The recent correspondence (21 feb ff.) between Professor Ferguson and a number of sceptics published on the Lockdown Sceptics website (LDS) is a striking illustration of the difficulties we face in recovering from the trials of the last year.
Ferguson and his correspondents are to be applauded for their willingness to engage with each other. I have not sent this piece to Prof. Ferguson but should he read it, I welcome his comments.
From Professor Ferguson’s point of view sceptical questioning takes misinformation as fact, persistently ignores mainstream scientific opinion, indulges in “cherry-picking” and is at times an “ideologically motivated rhetorical rant.” (pub. LDS 23/2)
From a sceptical point of view Prof. Ferguson persistently avoids answering straightforward questions and addressing the failures of his methods.
Professor Ferguson has a PhD in theoretical physics. He is a scientist. He writes, “my views are driven by the data and analysis of it.” (pub. LDS 23/2) From a scientific point of view it is unremarkable that he does not directly address the specific examples of distress mentioned by his correspondents. As has often been noted, the plural of anecdote is not data. However it is equally true, though less often noted, that the plural of datum is not reality.
Both science and modelling necessarily involves selection. Confronted with the infinitude of phenomena and data the scientist or modeller must choose the data sets to which he or she will pay attention.
As Ferguson writes, “All models are simplified representations of reality and different scientists interpret data differently and make different assumptions.” (pub. LDS 24/2) Therein lies the rub. One man’s data set is another man’s cherry-picking. All science is cherry-picking. It must be cherry-picking. “A simplified representation” ultimately means “a version of cherry-picking for which I can provide a reasonable argument.”
Many disputes, such as this one, will remain unresolved until there is a recognition by all parties that not only are different "cherries", aka data sets, picked but it is very difficult to pick the best cherries and impossible to prove that one has done so.
Professor Ferguson would demonstrate great moral courage and admirable scientific rigour if he were to address directly the specific data sets cited by sceptics (e.g. Florida v. UK) and the questions which were posed to him by Winton and the anonymous sceptic in spite of the fact that he believes that his questioners are ideologically motivated or in some other way deluded. He could also show precisely how and why he believes their questions to be misguided.
Whether or not he does so, the inescapable question to be addressed by all parties is:
“What are the criteria by means of which we select our data-sets?”
This is a moral, practical and philosophical question. It cannot be answered by science alone.
A further complication
A further complication arises from the fact that science concerns itself with accurate measurement, testing and comparison. It is not unreasonable to say that science favours that which can be measured and has a tendency to avoid engaging with that which cannot be measured. However sometimes what matters lies outwith that which can be measured. This is obviously true in aesthetics or matters of the heart, but significantly it can also be the case in fields wherein science may initially appear to offer a solution.
Let us pause and consider a non-covid example from thirty years ago. Jerry Mander (In the Absence of the Sacred, p 258ff.) writes of a case that illustrates how scientific, objective information based categorical metrics is not just inadequate but actually destructive as compared to understanding founded in lived engagement with the relevant context. He cites the work of Dr Milton M. Freeman of University of Alberta on caribou hunting on the Canadian Ellesmere Islands:
“Canadian wildlife managers told the Inuit that they should hunt only large and/or male caribou, and only a few animals from each herd. The Inuit argued that this practice contradicted their traditional relationship with the animals and would destroy the caribou herds, but their pleas were ignored. The result was as the Inuit predicted. Though their new limit was only twenty-six kills per year – far less than the Inuit had hunted before – the formerly abundant population dropped sharply.”
The old males were not supernumerary as the managers believed. The Inuit knew that the older animals were stronger and more experienced at finding food in the harsh environment. Mander continues to quote Freeman,
“behavioural knowledge of the species was the critical point of the Inuit position, contrasted with an inexact quantitative perspective proposed by the game management service ... The sum total of the [indigenous] community’s empirically based knowledge is awesome in breadth and detail, and often stands in marked contrast to the attenuated data available from scientific studies of these same populations.”
To conclude, Mander writes,
“Professor Freeman argues that the main problem with Western wildlife biology, as with most scientific interventions in age-old economic management systems, is that the basic operating assumptions are inappropriate to the situation at hand. For example, says Freeman, most Western biologists – college trained, usually white, and usually lacking direct knowledge of the environment or cultural group they are researching – will tend to view wildlife as a resource, and the harvesting of animals as strictly an economic activity. … The biologist essentially acts as a resource manager, like a corporate functionary, whose goal is to maximize production and contribute to profit. No effort is made to become sensitive to alternative views stemming from native traditions and culture.”
It is at least plausible that Ferguson and his colleagues and friends may suffer from limitations similar to those of the wildlife managers referred to by Professor Freeman. Even if they live in physical proximity to those for whom they make the rules, their lived-world experience of academia and modelling, of guaranteed income and middle-class comfort is as distant from those who suffer most from lockdowns as the managers were from the Inuit hunters. High status scientists advising the Government lack the lived experience of struggling for a living, of coping with young children in cramped accommodation without outside space for months on end, and of trying to educate children whilst lacking further education oneself.
It would be hard to accept that scientific measurements, however careful or accurate, are not fit for the task they appear to have addressed. It would require great courage to acknowledge that the projections generated by “the science” are misleading and counter-productive because they cannot capture crucial yet immeasurable existential elements.
How can the scientific mind approach the notion that there is a way of knowing which in some circumstances is more insightful than data-driven, evidence-based reasoning? In our rightful appreciation of the benefits of science we can overlook the fact that it is a very, very slow means of accumulating knowledge. I have a GP friend who was so well-respected by his local hospital that the consultants hosted a dinner for him on his retirement. His decades of experience had endowed him with an insight which far outstripped the crude protocols proposed by “evidence-based” research papers which regularly landed on his desk. This sort of insight cannot be codified or quantified, it may even be completely invisible to the novice, but experienced practitioners in medicine and science know it has a value and a meaning far beyond what can be attained by measurement or modelling.
Scientists frequently over-estimate the quality and veracity of the knowledge they have to date. We look back and laugh at notions such as phlogiston but the far more recent story of Barry Marshall’s struggle to get his findings about helicobacter pylori accepted should remind us that there is no good reason to assume that we are not harbouring many similar scientific delusions today.
What can we do?
The anonymous correspondent who wrote about the "relatively rubbish" life of her twins at the moment, about the suicide of their friend, and about the lack of laptops among deprived children of Stratford-on-Avon was doing more than recounting anecdotes. She was offering fragments of a different way of knowing, insights from a richer totality of life experience than can be captured in abstract scientific data. The challenge to scientists is to be willing to be open to this other mode of understanding which augments their own.
The insights of sceptics is that none of us really live in the uncertain world of ‘R’ numbers although far too many have allowed their minds to be captured by such abstract concepts. Each of us really lives in an immediate, physical context and in some cases our existential connection with other people in those contexts demands of us acts of compassion and bravery that are not congruent with the over-extension of over-cautious restrictions.
If scientists like Prof. Ferguson are willing to embrace a more situated and experiential mode of understanding the world - as the rest of us have had to learn about epidemiology - we have a chance to bring our society together and heal our wounds.
The Covid vaccination programme maybe a magnificent scientific achievement or a cynical and expensive excuse to exit a failed strategy. Either way, it is time to move on. That requires humility. It requires scientists to step far from their comfort zone. It requires politicians who imagine they know what they are talking about to learn again to listen. It requires sceptics to practise empathy. There is no simple protocol to follow. It requires all of us to continue to be willing to help when it is far, far easier to dismiss and condemn.
It is time to mourn our losses, to learn how to make better decisions in the future, to let go of fear and to acknowledge that we cannot eliminate risk, nor death, from life. When we remember we are going to die it may make us all a little braver while we live.
This is an excellent article but I think there is a morefundamental problem with Ferguson and the like. It is in the statement that “All models are simplified representations of reality...". IMO that is a category error. Models are NOT reality. They never can be. We are not living (yet) in the world of Quantum computing and DEVS. There is far more complexity in the real world than can ever be captured in a model, even if it did not contain such egregious errors as ignoring seasonality. That is one of the known omissions, then there are the known unknowns and the unknown unknowns... But the modellers both of Covid and Climate Change believe implicitly in their models and treat the outputs as…
I think there's a further aspect to this tale that your piece doesn't quite capture. It is that the core of the supposedly-state-of-the-art epidemiological model used by Ferguson is bafflingly simplistic and appears to be based on a childlike representation of the spread of the lurgy. It does not allow for gradations of infection and immunity, for example - you're either immune or you're completely susceptible.
To an interested lay-person outside the field, immunologists seem to be telling us that the reality is much different from this (that shop workers and doctors build up silent resistance to pathogens due to repeated exposure, for example) but their knowledge is not incorporated in the models, and in discussions they hold their…