OUR PROPER STUDY SHOULD BE THE HUMAN INTERPRETERS OF DATA
1st September, 2020
John Azzolini - Introduction to Systems Engineering Practices, Public Domain,
As with anything that comes to hand, Covid is shamelessly and opportunistically conscripted by the liberal Remainer press to settle scores and political grudges or even to make the political weather. This means that it’s bound to be used to convict the hated Leaver, Boris Johnson of incompetence, insouciance and the usual Tory callousness. This being the case it was either brave or foolhardy of Professor Neil Scolding, a clinical neuroscientist at Bristol University, to write in the Spectator that “....even if the UK should ultimately prove to have done ‘badly’, the idea that the Prime Minister must be therefore responsible is unsustainable.” and that “some good epidemiologists may – just may – be able to tell us which countries fared worst.” in “perhaps twelve months.”
Professor Scolding lists a litany of over twenty factors that should be properly considered that explain why this is true. It includes, amongst other things, the difference between numbers of deaths versus deaths per head of population, assumed and suspected diagnosis versus proved diagnosis, died ‘of’ versus died ‘with’, inclusive of nursing home or home deaths versus exclusive of the same. Then there are DHS versus ONS methods of calculating excess deaths, the time-frame used, the different stages in the progress of the pandemic different countries are at (at the time of writing some are still on the rising trajectory and some who did well at first and are now doing badly), the trustworthiness of reports from countries the like of Russia, China (who changed reporting criteria several times), Iran, and war-torn Syria and Yemen. Next there is whether the inability of African countries to afford proper testing deflates their figures. Finally there is population density, a country’s age and racial susceptibility profile, the number of international airport and major metropolitan hubs it has and the presence and frequency of multi-generational homes. The Professor suggests, perhaps, that the distinctive nature of a given country explains why what happened was predictable and unavoidable.
This leads us to observe that, in the case of hostile reporting, there are usually two factors. Firstly there is a mountain of ‘big’ data and secondly there are those who interpret it; the readers of the runes. The former is a set of raw impersonal mathematical statistics, the latter living, thinking, choosing human persons. The question is, confronted with the two, which is the most interesting and significant to study - the ‘science’ or the human? This might lead us to remember the two cultures debate of the late 1950s between CP Snow and FR Leavis over the pre-eminence of science or the humanities and which should encompass which. For what it’s worth I believe this is a mutually exclusive zero-sum game with only one possible winner on which we must take sides. Leavis may have conducted the debate in a rude and irascible manner but he was right. Science emerges from the human person and so that person must always wield science lest crackpots think it can wield him. Such crackpots might believe in the AI, machine-learning, algorithmic wet dream in which inhuman number-crunching solves all of our problems. They might even wish to dethrone, demote or even eliminate that which is the most sophisticated thing of all - the human. This is the definition of anti-humanism whereby the technocrats actually see the replacement of redundant and surplus to requirements human persons as an advance. This is also a definition of insanity.
So far, though, the machines have not won and humans are still in the driving seat. Which brings us back to the question - what should the really interesting study be - that of the data or that of the human persons who interpret it? Confronted by Professor Scolding’s vast array of variables how people interpret them tells us primarily something about the human persons that they are and the moral and political dispositions they have. We should be interested in the living person, not the dry facts he appears to be pointing at away from himself. In a sense those facts mean nothing or anything until read according to what the definition of scientific enquiry is supposed to be - a dispassionate and distanced lack of pre-judice of the measured kind that the professor is advocating. To do anything less is to treat scientific data un-scientifically which is ironic to say the least.
When an interpreter draws from the sea of data a heavily biased political interpretation, we can probably conclude that the biased predisposition to find damaging political conclusions was in him rather than in the data itself. This might mean that he simply uses the data as a means to the end of projecting and confirming his prejudices. To understand this we can enlist a legal and a theological idea. In legal circles malicious intent to commit crime is described as mens rea or ‘evil mind’. If a political commentator comes to data with a mind evilly disposed towards Boris Johnson he is bound to make that data proof of Boris Johnson’s failings. What is interesting is why that mind is thus evilly disposed. Why is the commentator endowed with the default liberal mind? In theological terms we can perhaps see the interpreter acting as a kind of prime mover in that the bad things in his mind will be made by him to emerge from the data in a kind of self-fulfilling prophecy that was bound to happen even before he read the data. One can even impute a kind of determinism here. The data thus becomes nothing more than a medium through which an interpreter expresses and reveals his own prejudice.
It is desirable that the human is always pre-eminent over the scientific and, that being achieved and sustained, we should never forget to scrutinise the human person involved and not be distracted by sleights of hand that might lead us to think that ultimate reality resides in lists of digitised binary data. It can be serviceable but, far too frequently, we make it in our own image.
Scolding, whose article is entitled 'Don't rush to judegment on Boris's handling of Covid-19' points up nicely how proper science must not be rushed and how the motives of those who wish to rush in this manner are likely to be far from scientific.