Against many risks, excess precaution is a mistake, doing more harm than good by preventing innovation — but not in the case of threats that can explode exponentially from small beginnings. Apart from Singapore, which built a special hospital after Sars, and South Korea, which geared up to test people and gather data on a huge scale, the world had done far too little to get ready for this possibility. Everything now is catch-up. And we are trying to catch the fastest runner of them all: an exponential curve.

The number of cases outside China has increased tenfold roughly every ten days. If it continues at this rate, in two months the virus will have infected 100 million people. A vast experiment is happening as different countries try different strategies, with some in Asia having notably more success than others in Europe. But to stop the virus gaining a permanent foothold in the human population will require every single one of those experiments to work.

In the long run, we will get through this. Effective drugs will be found. Methods for keeping the very sick from dying will improve. The milder forms of the virus will probably out-compete the harsher versions. A vaccine may yet work — though it has come as quite a shock to find out just how little vaccine development has improved in recent decades. Genomics allows us to read viral genomes in a flash and make messenger-RNA vaccines, in which the body manufactures the viral proteins, doing away with the need to put killed or attenuated viruses into the body. None the less, a scientific paper that I failed to read last year warned: ‘The current state of vaccine development is an expensive, slow and laborious process, costing billions of dollars, taking decades, with less than a 10 per cent rate of success.’