published only because the editors and selected reviewers believe in the message, or because of their potential newsworthiness. As examples, let me choose papers by well-established professionals who have long records of significant work beyond the papers discussed below and I emphasize the problem is with the peer review and editorial system, not the authors of the papers. Casey and Myers’ paper on barndoor skate (Casey and Myers 1998) argued that these skates were headed towards extinction. Analysis by others more familiar with the data showed that the survey data came from areas that are not the core of the range of the species (Kulka et al. 2002), and subsequent evaluation of the status of barndoor skate in New England by the National Marine Fisheries Service concluded they were not overfished (Boelke et al. 2005), hardly headed towards extinction. A review by the Canadian Department of Fisheries and Oceans removed barndoor skates from a list of species that are threatened or endangered. The original paper has not been withdrawn and continues to frequently cited as an example of near extinction of marine species. Myers and Worm published a paper in Nature (Myers and Worm 2003) which made the front page of major national newspapers, purporting to show that large pelagic fish stocks around the world had declined rapidly and by the 1980s were at less than 10% of their historic abundance. Widely cited in the scientific and popular literature, this paper raised a furor among many scientists specializing in pelagic fisheries who knew the same data, knew it was being misinterpreted, and knew there was a large body of other data that contradicted Myers and Worm’s results. At least three independent critiques of the paper subsequently have been published: Walters (2003), Hampton et al. (2005), and Polacheck (2006). The critics are not the “old guard defending their turf,” because it is not as if no one had noticed that the catch-per-unit effort data Myers and Worm used had declined. Rather these critics have themselves long been arguing that some of these fisheries are now depleted and overfished. What they criticized was Myers and Worm’s analysis, their highly selective use of data, and specific conclusions about the extent and timing of depletion of these stocks, not their concern about overexploitation. Conover and Munch (2002) published a highly cited paper in Science showing experimental evidence that size-selective fishing could induce growth changes in fish stocks and suggested this was a mechanism that could lead to collapse of fish stocks. The article never looked at actual fisheries data to ask if the laboratory selection regime imposed resembled what happens in fisheries, nor did they look at the vast body of fisheries data which shows that fish more commonly grow faster, not slower, when fishing pressure is high. A paper in Science (Roberts et al. 2001) purported to show an example of how a marine protected area (MPA) increased yields outside the protected area, when in fact the abundance of fish outside the protected area increased within one year of the establishment of the MPA. Any competent peer reviewer would have seen the flaw in this logic—the theory of MPA impacts on adjoining areas requires at least a generation for abundance to build inside reserves and recruitment to spill out (Hilborn 2002). The displacement of fishing effort from inside to outside the reserve should initially cause abundance outside to decrease, so the increasing abundance outside the reserve after MPA establishment must have been due to an uncontrolled effect. These four examples illustrate a failure of the peer review system and lack of the basic skepticism needed in science, and are unfortunately but a few of the many papers now appearing with similar sensational but unsubstantiated headlines. The people who knew the data used in the Casey and Myers paper and the Myers and Worm paper clearly were not involved in the review process, or the editors chose to ignore their opinions. The complete absence of skepticism by the reviewers of Ray Hilborn

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.