In a post two weeks ago I raised the issue of “settled science”. The concept was there had to be things you were not persistently checking. Obviously, you do not want everyone wasting time rechecking the melting point of benzoic acid, but fundamental theory is different. Who settles that, and how? What is sufficient to say, “We know it must be that!” In my opinion, admittedly biased, there really is something rotten in the state of science. Once upon a time, namely in the 19th century, there were many mistakes made by scientists, but they were sorted out by vigorous debate. The Solvay conferences continued that tradition in the 1920s for quantum mechanics, but something went wrong in the second half of the twentieth century. A prime example occurred in 1952, when David Bohm decided the mysticism inherent in the Copenhagen Interpretation of quantum mechanics required a causal interpretation, and he published a paper in the Physical Review. He expected a storm of controversy and he received – silence. What had happened was that J. Robert Oppenheimer, previously Bohm’s mentor, had called together a group of leading physicists to find an error in the paper. When they failed, Oppenheimer told the group, “If we cannot disprove Bohm we can all agree to ignore him”. Some physicists are quite happy to say Bohm is wrong; they don’t actually know what Bohm said, but they know he is wrong. (https://www.infinitepotential.com/david-bohm-silenced-by-his-colleagues/ ) If that were one isolated example, that would be wrong, but not exactly a crisis. Unfortunately, it is not an isolated case. We cannot know how bad the problem is because we cannot sample it properly.
A complicating issue is how science works. There are millions of scientific papers produced every year. Thanks to time constraints, very few are read by several people. The answer to that would be to publish in-depth reviews, but nobody appears to publish logic analysis reviews. I believe science can be “settled” by quite unreasonable means. As an example, my career went “off the standard rails” with my PhD thesis.
My supervisor’s projects would not work, so I selected my own. There was a serious debate at the time whether strained systems could delocalize their electrons into adjacent unsaturation in the same way double bonds did. My results showed they did not, but it became obvious that cyclopropane stabilized adjacent positive charge. Since olefins do this by delocalizing electrons, it was decided that cyclopropane did that too. When the time came for my supervisor to publish, he refused to publish the most compelling results, despite suggesting this sequence of experiments was his only contribution, because the debate was settling down to the other side. An important part of logic must now be considered. Suppose we can say, if theory A is correct, then we shall see Q. If we see Q we can say that the result is consistent with A, but we cannot say that theory B would not predict Q also. So the question is, is there an alternative?
The answer is yes. The strain arises from orbitals containing electrons being bent inwards towards the centre of the system, hence coming closer to each other. Electrons repel each other. But it also becomes obvious that if you put positive charge adjacent to the ring, that charge will attract the electrons and override the repulsion on and attract electrons moving towards the positive charge. That lowers the energy, and hence stabilizes the system. I actually used an alternative way of looking at it: If you move charge, by bending the orbitals, you should generate a polarization field, and that stabilizes the positive charge. So why look at it like that? Because if the cause of a changing electric field is behind a wall, say, you cannot tell the difference between charge moving or of charge being added. Since the field contains the energy the two approaches give the same strain energy but by considering an added pseudocharge it was easy to put numbers on effects.
However, the other side won, by “proving” delocalization through molecular orbital theory, which, as an aside, assumes it is delocalized. Aristotle had harsh words for people who prove what they assume after a tortuous path. As another aside, the same quantum methodology proved the stability of “polywater” – where your water could turn into a toffee-like consistency. A review came out, and confirmed the “other side” by showing numerous examples of where the strained ring stabilized positive charge. It also it ignored everything that contradicted it.
Much later I wrote a review that showed this first one had ignored up to sixty different types of experimental result that contradicted the conclusion. That was never published by a journal – the three reasons for rejection, in order, were: not enough pictures and too many mathematics; this is settled; and some other journals that said “We do not publish logic analyses”.
I most certainly do not want this post to simply turn into a general whinge, so I shall stop there other than to say I could make five other similar cases that I know about. If that happens to or comes to the attention of one person, how general is it? Perhaps a final comment might be of interest. As those who have followed my earlier posts may know, I concluded that the argument that the present Nobel prize winners in physics found violations of Bell’s Inequality is incorrect in logic. (Their procedure violates Noether’s theorem.) When the prize was announced, I sent a polite communication to the Swedish Academy of Science, stating one reason why the logic was wrong, and asking them, if I had missed something, to inform me where I was wrong. So far, over five weeks later, no response. It appears no response might be the new standard way of dealing with those who question the standard approach.