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
175 RUDERI ER: T IIK I'A I.I ACY O l PEER REVIEW systematic differences between the six clock rates must then eventually assert themselves by difference in clock readings and/or the rate corrections they entail. Absolute clock accuracy is not directly involved. Report M. This referee appears to be the same as referee D. His refusal to discuss the issues despite the reply (E) to his initial objections and the revisions initiated by them, or to provide any attempt at further falsification, is an example of arbitrary rejection. Report N. This rejection has two main errors: (i) The offered explanation is nowhere based on a one-way anisotropy in the speed of light, as claimed by the referee. Isotropy is assumed to be constant throughout, as discussed (p. 389). (ii) His formal theory is incorrect and hence inappropriate, as detailed in P, due to his confusion of one-way travel time T as a general time coordinate. This is affirmed by his statement "T = t is the coordinate time in the frame of the stationary clock." The meaning of t is made clear at equation (1) (p. 388) early in the paper, contrary to his statement further below, "t is not made clear until p. 12" (p. 394), at which point he could have ascertained his error. Also see P. The last comment is a misconstrued version of objection (b) of D which was evaluated and inserted in the manuscript, as discussed in P. 5. GENERAL COMMENTS The immediate consideration, which precedes any final judgement, is the rejections for the wrong reasons. The four reviews of the three referees plus the one satisfactory response yield a minimum error rate of 3/4 for the review history. This is intolerably high for a presumed scientific process involving technical matters which are inherently resolvable. Such a large deviation from the above-chance accuracy of the average review process in physics' 7 8' indicates that there is something radically wrong with the way manuscripts are now judged. We can of course arbitrarily assign errors in any isolated instance to indifference, laxity, chance, prejudice, politics, author status or even malpractice' 145 ), but this would involve only errors of similar arbitrariness. The availability of case histories presents the opportunity to seek the root causes of erroneous rejection. In this case the most prominent element common to the rejections is the pre-occupation with theory. This is undoubtedly due to the unconventional form of the Doppler effect applied since a travel time formalism is not the common text-book explanation. Nevertheless, the travel time form applied, equation (13) (p. 391), is justified for constant radial and transverse velocities and is more rigorously affirmed to the required precision in the follow-up paper (p.417). The necessity for an unconventional formalism is obvious in retrospect: (i) The use of a one-way coordination signal automatically prescribes one-way propagation theory. However, this little known area has still not entered the main stream of physics, (ii) The constancy required by conventional theory in the rates of (uncoordinated) worldwide atomic clocks in the geocentric frame independent of Earth's rotation (p. 385) obscures the possibility that other effects of clock rotation may exist. These combined with the supposed published resolution of the Cannon-Jensen finding largely account for the high error rate in this case history vis-a-vis the average. This preoccupation with conformance to conventional theory thereby suggests the
