The Primacy of Doubt - Tim Palmer *****

This is quite possibly the best popular science book I've ever read (and I've read many hundreds). To describe what Tim Palmer, a physicist turned meteorologist, does in simple terms does not do it justice. But essentially he explores the nature of (mathematically) chaotic systems and shows how we can deal better with uncertainty, even using his expertise to propose a different way to look at the lack of local reality in quantum physics.

This is interesting stuff anyway, but what is astounding is the way that Palmer rattles through a series of topics that are quite difficult to get your head around and, in several diverse cases, gives the most approachable explanation of the topic I've ever seen.

I'm not saying this book is an easy read, by the way. You do have to think about what you are reading, and I had to go back over a couple of sections to make sure it sunk in. But it is so rewarding of the effort.

In terms of this broad enlightening nature, the first of the three sections in the book stands out head and shoulders above the rest. Palmer starts by exploring chaos and gives the best explanation of the behaviour of chaotic systems, state space and attractors I've come across. Then he throws in Cantor sets, then shows the relationship of weather forecasts to all this, and introduces p-adic numbers (arguably the only bit that could have been better explained). He then shows graphically (literally, not metaphorically) how the introduction of noise can make models of chaotic systems work better. Finally in this section, he takes on quantum uncertainty, with one of the only explanations of the use of Bell's inequality I've ever seen that is at least vaguely comprehensible.

I don't usually go into that much detail in a review, but just wanted to show how much is crammed into the first 80 or so pages.

In the second section, Palmer addresses the use of Monte Carlo methods and ensembles in making at least partly successful predictions of chaotic systems, such as the weather, the climate and pandemics. Usually, the applications of the theory are the most interesting bits of a book, but somehow this isn't quite as engaging as the theory in the first section, though things really liven up when we get onto economics, and how economists are stuck in the fairly useless state meteorologists were before the great storm of 1987, when they used single-run forecasts, rather than ensembles. He also shows fairly bluntly that economists have failed in the development of the kind of models that can handle a chaotic system like the economy.

Finally, in the third section, Palmer addresses the big picture. He starts with an alternative interpretation of quantum theory that effectively enables hidden variables, using an approach that he describes as involving 'counterfactual indefiniteness', a concept he calls the 'cosmological invariant set' and invariant set theory. How much this will appeal probably depends how you feel about quantum interpretations, or get worried about the idea that until a quantum system interacts with the outside world it doesn't have real values for things like the location of particles. This part felt a bit hand-wavy, partly, I think because it needed too much of the mathematics behind it (which we sensibly don't see) to get a handle on it.

To end this more speculative section, Palmer takes on things like consciousness, free will and God - not bad going for a relatively short book. Finishing The Primacy of Doubt is like getting off one of those exciting roller coaster rides, when your immediate inclination is to think 'I want to do that again, but I'll have a bit of a break first.' I will be reading this book again, without doubt. Remarkable.

Hardback:

Using these links earns us commission at no cost to you
Review by Brian Clegg - See all of Brian's online articles or subscribe to a weekly digest for free here

Mathematics with Love – Mary Stopes-Roe *****

Admittedly it’s early days (this review is written in January), but this, for me, is the surprise hit of the year so far! I approached this book with trepidation, but found it absolutely delightful. It is described on the cover as the “courtship correspondence of Barnes Wallis, inventor of the bouncing bomb”, and contains a series of letters between Wallis and his cousin and eventual wife Molly Bloxham, along with some useful annotation by their daughter, Mary. The courtship itself is not without difficulties, as Wallis was 18 years older than the 17-year-old Molly at the start of the correspondence, and her father, not surprisingly, wasn’t too pleased about the interest of such an elderly suitor, but that isn’t the only reason the letters are interesting – it’s also because of maths, and Wallis’s position in the UK as the engineering hero of the Second World War. (Incidentally, it seemed very strange to see letters addressed to “Barnes” – I had always assumed Barnes Wallis was a ...

Why Nobody Understands Quantum Physics - Frank Verstraete and Céline Broeckaert **

It's with a heavy heart that I have to say that I could not get on with this book. The structure is all over the place, while the content veers from childish remarks to unexplained jargon. Frank Versraete is a highly regarded physicist and knows what he’s talking about - but unfortunately, physics professors are not always the best people to explain physics to a general audience and, possibly contributed to by this being a translation, I thought this book simply doesn’t work. A small issue is that there are few historical inaccuracies, but that’s often the case when scientists write history of science, and that’s not the main part of the book so I would have overlooked it. As an example, we are told that Newton's apple story originated with Voltaire. Yet Newton himself mentioned the apple story to William Stukeley in 1726. He may have made it up - but he certainly originated it, not Voltaire. We are also told that ‘Galileo discovered the counterintuitive law behind a swinging o...

On Time - Jim Al-Khalili *****

By a small margin Jim Al-Khalili's The World According to Physics is my favourite of his books, both because the 'three pillars' concept at its heart works so well - but also because it's short. On Time is somewhat heavier, but thankfully not a wrist-buster. It might seem at first sight odd that a book on a subset of physics is considerably longer than one covering the whole thing, but time is one of those subjects that inevitably involves considerable complexities and needs depth. As Al-Khalili establishes, it's hard to think of any other scientific concept where our perception of it - what he refers to as manifest time or psychological time - is so different from the view from physics. Some physicists go so far as to suggest that time doesn't exist at all. In the book, Al-Khalili sets out (to my mind quite successfully) both to explain why they think this is the case, but also why it's not the best interpretation of time. I'd certainly note that I...

zebedee23 July 2024 at 04:06
The pandemic chapter has a few errors. I can't remember ever seeing a SIR model in which the R included the dead. If exponential growth was rapid I would be very happy with the balance of my savings account. Exponential growth of epidemics is debunked by Farr's Law of 1840. In 1927 Kermack and McKendrick showed logistic growth.

Now to read the rest of the book.
zebedee27 July 2024 at 15:58
I've finally finished the book and think it is a very worthwhile read that has some significant omissions.

He discusses CovidSim which had its source code released after John Carmack (of Doom fame) had made its output for multi-threaded runs deterministic thus allowing regression tests. A regression tests fails for one of two reasons - either you introduce a bug or fix a bug. What it didn't have, and what the book doesn't mention about this or any other code, is validation tests. Such tests are used by commercial engineering codes to assure their customers that the implementation of the physical models do not have bugs. The author expresses doubt in some of the results of the models but never the model implementation.

Continuing with epidemiology the standard SIR model has been extended to be stochastic in a number of ways. Several of which are by making the reproduction rate normally distributed. This is the usual lazy assumption by modellers and is debunked by a derivation from first principles that shows it is logit-normally distributed. The author references various models in which he has added noise but never mentions which distribution was used and how that was determined which raises questions over the subsequent conclusions that he draws.

He seems to rate the Stern Review which I believe to have been a waste of tax payers money on the assumption that its software was seemingly never made available to be updated. The discussion of cost-loss models for tying economics to climate reminded me of Pascal's wager.

Search This Blog