Science book reviews & recommendations

Speakable and Unspeakable In Quantum Mechanics By John S. Bell 2nd Edition ( Cambridge University Press ) with Introduction by Alain Aspect. I have a digital copy via eBooks.com at $56 AUD.

This book is a masterpiece but also not for the feint-hearted. Alas you need to know a good slab of QM at tertiary level to follow some of the content. One can still skim over the denser mathematical bits and still have a good read though. Most of the difficulty for me at least is that in close detail QM is really screwy, and the more you look the screwier it becomes. If nothing else this book defines the extra screwiness more precisely, but built upon the already non-intuitive baseline that 'ordinary' QM is. Here 'screwy' is with reference to the expectations we learn simply by being of the scale we exist at. Which historically is obviously going to be what we have immediately understood and before we delved into other scales, both smaller and bigger.

A brief answer is that what we see and experience at human, everyday scale ( call that classical physics ) is but an emergent scheme of behaviour of a vast number of far smaller objects which do not resemble as conglomerates their machinations in individual instances. Far from it. The word 'quantum' has several related meanings in context :

- the first simply means small. Exactly how small one can debate, indeed so called 'mesoscale' systems have pushed the scale up a couple of orders in recent times.

- the second meaning is discontinuous which refer generally to some measurable properties like energy levels, spin directions and particulate entities. Thus one might have some integral multiple of photons but never 3.5 photons.

- there is a third less evident meaning and that is the statistical nature of predictions and measurements. Any system, even of a single particle, is deemed as a ( specially mathematically defined ) combination of base states. Any one of these base states we are not actually able to prepare in the real world to infinitesimally exact degree. But we use the concept of them and posit that a real system instance is a blend of these unknowable states. When we measure we select one of those states from it's spectrum for a single detection instance. There is some tedious regulation in the rules of applying this logic, but within lies a nasty rub which is explained in detail in the book.

- hand in hand with the above is a queer requirement : measuring one type of quantity may affect how we measure another type of quantity ! So when measuring we select from a distribution of posibilities along one spectrum, but inevitably alter the range of possibilities along some other spectrum. A priori one wouldn't expect these paired conjugate quantities to be related in this manner. Why would the measurement of position affect what we meaure for momentum ? Seriously what has time got to do with energy ? Etc. There are theoretical narratives which 'explain' this nature ie. the Heisenberg uncertainty stuff and it is coherent and useful to do so. But like an internet troll the cognitive discomfort never goes away willingly. Again a clash with classical 'natural' ideas.

Up to the early 1900's it became apparent that - confirmed by measurement and example - matter and energy did have divisions ( eg. atoms, electrons, photons ) but the expectation was these tiny components would have classical mechanical rules ie. Newton's Laws et al. In series of breakthrough experiments this was fairly rapidly buried, followed by a period of to & fro muddling, until the mid to late 1920's whence emerged a consistent and dramatically accurate theory. A key question that was answered, which classical physics failed at, was explaining why matter was stable at all. There were special relativistic additions to that description in the early 1930's, Dirac was pivotal here, and so more or less the current QM scheme was formed. Sure we have added other force/field types, found a wondrous plethora of really small particles, then Feynman et al weighed in with a very useful formalism, but most QM as applied currently does not really address ( or need to address ) what is presented in this book. That is one can go a long way with, for instance, technology involving crystal structures without straying much from the prescription formed prior to say, WWII. None of the new electronics past about 1960 could have been possible if our base understanding via QM was wrong.

Hence what this book deals with is the QM aspects not readily apparent in most scenarios of study. One has to set up the experimental circumstances quite carefully to demonstrate these features. These features are shown in special corners of reality that bring out or challenge the notion that particles are self contained things ie. they have their own 'personal numbers' in and of themselves. This brings on the topic of 'hidden variables' and 'entanglement' which was the bulk of Mr Bell's focus in this topic. He did do other things in his career BTW. He spent much of his time at CERN. He died too soon, as they say, of a massive stroke.

Alain Aspect was John Bell's intellectual twin here. Some key elements of what John proposed were followed up and validated by Mr Aspect's groundbreaking experiments. The short answer is that Einstein was right when he implied ( with Boris Podolsky & Nathan Rosen ) that there is a problem with what we have usually understood as the normal cause & effect structure of spacetime. Making a selection of measurement mode at one interaction vertex can affect at faster than light speed the availability of measurement outcomes at another. One can also validly claim that it is the measurement devices which are entangled too. But the ultimate practical joke here is that this may only be demonstrated statistically. A single measurement outcome does not and cannot prove this. One can only say that on average superluminal effects must be transmitted for some bulk results to make sense. So there is no opportunity for a faster-than-light signalling system here, although some have proposed that either interception and/or error detection schemes might be devised. There is even more tedious logic regulation here again but is especially necessary in order to not conflict with actual findings.

I won't outline further detail on the topic here. It is terribly hard to summarise without losing important essences alas. In fact I have already done that here. To correctly go into further detail would be equivalent to reading the whole book anyway ! However the included preface from the first edition has helpfully indicated what papers may be easiest and thus perhaps tackled first.

{ BTW Mr Bertlmann is a real person. He was a colleague of Bell at CERN. Bell used his eclectic sock choice behaviour as a very useful analogy for the measurement strategies devised to show the QM weirdness. }

Cheers, Mike.