The Universe in a Box - Andrew Pontzen ****

Our attention was drawn to this book’s existence by a helpful nudge on Twitter following the review of Romeel Davé’s Simulating the Cosmos a few weeks ago. I gave that book four stars, as regards its appeal to a general audience, while saying that my own personal rating was a full five stars. That’s because computer simulation is, in a sense, my own specialist subject, and I thought Davé did a great job of explaining how cosmological simulations work, and what they can tell us about the universe. At the same time, I acknowledged that not everyone is going to be enough of a space or computer geek to care about such questions in the first place.

Pontzen’s book deals with essentially the same subject, but approaches it from such a different angle that there’s very little overlap between the two. I’m giving this one four stars, too, but for pretty much the opposite reason. Personally, I found it frustratingly tangential to the subject it was ostensibly about, never really getting to grips with the nitty-gritty of simulations and constantly drifting onto bigger, more philosophical, questions. But I can see that Pontzen’s approach may well have more appeal for general readers. In any case, the book’s back cover is splattered with praise from the likes of Philip Pullman, Hannah Fry and Jim Al-Khalili, so I’m sure it can withstand a tad less than total enthusiasm on my part.

The fact is, reading a book is a subjective experience. Coming to it so soon after the one by Davé, who tackled the field of cosmological simulation so directly and incisively, Pontzen’s offering can seem rambling and waffly, never quite hitting the target. Yet the various sidetracks he takes, into surrounding areas like the history of computing, weather forecasting and artificial intelligence, might have seemed like welcome context-setting if I’d been less familiar with the subject to start with.

And although I tend to use the word ‘philosophy’ as a pejorative, some of his remarks in this area are genuinely intriguing, such as when he says ‘a simulation doesn’t have to be literally true for it to upend our ideas about the universe’. That sounds like nonsense, but it’s completely true in the specific instance he’s talking about. This was back in the 1960s, when Beatrice Tinsley found it was impossible to create a galaxy simulation that produced a constant amount of light over billions of years. Her simulated galaxies looked nothing like real ones, but that wasn’t the point – she’d shown that the prevailing wisdom, that galaxies don’t change on cosmological timescales, had to be wrong.

There’s another reason I prefer Davé’s book over Pontzen’s, and that’s simply one of style. This is another area of subjectivity, of course, but I felt Davé’s approach – with plenty of diagrams and a chattily direct writing style – was better suited to a technical subject like this one. In contrast, Pontzen’s book has no illustrations at all, and a distinctly ‘literary’ writing style (sorry, but ‘literary’ is another pejorative word, at least when it’s used by me). Maybe I’m a bit backward in this respect, but it did mean I had to read some of his sentences several times before I got them. Here’s one that’s 46 words long, and actually makes a good point when you battle your way through it: ‘Disentangling what is a prediction from what is an assumption, what can be trusted from what cannot, takes an expertise of its own and can often be controversial; there are still a few experts out there who question whether simulations can tell us anything at all.’

When I reviewed Simulating the Cosmos, I made the point that astronomical computer simulation is something that’s hardly ever touched on in popular science writing. So the fact that we’ve suddenly got two brand new books about it can only be a good thing. Which of them you go for is purely a matter of taste. Anyone who’s already deeply into computers and astronomy, and maybe even thinking of taking up this sort of thing as a career, is probably going to prefer Davé’s book. But everyone else – the great majority, in other words – should probably go for Pontzen’s.

Hardback:   

Kindle:  

Using these links earns us commission at no cost to you

Review by Andrew May - See more reviews or subscribe to a weekly email free here