Here we go again, fighting the last war. Because governments are perceived to have moved too slowly to ban flights when the delta variant arose in India, we jumped into action this time, punishing the poor South Africans for their molecular vigilance. But nothing was going to stop the delta going global, and the latest set of government measures to stop the spread of the new omicron variant are about as likely to succeed as the Maginot line was to stop General Guderian’s tanks. The cat is already out of the bag. Just because we can take action does not make it the right thing to do.

This pandemic has mocked public-health experts. They told us to wash our hands and then realised it was spreading through the air. They told us masks were useless and then made them mandatory. They sent covid cases to ordinary hospitals where they infected patients.

Banning flights might have stopped the Wuhan variant of SARS-CoV-2 right at the start had the Chinese authorities not insisted until mid-January 2020, along with the World Health Organisation, that the only way to catch the virus was from an infected animal. Since then, the virus has evolved to be ever more infectious. Alpha spread twice as fast as the original Wuhan strain; delta twice as fast as alpha, and omicron looks like it may have doubled that speed again.

Worryingly, omicron may itself have emerged as a result of modern treatments offered to patients – in particular the use of polyclonal and monoclonal antibodies as therapy. There is evidence that its mutations are concentrated in parts of the spike gene where they help it to evade such antibodies, which both reduces the effectiveness of the treatments and hints at how the mutations arose. Maybe such mutations would not have occurred in a previous era.

Lockdowns are another example. Before the internet, and online retail, locking down the freedoms of healthy people was never an option. Yes, lockdowns slowed the spread but at severe psychological, economic and human cost, and worryingly, one of those costs may have been to prevent the virus evolving to be milder. In the spring of 2020, a strain that caused a severe case of covid was more likely to be passed on (because the victim went to hospital) than a mild case (where the victim stayed locked down at home). Much the same happened in 1918; severe flu cases got evacuated from the trenches to crowded hospitals, while mild cases did not, and in August 1918 the virus became more deadly.

It is no coincidence that there are about 200 kinds of virus that cause the common cold, yet none are dangerous. The perfect recipe for a respiratory virus is to stay in the cool cells of the nose and throat lining and not cause systemic illness that would cause the host to cancel his or her plans. Unlike viruses transmitted by insects, sex or water, respiratory viruses generally do evolve to be mild but highly infectious.

In the 1918 flu pandemic or the “Russian flu” of 1889-90 (which some biologists think was a coronavirus), there were two waves of deaths, then the pathogen settled down to be endemic and mild. I fear – though of course I might be wrong – that our policies this time have saved lives at first but delayed a similar taming of the virus. Evolution is not just mutation: it’s mutation plus selection caused by competition between strains for susceptible hosts.

With luck omicron will prove to be not only more infectious but also milder than delta. According to the doctor who diagnosed it, omicron causes mild symptoms: “It presents mild disease with symptoms being sore muscles and tiredness for a day or two not feeling well…They might have a slight cough.”

That, plus the effect of the vaccines means that Britain’s policy of opening up society in July, defying the modellers’ apocalyptic obsessions, proved sensible. The virus did not spiral out of control, or overwhelm the NHS, but a series of small waves came and went, as society inched towards an endemic truce with the enemy.

The knowledge that we possess about this virus is truly extraordinary. Compared with even two decades ago, we can read its genome, trace its ancestry, map its mutations, predict its characteristics and understand its biochemistry in stunning detail. But this has profited us little. Our ability to stop it in its tracks – vaccines aside – is barely above the mediaeval. Would the course of pandemic have been better or worse if we could not take tests, model curves or differentiate mutant strains? I am not sure. Like Cassandra we are cursed to see the truth, but not be able to act on it.

Matt Ridley is the co-author with Alina Chan of Viral: the search for the origin of Covid-19