sharp@kpnoa.UUCP (05/15/84)
<> I have better things to do than discuss fringe science, but I've been feeling depressed, and there's nothing better than a good argument ! That I am posting this now should not be taken to imply a commitment to continue discussing such claims. This is in response to cosivax!dzd, whose lines are quoted below. Because he posted the same thing three times, I am limiting my quotations. >I would like to cite three references on several sides of these >issues: > > 1) A. Eddington, "Universal Theory" <why fool around?> > > 2) Kantor, "Information Mechanics" > > 3) ?, "Cosmology, Physics and Philosophy" 1) I think you mean "Fundamental Theory", Cambridge University Press, 1953, whose publication was supervised by E.T.Whittaker. For a fuller discussion of Eddington's ideas, and more of his actual notes and ideas, see "The Development & Meaning of Eddington's `Fundamental Theory'", by Noel B. Slater, published by Cambridge University Press in 1957. Eddington clearly intended that his work be considered as an epistemology - i.e. a theory of knowledge - rather than a straight scientific treatise. In it, he uses various broad principles, particularly concepts from quantum mechanics, Newtonian mechanics, and relativity, to derive numerical values for the fundamental physical constants. Early on, he amended his calculations so that they would agree with the observations, usually by complicating his arguments until the (mainly) dimensional derivations fitted. That is, there were more than enough "spare" adjustable arguments to get anything right. By contrast, if the bending of light around the sun had not agreed with Einstein's General Relativity, the theory would have been doomed - IT has no way out of disagreements. Eddington claimed epistemology: if we feel it necessary, as physicists usually do, to insist on accurate NUMERICAL agreement as well, then we are denying that claim. It is clear that Eddington's conjectures have sparked a lot of argument and calculation, and have done much good for physics. However, the very great arbitrariness evident in all the calculations have led most workers in what is, after all, fundamentally a QUANTITATIVE discipline, concerned with getting the numbers right, to dismiss Eddington's personal approach as sterile. > >Notice that under all variation of big bang theories, current >universe is of finite size. This point seems to me to be >under-appreciated by the "standard" cosmologists nowadays. I don't understand how it is that we "under-appreciate" this point. I would personally appreciate clarification of this attack. 3) I am not familiar with any work of this title. I also do not know where the remarks made by dzd about metrology & finite universes fit into Eddington. >Finally, there is reference 2) which came out about four years >ago and has been completely ignored as far as I can find out. It >is either completely crackpot or the **ANSWER**. It proceeds from ..... Got it in one. A simple search has turned up no information about Frederick W. Kantor, and his book (published in 1977 by Wiley Interscience) contains no biography. The work is based on a peculiarly personal definition of "information". Since a system is in only one state out of all possible states it could occupy, the "information" of that state is defined to be log2 of the number of all possible alternatives (Definition 1, page 37). Let's start at the beginning: Theorem 1 states that "information" is conserved in an isolated non-relativistic quantum mechanical (NRQM) system. The argument is based on NRQM, and is essentially the argument that the number of possible QM states depends on the energy, which cannot change because the system is isolated. Fair enough. Kantor then states that he considers this conservation of information to be more fundamental than the NRQM from which he derived it. But, for example, if I change one of the distinguishable QM states into another, different, but still distinguishable, state, then the information has not changed (total no. of states, remember ?), but the QM system certainly has. In this sense, the QM theory is MORE restrictive than the conservation principle, and therefore more likely to be fundamental. Theorem 2: all information in an electromagnetic system is representable by photon state occupation numbers without phase. Now, normally in QM and EM, the phase of photons is very relevant, so this is an interesting statement - the most fundamental property, Kantor's information, is independent of phase. The argument goes: Take a detector, which assigns each photon to its state (this is unambiguous, since no position measurement) But, energy & time obey an uncertainty relation, so to do this requires infinite time. But, infinite time => infinite phase uncertainty. Removing all energy by this measurement removes all the information from the system. Therefore, all information is representable without phase. Can you spot the fallacy ? If I remove properties from a system without recording some of them, and I have no properties left, then I have recorded all of the properties. I have phrased it this way to emphasise the distinction between "record" and "remove". At this point, (only page 41, of which 34 are a "summary introduction") things get more technical. By postulating a box containing two photons, Kantor derives E=mc^2. This box has internally contradictory properties, and the "proof" is no more than using de Broglie's mass/wavelength relation and the Doppler shift. Standard stuff. Is this theory worthwhile, anyway ? In other words, does it tell us something we didn't know before ? Well, it certainly claims to: section five is a list of interesting predictions and experiments. A great many of them have to do with esoteric particle properties, which I have neither the time nor the inclination to investigate. Two others are particularly interesting: predicted values for Hubble's constant and for the density of the Universe. Translating Kantor's units, H is about 29 km/s/Mpc - current "best guesses" are 50-100, with no observers who study this topic even approaching this low. The density is given as 3.38 x 10^-27 kg/m^3, and this is a little tricky to compare with observations, which depend on H as well. I estimate that it corresponds to twice the closure density for the predicted H. Current estimates are generally independent of H, because of their derivation, and reach maybe .5 to .7 of closure. It seems that the theory breaks down early on, and disagrees with observation anyway. Perhaps it's been ignored for 7 years simply because it is of no use ? :-) Sharp's theorem: When an eminent and respected scientist goes crazy, he is at least plausible. When an unknown fringe scientist goes crazy, he's very probably crazy. >BTW: If you think this question is dumb, remember that modern >scientific cosmology began when somebody [Hubble?] asked: > "Why is the sky dark at night?" Interesting, if a trifle out. It is perhaps possible to argue that the first time someone tried to ANSWER this was the beginning of cosmology. It seems to have been Kepler, around 1610, although it is often known as Olbers' paradox, after Heinrich Olbers, who in 1823 presented a completely erroneous answer. See the excellent chapter 12 of E.R.Harrison's book "Cosmology" (Cambridge University Press, 1981) OK, that should be enough for now ! I should emphasise that quite a few "standard scientists" like myself are quite willing to entertain unusual ideas. They need to be plausible, have numerical, testable predictions, and be clearly separate from philosophy in order to be considered, however. (Note that philosophy is something I enjoy, but physics is about observables NOT intangibles.) -- Nigel Sharp National Optical Astronomy Observatories Tucson, Arizona (602) 325-9273 UUCP: {akgua,allegra,arizona,decvax,hao,ihnp4,lbl-csam,seismo}!noao!sharp ARPA: noao!sharp@lbl-csam.arpa