Braun Tibor, Schubert András (szerk.): Szakértői bírálat (peer review) a tudományos kutatásban : Válogatott tanulmányok a téma szakirodalmából (A MTAK Informatikai És Tudományelemzési Sorozata 7., 1993)
MARTIN RUDERFER: The Fallacy of Peer Review: Judgement without Science and a Case History
176 RUDER I ER: T IIK I'A I.I ACY O l PEER REVIEW following working hypothesis, termed here the Innovation Theorem: The probable delay in acceptance of an innovation increases with its departure from the conventional norm. The history of innovation in science, technology and other areas, as politics and religion, generally support such an hypothesis. The familiar.widespread hesitancy to adopt radical ideas inexorably points to a deeply ingrained property of the average human mind. Resistance to innovation in science has been sporadically noted but was never systematically considered until 1961 by Barber 125 1. Applied to the review process it suggests that rejection error rate must increase for manuscripts which are rated on an increasing scale from evolutionary to revolutionary. The enormous success of the scientific method is primarily due to its ability to negate resistance to change by the test of experience for resolving the conflicts precipitated by dogma, tradition, preconceptions and the like. Only one conclusive experience has often been sufficient to overcome the ingrained resistance to a new approach, e.g. the 1919 eclipse observations led rapidly to the wide acceptance of Einstein's theory; Hertz's experiment quickly overcame the deep 25-year rejection of Maxwell's theory 126 3; and the discovery of the mid-Atlantic ridge soon settled the theoretical objection to Wegener's continental displacement mechanism despite the other supporting evidence he had collected 127 1. In other cases, preoccupation with theory has delayed consideration of well-founded experiment, e.g. Ohm 1 2 • 25 1. The referees similarly neglected the observational aspects that may have modified their preoccupation with theory: (i) The equalisation of clock rates was too significant to summarily ignore (pp. 396-7). (ii) The clarification of the one-way effects reported by Sadeh et al 12 8' —a diurnal variation with a superposed sunrise effect and a variation with clock separation — were based only on conventional theory, (iii) The extension to the solar frame for which a laboratory experiment was proposed (p. 401) was based on the verified Sagnac effect. This same preoccupation with theory manifested itself in the five SST reviews of the follow-up paper, but the weight of added evidence tempered recommendation of outright rejection. 6. CONFIRMATION OF THE REJECTED PAPER In the hindsight of the delayed acceptance of innovation, as for Maxwell, Wegener and Ohm, it is difficult to comprehend why their innovations were not at first provisionally accepted instead of being firmly rejected. The follow-up 119 1 of the rejected paper supports the working hypothesis that initial rejection of valid innovation relates to its departure from accepted ideas. The present theory of relatively rotating clocks stems from Einstein's 1905 introduction of special relativity. This is widely assumed to explain the Hafele-Keating experiment comparing two relatively rotating clocks. Atomic timekeeping and the long known Sagnac effect also involve rotating clocks. (Although the Sagnac effect has been expressly confirmed only with oppositely directed light rays in a rotating mirror system, Ives rigorously showed 129 1 that the substitution of rotating clocks for mirrors and light rays gives identical results.) The difference between the three techniques is trivial: differential rotation derives from fixed atomic clocks at different latitudes in atomic
