The Rt Hon Sir Oliver Letwin has written one of the most important books of the year. But just because it’s important doesn’t mean it’s enjoyable. It’s important in the same way that an injection or life insurance is important.

Apocalypse How? is set in a gleaming, hyperconnected, smart-car-driving 2037. An extreme space weather event knocks out the national grid and our internet-dependent society, which was running so smoothly five minutes ago, is paralysed for days. Hospitals overflow, the BBC goes off air, smart cars go dumb, the elderly freeze and Google Maps stops working (help!). There’s not even any social media for people to turn to in order to complain about it all.

In presenting this scenario, Letwin is making a simple argument: as we become more interconnected in all aspects of our economy and society, we also become less resilient. In case of a black swan event – a mega-hack, a change in space weather, maybe even the emergence of some deadly new virus – we’ll be in serious trouble. You probably knew this anyway. Don’t tell me dark thoughts haven’t entered your head recently concerning coronavirus, baseball bats, tins of beans and so on. Indeed, the timing of Apocalypse How? could hardly be better, or worse.

Letwin isn’t negative about the networked society, since he accepts that it’s both inevitable and generally beneficial. The problem is with governments, which don’t prepare properly for high-risk, low-probability events. There are understandable reasons for this, but Letwin says that things need to change. If, for some reason, you enjoy the political-memoir-describing-what-needs-to-be-done-but-none-of-which-happened-when-the-author-was-in-power subgenre, then you are in for a treat. For four years Letwin ran the Cabinet Office.

What’s unusual about this book is the style: chapters alternate between descriptions of the imagined catastrophe unfolding in 2037 and straight non-fiction analysis. I tried a similar combination once and my editor told me to stop it immediately. Mixing fiction and non-fiction is like spreading jam and marmalade together on toast. It takes a Salman Rushdie to make a pioneering format work, and Salman Rushdie Oliver Letwin ain’t.

The non-fiction chapters are decent and the argument is compelling, as you’d expect. Letwin is known in political circles as an unusually thoughtful minister and he provides useful insider insights into how governments think. You’d assume that someone in government is dealing with the problems he brings up. Wrong, says Letwin. No one is. That admission, from a former high-up, is worth the entry fee alone. But his fictionalised world is far harder to like – not because it’s wild or speculative or unrealistic (fifteen to twenty years is the typical range for a futurist – close enough to be relevant to the present day, far away enough to permit enjoyable flights of fancy) but because it’s so unerringly, tediously believable. Much of the action set in 2037 involves COBRA meetings, with what’s left of the government trying to establish what’s going on. These potentially dramatic moments mostly consist of procedure, stuffy formality, what’s-the-MoD’s-perspective questions and laboriously hedged language. It will be exactly like this of course, but that doesn’t mean I want to read about it. An example:

‘Prime Minister, would you like me to continue with a brief assessment of the likely consequences at this point?’
‘Yes please, Simon. We cannot be sure that your analysis will prove correct; but at least it fits the facts that we know so far. If that is correct, what will happen next?’

Then comes this, a few pages later:

‘Jan, I wonder whether CCS could at this stage give us some indication of how the recovery of the grid will be managed, and how long it will take?’

On and on it goes. Do you remember this line in the film Terminator 2? ‘The Skynet Funding Bill is passed. The system goes on-line August 4th, 1997. Human decisions are removed from strategic defense.’ Wallop! If you’re going to do near-future fiction, then make it fast-paced and snappy, man! That’s the joy of it. Ever notice how, in films, decisions actually get made in meetings and people end phone calls without saying ‘bye, bye, byebyebye’? It’s not about accuracy, but pace and narrative momentum. (That said, in Letwin’s 2037, health care and social care are still not integrated, and I appreciated his attention to detail here.)

But fictionalised futures do have their uses. Dotted throughout his tale of 2037 are references to ‘historic’ decisions made in the 2020s – in other words, around now – that led us to the meltdown. These references are about as subtle as one of his theoretical meteor strikes, but they relate to decisions that will be taken fairly soon, about, for example, moving to a cashless economy (in 2027, apparently) and shifting all communications to 5G with no alternative (in the early 2020s). Politicians considerably less thoughtful than Letwin, some of them currently senior government ministers, will be making these decisions. His vital message is that in a hyperconnected world they need to spend more on fall-back options. We must all hope they make it through enough of the book to hear it.