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
189 RUDER I ER: T IIK I' A I.I ACY O l PEER REVIEW applies only to the synchronisation of any two remote time scales irrespective of their absolute calibration. I make this clear in the top paragraph on page 22 [p.402], I trust that review [C] will not influence any further review by being given weight as a prior rejection. The comments of referee [D] are more relevant but, unfortunately, reveal a misunderstanding of portions of the MS. The following comments correspond to his headings. (a) My paper does not "derive a number of well-known effects", but deals with effects that have not been systematically exploited heretofore. The only truly "one-way" effects that are well-known are the conventional Doppler shift and aberration relations. The effects of n, dn/dt and the double-disc Sagnac experiment which I discuss have not appeared elsewhere to my knowledge — and I have been specifically searching for such one-way effects in the literature for over 15 years. Perhaps they have been mentioned in some remote place; if so, they certainly are not "well known". The only well-known refractive effect in the literature related to time dilation is the refinement by Lorentz, confirmed by Zeeman, of the Fizeau and related experiments, e.g. D.A. Evans, Int. J. Theor. Phys., 2, 313 (1969). This is a two-way effect. The one-way refractive effects I discuss disappear in a two-way measurement. The referee appears to be unfamiliar with phase modulation theory and its accompanying concept of instantaneous frequency, which is the physical origin of the frequency shift Af that I derive. This accounts for his confusion regarding the frequency and travel time variables. Note that I verify the form of the Doppler effect I derive by showing its consistency with the conventional form by the examples I give for recessional and transverse motion. What the referee's remarks do indicate, however, is that the derivation is too succinct. Accordingly, I have revised page 7 and have replaced them with pages 7 (rev.) and 7a (rev.) [pp. 390-1]. (b) I neglect the effect of Sun's gravitational potential (as well as those of other solar bodies) because they are negligible here. I have therefore added a paragraph (p.19a) [p. 400) to dispose of this objection explicitly. (c) I have followed Cannon and Jensen's use of clock "reproducibility" which is appropriate in my application. The referee's concern with absolute calibration of proper time to SI has been extensively discussed in your 6 February issue (191, 489491). Although this may be pertinent to Cannon and Jensen's theory, it is irrelevant to the testing of the one-way synchronisation effects I discuss. I spent considerable space explaining this in the Discussion section. Why has [D] overlooked this? I suspect he did not read this far. To illustrate that the "inherent accuracy" of clocks is irrelevant, consider the synchronisation of any watch to any wall clock. To do this, it is only necessary to periodically reset the time on the watch to that of the wall clock. The time difference between the two is then only dependent on the resetting precision and is not directly related to the absolute precision of the two clocks. This is precisely what is done by the time laboratories in synchronising the UTCi to UTC, as has been noted in ref. (6), in the adjacent article by Allan, et al, and in my MS. Only the difference between these is required to test the predicted one-way Doppler effects , as I discussed. Paradoxically, [D ] also notes that "the UTC; time scales are periodically adjusted to coordinate them with
