The COVID diaries: Diary of an astrophysicist
A short essay I wrote in August 2020 as part of The COVID Diaries, a project by Imperial College London’s Science Communication Unit which collected reflections from Science Communication Master students, Imperial staff and other contributors on the theme of life under COVID1-19.
We thought we could help. We thought we’d waltz in with a little machine learning magic. We thought we could save the day.
After all, we were the heirs of scientists like Fermi and Oppenheimer, who led the Manhattan Project. We got into physics excited by the Moon landings, decrypted the leftover radiation from the Big Bang, and can hold our own when pitted against the AI overlords of Silicon Valley.
We went in, and learnt about the R number, SEIRS models for infectious disease dynamics, agent-based models, excess deaths. Some of us fired up their Python and coded their own fits to the data from scratch. We were of a breed undaunted by non-linear differential equations, steeped in probabilistic programming, comfortable with numerical General Relativity. How difficult could a little epidemiology modelling be? It wouldn’t take long for us to clear up the confusion, come up with firm predictions, steel the hand of Government and lead society out of the greatest public health crisis anyone could remember since the Spanish Flu.
Then it dawned on us.
The data were not what we thought they were. Unlike the dependable, well-calibrated readings of the luminosity of galaxies billions of light years away, the number of positive cases and even the number of deaths were incomplete, unreliable, and often mis-reported.
The models our epidemiology colleagues had worked on for decades were far more sophisticated than anything a bunch of physicists could come up with in an afternoon spent brainstorming on zoom. Far more.
The human beings whose behaviour we needed to influence to stop the virus from spreading were far less predictable than the identical elementary particles whose compliance with the laws of physics we were used to taking for granted.
But the penny dropped only when we finally realized what should have been apparent from the very beginning, and somehow had completely missed. This wasn’t just about the science. This perhaps wasn’t even mostly about the science, no matter how many times the Government insisted they were being led by it – not just by any science, mind you, but “the very best science in the world”. The science was only one of the ingredients in an unpalatable cocktail of politics, policy, science communication, social media and fake news that polarized people’s reactions between the two extremes of existential panic and unconcerned dismissal.
No, we physicists could not help, at least not in the way we had imagined — and indeed some of the noise we created only made matters worse by making it look as if “the scientists” were split, unreliable, self-contradictory. Unwittingly, we gave ammo to those who had their reasons to dismiss COVID as a Chinese hoax, “nothing worse than the flu”, a fake pandemic created by left-wing Twitterati to advance their liberal agenda.
We did discover a few things about the virus, in the end – but it didn’t take any Markov Chain Monte Carlo or differential calculus: that, just like dark matter, the virus wasn’t any less real for being imperceptible to our senses; that, just like unconscious bias, it disproportionally affected the BAME community; that, just like austerity, it ravaged poorer communities most.
As citizens, we learnt the painful way that science is not about truth, but about understanding and managing uncertainty; that science isn’t a straight highway speeding from one glorious fact to the next, but a tortuous path in an overgrown forest with many dead ends, where the sunny uplands of discovery are not always in sight; that science and technology, as advanced as they are in the 21st century, don’t have all the answers, and the technocratic control on our lives we were lulled into believing we had is an illusion.
As scientists, perhaps the hardest lesson is one that many of us, who skipped the optional classes on history of science at uni, are still catching up on, over 60 years after C.P. Snow’s “The Two Cultures” Rede lecture. Science is a human activity, and for all the mathematical rigour of its tools and almost god-like powers yielded by its applications, it remains entangled in all of the shortcomings of human nature: greed, hubris, self-interest, narcissism, brinkmanship, unethical behaviour, sexism, prejudice, sometimes even fraud. That science works, despite the weaknesses of its unworthy standard bearers, is perhaps the greatest testament to its uncannily self-correcting nature. Paraphrasing the mathematical physicist Eugene Wigner, the unreasonable effectiveness of science is the larger conundrum if one is to understand how is it that an unremarkable biped, armed with nothing else than an opposing thumb and an anomalously large cerebral cortex, came to dominate the Earth.
One day, hopefully soon, COVID will be vanquished. Science will play a big part in that victory, no doubt, but rebuilding our world, its economy and its inter-connectedness, reshaping our society so that resources are more equally shared, finding prosperity for all while respecting the ever more fragile ecosystems of our planet, preserving the Earth for future generations to enjoy will require more than science – it will require all of our resources of imagination, generosity of spirit, creativity and empathy. And no, there won’t be an app for that.
Are we up to the task? Will we raise to the challenge of Jonas Salk, who gave away for free his polio vaccine in 1955 with the disarmingly simple question “Could you patent the sun?”. Only if we follow Salk’s admonition to become “good ancestors” to our future descendants will all the suffering that COVID has bestowed upon us be somewhat redeemed.