For anyone who missed it, this is a remarkable article by Dominic Cummings, outlining his concerns regarding the structure of our governmental institutions, – and their ability to operate successfully on the scale we now demand of them – some of his own anecdotal experiences of what was actually taking place in the DfE building during the years of Gove-mania, and the vital (and ironic) role he feels education has to play in addressing many of the causes of the dysfunction he says he witnessed:
It’s 22,000 words, so not a quick read, but for anyone interested in education, worth it!
Here are a few of my highlights:
B) Their education and training is such that almost nobody has the skills needed to cope with the complexity they face or even to understand the tools (such as Palantir) that might help them. Political ‘experts’ are usually hopeless at predictions and routinely repeat the same sorts of errors without being forced to learn. While our ancestor chiefs understood bows, horses, and agriculture, our contemporary chiefs (and those in the media responsible for scrutiny of decisions) generally do not understand their equivalents, and are often less experienced in managing complex organisations than their predecessors.
The selection, education, and training, of those making crucial decisions about our civilisation are between inadequate and disastrous. The institutions they work in are generally dysfunctional.
Second, our education and training. The education of the majority even in rich countries is between awful and mediocre. In England, few are well-trained in the basics of extended writing or mathematical and scientific modelling and problem-solving. Less than 10 percent per year leave school with formal training in basics such as exponential functions, ‘normal distributions’ (‘the bell curve’), and conditional probability. Only about 2-3 percent are taught about matrices and ‘complex numbers’ (which many children can grasp between the age of 10-14 but they are not given the chance unless they do Further Maths A Level). Less than one percent learn hard skills necessary to grasp how the ‘unreasonable effectiveness of mathematics’ provides the language of nature and a foundation for our scientific civilisation. Only a small subset of that fraction of one percent then study trans-disciplinary issues concerning vital complex systems the failure of which cause chaos.
This small subset has approximately zero overlap with powerful decision-makers. Generally, these people are badly or narrowly educated and trained. Courses offered by elite universities are thought to prepare future leaders well but are clearly inadequate and in some ways are damaging (see below). Those who scramble to the apex of power are sometimes relatively high scorers in tests of verbal ability (like Cameron) but are rarely high scorers in tests of mathematical ability or have good problem-solving skills in cognitively hard areas such as physics or computer science.
MPs and officials have to make constant forecasts but have little idea about how to make them, how the statistics and computer models underlying these forecasts work, or how to judge the reliability of their own views. A recent survey of 100 MPs by the Royal Statistical Society found that only 40% of MPs correctly answered a simple probability question (much simpler than the type of problem they routinely opine on): ‘what is the probability of getting two heads from flipping a fair coin twice?’ Despite their failures on a beginner question, about three-quarters nevertheless said they are confident in their ability to deal with numbers. Issues such as ‘how financial models contributed to the 2008 crisis’ or ‘intelligence and genetics’ cannot be understood in even a basic way without some statistical knowledge, such as normal distribution and standard deviation, yet most MPs do not understand much simpler concepts than these. They also have little knowledge of evolutionary systems (biological or cultural), and little understanding of technology. (How many of those at a senior level dealing with Ebola discussions or financial market disasters recently have any idea about the topology of ‘scale free networks’, cf. HERE? The basic concepts, as opposed to detailed modelling, are not hard to grasp but they do not appear in the typical education of ministers or senior officials.)
A mismatch on a scale of 104 between the experience of MPs and the responsibilities of ministers.Further, ministers have little experience in well-managed complex organisations and their education and training does not fill this huge gap. Even most of the ones who have good motives – and there are many, though they struggle to advance – have a fundamental problem of scale. The apex of the political system is full of people who have never managed employees on the scale of 102 people or budgets on the scale of 106 pounds, yet their job is to reshape bureaucracies on scales of 104 (DfE) – 106 (NHS) employees and 1010-1011 pounds. The scale of their experience of management is therefore often at least 104 off from what they are trying to control. Some unusual people can make jumps like this. Most cannot. For example, Cameron never worked in a highly functioning entity before suddenly acquiring large responsibilities – he went straight from PPE to Conservative Central Office – and never had responsibility for anything on a significant scale so he could not acquire the experience that he so needs now (and, perhaps more importantly, he has never understood how unprepared he and his gang were). The only minister in the DfE team 2010-14 who had significant experience of dealing with budgets on a scale of £108-109 was Nash – unsurprisingly, he was the most effective minister at dealing with DfE budgets / capital / property deals and so on. John Holland, the inventor of ‘genetic algorithms’, points out that ‘changes of three orders of magnitude or more usually require a new science’. It should be no surprise that politics is a story of repeated administrative failure.
Many of these problems can be seen particularly starkly in those who did courses like Politics, Philosophy, and Economics (PPE). PPE is treated as a cross-disciplinary course suitable to educate future leaders. It is failing. Part of the reason for this is that the conventional economics that is taught often gives students a greatly misplaced confidence in their understanding of the world. They are taught to treat some economic theories as if they are similar to physical theories, and there is often spurious precision involving mathematical models but no explanation of the conceptual problems with these models, or the critique of them by physical scientists. I have watched many PPE graduates give presentations of forecasts, complete with decimal points, of economic numbers years into the future, then dismiss arrogantly those who point out the repeated failure of such predictions. PPE also teaches nothing about project management in complex organisations so they have little feel for how decisions will ripple through systems (including bureaucracies) into the real world.
At its worst, therefore, students leave university for politics and the civil service with degrees that reward verbal fluency, some fragments of philosophy, little knowledge of maths or science, and confidence in a sort of arrogant bluffing combined with ignorance about how to get anything done. They think they are prepared to ‘run the country’ but many cannot run their own diaries. In the absence of relevant experience, people naturally resort to destructive micromanagement rather than trusting to Auftragstaktik (give people a goal and let them work out the means rather than issue detailed instructions) which requires good training of junior people. This combination of arrogant incompetence is very widespread in Westminster and responsible for many problems. When such people surround themselves largely or solely with advisers who are very similar to themselves, we know from large amounts of research that the odds are high that groupthink will make these errors and problems even worse.
NB. These gaps in education and training are not a ‘natural’ product of the concepts’ difficulty but because of deep flaws in a) school and university education and b) training programmes