On the referendum #33: High performance government, ‘cognitive technologies’, Michael Nielsen, Bret Victor, & ‘Seeing Rooms’

‘People, ideas, machines — in that order!’ Colonel Boyd.

‘The main thing that’s needed is simply the recognition of how important seeing is, and the will to do something about it.’ Bret Victor.

‘[T]he transfer of an entirely new and quite different framework for thinking about, designing, and using information systems … is immensely more difficult than transferring technology.’ Robert Taylor, one of the handful most responsible for the creation of the internet and personal computing, and in inspiration to Bret Victor.

‘[M]uch of our intellectual elite who think they have “the solutions” have actually cut themselves off from understanding the basis for much of the most important human progress.’ Michael Nielsen, physicist. 

Introduction

This blog looks at an intersection of decision-making, technology, high performance teams and government. It sketches some ideas of physicist Michael Nielsen about cognitive technologies and of computer visionary Bret Victor about the creation of dynamic tools to help understand complex systems and ‘argue with evidence’, such as ‘tools for authoring dynamic documents’, and ‘Seeing Rooms’ for decision-makers — i.e rooms designed to support decisions in complex environments. It compares normal Cabinet rooms, such as that used in summer 1914 or October 1962, with state-of-the-art Seeing Rooms. There is very powerful feedback between: a) creating dynamic tools to see complex systems deeper (to see inside, see across time, and see across possibilities), thus making it easier to work with reliable knowledge and interactive quantitative models, semi-automating error-correction etc, and b) the potential for big improvements in the performance of political and government decision-making.

It is relevant to Brexit and anybody thinking ‘how on earth do we escape this nightmare’ but 1) these ideas are not at all dependent on whether you support or oppose Brexit, about which reasonable people disagree, and 2) they are generally applicable to how to improve decision-making — for example, they are relevant to problems like ‘how to make decisions during a fast moving nuclear crisis’ which I blogged about recently, or if you are a journalist ‘what future media could look like to help improve debate of politics’. One of the tools Nielsen discusses is a tool to make memory a choice by embedding learning in long-term memory rather than, as it is for almost all of us, an accident. I know from my days working on education reform in government that it’s almost impossible to exaggerate how little those who work on education policy think about ‘how to improve learning’.

Fields make huge progress when they move from stories (e.g Icarus)  and authority (e.g ‘witch doctor’) to evidence/experiment (e.g physics, wind tunnels) and quantitative models (e.g design of modern aircraft). Political ‘debate’ and the processes of government are largely what they have always been — largely conflict over stories and authorities where almost nobody even tries to keep track of the facts/arguments/models they’re supposedly arguing about, or tries to learn from evidence, or tries to infer useful principles from examples of extreme success/failure. We can see much better than people could in the past how to shift towards processes of government being ‘partially rational discussion over facts and models and learning from the best examples of organisational success‘. But one of the most fundamental and striking aspects of government is that practically nobody involved in it has the faintest interest in or knowledge of how to create high performance teams to make decisions amid uncertainty and complexity. This blindness is connected to another fundamental fact: critical institutions (including the senior civil service and the parties) are programmed to fight to stay dysfunctional, they fight to stay closed and avoid learning about high performance, they fight to exclude the most able people.

I wrote about some reasons for this before the referendum (cf. The Hollow Men). The Westminster and Whitehall response was along the lines of ‘natural party of government’, ‘Rolls Royce civil service’ blah blah. But the fact that Cameron, Heywood (the most powerful civil servant) et al did not understand many basic features of how the world works is why I and a few others gambled on the referendum — we knew that the systemic dysfunction of our institutions and the influence of grotesque incompetents provided an opportunity for extreme leverage. 

Since then, after three years in which the parties, No10 and the senior civil service have imploded (after doing the opposite of what Vote Leave said should happen on every aspect of the negotiations) one thing has held steady — Insiders refuse to ask basic questions about the reasons for this implosion, such as: ‘why Heywood didn’t even put together a sane regular weekly meeting schedule and ministers didn’t even notice all the tricks with agendas/minutes etc’, how are decisions really made in No10, why are so many of the people below some cognitive threshold for understanding basic concepts (cf. the current GATT A24 madness), what does it say about Westminster that both the Adonis-Remainers and the Cash-ERGers have become more detached from reality while a large section of the best-educated have effectively run information operations against their own brains to convince themselves of fairy stories about Facebook, Russia and Brexit…

It’s a mix of amusing and depressing — but not surprising to me — to hear Heywood explain HERE how the British state decided it couldn’t match the resources of a single multinational company or a single university in funding people to think about what the future might hold, which is linked to his failure to make serious contingency plans for losing the referendum. And of course Heywood claimed after the referendum that we didn’t need to worry about the civil service because on project management it has ‘nothing to learn’ from the best private companies. The elevation of Heywood in the pantheon of SW1 is the elevation of the courtier-fixer at the expense of the thinker and the manager — the universal praise for him recently is a beautifully eloquent signal that those in charge are the blind leading the blind and SW1 has forgotten skills of high value, the skills of public servants such as Alanbrooke or Michael Quinlan.

This blog is hopefully useful for some of those thinking about a) improving government around the world and/or b) ‘what comes after the coming collapse and reshaping of the British parties, and how to improve drastically the performance of critical institutions?’

Some old colleagues have said ‘Don’t put this stuff on the internet, we don’t want the second referendum mob looking at it.’ Don’t worry! Ideas like this have to be forced down people’s throats practically at gunpoint. Silicon Valley itself has barely absorbed Bret Victor’s ideas so how likely is it that there will be a rush to adopt them by the world of Blair and Grieve?! These guys can’t tell the difference between courtier-fixers and people with models for truly effective action like General Groves (HERE). Not one in a thousand will read a 10,000 word blog on the intersection of management and technology and the few who do will dismiss it as the babbling of a deluded fool, they won’t learn any more than they learned from the 2004 referendum or from Vote Leave. And if I’m wrong? Great. Things will improve fast and a second referendum based on both sides applying lessons from Bret Victor would be dynamite.

NB. Bret Victor’s project, Dynamic Land, is a non-profit. For an amount of money that a government department like the Department for Education loses weekly without any minister realising it’s lost (in the millions per week in my experience because the quality of financial control is so bad), it could provide crucial funding for Victor and help itself. Of course, any minister who proposed such a thing would be told by officials ‘this is illegal under EU procurement law and remember minister that we must obey EU procurement law forever regardless of Brexit’ — something I know from experience officials say to ministers whether it is legal or not when they don’t like something. And after all, ministers meekly accepted the Kafka-esque order from Heywood to prioritise duties of goodwill to the EU under A50 over preparations to leave A50, so habituated had Cameron’s children become to obeying the real deputy prime minister…

Below are 4 sections:

1. The value found in intersections of fields
2. Some ideas of Bret Victor
3. Some ideas of Michael Nielsen
4. A summary

*

1. Extreme value is often found in the intersection of fields

The legendary Colonel Boyd (he of the ‘OODA loop’) would shout at audiences ‘People, ideas, machines — in that order.‘ Fundamental political problems we face require large improvements in the quality of all three and, harder, systems to integrate all three. Such improvements require looking carefully at the intersection of roughly five entangled areas of study. Extreme value is often found at such intersections.

• Explore what we know about the selection, education and training of people for high performance (individual/team/organisation) in different fields. We should be selecting people much deeper in the tails of the ability curve — people who are +3 (~1:1,000) or +4 (~1:30,000) standard deviations above average on intelligence, relentless effort, operational ability and so on (now practically entirely absent from the ’50 most powerful people in Britain’). We should  train them in the general art of ‘thinking rationally’ and making decisions amid uncertainty (e.g Munger/Tetlock-style checklists, exercises on SlateStarCodex blog). We should train them in the practical reasons for normal ‘mega-project failure’ and case studies such as the Manhattan Project (General Groves), ICBMs (Bernard Schriever), Apollo (George Mueller), ARPA-PARC (Robert Taylor) that illustrate how the ‘unrecognised simplicities’ of high performance bring extreme success and make them work on such projects before they are responsible for billions rather than putting people like Cameron in charge (after no experience other than bluffing through PPE then PR). NB. China’s leaders have studied these episodes intensely while American and British institutions have actively ‘unlearned’ these lessons.
• Explore the frontiers of the science of prediction across different fields from physics to weather forecasting to finance and epidemiology. For example, ideas from physics about early warning systems in physical systems have application in many fields, including questions like: to what extent is it possible to predict which news will persist over different timescales, or predict wars from news and social media? There is interesting work combining game theory, machine learning, and Red Teams to predict security threats and improve penetration testing (physical and cyber). The Tetlock/IARPA project showed dramatic performance improvements in political forecasting are possible, contra what people such as Kahneman had thought possible. A recent Nature article by Duncan Watts explained fundamental problems with the way normal social science treats prediction and suggested new approaches — which have been almost entirely ignored by mainstream economists/social scientists. There is vast scope for applying ideas and tools from the physical sciences and data science/AI — largely ignored by mainstream social science, political parties, government bureaucracies and media — to social/political/government problems (as Vote Leave showed in the referendum, though this has been almost totally obscured by all the fake news: clue — it was not ‘microtargeting’).
• Explore technology and tools. For example, Bret Victor’s work and Michael Nielsen’s work on cognitive technologies. The edge of performance in politics/government will be defined by teams that can combine the ancient ‘unrecognised simplicities of high performance’ with edge-of-the-art technology. No10 is decades behind the pace in old technologies like TV, doesn’t understand simple tools like checklists, and is nowhere with advanced technologies.
• Explore the frontiers of communication (e.g crisis management, applied psychology). Technology enables people to improve communication with unprecedented speed, scale and iterative testing. It also allows people to wreak chaos with high leverage. The technologies are already beyond the ability of traditional government centralised bureaucracies to cope with. They will develop rapidly such that most such centralised bureaucracies lose more and more control while a few high performance governments use the leverage they bring (c.f China’s combination of mass surveillance, AI, genetic identification, cellphone tracking etc as they desperately scramble to keep control). The better educated think that psychological manipulation is something that happens to ‘the uneducated masses’ but they are extremely deluded — in many ways people like FT pundits are much easier to manipulate, their education actually makes them more susceptible to manipulation, and historically they are the ones who fall for things like Russian fake news (cf. the Guardian and New York Times on Stalin/terror/famine in the 1930s) just as now they fall for fake news about fake news. Despite the centrality of communication to politics it is remarkable how little attention Insiders pay to what works — never mind the question ‘what could work much better?’.  The fact that so much of the media believes total rubbish about social media and Brexit shows that the media is incapable of analysing the intersection of politics and technology but, although it is obviously bad that the media disinforms the public, the only rational planning assumption is that this problem will continue and even get worse. The media cannot explain either the use of TV or traditional polling well, these have been extremely important for over 70 years, and there is no trend towards improvement so a sound planning assumption is surely that the media will do even worse with new technologies and data science. This will provide large opportunities for good and evil. A new approach able to adapt to the environment an order of magnitude faster than now would disorient political opponents (desperately scrolling through Twitter) to such a degree — in Boyd’s terms it would ‘collapse their OODA loops’ — that it could create crucial political space for focus on the extremely hard process of rewiring government institutions which now seems impossible for Insiders to focus on given their psychological/operational immersion in the hysteria of 24 hour rolling news and the constant crises generated by dysfunctional bureaucracies.
• Explore how to re-program political/government institutions at the apex of decision-making authority so that a) people are more incentivised to optimise things we want them to optimise, like error-correction and predictive accuracy, and less incentivised to optimise bureaucratic process, prestige, and signalling as our institutions now do; b) institutions are incentivised to build high performance teams rather than make this practically illegal at the apex of government; and c) we have ‘immune systems’ based on decentralisation and distributed control to minimise the inevitable failures of even the best people and teams.

Example 1: Red Teams and pre-mortems can combat groupthink and normal cognitive biases but they are practically nowhere in the formal structure of governments. There is huge scope for a Parliament-mandated small and extremely elite Red Team operating next to, and in some senses above, the Cabinet Office to ensure diversity of opinions, fight groupthink and other standard biases, make sure lessons are learned and so on. Cost: a few million that it would recoup within weeks by stopping blunders.

Example 2: prediction tournaments/markets could improve policy and project management, with people able to ‘short’ official delivery timetables — imagine being able to short Grayling’s transport announcements, for example. In many areas new markets could help — e.g markets to allow shorting of house prices to dampen bubbles, as Chris Dillow and others have suggested. The way in which the IARPA/Tetlock work has been ignored in SW1 is proof that MPs and civil servants are not actually interested in — or incentivised to be interested in — who is right, who is actually an ‘expert’, and so on. There are tools available if new people do want to take these things seriously. Cost: a few million at most, possibly thousands, that it would recoup within a year by stopping blunders.

Example 3: we need to consider projects that could bootstrap new international institutions that help solve more general coordination problems such as the risk of accidental nuclear war. The most obvious example of a project like this I can think of is a manned international lunar base which would be useful for a) basic science, b) the practical purposes of building urgently needed near-Earth infrastructure for space industrialisation, and c) to force the creation of new practical international institutions for cooperation between Great Powers. George Mueller’s team that put man on the moon in 1969 developed a plan to do this that would have been built by now if their plans had not been tragically abandoned in the 1970s. Jeff Bezos is explicitly trying to revive the Mueller vision and Britain should be helping him do it much faster. The old institutions like the UN and EU — built on early 20th Century assumptions about the performance of centralised bureaucracies — are incapable of solving global coordination problems. It seems to me more likely that institutions with qualities we need are much more likely to emerge out of solving big problems than out of think tank papers about reforming existing institutions. Cost = 10s/100s of billions, return = trillions, or near infinite if shifting our industrial/psychological frontiers into space drastically reduces the chances of widespread destruction.

A) Some fields have fantastic predictive models and there is a huge amount of high quality research, though there is a lot of low-hanging fruit in bringing methods from one field to another.

B) We know a lot about high performance including ‘systems management’ for complex projects but very few organisations use this knowledge and government institutions overwhelmingly try to ignore and suppress the knowledge we have.

C) Some fields have amazing tools for prediction and visualisation but very few organisations use these tools and almost nobody in government (where colour photocopying is a major challenge).

D) We know a lot about successful communication but very few organisations use this knowledge and most base action on false ideas. E.g political parties spend millions on spreading ideas but almost nothing on thinking about whether the messages are psychologically compelling or their methods/distribution work, and TV companies spend billions on news but almost nothing understanding what science says about how to convey complex ideas — hence why you see massively overpaid presenters like Evan Davis babbling metaphors like ‘economic takeoff’ in front of an airport while his crew films a plane ‘taking off’, or ‘the economy down the plughole’ with pictures of — a plughole.

E) Many thousands worldwide are thinking about all sorts of big government issues but very few can bring them together into coherent plans that a government can deliver and there is almost no application of things like Red Teams and prediction markets. E.g it is impossible to describe the extent to which politicians in Britain do not even consider ‘the timetable and process for turning announcement X into reality’ as something to think about — for people like Cameron and Blair the announcement IS the only reality and ‘management’ is a dirty word for junior people to think about while they focus on ‘strategy’. As I have pointed out elsewhere, it is fascinating that elite business schools have been collecting billions in fees to teach their students WRONGLY that operational excellence is NOT a source of competitive advantage, so it is no surprise that politicians and bureaucrats get this wrong.

But I can see almost nobody integrating the very best knowledge we have about A+B+C+D with E and I strongly suspect there are trillion dollar bills lying on the ground that could be grabbed for trivial cost — trillion dollar bills that people with power are not thinking about and are incentivised not to think about. I might be wrong but I would remind readers that Vote Leave was itself a bet on this proposition being right and I think its success should make people update their beliefs on the competence of elite political institutions and the possibilities for improvement.

Here I want to explore one set of intersections — the ideas of Bret Victor and Michael Nielsen.

*

2. Bret Victor: Cognitive technologies, dynamic tools, interactive quantitative models, Seeing Rooms — making it as easy to insert facts, data, and models in political discussion as it is to insert emoji 

In the 1960s visionaries such as Joseph Licklider, Robert Taylor and Doug Engelbart developed a vision of networked interactive computing that provided the foundation not just for new technologies (the internet, PC etc) but for whole new industries. Licklider, Sutherland,Taylor et al provided a model (ARPA) for how science funding can work. Taylor provided a model (PARC) of how to manage a team of extremely talented people who turned a profound vision into reality. The original motivation for the vision of networked interactive computing was to help humans make good decisions in a complex world — or, ‘augmenting human intelligence’ and ‘man-machine symbiosis’. This story shows how to make big improvements in the world with very few resources if they are structured right: PARC involved ~25 key people and tens of millions over roughly a decade and generated trillions of dollars in value. If interested in the history and the super-productive processes behind the success of ARPA-PARC read THIS.

It’s fascinating that in many ways the original 1960s Licklider vision has still not been implemented. The Silicon Valley ecosystem developed parts of the vision but not others for complex reasons I don’t understand (cf. The Future of Programming). One of those who is trying to implement parts of the vision that have not been implemented is Bret Victor. Bret Victor is a rare thing: a genuine visionary in the computing world according to some of those ‘present at the creation’ of ARPA-PARC such as Alan Kay. His ideas lie at critical intersections between fields sketched above. Watch talks such as Inventing on Principle and Media for Thinking the Unthinkable and explore his current project, Dynamic Land in Berkeley.

Victor has described, and now demonstrates in Dynamic Land, how existing tools fail and what is possible. His core principle is that creators need an immediate connection to what they are creating. Current programming languages and tools are mostly based on very old ideas before computers even had screens and there was essentially no interactivity — they date from the era of punched cards. They do not allow users to interact dynamically. New dynamic tools enable us to think previously unthinkable thoughts and allow us to see and interact with complex systems: to see inside, see across time, and see across possibilities.

I strongly recommend spending a few days exploring his his whole website but I will summarise below his ideas on two things:

1. His ideas about how to build new dynamic tools for working with data and interactive models.
2. His ideas about transforming the physical spaces in which teams work so that dynamic tools are embedded in their environment — people work inside a tool.

Applying these ideas would radically improve how people make decisions in government and how the media reports politics/government.

Language and writing were cognitive technologies created thousands of years ago which enabled us to think previously unthinkable thoughts. Mathematical notation did the same over the past 1,000 years. For example, take a mathematics problem described by the 9th Century mathematician al-Khwarizmi (who gave us the word algorithm):

Once modern notation was invented, this could be written instead as:

x² + 10x = 39

Michael Nielsen uses a similar analogy. Descartes and Fermat demonstrated that equations can be represented on a diagram and a diagram can be represented as an equation. This was a new cognitive technology, a new way of seeing and thinking: algebraic geometry. Changes to the ‘user interface’ of mathematics were critical to its evolution and allowed us to think unthinkable thoughts (Using Artificial Intelligence to Augment Human Intelligence, see below).

Similarly in the 18th Century, there was the creation of data graphics to demonstrate trade figures. Before this, people could only read huge tables. This is the first data graphic:

The Jedi of data visualisation, Edward Tufte, describes this extraordinary graphic of Napoleon’s invasion of Russia as ‘probably the best statistical graphic ever drawn’. It shows the losses of Napoleon’s army: from the Polish-Russian border, the thick band shows the size of the army at each position, the path of Napoleon’s winter retreat from Moscow is shown by the dark lower band, which is tied to temperature and time scales (you can see some of the disastrous icy river crossings famously described by Tolstoy). NB. The Cabinet makes life-and-death decisions now with far inferior technology to this from the 19th Century (see below).

If we look at contemporary scientific papers they represent extremely compressed information conveyed through a very old fashioned medium, the scientific journal. Printed journals are centuries old but the ‘modern’ internet versions are usually similarly static. They do not show the behaviour of systems in a visual interactive way so we can see the connections between changing values in the models and changes in behaviour of the system. There is no immediate connection. Everything is pretty much the same as a paper and pencil version of a paper. In Media for Thinking the Unthinkable, Victor shows how dynamic tools can transform normal static representations so systems can be explored with immediate feedback. This dramatically shows how much more richly and deeply ideas can be explored. With Victor’s tools we can interact with the systems described and immediately grasp important ideas that are hidden in normal media.

Picture: the very dense writing of a famous paper (by chance the paper itself is at the intersection of politics/technology and Watts has written excellent stuff on fake news but has been ignored because it does not fit what ‘the educated’ want to believe)

Picture: the same information presented differently. Victor’s tools make the information less compressed so there’s less work for the brain to do ‘decompressing’. They not only provide visualisations but the little ‘sliders’ over the graphics are to drag buttons and interact with the data so you see the connection between changing data and changing model. A dynamic tool transforms a scientific paper from ‘pencil and paper’ technology to modern interactive technology.

Victor’s essay on climate change

Victor explains in detail how policy analysis and public debate of climate change could be transformed. Leave aside the subject matter — of course it’s extremely important, anybody interested in this issue will gain from reading the whole thing and it would be great material for a school to use for an integrated science / economics / programming / politics project, but my focus is on his ideas about tools and thinking, not the specific subject matter.

Climate change is a great example to consider because it involves a) a lot of deep scientific knowledge, b) complex computer modelling which is understood in detail by a tiny fraction of 1% (and almost none of the social science trained ‘experts’ who are largely responsible for interpreting such models for politicians/journalists, cf HERE for the science of this), c) many complex political, economic, cultural issues, d) very tricky questions about how policy is discussed in mainstream culture, and e) the problem of how governments try to think about and act on important, complex, and long-term problems. Scientific knowledge is crucial but it cannot by itself answer the question: what to do? The ideas BV describes to transform the debate on climate change apply generally to how we approach all important political issues.

In the section Languages for technical computing, BV describes his overall philosophy (if you look at the original you will see dynamic graphics to help make each point but I can’t make them play on my blog — a good example of the failure of normal tools!):

‘The goal of my own research has been tools where scientists see what they’re doing in realtime, with immediate visual feedback and interactive exploration. I deeply believe that a sea change in invention and discovery is possible, once technologists are working in environments designed around:

• ubiquitous visualization and in-context manipulation of the system being studied;
• actively exploring system behavior across multiple levels of abstraction in parallel;
• visually investigating system behavior by transforming, measuring, searching, abstracting;
• seeing the values of all system variables, all at once, in context;
• dynamic notations that embed simulation, and show the effects of parameter changes;
• visually improvising special-purpose dynamic visualizations as needed.’

He then describes how the community of programming language developers have failed to create appropriate languages for scientists, which I won’t go into but which is fascinating.

He then describes the problem of how someone can usefully get to grips with a complex policy area involving technological elements.

‘How can an eager technologist find their way to sub-problems within other people’s projects where they might have a relevant idea? How can they be exposed to process problems common across many projects?… She wishes she could simply click on “gas turbines”, and explore the space:

• What are open problems in the field?
• Who’s working on which projects?
• What are the fringe ideas?
• What are the process bottlenecks?
• What dominates cost? What limits adoption?
• Why make improvements here? How would the world benefit?

‘None of this information is at her fingertips. Most isn’t even openly available — companies boast about successes, not roadblocks. For each topic, she would have to spend weeks tracking down and meeting with industry insiders. What she’d like is a tool that lets her skim across entire fields, browsing problems and discovering where she could be most useful…

‘Suppose my friend uncovers an interesting problem in gas turbines, and comes up with an idea for an improvement. Now what?

• Is the improvement significant?
• Is the solution technically feasible?
• How much would the solution cost to produce?
• How much would it need to cost to be viable?
• Who would use it? What are their needs?
• What metrics are even relevant?

‘Again, none of this information is at her fingertips, or even accessible. She’d have to spend weeks doing an analysis, tracking down relevant data, getting price quotes, talking to industry insiders.

‘What she’d like are tools for quickly estimating the answers to these questions, so she can fluidly explore the space of possibilities and identify ideas that have some hope of being important, feasible, and viable.

‘Consider the Plethora on-demand manufacturing service, which shows the mechanical designer an instant price quote, directly inside the CAD software, as they design a part in real-time. In what other ways could inventors be given rapid feedback while exploring ideas?’

Victor then describes a public debate over a public policy. Ideas were put forward. Everybody argued.

‘Who to believe? The real question is — why are readers and decision-makers forced to “believe” anything at all? Many claims made during the debate offered no numbers to back them up. Claims with numbers rarely provided context to interpret those numbers. And never — never! — were readers shown the calculations behind any numbers. Readers had to make up their minds on the basis of hand-waving, rhetoric, bombast.’

And there was no progress because nobody could really learn from the debate or even just be clear about exactly what was being proposed. Sound familiar?!! This is absolutely normal and Victor’s description applies to over 99% of public policy debates.

Victor then describes how you can take the policy argument he had sketched and change its nature. Instead of discussing words and stories, DISCUSS INTERACTIVE MODELS. 

Here you need to click to the original to understand the power of what he is talking about as he programs a simple example.

‘The reader can explore alternative scenarios, understand the tradeoffs involved, and come to an informed conclusion about whether any such proposal could be a good decision.

‘This is possible because the author is not just publishing words. The author has provided a model — a set of formulas and algorithms that calculate the consequences of a given scenario… Notice how the model’s assumptions are clearly visible, and can even be adjusted by the reader.

‘Readers are thus encouraged to examine and critique the model. If they disagree, they can modify it into a competing model with their own preferred assumptions, and use it to argue for their position. Model-driven material can be used as grounds for an informed debate about assumptions and tradeoffs.

‘Modeling leads naturally from the particular to the general. Instead of seeing an individual proposal as “right or wrong”, “bad or good”, people can see it as one point in a large space of possibilities. By exploring the model, they come to understand the landscape of that space, and are in a position to invent better ideas for all the proposals to come. Model-driven material can serve as a kind of enhanced imagination.‘

Victor then looks at some standard materials from those encouraging people to take personal action on climate change and concludes:

‘These are lists of proverbs. Little action items, mostly dequantified, entirely decontextualized. How significant is it to “eat wisely” and “trim your waste”? How does it compare to other sources of harm? How does it fit into the big picture? How many people would have to participate in order for there to be appreciable impact? How do you know that these aren’t token actions to assauge guilt?

‘And why trust them? Their rhetoric is catchy, but so is the horrific “denialist” rhetoric from the Cato Institute and similar. When the discussion is at the level of “trust me, I’m a scientist” and “look at the poor polar bears”, it becomes a matter of emotional appeal and faith, a form of religion.

‘Climate change is too important for us to operate on faith. Citizens need and deserve reading material which shows context — how significant suggested actions are in the big picture — and which embeds models — formulas and algorithms which calculate that significance, for different scenarios, from primary-source data and explicit assumptions.’

Even the supposed ‘pros’ — Insiders at the top of research fields in politically relevant areas — have to scramble around typing words into search engines, crawling around government websites, and scrolling through PDFs. Reliable data takes ages to find. Reliable models are even harder to find. Vast amounts of useful data and models exist but they cannot be found and used effectively because we lack the tools.

‘Authoring tools designed for arguing from evidence’

Why don’t we conduct public debates in the way his toy example does with interactive models? Why aren’t paragraphs in supposedly serious online newspapers written like this? Partly because of the culture, including the education of those who run governments and media organisations, but also because the resources for creating this sort of material don’t exist.

‘In order for model-driven material to become the norm, authors will need data, models, tools, and standards…

‘Suppose there were good access to good data and good models. How would an author write a document incorporating them? Today, even the most modern writing tools are designed around typing in words, not facts. These tools are suitable for promoting preconceived ideas, but provide no help in ensuring that words reflect reality, or any plausible model of reality. They encourage authors to fool themselves, and fool others…

‘Imagine an authoring tool designed for arguing from evidence. I don’t mean merely juxtaposing a document and reference material, but literally “autocompleting” sourced facts directly into the document. Perhaps the tool would have built-in connections to fact databases and model repositories, not unlike the built-in spelling dictionary. What if it were as easy to insert facts, data, and models as it is to insert emoji and cat photos?

‘Furthermore, the point of embedding a model is that the reader can explore scenarios within the context of the document. This requires tools for authoring “dynamic documents” — documents whose contents change as the reader explores the model. Such tools are pretty much non-existent.’

These sorts of tools for authoring dynamic documents should be seen as foundational technology like the integrated circuit or the internet.

‘Foundational technology appears essential only in retrospect. Looking forward, these things have the character of “unknown unknowns” — they are rarely sought out (or funded!) as a solution to any specific problem. They appear out of the blue, initially seem niche, and eventually become relevant to everything.

‘They may be hard to predict, but they have some common characteristics. One is that they scale well. Integrated circuits and the internet both scaled their “basic idea” from a dozen elements to a billion. Another is that they are purpose-agnostic. They are “material” or “infrastructure”, not applications.’

Victor ends with a very potent comment — that much of what we observe is ‘rearranging  app icons on the deck of the Titanic’. Commercial incentives drive people towards trying to create ‘the next Facebook’ — not fixing big social problems. I will address this below.

If you are an arts graduate interested in these subjects but not expert (like me), here is an example that will be more familiar… If you look at any big historical subject, such as ‘why/how did World War I start?’ and examine leading scholarship carefully, you will see that all the leading books on such subjects provide false chronologies and mix facts with errors such that it is impossible for a careful reader to be sure about crucial things. It is routine for famous historians to write that ‘X happened because Y’ when Y happened after X. Part of the problem is culture but this could potentially be improved by tools. A very crude example: why doesn’t Kindle make it possible for readers to log factual errors, with users’ reliability ranked by others, so authors can easily check potential errors and fix them in online versions of books? Even better, this could be part of a larger system to develop gold standard chronologies with each ‘fact’ linked to original sources and so on. This would improve the reliability of historical analysis and it would create an ‘anti-entropy’ ratchet — now, entropy means that errors spread across all books on a subject and there is no mechanism to reverse this…

 

‘Seeing Rooms’: macro-tools to help make decisions

Victor also discusses another fundamental issue: the rooms/spaces in which most modern work and thinking occurs are not well-suited to the problems being tackled and we could do much better. Victor is addressing advanced manufacturing and robotics but his argument applies just as powerfully, perhaps more powerfully, to government analysis and decision-making.

Now, ‘software based tools are trapped in tiny rectangles’. We have very sophisticated tools but they all sit on computer screens on desks, just as you are reading this blog.

In contrast, ‘Real-world tools are in rooms where workers think with their bodies.’ Traditional crafts occur in spatial environments designed for that purpose. Workers walk around, use their hands, and think spatially. ‘The room becomes a macro-tool they’re embedded inside, an extension of the body.’ These rooms act like tools to help them understand their problems in detail and make good decisions.

Picture: rooms designed for the problems being tackled

The wave of 3D printing has developed ‘maker rooms’ and ‘Fab Labs’ where people work with a set of tools that are too expensive for an individual. The room is itself a network of tools. This approach is revolutionising manufacturing.

Why is this useful?

‘Modern projects have complex behavior… Understanding requires seeing and the best seeing tools are rooms.’ This is obviously particularly true of politics and government.

Here is a photo of a recent NASA mission control room. The room is set up so that all relevant people can see relevant data and models at different scales and preserve a common picture of what is important. NASA pioneered thinking about such rooms and the technology and tools needed in the 1960s.

Here are pictures of two control rooms for power grids.

Here is a panoramic photo of the unified control centre for the Large Hadron Collider – the biggest of ‘big data’ projects. Notice details like how they have removed all pillars so nothing interrupts visual communication between teams.

Now contrast these rooms with rooms from politics.

Here is the Cabinet room. I have been in this room. There are effectively no tools. In the 19th Century at least Lord Salisbury used the fireplace as a tool. He would walk around the table, gather sensitive papers, and burn them at the end of meetings. The fire is now blocked. The only other tool, the clock, did not work when I was last there. Over a century, the physical space in which politicians make decisions affecting potentially billions of lives has deteriorated.

British Cabinet room practically as it was July 1914

Here are JFK and EXCOM making decisions during the Cuban Missile Crisis that moved much faster than July 1914, compressing decisions leading to the destruction of global civilisation potentially into just minutes.

Here is the only photo in the public domain of the room known as ‘COBRA’ (Cabinet Office Briefing Room) where a shifting set of characters at the apex of power in Britain meet to discuss crises.

Notice how poor it is compared to NASA, the LHC etc. There has clearly been no attempt to learn from our best examples about how to use the room as a tool. The screens at the end are a late add-on to a room that is essentially indistinguishable from the room in which Prime Minister Asquith sat in July 1914 while doodling notes to his girlfriend as he got bored. I would be surprised if the video technology used is as good as what is commercially available cheaper, the justification will be ‘security’, and I would bet that many of the decisions about the operation of this room would not survive scrutiny from experts in how to construct such rooms.

I have not attended a COBRA meeting but I’ve spoken to many who have. The meetings, as you would expect looking at this room, are often normal political meetings. That is:

• aims are unclear,
• assumptions are not made explicit,
• there is no use of advanced tools,
• there is no use of quantitative models,
• discussions are often dominated by lawyers so many actions are deemed ‘unlawful’ without proper scrutiny (and this device is routinely used by officials to stop discussion of options they dislike for non-legal reasons),
• there is constant confusion between policy, politics and PR then the cast disperses without clarity about what was discussed and agreed.

Here is a photo of the American equivalent – the Situation Room.

It has a few more screens but the picture is essentially the same: there are no interactive tools beyond the ability to speak and see someone at a distance which was invented back in the 1950s/1960s in the pioneering programs of SAGE (automated air defence) and Apollo (man on the moon). Tools to help thinking in powerful ways are not taken seriously. It is largely the same, and decisions are made the same, as in the Cuban Missile Crisis. In some ways the use of technology now makes management worse as it encourages Presidents and their staff to try to micromanage things they should not be managing, often in response to or fear of the media.

Individual ministers’ officers are also hopeless. The computers are old and rubbish. Even colour printing is often a battle. Walls are for kids’ pictures. In the DfE officials resented even giving us paper maps of where schools were and only did it when bullied by the private office. It was impossible for officials to work on interactive documents. They had no technology even for sharing documents in a way that was then (2011) normal even in low-performing organisations. Using GoogleDocs was ‘against the rules’. (I’m told this has slightly improved.) The whole structure of ‘submissions’ and ‘red boxes’ is hopeless. It is extremely bureaucratic and slow. It prevents serious analysis of quantitative models. It reinforces the lack of proper scientific thinking in policy analysis. It guarantees confusion as ministers scribble notes and private offices interpret rushed comments by exhausted ministers after dinner instead of having proper face-to-face meetings that get to the heart of problems and resolve conflicts quickly. The whole approach reinforces the abject failure of the senior civil service to think about high performance project management.

Of course, most of the problems with the standards of policy and management in the civil service are low or no-tech problems — they involve the ‘unrecognised simplicities’ that are independent of, and prior to, the use of technology — but all these things negatively reinforce each other. Anybody who wants to do things much better is scuppered by Whitehall’s entangled disaster zone of personnel, training, management, incentives and tools.

*

Dynamic Land: ‘amazing’

I won’t go into this in detail. Dynamic Land is in a building in Berkeley. I visited last year. It is Victor’s attempt to turn the ideas above into a sort of living laboratory. It is a large connected set of rooms that have computing embedded in surfaces. For example, you can scribble equations on a bit of paper, cameras in the ceiling read your scribbles automatically, turn them into code, and execute them — for example, by producing graphics. You can then physically interact with models that appear on the table or wall while the cameras watch your hands and instantly turn gestures into new code and change the graphics or whatever you are doing. Victor has put these cutting edge tools into a space and made it open to the Berkeley community. This is all hard to explain/understand because you haven’t seen anything like it even in sci-fi films (it’s telling the media still uses the 15 year-old Minority Report as its sci-fi illustration for such things).

This video gives a little taste. I visited with a physicist who works on the cutting edge of data science/AI. I was amazed but I know nothing about such things — I was interested to see his reaction as he scribbled gravitational equations on paper and watched the cameras turn them into models on the table in real-time, then he changed parameters and watched the graphics change in real-time on the table (projected from the ceiling): ‘Ohmygod, this is just obviously the future, absolutely amazing.’ The thought immediately struck us: imagine the implications of having policy discussions with such tools instead of the usual terrible meetings. Imagine discussing HS2 budgets or possible post-Brexit trading arrangements with the models running like this for decision-makers to interact with.

Video of Dynamic Land: the bits of coloured paper are ‘code’, graphics are projected from the ceiling

 

*

3. Michael Nielsen and cognitive technologies

Connected to Victor’s ideas are those of the brilliant physicist, Michael Nielsen. Nielsen wrote the textbook on quantum computation and a great book, Reinventing Discovery, on the evolution of the scientific method. For example, instead of waiting for the coincidence of Grossmann helping out Einstein with some crucial maths, new tools could create a sort of ‘designed serendipity’ to help potential collaborators find each other.

In his essay Thought as a Technology, Nielsen describes the feedback between thought and interfaces:

‘In extreme cases, to use such an interface is to enter a new world, containing objects and actions unlike any you’ve previously seen. At first these elements seem strange. But as they become familiar, you internalize the elements of this world. Eventually, you become fluent, discovering powerful and surprising idioms, emergent patterns hidden within the interface. You begin to think with the interface, learning patterns of thought that would formerly have seemed strange, but which become second nature. The interface begins to disappear, becoming part of your consciousness. You have been, in some measure, transformed.’

He describes how normal language and computer interfaces are cognitive technologies:

‘Language is an example of a cognitive technology: an external artifact, designed by humans, which can be internalized, and used as a substrate for cognition. That technology is made up of many individual pieces – words and phrases, in the case of language – which become basic elements of cognition. These elements of cognition are things we can think with…

‘In a similar way to language, maps etc, a computer interface can be a cognitive technology. To master an interface requires internalizing the objects and operations in the interface; they become elements of cognition. A sufficiently imaginative interface designer can invent entirely new elements of cognition… In general, what makes an interface transformational is when it introduces new elements of cognition that enable new modes of thought. More concretely, such an interface makes it easy to have insights or make discoveries that were formerly difficult or impossible. At the highest level, it will enable discoveries (or other forms of creativity) that go beyond all previous human achievement.’

Nielsen describes how powerful ways of thinking among mathematicians and physicists are hidden from view and not part of textbooks and normal teaching.

‘The reason is that traditional media are poorly adapted to working with such representations… If experts often develop their own representations, why do they sometimes not share those representations? To answer that question, suppose you think hard about a subject for several years… Eventually you push up against the limits of existing representations. If you’re strongly motivated – perhaps by the desire to solve a research problem – you may begin inventing new representations, to provide insights difficult through conventional means. You are effectively acting as your own interface designer. But the new representations you develop may be held entirely in your mind, and so are not constrained by traditional static media forms. Or even if based on static media, they may break social norms about what is an “acceptable” argument. Whatever the reason, they may be difficult to communicate using traditional media. And so they remain private, or are only discussed informally with expert colleagues.’

If we can create interfaces that reify deep principles, then ‘mastering the subject begins to coincide with mastering the interface.’ He gives the example of Photoshop which builds in many deep principles of image manipulation.

‘As you master interface elements such as layers, the clone stamp, and brushes, you’re well along the way to becoming an expert in image manipulation… By contrast, the interface to Microsoft Word contains few deep principles about writing, and as a result it is possible to master Word‘s interface without becoming a passable writer. This isn’t so much a criticism of Word, as it is a reflection of the fact that we have relatively few really strong and precise ideas about how to write well.’

He then describes what he calls ‘the cognitive outsourcing model’: ‘we specify a problem, send it to our device, which solves the problem, perhaps in a way we-the-user don’t understand, and sends back a solution.’ E.g we ask Google a question and Google sends us an answer.

This is how most of us think about the idea of augmenting the human intellect but it is not the best approach. ‘Rather than just solving problems expressed in terms we already understand, the goal is to change the thoughts we can think.’

‘One challenge in such work is that the outcomes are so difficult to imagine. What new elements of cognition can we invent? How will they affect the way human beings think? We cannot know until they’ve been invented.

‘As an analogy, compare today’s attempts to go to Mars with the exploration of the oceans during the great age of discovery. These appear similar, but while going to Mars is a specific, concrete goal, the seafarers of the 15th through 18th centuries didn’t know what they would find. They set out in flimsy boats, with vague plans, hoping to find something worth the risks. In that sense, it was even more difficult than today’s attempts on Mars.

‘Something similar is going on with intelligence augmentation. There are many worthwhile goals in technology, with very specific ends in mind. Things like artificial intelligence and life extension are solid, concrete goals. By contrast, new elements of cognition are harder to imagine, and seem vague by comparison. By definition, they’re ways of thinking which haven’t yet been invented. There’s no omniscient problem-solving box or life-extension pill to imagine. We cannot say a priori what new elements of cognition will look like, or what they will bring. But what we can do is ask good questions, and explore boldly.

In another essay, Using Artificial Intelligence to Augment Human Intelligence, Nielsen points out that breakthroughs in creating powerful new cognitive technologies such as musical notation or Descartes’ invention of algebraic geometry are rare but ‘modern computers are a meta-medium enabling the rapid invention of many new cognitive technologies‘ and, further, AI will help us ‘invent new cognitive technologies which transform the way we think.’

Further, historically powerful new cognitive technologies, such as ‘Feynman diagrams’, have often appeared strange at first. We should not assume that new interfaces should be ‘user friendly’. Powerful interfaces that repay mastery may require sacrifices.

‘The purpose of the best interfaces isn’t to be user-friendly in some shallow sense. It’s to be user-friendly in a much stronger sense, reifying deep principles about the world, making them the working conditions in which users live and create. At that point what once appeared strange can instead becomes comfortable and familiar, part of the pattern of thought…

‘Unfortunately, many in the AI community greatly underestimate the depth of interface design, often regarding it as a simple problem, mostly about making things pretty or easy-to-use. In this view, interface design is a problem to be handed off to others, while the hard work is to train some machine learning system.

‘This view is incorrect. At its deepest, interface design means developing the fundamental primitives human beings think and create with. This is a problem whose intellectual genesis goes back to the inventors of the alphabet, of cartography, and of musical notation, as well as modern giants such as Descartes, Playfair, Feynman, Engelbart, and Kay. It is one of the hardest, most important and most fundamental problems humanity grapples with.

‘As discussed earlier, in one common view of AI our computers will continue to get better at solving problems, but human beings will remain largely unchanged. In a second common view, human beings will be modified at the hardware level, perhaps directly through neural interfaces, or indirectly through whole brain emulation.

‘We’ve described a third view, in which AIs actually change humanity, helping us invent new cognitive technologies, which expand the range of human thought. Perhaps one day those cognitive technologies will, in turn, speed up the development of AI, in a virtuous feedback cycle:

‘It would not be a Singularity in machines. Rather, it would be a Singularity in humanity’s range of thought… The long-term test of success will be the development of tools which are widely used by creators. Are artists using these tools to develop remarkable new styles? Are scientists in other fields using them to develop understanding in ways not otherwise possible?’

I would add: are governments using these tools to help them think in ways we already know are more powerful and to explore new ways of making decisions and shaping the complex systems on which we rely?

Nielsen also wrote this fascinating essay ‘Augmenting long-term memory’. This involves a computer tool (Anki) to aid long-term memory using ‘spaced repetition’ — i.e testing yourself at intervals which is shown to counter the normal (for most people) process of forgetting. This allows humans to turn memory into a choice so we can decide what to remember and achieve it systematically (without a ‘weird/extreme gift’ which is how memory is normally treated). (It’s fascinating that educated Greeks 2,500 years ago could build sophisticated mnemonic systems allowing them to remember vast amounts while almost all educated people now have no idea about such techniques.)

Connected to this, Nielsen also recently wrote an essay teaching fundamentals of quantum mechanics and quantum computers — but it is an essay with a twist:

‘[It] incorporates new user interface ideas to help you remember what you read… this essay isn’t just a conventional essay, it’s also a new medium, a mnemonic medium which integrates spaced-repetition testing. The medium itself makes memory a choice… This essay will likely take you an hour or two to read. In a conventional essay, you’d forget most of what you learned over the next few weeks, perhaps retaining a handful of ideas. But with spaced-repetition testing built into the medium, a small additional commitment of time means you will remember all the core material of the essay. Doing this won’t be difficult, it will be easier than the initial read. Furthermore, you’ll be able to read other material which builds on these ideas; it will open up an entire world…

‘Mastering new subjects requires internalizing the basic terminology and ideas of the subject. The mnemonic medium should radically speed up this memory step, converting it from a challenging obstruction into a routine step. Frankly, I believe it would accelerate human progress if all the deepest ideas of our civilization were available in a form like this.’

This obviously has very important implications for education policy. It also shows how computers could be used to improve learning — something that has generally been a failure since the great hopes at PARC in the 1970s. I have used Anki since reading Nielsen’s blog and I can feel it making a big difference to my mind/thoughts — how often is this true of things you read? DOWNLOAD ANKI NOW AND USE IT!

We need similarly creative experiments with new mediums that are designed to improve  standards of high stakes decision-making.

*

4. Summary

We could create systems for those making decisions about m/billions of lives and b/trillions of dollars, such as Downing Street or The White House, that integrate inter alia:

• Cognitive toolkits compressing already existing useful knowledge such as checklists for rational thinking developed by the likes of Tetlock, Munger, Yudkowsky et al.
• A Nielsen/Victor research program on ‘Seeing Rooms’, interface design, authoring tools, and cognitive technologies. Start with bunging a few million to Victor immediately in return for allowing some people to study what he is doing and apply it in Whitehall, then grow from there.
• An alpha data science/AI operation — tapping into the world’s best minds including having someone like David Deutsch or Tim Gowers as a sort of ‘chief rationalist’ in the Cabinet (with Scott Alexander as deputy!) — to support rational decision-making where this is possible and explain when it is not possible (just as useful).
• Tetlock/Hanson prediction tournaments could easily and cheaply be extended to consider ‘clusters’ of issues around themes like Brexit to improve policy and project management.
• Groves/Mueller style ‘systems management’ integrated with the data science team.
• Legally entrenched Red Teams where incentives are aligned to overcoming groupthink and error-correction of the most powerful. Warren Buffett points out that public companies considering an acquisition should employ a Red Team whose fees are dependent on the deal NOT going ahead. This is the sort of idea we need in No10.

Researchers could see the real operating environment of decision-makers at the apex of power, the sort of problems they need to solve under pressure, and the constraints of existing centralised systems. They could start with the safe level of ‘tools that we already know work really well’ — i.e things like cognitive toolkits and Red Teams — while experimenting with new tools and new ways of thinking.

Hedge funds like Bridgewater and some other interesting organisations think about such ideas though without the sophistication of Victor’s approach. The world of MPs, officials, the Institute for Government (a cheerleader for ‘carry on failing’), and pundits will not engage with these ideas if left to their own devices.

This is not the place to go into how to change this. We know that the normal approach is doomed to produce the normal results and normal results applied to things like repeated WMD crises means disaster sooner or later. As Buffett points out, ‘If there is only one chance in thirty of an event occurring in a given year, the likelihood of it occurring at least once in a century is 96.6%.’ It is not necessary to hope in order to persevere: optimism of the will, pessimism of the intellect…

*

A final thought…

A very interesting comment that I have heard from some of the most important scientists involved in the creation of advanced technologies is that ‘artists see things first’ — that is, artists glimpse possibilities before most technologists and long before most businessmen and politicians.

Pixar came from a weird combination of George Lucas getting divorced and the visionary Alan Kay suggesting to Steve Jobs that he buy a tiny special effects unit from Lucas, which Jobs did with completely wrong expectations about what would happen. For unexpected reasons this tiny unit turned into a huge success — as Jobs put it later, he was ‘sort of snookered’ into creating Pixar. Now Alan Kay says he struggles to get tech billionaires to understand the importance of Victor’s ideas.

The same story repeats: genuinely new ideas that could create huge value always seem so odd that almost all people in almost all organisations cannot see new possibilities. If this is true in Silicon Valley, how much more true is it in Whitehall or Washington… 

If one were setting up a new party in Britain, one could incorporate some of these ideas. This would of course also require recruiting very different types of people to the norm in politics. The closed nature of Westminster/Whitehall combined with first-past-the-post means it is very hard to solve the coordination problem of how to break into this system with a new way of doing things. Even those interested in principle don’t want to commit to a 10-year (?) project that might get them blasted on the front pages. Vote Leave hacked the referendum but such opportunities are much rarer than VC-funded ‘unicorns’. On the other hand, arguably what is happening now is a once in 50 or 100 year crisis and such crises also are the waves that can be ridden to change things normally unchangeable. A second referendum in 2020 is quite possible (or two referendums under PM Corbyn, propped up by the SNP?) and might be the ideal launchpad for a completely new sort of entity, not least because if it happens the Conservative Party may well not exist in any meaningful sense (whether there is or isn’t another referendum). It’s very hard to create a wave and it’s much easier to ride one. It’s more likely in a few years you will see some of the above ideas in novels or movies or video games than in government — their pickup in places like hedge funds and intelligence services will be discrete — but you never know…

*

Ps. While I have talked to Michael Nielsen and Bret Victor about their ideas, in no way should this blog be taken as their involvement in anything to do with my ideas or plans or agreement with anything written above. I did not show this to them or even tell them I was writing about their work, we do not work together in any way, I have just read and listened to their work over a few years and thought about how their ideas could improve government.

Further Reading

If interested in how to make things work much better, read this (lessons for government from the Apollo project) and this (lessons for government from ARPA-PARC’s creation of the internet and PC).

Links to recent reports on AI/ML.

What You Should Know About Megaprojects and Why, Flyvbjerg, 2014.

On the referendum #30: Genetics, genomics, predictions & ‘the Gretzky game’ — a chance for Britain to help the world

Britain could contribute huge value to the world by leveraging existing assets, including scientific talent and how the NHS is structured, to push the frontiers of a rapidly evolving scientific field — genomic prediction — that is revolutionising healthcare in ways that give Britain some natural advantages over Europe and America. We should plan for free universal ‘SNP’ genetic sequencing as part of a shift to genuinely preventive medicine — a shift that will lessen suffering, save money, help British advanced technology companies in genomics and data science/AI, make Britain more attractive for scientists and global investment, and extend human knowledge in a crucial field to the benefit of the whole world.

‘SNP’ sequencing means, crudely, looking at the million or so most informative markers or genetic variants without sequencing every base pair in the genome. SNP sequencing costs ~$50 per person (less at scale), whole genome sequencing costs ~$1,000 per person (less at scale). The former captures most of the predictive power now possible at 1/20th of the cost of the latter.

*

Background: what seemed ‘sci fi’ ~2010-13 is now reality

In my 2013 essay on education and politics, I summarised the view of expert scientists on genetics (HERE between pages 49-51, 72-74, 194-203). Although this was only a small part of the essay most of the media coverage focused on this, particularly controversies about IQ.

Regardless of political affiliation most of the policy/media world, as a subset of ‘the educated classes’ in general, tended to hold a broadly ‘blank slate’ view of the world mostly uninformed by decades of scientific progress. Technical terms like ‘heritability’, which refers to the variance in populations, caused a lot of confusion.

When my essay hit the media, fortunately for me the world’s leading expert, Robert Plomin, told hacks that I had summarised the state of the science accurately. (I never tried to ‘give my views on the science’ as I don’t have ‘views’ — all people like me can try to do with science is summarise the state of knowledge in good faith.) Quite a lot of hacks then spent some time talking to Plomin and some even wrote about how they came to realise that their assumptions about the science had been wrong (e.g Gaby Hinsliff).

Many findings are counterintuitive to say the least. Almost everybody naturally thinks that ‘the shared environment’ in the form of parental influence ‘obviously’ has a big impact on things like cognitive development. The science says this intuition is false. The shared environment is much less important than we assume and has very little measurable effect on cognitive development: e.g an adopted child who does an IQ test in middle age will show on average almost no correlation with the parents who brought them up (genes become more influential as you age). People in the political world assumed a story of causation in which, crudely, wealthy people buy better education and this translates into better exam and IQ scores. The science says this story is false. Environmental effects on things like cognitive ability and education achievement are almost all from what is known as the ‘non-shared environment’ which has proved very hard to pin down (environmental effects that differ for children, like random exposure to chemicals in utero). Further, ‘The case for substantial genetic influence on g [g = general intelligence ≈ IQ] is stronger than for any other human characteristic’ (Plomin) and g/IQ has far more predictive power for future education than class does. All this has been known for years, sometimes decades, by expert scientists but is so contrary to what well-educated people want to believe that it was hardly known at all in ‘educated’ circles that make and report on policy.

Another big problem is that widespread ignorance about genetics extends to social scientists/economists, who are much more influential in politics/government than physical scientists. A useful heuristic is to throw ~100% of what you read from social scientists about ‘social mobility’ in the bin. Report after report repeats the same clichés, repeats factual errors about genetics, and is turned into talking points for MPs as justification for pet projects. ‘Kids who can read well come from homes with lots of books so let’s give families with kids struggling to read more books’ is the sort of argument you read in such reports without any mention of the truth: children and parents share genes that make them good at and enjoy reading, so causation is operating completely differently to the assumptions. It is hard to overstate the extent of this problem. (There are things we can do about ‘social mobility’, my point is Insider debate is awful.)

A related issue is that really understanding the science requires serious understanding of statistics and, now, AI/machine learning (ML). Many social scientists do not have this training. This problem will get worse as data science/AI invades the field. 

A good example is ‘early years’ and James Heckman. The political world is obsessed with ‘early years’ such as Sure Start (UK) and Head Start (US). Politicians latch onto any ‘studies’ that seem to justify it and few have any idea about the shocking state of the studies usually quoted to justify spending decisions. Heckman has published many papers on early years and they are understandably widely quoted by politicians and the media. Heckman is a ‘Nobel Prize’ winner in economics. One of the world’s leading applied mathematicians, Professor Andrew Gelman, has explained how Heckman has repeatedly made statistical errors in his papers but does not correct them: cf. How does a Nobel-prize-winning economist become a victim of bog-standard selection bias?  This really shows the scale of the problem: if a Nobel-winning economist makes ‘bog standard’ statistical errors that confuse him about studies on pre-school, what chance do the rest of us in the political/media world have?

Consider further that genomics now sometimes applies very advanced mathematical ideas such as ‘compressed sensing’. Inevitably few social scientists can judge such papers but they are overwhelmingly responsible for interpreting such things for ministers and senior officials. This is compounded by the dominance of social scientists in Whitehall units responsible for data and evidence. Many of these units are unable to provide proper scientific advice to ministers (I have had personal experience of this in the Department for Education). Two excellent articles by Duncan Watts recently explained fundamental problems with social science and what could be done (e.g a much greater focus on successful prediction) but as far as I can tell they have had no impact on economists and sociologists who do not want to face their lack of credibility and whose incentives in many ways push them towards continued failure (Nature paper HERE, Science paper HERE — NB. the Department for Education did not even subscribe to the world’s leading science journals until I insisted in 2011).

1) The problem that the evidence for early years is not what ministers and officials think it is is not a reason to stop funding but I won’t go into this now. 2) This problem is incontrovertible evidence, I think, of the value of an alpha data science unit in Downing Street, able to plug into the best researchers around the world, and ensure that policy decisions are taken on the basis of rational thinking and good science or, just as important, everybody is aware that they have to make decisions in the absence of this. This unit would pay for itself in weeks by identifying flawed reasoning and stopping bad projects, gimmicks etc. Of course, this idea has no chance with those now at the top of Government and the Cabinet Office would crush such a unit as it would threaten the traditional hierarchy. One of the  arguments I made in my essay was that we should try to discover useful and reliable benchmarks for what children of different abilities are really capable of learning and build on things like the landmark Study of Mathematically Precocious Youth. This obvious idea is anathema to the education policy world where there is almost no interest in things like SMPY and almost everybody supports the terrible idea that ‘all children must do the same exams’ (guaranteeing misery for some and boredom/time wasting for others). NB. Most rigorous large-scale educational RCTs are uninformative. Education research, like psychology, produces a lot of what Feynman called ‘cargo cult science’.

Since 2013, genomics has moved fast and understanding in the UK media has changed probably faster in five years than over the previous 35 years. As with the complexities of Brexit, journalists have caught up with reality much better than MPs. It’s still true that almost everything written by MPs about ‘social mobility’ is junk but you could see from the reviews of Plomin’s recent book, Blueprint, that many journalists have a much better sense of the science than they did in 2013. Rare good news, though much more progress is needed…

*

What’s happening now?

In 2013 it was already the case that the numbers on heritability derived from twin and adoption studies were being confirmed by direct inspection of DNA — therefore many of the arguments about twin/adoption studies were redundant — but this fact was hardly known.

I pointed out that the field would change fast. Both Plomin and another expert, Steve Hsu, made many predictions around 2010-13 some of which I referred to in my 2013 essay. Hsu is a physics professor who is also one of the world’s leading researchers on genomics. 

Hsu predicted that very large samples of DNA would allow scientists over the next few years to start identifying the actual genes responsible for complex traits, such as diseases and intelligence, and make meaningful predictions about the fate of individuals. Hsu gave estimates of the sample sizes that would be needed. His 2011 talk contains some of these predictions and also provides a physicist’s explanation of ‘what is IQ measuring’. As he said at Google in 2011, the technology is ‘right on the cusp of being able to answer fundamental questions’ and ‘if in ten years we all meet again in this room there’s a very good chance that some of the key questions we’ll know the answers to’. His 2014 paper explains the science in detail. If you spend a little time looking at this, you will know more than 99% of high status economists gabbling on TV about ‘social mobility’ saying things like ‘doing well on IQ tests just proves you can do IQ tests’.

In 2013, the world of Westminster thought this all sounded like science fiction and many MP said I sounded like ‘a mad scientist’. Hsu’s predictions have come true and just five years later this is no longer ‘science fiction’. (Also NB. Hsu’s blog was one of the very few places where you would have seen discussion of CDOs and the 2008 financial crash long BEFORE it happened. I have followed his blog since ~2004 and this from 2005, two years before the crash started, was the first time I read about things like ‘synthetic CDOs’: ‘we have yet another ill-understood casino running, with trillions of dollars in play’. The quant-physics network had much better insight into the dynamics behind the 2008 Crash than high status mainstream economists like Larry Summers responsible for regulation.)

His group and others have applied machine learning to very large genetic samples and built predictors of complex traits. Complex traits like general intelligence and most diseases are ‘polygenic’ — they depend on many genes each of which contributes a little (unlike diseases caused by a single gene). 

‘There are now ~20 disease conditions for which we can identify, e.g, the top 1% outliers with 5-10x normal risk for the disease. The papers reporting these results have almost all appeared within the last year or so.’

For example, the height predictor ‘captures nearly all of the predicted SNP heritability for this trait — actual heights of most individuals in validation tests are within a few cm of predicted heights.’ Height is similar to IQ — polygenic and similar heritability estimates.

These predictors have been validated with out-of-sample tests. They will get better and better as more and more data is gathered about more and more traits. 

This enables us to take DNA from unborn embryos, do SNP genetic sequencing costing ~$50, and make useful predictions about the odds of the embryo being an outlier for diseases like atrial fibrillation, diabetes, breast cancer, or prostate cancer. NB. It is important that we do not need to sequence the whole genome to do this (see below). We will also be able to make predictions about outliers in cognitive abilities (the high and low ends). (My impression is that predicting Alzheimers is still hampered by a lack of data but this will improve as the data improves.)

There are many big implications. This will obviously revolutionise IVF. ~1 million IVF embryos per year are screened worldwide using less sophisticated tests. Instead of picking embryos at random, parents will start avoiding outliers for disease risks and cognitive problems. Rich people will fly to jurisdictions offering the best services.

Forensics is being revolutionised. First, DNA samples can be used to give useful physical descriptions of suspects because you can identify ethnic group, height, hair colour etc. Second, ‘cold cases’ are now routinely being solved because if a DNA sample exists, then the police can search for cousins of the perpetrator from public DNA databases, then use the cousins to identify suspects. Every month or so now in America a cold case murder is solved and many serial killers are being found using this approach — just this morning I saw what looks to be another example just announced, a murder of an 11 year-old in 1973. (Some companies are resisting this development but they will, I am very confident, be smashed in court and have their reputations trashed unless they change policy fast. The public will have no sympathy for those who stand in the way.)

Hsu recently attended a conference in the UK where he presented some of these ideas to UK policy makers. He wrote this blog about the great advantages the NHS has in developing this science. 

‘The UK could become the world leader in genomic research by combining population-level genotyping with NHS health records… The US private health insurance system produces the wrong incentives for this kind of innovation: payers are reluctant to fund prevention or early treatment because it is unclear who will capture the ROI [return on investment]… The NHS has the right incentives, the necessary scale, and access to a deep pool of scientific talent. The UK can lead the world into a new era of precision genomic medicine. 

‘NHS has already announced an out-of-pocket genotyping service which allows individuals to pay for their own genotyping and to contribute their health + DNA data to scientific research. In recent years NHS has built an impressive infrastructure for whole genome sequencing (cost ~$1k per individual) that is used to treat cancer and diagnose rare genetic diseases. The NHS subsidiary Genomics England recently announced they had reached the milestone of 100k whole genomes…

‘At the meeting, I emphasized the following:

1. NHS should offer both inexpensive (~$50) genotyping (sufficient for risk prediction of common diseases) along with the more expensive $1k whole genome sequencing. This will alleviate some of the negative reaction concerning a “two-tier” NHS, as many more people can afford the former.

2. An in-depth analysis of cost-benefit for population wide inexpensive genotyping would likely show a large net cost savings: the risk predictors are good enough already to guide early interventions that save lives and money. Recognition of this net benefit would allow NHS to replace the $50 out-of-pocket cost with free standard of care.’ (Emphasis added)

NB. In terms of the short-term practicalities it is important that whole genome sequencing costs ~$1,000 (and falling) but is not necessary: a version 1/20th of the cost, looking just at the most informative genetic variants, captures most of the predictive benefits. Some have incentives to distort this, such as companies like Illumina trying to sell expensive machines for whole genome sequencing, which can distort policy — let’s hope officials are watching carefully. These costs will, obviously, keep falling.

This connects to an interesting question… Why was the likely trend in genomics clear ~2010 to Plomin, Hsu and others but invisible to most? Obviously this involves lots of elements of expertise and feel for the field but also they identified FAVOURABLE EXPONENTIALS. Here is the fall in the cost of sequencing a genome compared to Moore’s Law, another famous exponential. The drop over ~18 years has been a factor of ~100,000. Hsu and Plomin could extrapolate that over a decade and figure out what would be possible when combined with other trends they could see. Researchers are already exploring what will be possible as this trend continues.

Identifying favourable exponentials is extremely powerful. Back in the early 1970s, the greatest team of computer science researchers ever assembled (PARC) looked out into the future and tried to imagine what could be possible if they brought that future back to the present and built it. They were trying to ‘compute in the future’. They created personal computing. (Chart by Alan Kay, one of the key researchers — he called it ‘the Gretzky game’ because of Gretzky’s famous line ‘I skate to where the puck is going to be, not where it has been.’ The computer is the Alto, the first personal computer that stunned Steve Jobs when he saw a demo. The sketch on the right is of children using a tablet device that Kay drew decades before the iPad was launched.)

Hopefully the NHS and Department for Health will play ‘the Gretzky game’, take expert advice from the likes of Plomin and Hsu and take this opportunity to make the UK a world leader in one of the most important frontiers in science.

• We can imagine everybody in the UK being given valuable information about their health for free, truly preventive medicine where we target resources at those most at risk, and early (even in utero) identification of risks.
• This would help bootstrap British science into a stronger position with greater resources to study things like CRISPR and the next phase of this revolution — editing genes to fix problems, where clinical trials are already showing success.
• It would also give a boost to British AI/data science companies — the laws, rules on data etc should be carefully shaped to ensure that British companies (not Silicon Valley or China) capture most of the financial value (though everybody will gain from the basic science).
• These gains would have positive feedback effects on each other, just as investment in basic AI/ML research will have positive feedback effects in many industries.
• I have argued many times for the creation of a civilian UK ‘ARPA’ — a centre for high-risk-high-payoff research that has been consistently blocked in Whitehall (see HERE for an account of how ARPA-PARC created the internet and personal computing). This fits naturally with Britain seeking to lead in genomics/AI. Thinking about this is part of a desperately needed overall investigation into the productivity of the British economy and the ecosystem of universities, basic science, venture capital, startups, regulation (data, intellectual property etc) and so on.

There will also be many controversies and problems. The ability to edit genomes — and even edit the germline with ‘gene drives’ so all descendants have the same copy of the gene — is a Promethean power implying extreme responsibilities. On a mundane level, embracing new technology is clearly hard for the NHS with its data infrastructure. Almost everyone I speak to using the NHS has had similar problems that I have had — nightmares with GPs, hospitals, consultants et al being able to share data and records, things going missing, etc. The NHS will be crippled if it can’t fix this, but this is another reason to integrate data science as a core ‘utility’ for the NHS.

On a political note…

Few scientists and even fewer in the tech world are aware of the EU’s legal framework for regulating technology and the implications of the recent Charter of Fundamental Rights (the EU’s Charter, NOT the ECHR) which gives the Commission/ECJ the power to regulate any advanced technology, accelerate the EU’s irrelevance, and incentivise investors to invest outside the EU. In many areas, the EU regulates to help the worst sort of giant corporate looters defending their position against entrepreneurs. Post-Brexit Britain will be outside this jurisdiction and able to make faster and better decisions about regulating technology like genomics, AI and robotics. Prediction: just as Insiders now talk of how we ‘dodged a bullet’ in staying out of the euro, within ~10 years Insiders will talk about being outside the Charter/ECJ and the EU’s regulation of data/AI in similar terms (assuming Brexit happens and UK politicians even try to do something other than copy the EU’s rules).

China is pushing very hard on genomics/AI and regards such fields as crucial strategic ground for its struggle for supremacy with America. America has political and regulatory barriers holding it back on genomics that are much weaker here. Britain cannot stop the development of such science. Britain can choose to be a backwater, to ignore such things and listen to MPs telling fairy stories while the Chinese plough ahead, or it can try to lead. But there is no hiding from the truth and ‘for progress there is no cure’ (von Neumann). We will never be the most important manufacturing nation again but we could lead in crucial sub-fields of advanced technology. As ARPA-PARC showed, tiny investments can create entire new industries and trillions of dollars of value.

Sadly most politicians of Left and Right have little interest in science funding with tremendous implications for future growth, or the broader question of productivity and the ecosystem of science, entrepreneurs, universities, funding, regulation etc, and we desperately need institutions that incentivise politicians and senior officials to ‘play the Gretzky game’. The next few months will be dominated by Brexit and, hopefully, the replacement of the May/Hammond government. Those thinking about the post-May landscape and trying to figure out how to navigate in uncharted and turbulent waters should focus on one of the great lessons of politics that is weirdly hard for many MPs to internalise: the public rewards sustained focus on their priorities!

One of the lessons of the 2016 referendum (that many Conservative MPs remain desperate not to face) is the political significance of the NHS. The concept described above is one of those concepts in politics that maximises positive futures for the force that adopts it because it draws on multiple sources of strength. It combines, inter alia, all the political benefits of focus on the NHS, helping domestic technology companies, incentivising global investment, doing something that shows the world that Britain is (contra the May/Hammond outlook) open to science and high skilled immigrants, it is based on intrinsic advantages that Europe and America will find hard to overcome over a decade, it supplies (NB. MPs/spads) a never-ending string of heart-wrenching good news stories, and, very rarely in SW1, those pushing it would be seen as leading something of global importance. It will, therefore, obviously be rejected by a section of Conservative MPs who much prefer to live in a parallel world, who hate anything to do with science and who are ignorant about how new industries and wealth are really created. But for anybody trying to orient themselves to reality, connect themselves to sources of power, and thinking ‘how on earth could we clamber out of this horror show’, it is an obvious home run…

NB. It ought to go without saying that turning this idea into a political/government success requires focus on A) the NHS, health, science, NOT getting sidetracked into B) arguments about things like IQ and social mobility. Over time, the educated classes will continue to be dragged to more realistic views on (B) but this will be a complex process entangled with many hysterical episodes. (A) requires ruthless focus…

Please leave comments, fix errors below. I have not shown this blog in draft to Plomin or Hsu who obviously are not responsible for my errors.

Further reading

Plomin’s excellent new book, Blueprint. I would encourage journalists who want to understand this subject to speak to Plomin who works in London and is able to explain complex technical subjects to very confused arts graduates like me.

On the genetic architecture of intelligence and other quantitative traits, Hsu 2014.

Cf. this thread by researcher Paul Pharaoh on breast cancer.

Hsu blogs on genomics.

Some recent developments with AI/ML, links to papers.

On how ARPA-PARC created the modern computer industry and lessons for high-risk-high-payoff science research.

My 2013 essay.

#29 On the referendum & #4c on Expertise: On the ARPA/PARC ‘Dream Machine’, science funding, high performance, and UK national strategy

Post-Brexit Britain should be considering the intersection of 1) ARPA/PARC-style science research and ‘systems management’ for managing complex projects with 2) the reform of government institutions so that high performance teams — with different education/training (‘Tetlock processes’) and tools (including data science and visualisations of interactive models of complex systems) — can make ‘better decisions in a complex world’.  

This paper examines the ARPA/PARC vision for computing and the nature of the two organisations. In the 1960s visionaries such as Joseph Licklider, Robert Taylor and Doug Engelbart developed a vision of networked interactive computing that provided the foundation not just for new technologies but for whole new industries. Licklider, Sutherland, Taylor et al provided a model (ARPA) for how science funding can work. Taylor provided a model (PARC) of how to manage a team of extremely talented people who turned a profound vision into reality. The original motivation for the vision of networked interactive computing was to help humans make good decisions in a complex world.

This story suggests ideas about how to make big improvements in the world with very few resources if they are structured right. From a British perspective it also suggests ideas about what post-Brexit Britain should do to help itself and the world and how it might be possible to force some sort of ‘phase transition’ on the rotten Westminster/Whitehall system.

For the PDF of the paper click HERE. Please correct errors with page numbers below. I will update it after feedback.

Further Reading

The Dream Machine.

Dealers of Lightning.

‘Sketchpad: A man-machine graphical communication system’, Ivan Sutherland 1963.

Oral history interview with Sutherland, head of ARPA’s IPTO division 1963-5.

This link has these seminal papers:

• Man-Computer Symbiosis, Licklider (1960)
• The computer as a communications device, Licklider & Taylor (1968)

Watch Alan Kay explain how to invent the future to YCombinator classes HERE and HERE.  

HERE for Kay quotes from emails with Bret Victor.

HERE for Kay’s paper on PARC, The Power of the Context.

Kay’s Early History of Smalltalk.

HERE for a conversation between Kay and Engelbart.

Alan Kay’s tribute to Ted Nelson at “Intertwingled” Fest (an Alto using Smalltalk).

Personal Distributed Computing: The Alto and Ethernet Software1, Butler Lampson. 

You and Your Research, Richard Hamming.

AI nationalism, essay by Ian Hogarth. This concerns implications of AI for geopolitics.

Drones go to work, Chris Anderson (one of the pioneers of commercial drones). This explains the economics of the drone industry.

Meditations on Moloch, Scott Alexander. This is an extremely good essay in general about deep problems with our institutions.

Intelligence Explosion Microeconomics, Yudkowsky.

Autonomous technology and the greater human good. Omohundro.

Can intelligence explode? Hutter.

For the issue of IQ, genetics and the distribution of talent (and much much more), cf. Steve Hsu’s brilliant blog.

Bret Victor.

Michael Nielsen.

For some pre-history on computers, cf. The birth of computational thinking (some of the history of computing devices before the Turing/von Neumann revolution) and The crisis of mathematical paradoxes, Gödel, Turing and the basis of computing (some of the history of ideas about mathematical foundations and logic such as the famous papers by Gödel and Turing in the 1930s)

Part I of this series of blogs is HERE.

Part II on the emergence of ‘systems management’, how George Mueller used it to put man on the moon, and a checklist of how successful management of complex projects is systematically different to how Whitehall works is HERE.

Effective action #4a: ‘Expertise’ from fighting and physics to economics, politics and government

‘We learn most when we have the most to lose.’ Michael Nielsen, author of the brilliant book Reinventing Discovery.

‘There isn’t one novel thought in all of how Berkshire [Hathaway] is run. It’s all about … exploiting unrecognized simplicities…Warren [Buffett] and I aren’t prodigies.We can’t play chess blindfolded or be concert pianists. But the results are prodigious, because we have a temperamental advantage that more than compensates for a lack of IQ points.’ Charlie Munger,Warren Buffett’s partner.

I’m going to do a series of blogs on the differences between fields dominated by real expertise (like fighting and physics) and fields dominated by bogus expertise (like macroeconomic forecasting, politics/punditry, active fund management).

Fundamental to real expertise is 1) whether the informational structure of the environment is sufficiently regular that it’s possible to make good predictions and 2) does it allow high quality feedback and therefore error-correction. Physics and fighting: Yes. Predicting recessions, forex trading and politics: not so much. I’ll look at studies comparing expert performance in different fields and the superior performance of relatively very simple models over human experts in many fields.

This is useful background to consider a question I spend a lot of time thinking about: how to integrate a) ancient insights and modern case studies about high performance with b) new technology and tools in order to improve the quality of individual, team, and institutional decision-making in politics and government.

I think that fixing the deepest problems of politics and government requires a more general and abstract approach to principles of effective action than is usually considered in political discussion and such an approach could see solutions to specific problems almost magically appear, just as you see happen in a very small number of organisations — e.g Mueller’s Apollo program (man on the moon), PARC (interactive computing), Berkshire Hathaway (most successful investors in history), all of which have delivered what seems almost magical performance because they embody a few simple, powerful, but largely unrecognised principles. There is no ‘solution’ to the fundamental human problem of decision-making amid extreme complexity and uncertainty but we know a) there are ways to do things much better and b) governments mostly ignore them, so there is extremely valuable low-hanging fruit if, but it’s a big if, we can partially overcome the huge meta-problem that governments tend to resist the institutional changes needed to become a learning system.

This blog presents some basic background ideas and examples…

*

Extreme sports: fast feedback = real expertise 

In the 1980s and early 1990s, there was an interesting case study in how useful new knowledge jumped from a tiny isolated group to the general population with big effects on performance in a community. Expertise in Brazilian jiu-jitsu was taken from Brazil to southern California by the Gracie family. There were many sceptics but they vanished rapidly because the Gracies were empiricists. They issued ‘the Gracie challenge’.

All sorts of tough guys, trained in all sorts of ways, were invited to come to their garage/academy in Los Angeles to fight one of the Gracies or their trainees. Very quickly it became obvious that the Gracie training system was revolutionary and they were real experts because they always won. There was very fast and clear feedback on predictions. Gracie jiujitsu quickly jumped from an LA garage to TV. At the televised UFC 1 event in 1993 Royce Gracie defeated everyone and a multi-billion dollar business was born.

People could see how training in this new skill could transform performance. Unarmed combat changed across the world. Disciplines other than jiu jitsu have had to make a choice: either isolate themselves and not compete with jiu jitsu or learn from it. If interested watch the first twenty minutes of this documentary (via professor Steve Hsu, physicist, amateur jiu jitsu practitioner, and predictive genomics expert).

Video: Jiu Jitsu comes to Southern California

Royce Gracie, UFC 1 1993 

 

Flow, deep in the zone

Another field where there is clear expertise is extreme skiing and snowboarding. One of the leading pioneers, Jeremy Jones, describes how he rides ‘spines’ hurtling down the side of mountains:

‘The snow is so deep you need to use your arms and chest to swim, and your legs to ride. They also collapse underfoot, so you’re riding mini-avalanches and dodging slough slides. Spines have blind rollovers, so you can’t see below. Or to the side. Every time the midline is crossed, it’s a leap into the abyss. Plus, there’s no way to stop and every move is amplified by complicated forces. A tiny hop can easily become a twenty-foot ollie. It’s the absolute edge of chaos. But the easiest way to live in the moment is to put yourself in a situation where there’s no other choice. Spines demand that, they hurl you deep into the zone.’ Emphasis added.

Video: Snowboarder Jeremy Jones

What Jones calls ‘the zone’ is also known as ‘flow‘ — a particular mental state, triggered by environmental cues, that brings greatly enhanced performance. It is the object of study in extreme sports and by the military and intelligence services: for example DARPA is researching whether stimulating the brain can trigger ‘flow’ in snipers.

Flow — or control on ‘the edge of chaos’ where ‘every move is amplified by complicated forces’ — comes from training in which people learn from very rapid feedback between predictions and reality. In ‘flow’, brains very rapidly and accurately process environmental signals and generate hypothetical scenarios/predictions and possible solutions based on experience and training. Jones’s performance is inseparable from developing this fingertip feeling. Similarly, an expert fireman feels the glow of heat on his face in a slightly odd way and runs out of the building just before it collapses without consciously knowing why he did it: his intuition has been trained to learn from feedback and make predictions. Experts operating in ‘flow’ do not follow what is sometimes called the ‘rational model’ of decision-making in which they sequentially interrogate different options — they pattern-match solutions extremely quickly based on experience and intuition.

The video below shows extreme expertise in a state of ‘flow’ with feedback on predictions within milliseconds. This legendary ride is so famous not because of the size of the wave but its odd, and dangerous, nature. If you watch carefully you will see what a true expert in ‘flow’ can do: after committing to the wave Hamilton suddenly realises that unless he reaches back with the opposite hand to normal and drags it against the wall of water behind him, he will get sucked up the wave and might die. (This wave had killed someone a few weeks earlier.) Years of practice and feedback honed the intuition that, when faced with a very dangerous and fast moving problem, almost instantly (few seconds maximum) pattern-matched an innovative solution.

Video: surfer Laird Hamilton in one of the greatest ever rides

 

The faster the feedback cycle, the more likely you are to develop a qualitative improvement in speed that destroys an opponent’s decision-making cycle. If you can reorient yourself faster to the ever-changing environment than your opponent, then you operate inside their ‘OODA loop’ (Observe-Orient-Decide-Act) and the opponent’s performance can quickly degrade and collapse.

This lesson is vital in politics. You can read it in Sun Tzu and see it with Alexander the Great. Everybody can read such lessons and most people will nod along. But it is very hard to apply because most political/government organisations are programmed by their incentives to prioritise seniority, process and prestige over high performance and this slows and degrades decisions. Most organisations don’t do it. Further, political organisations tend to make too slowly those decisions that should be fast and too quickly those decisions that should be slow — they are simultaneously both too sluggish and too impetuous, which closes off favourable branching histories of the future.

Video: Boxer Floyd Mayweather, best fighter of his generation and one of the quickest and best defensive fighters ever

The most extreme example in extreme sports is probably ‘free soloing’ — climbing mountains without ropes where one mistake means instant death. If you want to see an example of genuine expertise and the value of fast feedback then watch Alex Honnold.

Video: Alex Honnold ‘free solos’ El Sendero Luminoso (terrifying)

Music is similar to sport. There is very fast feedback, learning, and a clear hierarchy of expertise.

Video: Glenn Gould playing the Goldberg Variations (slow version)

Our culture treats expertise/high performance in fields like sport and music very differently to maths/science education and politics/government. As Alan Kay observes, music and sport expertise is embedded in the broader culture. Millions of children spend large amounts of time practising hard skills. Attacks on them as ‘elitist’ don’t get the same damaging purchase as in other fields and the public don’t mind about elite selection for sports teams or orchestras.

‘Two ideas about this are that a) these [sport/music] are activities in which the basic act can be seen clearly from the first, and b) are already part of the larger culture. There are levels that can be seen to be inclusive starting with modest skills. I think a very large problem for the learning of both science and math is just how invisible are their processes, especially in schools.’ Kay 

When it comes to maths and science education, the powers-that-be (in America and Britain) try very hard and mostly successfully to ignore the question: where are critical thresholds for valuable skills that develop true expertise. This is even more a problem with the concept of ‘thinking rationally’, for which some basic logic, probability, and understanding of scientific reasoning is a foundation. Discussion of politics and government almost totally ignores the concept of training people to update their opinions in response to new evidence — i.e adapt to feedback. The ‘rationalist community’ — people like Scott Alexander who wrote this fantastic essay (Moloch) about why so much goes wrong, or the recent essays by Eliezer Yudkowsky — are ignored at the apex of power. I will return to the subject of how to create new education and training programmes for elite decision-makers. It is a good time for UK universities to innovate in this field, as places like Stanford are already doing. Instead of training people like Cameron and Adonis to bluff with PPE, we need courses that combine rational thinking with practical training in managing complex projects. We need people who practice really hard making predictions in ways we know work well (cf. Tetlock) then update in response to errors.

*

A more general/abstract approach to reforming government

If we want to get much higher performance in government, then we need to think rigorously about: the selection of people and teams, their education and training, their tools, and the institutions (incentives and so on) that surround and shape them.

Almost all analysis of politics and government considers relatively surface phenomena. For example, the media briefly blasts headlines about Carillion’s collapse or our comical aircraft carriers but there is almost no consideration of the deep reasons for such failures and therefore nothing tends to happen — the media caravan moves on and the officials and ministers keep failing in the same ways. This is why, for example, the predicted abject failure of the traditional Westminster machinery to cope with Brexit negotiations has not led to self-examination and learning but, instead, mostly to a visible determination across both sides of the Brexit divide in SW1 to double down on long-held delusions.

Progress requires attacking the ‘system of systems’ problem at the right ‘level’. Attacking the problems directly — let’s improve policy X and Y, let’s swap ‘incompetent’ A for ‘competent’ B — cannot touch the core problems, particularly the hardest meta-problem that government systems bitterly fight improvement. Solving the explicit surface problems of politics and government is best approached by a more general focus on applying abstract principles of effective action. We need to surround relatively specific problems with a more general approach. Attack at the right level will see specific solutions automatically ‘pop out’ of the system. One of the most powerful simplicities in all conflict (almost always unrecognised) is: ‘winning without fighting is the highest form of war’. If we approach the problem of government performance at the right level of generality then we have a chance to solve specific problems ‘without fighting’ — or, rather, without fighting nearly so much and the fighting will be more fruitful.

This is not a theoretical argument. If you look carefully at ancient texts and modern case studies, you see that applying a small number of very simple, powerful, but largely unrecognised principles (that are very hard for organisations to operationalise) can produce extremely surprising results.

We have no alternative to trying. Without fundamental changes to government, we will lose our hourly game of Russian roulette with technological progress.

‘The combination of physics and politics could render the surface of the earth uninhabitable… [T]he ever accelerating progress of technology and changes in the mode of human life … gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue.’ John von Neumann

As Steve Hsu says: Pessimism of the Intellect, Optimism of the Will.

---

Ps. There is an interesting connection between the nature of counterfactual reasoning in the fast-moving world of extreme sports and the theoretical paper I posted yesterday on state-of-the-art AI. The human ability to interrogate stored representations of their environment with counter-factual questions is fundamental to the nature of intelligence and developing expertise in physical and mental skills. It is, for now, absent in machines.

On the referendum #23, a year after victory: ‘a change of perspective is worth 80 IQ points’ & ‘how to capture the heavens’

‘Just like all British governments, they will act more or less in a hand to mouth way on the spur of the moment, but they will not think out and adopt a steady policy.’ Earl Cromer, 1896.

‘Fascinating that the same problems recur time after time, in almost every program, and that the management of the program, whether it happened to be government or industry, continues to avoid reality.’ George Mueller, pioneer of systems management and head of the Apollo programme to put man on the moon.

Traditional cultures, those that all humans lived in until quite recently and which still survive in pockets, don’t realise that they are living inside a particular perspective. They think that what they see is ‘reality’. It is, obviously, not their fault. It is not because they are stupid. It is a historical accident that they did/do not have access to mental models that help more accurate thinking about reality.

Westminster and the other political cultures dotted around the world are similar to these traditional cultures. They think they they are living in ‘reality’. The MPs and pundits get up, read each other, tweet at each other, give speeches, send press releases, have dinner, attack, fuck or fight each other, do the same tomorrow and think ‘this is reality’. Like traditional cultures they are wrong. They are living inside a particular perspective that enormously distorts reality. 

They are trapped in thinking about today and their careers. They are trapped in thinking about incremental improvements. Almost nobody has ever been part of a high performance team responsible for a complex project. The speciality is a hot take to explain post facto what one cannot predict. They mostly don’t know what they don’t know. They don’t understand the decentralised information processing that allows markets to enable complex coordination. They don’t understand how scientific research works and they don’t value it. Their daily activity is massively constrained by the party and state bureaucracies that incentivise behaviour very different to what humanity needs to create long-term value. As Michael Nielsen (author of Reinventing Science) writes:

‘[M]uch of our intellectual elite who think they have “the solutions” have actually cut themselves off from understanding the basis for much of the most important human progress.’

Unlike traditional cultures, our modern political cultures don’t have the excuse of our hunter-gatherer ancestors. We could do better. But it is very very hard to escape the core imperatives that make big bureaucracies — public companies as well as state bureaucracies — so bad at learning. Warren Buffet explained decades ago how institutions actively fight against learning and fight to stay in a closed and vicious feedback loop:

‘My most surprising discovery: the overwhelming importance in business of an unseen force that we might call “the institutional imperative”. In business school, I was given no hint of the imperative’s existence and I did not intuitively understand it when I entered the business world. I thought then that decent, intelligence, and experienced managers would automatically make rational business decisions. But I learned the hard way that isn’t so. Instead rationality frequently wilts when the institutional imperative comes into play.

‘For example, 1) As if governed by Newton’s First Law, any institution will resist any change in its current direction. 2) … Corporate projects will materialise to soak up available funds. 3) Any business craving of the leader, however foolish, will quickly be supported by … his troops. 4) The behaviour of peer companies … will be mindlessly imitated.’

Almost nobody really learns from the world’s most successful investor about investing and how to run a successful business with good corporate governance. (People read what he writes but almost no investors choose to operate long-term like him, I think it is still true that not a single public company has copied his innovations with corporate governance like ‘no pay for company directors’, and governments have consistently rejected his and Munger’s advice about controlling the looting of public companies by management.) Almost nobody really learns how to do things better from the experience of dealing with this ‘institutional imperative’. We fail over and over again in the same way, trusting in institutions that are programmed to fail.

It is very very hard for humans to lift our eyes from today and to go out into the future and think about what could be done to bring the future back to the present. Like ants crawling around on the leaf, we political people only know our leaf.

Science has shown us a different way. Newton looked up from his leaf, looked far away from today, and created a new perspective — a new model of reality. It took an extreme genius to discover something like calculus but once discovered billions of people who are far from being geniuses can use this new perspective. Science advances by turning new ideas into standard ideas so each generation builds on the last.

Politics does the equivalent of constantly trying to reinvent children’s arithmetic and botching it. It does not build reliable foundations of knowledge. Archimedes is no longer cutting edge. Thucydides and Sun Tzu are still cutting edge. Even though Tetlock and others have shown how to start making similar progress with politics, our political cultures fiercely resist learning and fight ferociously to stay in closed and failing feedback loops.

In many ways our political culture has regressed as it has become more and more audio-visual and less and less literate. (Only 31% of US college graduates can read at a basic level. I’d guess it’s similar here. See end.) I’ve experimented with the way Jeff Bezos runs meetings at Amazon: i.e start the meeting with giving people a 5-10 page memo to read. Impossible in Westminster, nobody will sit and read like that! Officials have tried and failed for a year to get senior ministers to engage with complex written material about the EU negotiations. TV news dominates politics and is extremely low-bandwidth: it contains a few hundred words and rarely uses graphics properly. Evan Davis illustrates a comment about ‘going down the plughole’ with a picture of water down a plughole and Nick Robinson illustrates a comment about ‘the economy taking off’ with a picture of a plane taking off. The constant flow of bullshit from the likes of Robert Peston and Jon Snow dominates the medium because competition has been impossible until recently. BUT, although technology is making these charlatans less relevant (good) it also creates new problems and will not necessarily improve the culture.

Watching political news makes you dumber — switch it off and read books! If you work in it, either QUIT or go on holiday and come back determined to subvert it. How? Start with a previous blog which has some ideas, like tracking properly which people have a record of getting things right and wrong. Every editor I’ve suggested this to winces and says ‘impossible’. Insiders fear accountability and competition.

Today, the anniversary of the referendum, is a good day to forget the babble in the bubble and think about lessons from another project that changed the world, the famous ARPA/PARC team of the 1960s and 1970s.

*

ARPA/PARC and ‘capturing the heavens’: The best way to predict the future is to invent it

The panic over Sputnik brought many good things such as a huge increase in science funding. America also created the Advanced Research Projects Agency (ARPA, which later added ‘Defense’ and became DARPA). Its job was to fund high risk / high payoff technology development. In the 1960s and 1970s, a combination of unusual people and unusually wise funding from ARPA created a community that in turn invented the internet, or ‘the intergalactic network’ as Licklider originally called it, and the personal computer. One of the elements of this community was PARC, a research centre working for Xerox. As Bill Gates said, he and Steve Jobs essentially broke into PARC, stole their ideas, and created Microsoft and Apple.

The ARPA/PARC project has created over 35 TRILLION DOLLARS of value for society and counting.

The whole story is fascinating in many ways. I won’t go into the technological aspects. I just want to say something about the process.

What does a process that produces ideas that change the world look like?

One of the central figures was Alan Kay. One of the most interesting things about the project is that not only has almost nobody tried to repeat this sort of research but the business world has even gone out of its way to spread mis-information about it because it was seen as so threatening to business-as-usual.

I will sketch a few lessons from one of Kay’s pieces but I urge you to read the whole thing.

‘This is what I call “The power of the context” or “Point of view is worth 80 IQ points”. Science and engineering themselves are famous examples, but there are even more striking processes within these large disciplines. One of the greatest works of art from that fruitful period of ARPA/PARC research in the 60s and 70s was the almost invisible context and community that catalysed so many researchers to be incredibly better dreamers and thinkers. That it was a great work of art is confirmed by the world-changing results that appeared so swiftly, and almost easily. That it was almost invisible, in spite of its tremendous success, is revealed by the disheartening fact today that, as far as I’m aware, no governments and no companies do edge-of-the-art research using these principles.’

‘[W]hen I think of ARPA/PARC, I think first of good will, even before brilliant people… Good will and great interest in graduate students as “world-class researchers who didn’t have PhDs yet” was the general rule across the ARPA community.

‘[I]t is no exaggeration to say that ARPA/PARC had “visions rather than goals” and “funded people, not projects”. The vision was “interactive computing as a complementary intellectual partner for people pervasively networked world-wide”. By not trying to derive specific goals from this at the funding side, ARPA/PARC was able to fund rather different and sometimes opposing points of view.

‘The pursuit of Art always sets off plans and goals, but plans and goals don’t always give rise to Art. If “visions not goals” opens the heavens, it is important to find artistic people to conceive the projects.

‘Thus the “people not projects” principle was the other cornerstone of ARPA/PARC’s success. Because of the normal distribution of talents and drive in the world, a depressingly large percentage of organizational processes have been designed to deal with people of moderate ability, motivation, and trust. We can easily see this in most walks of life today, but also astoundingly in corporate, university, and government research. ARPA/PARC had two main thresholds: self-motivation and ability. They cultivated people who “had to do, paid or not” and “whose doings were likely to be highly interesting and important”. Thus conventional oversight was not only not needed, but was not really possible. “Peer review” wasn’t easily done even with actual peers. The situation was “out of control”, yet extremely productive and not at all anarchic.

‘”Out of control” because artists have to do what they have to do. “Extremely productive” because a great vision acts like a magnetic field from the future that aligns all the little iron particle artists to point to “North” without having to see it. They then make their own paths to the future. Xerox often was shocked at the PARC process and declared it out of control, but they didn’t understand that the context was so powerful and compelling and the good will so abundant, that the artists worked happily at their version of the vision. The results were an enormous collection of breakthroughs.

‘Our game is more like art and sports than accounting, in that high percentages of failure are quite OK as long as enough larger processes succeed… [I]n most processes today — and sadly in most important areas of technology research — the administrators seem to prefer to be completely in control of mediocre processes to being “out of control” with superproductive processes. They are trying to “avoid failure” rather than trying to “capture the heavens”.

‘All of these principles came together a little over 30 years ago to eventually give rise to 1500 Altos, Ethernetworked to: each other, Laserprinters, file servers and the ARPAnet, distributed to many kinds of end-users to be heavily used in real situations. This anticipated the commercial availability of this genre by 10-15 years. The best way to predict the future is to invent it.

‘[W]e should realize that many of the most important ARPA/PARC ideas haven’t yet been adopted by the mainstream. For example, it is amazing to me that most of Doug Engelbart’s big ideas about “augmenting the collective intelligence of groups working together” have still not taken hold in commercial systems. What looked like a real revolution twice for end-users, first with spreadsheets and then with Hypercard, didn’t evolve into what will be commonplace 25 years from now, even though it could have. Most things done by most people today are still “automating paper, records and film” rather than “simulating the future”. More discouraging is that most computing is still aimed at adults in business, and that aimed at nonbusiness and children is mainly for entertainment and apes the worst of television. We see almost no use in education of what is great and unique about computer modeling and computer thinking. These are not technological problems but a lack of perspective. Must we hope that the open-source software movements will put things right?

‘The ARPA/PARC history shows that a combination of vision, a modest amount of funding, with a felicitous context and process can almost magically give rise to new technologies that not only amplify civilization, but also produce tremendous wealth for the society. Isn’t it time to do this again by Reason, even with no Cold War to use as an excuse? How about helping children of the world grow up to think much better than most adults do today? This would truly create “The Power of the Context”.’

Note how this story runs contrary to how free market think tanks and pundits describe technological development. The impetus for most of this development came from government funding, not markets.

Also note that every attempt since the 1950s to copy ARPA and JASON (the semi-classified group that partly gave ARPA its direction) in the UK has been blocked by Whitehall. The latest attempt was in 2014 when the Cabinet Office swatted aside the idea. Hilariously its argument was ‘DARPA has had a lot of failures’ thus demonstrating extreme ignorance about the basic idea — the whole point is you must have failures and if you don’t have lots of failures then you are failing!

People later claimed that while PARC may have changed the world it never made any money for XEROX. This is ‘absolute bullshit’ (Kay). It made billions from the laser printer alone and overall Xerox made 250 times what they invested in PARC before they went bust. In 1983 they fired Bob Taylor, the manager of PARC and the guy who made it all happen.

‘They hated [Taylor] for the very reason that most companies hate people who are doing something different, because it makes middle and upper management extremely uncomfortable. The last thing they want to do is make trillions, they want to make a few millions in a comfortable way’ (Kay).

Someone finally listened to Kay recently. ‘YC Research’, the research arm of the world’s most successful (by far) technology incubator, is starting to fund people in this way. I am not aware of any similar UK projects though I know that a small network of people are thinking again about how something like this could be done here. If you can help them, take a risk and help them! Someone talk to science minister Jo Johnson but be prepared for the Treasury’s usual ignorant bullshit — ‘what are we buying for our money, and how can we put in place appropriate oversight and compliance?’ they will say!

Why is this relevant to the referendum?

As we ponder the future of the UK-EU relationship shaped amid the farce of modern Whitehall, we should think hard about the ARPA/PARC example: how a small group of people can make a huge breakthrough with little money but the right structure, the right ways of thinking, and the right motives.

Those of us outside the political system thinking ‘we know we can do so much better than this but HOW can we break through the bullshit?’ need to change our perspective and gain 80 IQ points.

This real picture is a metaphor for the political culture: ad hoc solutions that are either bad or don’t scale.

ARPA said ‘Let’s get rid of all the wires’. How do we ‘get rid of all the wires’ and build something different that breaks open the closed and failing political cultures? Winning the referendum was just one step that helps clear away dead wood but we now need to build new things.

The ARPA vision that aligned the artists ‘like little iron filings’ was:

‘Computers are destined to become interactive intellectual amplifiers for everyone in the world universally networked worldwide’ (Licklider).

We need a motivating vision aimed not at tomorrow but at changing the basic wiring of  the whole system, a vision that can align ‘the little iron filings’, and then start building for the long-term.

I will go into what I think this vision could be and how to do it another day. I think it is possible to create something new that could scale very fast and enable us to do politics and government extremely differently, as different to today as the internet and PC were to the post-war mainframes. This would enable us to build huge long-term value for humanity in a relatively short time (less than 20 years). To create it we need a process as well suited to the goal as the ARPA/PARC project was and incorporating many of its principles.

We must try to escape the current system with its periodic meltdowns and international crises. These crises move 500-1,000 times faster than that of summer 1914. Our destructive potential is at least a million-fold greater than it was in 1914. Yet we have essentially the same hierarchical command-and-control decision-making systems in place now that could not even cope with 1914 technology and pace. We have dodged nuclear wars by fluke because individuals made snap judgements in minutes. Nobody who reads the history of these episodes can think that this is viable long-term, and we will soon have another wave of innovation to worry about with autonomous robots and genetic engineering. Technology gives us no option but to try to overcome evolved instincts like destroying out-group competitors.

In a previous blog I outlined how the ‘systems management’ approach used to put man on the moon provides principles for a new approach.

*

Ironically, one of the very few people in politics who understood the sort of thinking needed was … Jean Monnet, the architect of the EEC/EU! Monnet understood how to step back from today and build institutions. He worked operationally to prepare the future:

‘If there was stiff competition round the centres of power, there was practically none in the area where I wanted to work – preparing the future.’

Monnet was one of the few people in modern politics who really deserve the label ‘genius’. The story of how he wangled the creation of his institutions through the daily chaos of post-war politics is a lesson to anybody who wants to get things done.

But the institutions he created are in many ways the opposite of what the world needs. Their core operating principle is perpetual centralisation of power in the hands of an all powerful bureaucracy (Commission) and Court (ECJ). Nothing that works well in the world works like this!

Thanks to the prominence of Farage the dominant story among educated people is that those who got us out of the EU want to take us back to the pre-1914 era of hostile competing nation states. Nothing could be further from the truth. The key people in Vote Leave wanted and want not just what is best for Britain but what is best for all humanity. We want more international cooperation, not less. The problem with the EU is not that it is about international cooperation but that it is so bad at it and actually undermines it.

Britain leaving forces those with power to ask: how can all European countries trade freely and cooperate without subscribing to Monnet’s bureaucratic centralism? This will help Europe in the long-term. To those who favour this bureaucratic centralism and uniformity, reflect on the different trajectories of Europe and China post-Renaissance. In Europe, regulatory competition (so Columbus could chase funding in Spain after rejection in Portugal) brought immense gains. In China, centrally directed uniformity led to centuries of stagnation. America’s model of competitive federalism created by the founding fathers has been a far more effective engine of civilisation, growth, and new knowledge than the Monnet-Delors Single Market model.

If Britain were to focus on science and education with huge resources and a new-found seriousness, then this regulatory diversity would help not just Britain but all Europe and the global science community. We could make Britain the best place in the world to be for those who can invent the future. Like Alan Kay and his colleagues, we could create whole new industries. We could call Jeff Bezos and say, ‘Ok Jeff, you want a permanent international manned moon base, let’s talk about who does what, but not with that old rocket technology.’ No country on earth funds science as well as we already know how it could be done — that is something for Britain to do that would create real long-term value for humanity, instead of the ‘punching above our weight’ and ‘special relationship’ bullshit that passes for strategy in London. How we change our domestic institutions is within our power and will have much much greater influence on our long-term future than whatever deal is botched together with Brussels. We have the resources. But can we break the system open? If we don’t then we’re likely to go down the path we were already going down inside the EU, like the deluded Norma Desmond in Sunset Boulevard claiming ‘I am big, it’s the pictures that got small.’

*

Vote Leave and ‘good will’

Although Vote Leave was enmeshed in a sort of collective lunacy we managed, barely, to fend it off from the inner working of the campaign. Much of my job (sadly) was just trying to maintain a cordon around the core team so they could deliver the campaign with as little disruption as possible. We managed this because among the core people we had great good will. The stories of the campaign focus on the lunacy, but the people who really made it work remember the goodwill.

A year ago tonight I was sitting alone in a room thinking ‘we’ve won, now…’ when the walls started rumbling. At first I couldn’t make it out then, as Tim Shipman tells the story in his definitive book on the campaign, I heard ‘Dom, Dom, DOM’ — the team had declared victory. I went next door…

Thanks to everybody who sacrificed something. As I said that night and as I said in my long blog on the campaign, I’ve been given credit I don’t deserve and which rightly belongs to others — Cleo Watson, Richard ‘Ricardo’ Howell, Brother Starkie, Oliver Lewis, Lord Suart et al. Now, let’s think about what should come next…

 

Watch Alan Kay explain how to invent the future HERE and HERE.

---

Ps. Kay also points out that the real computer revolution won’t happen until people fulfil the original vision of enabling children to use this powerful way of thinking:

‘The real printing revolution was a qualitative change in thought and argument that lagged the hardware inventions by almost two centuries. The special quality of computers is their ability to rapidly simulate arbitrary descriptions, and the real computer revolution won’t happen until children can learn to read, write, argue and think in this powerful new way. We should all try to make this happen much sooner than 200 or even 20 more years!’

Almost nobody in education policy is aware of the educational context for the ARPA/PARC project which also speaks volumes about the abysmal field of ‘education research/policy’.

* Re the US literacy statistic, cf. A First Look at the Literacy of America’s Adults in the 21st Century, National Assessment of Adult Literacy, U.S. Dept of Education, NCES 2006.

 

 

 

Complexity and Prediction Part V: The crisis of mathematical paradoxes, Gödel, Turing and the basis of computing

Before the referendum I started a series of blogs and notes exploring the themes of complexity and prediction. This was part of a project with two main aims: first, to sketch a new approach to education and training in general but particularly for those who go on to make important decisions in political institutions and, second, to suggest a new approach to political priorities in which progress with education and science becomes a central focus for the British state. The two are entangled: progress with each will hopefully encourage progress with the other.

I was working on this paper when I suddenly got sidetracked by the referendum and have just looked at it again for the first time in about two years.

The paper concerns a fascinating episode in the history of ideas that saw the most esoteric and unpractical field, mathematical logic, spawn a revolutionary technology, the modern computer. NB. a great lesson to science funders: it’s a great mistake to cut funding on theory and assume that you’ll get more bang for buck from ‘applications’.

Apart from its inherent fascination, knowing something of the history is helpful for anybody interested in the state-of-the-art in predicting complex systems which involves the intersection between different fields including: maths, computer science, economics, cognitive science, and artificial intelligence. The books on it are either technical, and therefore inaccessible to ~100% of the population, or non-chronological so it is impossible for someone like me to get a clear picture of how the story unfolded.

Further, there are few if any very deep ideas in maths or science that are so misunderstood and abused as Gödel’s results. As Alan Sokal, author of the brilliant hoax exposing post-modernist academics, said, ‘Gödel’s theorem is an inexhaustible source of intellectual abuses.’ I have tried to make clear some of these using the best book available by Franzen, which explains why almost everything you read about it is wrong. If even Stephen Hawking can cock it up, the rest of us should be particularly careful.

I sketched these notes as I tried to pull together the story from many different books. I hope they are useful particularly for some 15-25 year-olds who like chronological accounts about ideas. I tried to put the notes together in the way that I wish I had been able to read at that age. I tried hard to eliminate errors but they are inevitable given how far I am from being competent to write about such things. I wish someone who is competent would do it properly. It would take time I don’t now have to go through and finish it the way I originally intended to so I will just post it as it was 2 years ago when I got calls saying ‘about this referendum…’

The only change I think I have made since May 2015 is to shove in some notes from a great essay later that year by the man who wrote the textbook on quantum computers, Michael Nielsen, which would be useful to read as an introduction or instead, HERE.

As always on this blog there is not a single original thought and any value comes from the time I have spent condensing the work of others to save you the time. Please leave corrections in comments.

The PDF of the paper is HERE (amended since first publication to correct an error, see Comments).

 

‘Gödel’s achievement in modern logic is singular and monumental – indeed it is more than a monument, it is a land mark which will remain visible far in space and time.’  John von Neumann.

‘Einstein had often told me that in the late years of his life he has continually sought Gödel’s company in order to have discussions with him. Once he said to me that his own work no longer meant much, that he came to the Institute merely in order to have the privilege of walking home with Gödel.’ Oskar Morgenstern (co-author with von Neumann of the first major work on Game Theory).

‘The world is rational’, Kurt Gödel.

Specialist maths schools – some facts

The news reports that the Government will try to promote more ‘specialist maths schools’ similar to the King’s College and Exeter schools.

The idea for these schools came when I read about Perelman, the Russian mathematician who in 2003 suddenly posted on arXiv a solution to the Poincaré Conjecture, one of the most important open problems in mathematics. Perelman went to one of the famous Russian specialist maths schools that were set up by one of the most important mathematicians of the 20th Century, Kolmogorov.

I thought – a) given the fall in standards in maths and physics because of the corruption of the curriculum and exams started by the Tories and continued by Blair, b) the way in which proper teaching of advanced maths and physics is increasingly limited to a tiny number of schools many of which are private, and c) the huge gains for our civilisation from the proper education of the unusual small fraction of children who are very gifted in maths and physics, why not try to set up something similar.

Gove’s team therefore pushed the idea through the DfE. Dean Acheson, US Secretary of State, said, ‘I have long been the advocate of the heretical view that, whatever political scientists might say, policy in this country is made, as often as not, by the necessity of finding something to say for an important figure committed to speak without a prearranged subject.’ This is quite true (it also explains a lot about how Monnet created the ECSC and EEC). Many things that the Gove team did relied on this. We prepared the maths school idea and waited our chance. Sure enough, the word came through from Downing Street – ‘the Chancellor needs an announcement for the Budget, something on science’. We gave them this, he announced it, and bureaucratic resistance was largely broken.

If interested in some details, then look at pages 75ff of my 2013 essay for useful links. Other countries have successfully pursued similar ideas, including France for a couple of centuries and Singapore recently.

One of the interesting aspects of trying to get them going was the way in which a) the official ‘education world’ loathed not just the idea but also the idea about the idea – they hated thinking about ‘very high ability’ and specialist teaching; b) when I visited maths departments they all knew about these schools because university departments in the West employ a large number of people who were educated in these schools but they all said ‘we can’t help you with this even though it’s a good idea because we’d be killed politically for supporting “elitism” [fingers doing quote marks in the air], good luck I hope you succeed but we’ll probably attack you on the record.’ They mostly did.

The only reason why the King’s project happened is because Alison Wolf made it a personal crusade to defeat all the entropic forces that elsewhere killed the idea (with the exception of Exeter). Without her it would have had no chance. I found few equivalents elsewhere and where I did they were smashed by their VCs.

A few points…

1) Kolmogorov-type schools are a particular thing. They undoubtedly work. But they are aimed at a small fraction of the population. Given what the products of these schools go on to contribute to human civilisation they are extraordinarily cheap. They are also often a refuge for children who have a terrible time in normal schools. If they were as different to normal kids in a negative sense as they are in a positive sense then there would be no argument about whether they have ‘special needs’.

2) Don’t believe the rubbish in things like Gladwell’s book about maths and IQ. There is now very good data on this particularly in the form of the unprecedented SMPY multi-decade study. Even a short crude test at 11-13 gives very good predictions of who is likely to be very good at maths/physics. Further there is a strong correlation between performance at the top 1% / 1:1,000 / 1:10,000 level and many outcomes in later life such as getting a doctorate, a patent, writing a paper in Science and Nature, high income, health etc. The education world has been ~100% committed to rejecting the science of this subject though this resistance is cracking.

This chart shows the SMPY results (maths ability at 13) for the top 1% of maths ability broken down into quartiles 1-4: the top quartile of the top 1% clearly outperforms viz tenure, publication and patent rates.  

3) The arguments for Kolmogorov schools do not translate to arguments for selection in general – ie. they are specific to the subject. It is the structure of maths and the nature of the brain that allows very young people to make rapid progress. These features are not there for English, history and so on. I am not wading into the grammar school argument on either side – I am just pointing out a fact that the arguments for such maths schools are clear but should not be confused with the wider arguments over selection that involve complicated trade-offs. People on both sides of the grammar debate should, if rational, be able to support this policy.

4) These schools are not ‘maths hot houses’. Kolmogorov took the children to see  Shakespeare plays, music and so on. It is important to note that teaching English and other subjects is normal – other than you are obviously dealing with unusually bright children. If these children are not in specialist schools, then the solution is a) specialist maths teaching (including help from university-level mathematicians) and b) keeping other aspects of their education normal. Arguably the greatest mathematician in the world, Terry Tao, had wise parents and enjoyed this combination. So it is of course possible to educate such children without specialist schools but the risks are higher that either parents or teachers cock it up.

5) Extended wisely across Britain they could have big benefits not just for those children and elite universities but they could also play an important role in raising standards generally in their area by being a focus for high quality empirical training. One of the worst aspects of the education world is the combination of low quality training and resistance to experiments. This has improved since the Gove reforms but the world of education research continues to be dominated by what Feynman called ‘cargo cult science’.

6) We also worked with a physicist at Cambridge, Professor Mark Warner, to set up a project to improve the quality of 6th form physics. This project has been a great success thanks to his extraordinary efforts and the enthusiasm of young Cambridge physicists. Thousands of questions have been answered on their online platform from many schools. This project gives kids the chance to learn proper problem solving – that is the core skill that the corruption of the exam system has devalued and increasingly pushed into a ghetto of of private education. Needless to say the education world also was hostile to this project. Anything that suggests that we can do much much better is generally hated by all elements of the bureaucracy, including even elements such as the Institute of Physics that supposedly exist to support exactly this. A handful of officials helped us push through projects like this and of course most of them have since left Whitehall in disgust, thus does the system protect itself against improvement while promoting the worst people.

7) This idea connects to a broader idea. Kids anywhere in the state system should be able to apply some form of voucher to buy high quality advanced teaching from outside their school for a wide range of serious subjects from music to physics.

8) One of the few projects that the Gove team tried and failed to get going was to break the grip of GCSEs on state schools (Cameron sided with Clegg and although we cheated a huge amount through the system we hit a wall on this project). It is extremely wasteful for the system and boring for many children for them to be focused on existing exams that do not develop serious skills. Maths already has the STEP paper. There should be equivalents in other subjects at age 16. There is nothing that the bureaucracy will fight harder than this and it will probably only happen if excellent private schools decide to do it themselves and political pressure then forces the Government to allow state schools to do them.

Any journalists who want to speak to people about this should try to speak to Dan Abramson (the head of the King’s school), Alison Wolf, or Alexander Borovik (a mathematician at Manchester University who attended one of these schools in Russia).

It is hopeful that No10 is backing this idea but of course they will face determined resistance. It will only happen if at least one special adviser in the DfE makes it a priority and has the support of No10 so officials know they might as well fight about other things…

---

This is the most interesting comment probably ever left on this blog and it is much more interesting than the blog itself so I have copied it below. It is made by Borovik, mentioned above, who attended one of these schools in Russia and knows many who attended similar…

‘There is one more aspect of (high level) selective specialist mathematics education that is unknown outside the professional community of mathematicians.

I am not an expert on “gifted and talented” education. On the other hand, I spent my life surrounded by people who got exclusive academically selective education in mathematics and physics, whether it was in the Lavrentiev School in Siberia, or Lycée Louis-le-Grand in Paris, or Fazekas in Budapest, or Galatasaray Lisesi (aka Lycée de Galatasaray) in Istanbul — the list can be continued.

The schools have nothing in common, with the exception of being unique, each one in its own way.

I had research collaborators and co-authors from each of the schools that Ilisted above. Why was it so easy for us to find a common language?

Well, the explanation can be found in the words of Stanislas Dehaene, the leading researcher of neurophysiology of mathematical thinking:

“We have to do mathematics using the brain which evolved 30 000 years ago for survival in the African savanna.”

In humans, the speed of totally controlled mental operations is at most 16 bits per second. Standard school maths education trains children to work at that speed.

The visual processing module in the brain crunches 10,000,000,000 bits per second.

I offer a simple thought experiment to the readers who have some knowledge of school level geometry.

Imagine that you are given a triangle; mentally rotate it about the longest side. What is the resulting solid of revolution? Describe it. And then try to reflect: where the answer came from?

The best kept secret of mathematics: it is done by subconsciousness.

Mathematics is a language for communication with subconsciousness.

There are four conversants in a conversation between two mathematicians: two people and two their “inner”, “intuitive” brains.

When mathematicians talk about mathematics face-to-face, they
* frequently use language which is very fluid and informal;
* improvised on the spot;
* includes pauses (for a lay observer—very strange and awkwardly timed) for absorbtion of thought;
* has almost nothing in common with standardised mathematics “in print”.

Mathematician is trying to convey a message from his “intuitive brain” directly to his colleagues’ “intuitive brain”.

Alumni of high level specialist mathematics schools are “birds of feather” because they have been initiated into this mode of communication at the most susceptible age, as teenagers, at the peak of intensity of their socialisation / shaping group identity stream of self-actualisation.

In that aspect, mathematics is not much different from arts. Part of the skills that children get in music schools, acting schools, dancing school, and art schools is the ability to talk about music, acting, dancing, art with intuitive, subconscious parts of their minds — and with their peers, in a secret language which is not recognised (and perhaps not even registered) by uninitiated.

However, specialist mathematics schools form a continuous spectrum from just ordinary, with standard syllabus, but good schools with good maths teachers to the likes of Louis-le-Grand and Fazekas. My comments apply mostly to the top end of the spectrum. I have a feeling that the Green Paper is less ambitious and does not call for setting up mathematics boarding schools using Chetham’s School of Music as a model. However, middle tier maths school could also be very useful — if they are set up with realistic expectations, properly supported, and have strong connections with universities.’

A Borovik

 

 

Unrecognised simplicities of effective action #1: expertise and a quadrillion dollar business

‘The combination of physics and politics could render the surface of the earth uninhabitable.’ John von Neumann.

Introduction

This series of blogs considers:

• the difference between fields with genuine expertise, such as fighting and physics, and fields dominated by bogus expertise, such as politics and economic forecasting;
• the big big problem we face – the world is ‘undersized and underorganised’ because of a collision between four forces: 1) our technological civilisation is inherently fragile and vulnerable to shocks, 2) the knowledge it generates is inherently dangerous, 3) our evolved instincts predispose us to aggression and misunderstanding, and 4) there is a profound mismatch between the scale and speed of destruction our knowledge can cause and the quality of individual and institutional decision-making in ‘mission critical’ institutions – our institutions are similar to those that failed so spectacularly in summer 1914 yet they face crises moving at least ~10³ times faster and involving ~10⁶ times more destructive power able to kill ~10¹⁰ people;
• what classic texts and case studies suggest about the unrecognised simplicities of effective action to improve the selection, education, training, and management of vital decision-makers to improve dramatically, reliably, and quantifiably the quality of individual and institutional decisions (particularly 1) the ability to make accurate predictions and b) the quality of feedback);
• how we can change incentives to aim a much bigger fraction of the most able people at the most important problems;
• what tools and technologies can help decision-makers cope with complexity.

[I’ve tweaked a couple of things in response to this blog by physicist Steve Hsu.]

*

Summary of the big big problem

The investor Peter Thiel (founder of PayPal and Palantir, early investor in Facebook) asks people in job interviews: what billion (10⁹) dollar business is nobody building? The most successful investor in world history, Warren Buffett, illustrated what a quadrillion (10¹⁵) dollar business might look like in his 50th anniversary letter to Berkshire Hathaway investors.

‘There is, however, one clear, present and enduring danger to Berkshire against which Charlie and I are powerless. That threat to Berkshire is also the major threat our citizenry faces: a “successful” … cyber, biological, nuclear or chemical attack on the United States… The probability of such mass destruction in any given year is likely very small… Nevertheless, what’s a small probability in a short period approaches certainty in the longer run. (If there is only one chance in thirty of an event occurring in a given year, the likelihood of it occurring at least once in a century is 96.6%.) The added bad news is that there will forever be people and organizations and perhaps even nations that would like to inflict maximum damage on our country. Their means of doing so have increased exponentially during my lifetime. “Innovation” has its dark side.

‘There is no way for American corporations or their investors to shed this risk. If an event occurs in the U.S. that leads to mass devastation, the value of all equity investments will almost certainly be decimated.

‘No one knows what “the day after” will look like. I think, however, that Einstein’s 1949 appraisal remains apt: “I know not with what weapons World War III will be fought, but World War IV will be fought with sticks and stones.”’

Politics is profoundly nonlinear. (I have written a series of blogs about complexity and prediction HERE which are useful background for those interested.) Changing the course of European history via the referendum only involved about 10 crucial people controlling ~£10⁷  while its effects over ten years could be on the scale of ~10⁸ – 10⁹ people and ~£10¹²: like many episodes in history the resources put into it are extremely nonlinear in relation to the potential branching histories it creates. Errors dealing with Germany in 1914 and 1939 were costly on the scale of ~100,000,000 (10⁸) lives. If we carry on with normal human history – that is, international relations defined as out-groups competing violently – and combine this with modern technology then it is extremely likely that we will have a disaster on the scale of billions (10⁹) or even all humans (~10¹⁰). The ultimate disaster would kill about 100 times more people than our failure with Germany. Our destructive power is already much more than 100 times greater than it was then: nuclear weapons increased destructiveness by roughly a factor of a million.

Even if we dodge this particular bullet there are many others lurking. New genetic engineering techniques such as CRISPR allow radical possibilities for re-engineering organisms including humans in ways thought of as science fiction only a decade ago. We will soon be able to remake human nature itself. CRISPR-enabled ‘gene drives’ enable us to make changes to the germ-line of organisms permanent such that changes spread through the entire wild population, including making species extinct on demand. Unlike nuclear weapons such technologies are not complex, expensive, and able to be kept secret for a long time. The world’s leading experts predict that people will be making them cheaply at home soon – perhaps they already are. These developments have been driven by exponential progress much faster than Moore’s Law reducing the cost of DNA sequencing per genome from ~$10⁸ to ~$10³ in roughly 15 years.

It is already practically possible to deploy a cheap, autonomous, and anonymous drone with facial-recognition software and a one gram shaped-charge to identify a relevant face and blow it up. Military logic is driving autonomy. For example, 1) the explosion in the volume of drone surveillance video (from 71 hours in 2004 to 300,000 hours in 2011 to millions of hours now) requires automated analysis, and 2) jamming and spoofing of drones strongly incentivise a push for autonomy. It is unlikely that promises to ‘keep humans in the loop’ will be kept. It is likely that state and non-state actors will deploy low-cost drone swarms using machine learning to automate the ‘find-fix-finish’ cycle now controlled by humans. (See HERE for a video just released for one such program and imagine the capability when they carry their own communication and logistics network with them.)

In the medium-term, many billions are being spent on finding the secrets of general intelligence. We know this secret is encoded somewhere in the roughly 125 million ‘bits’ of information that is the rough difference between the genome that produces the human brain and the genome that produces the chimp brain. This search space is remarkably small – the equivalent of just 25 million English words or 30 copies of the King James Bible. There is no fundamental barrier to decoding this information and it is possible that the ultimate secret could be described relatively simply (cf. this great essay by physicist Michael Nielsen). One of the world’s leading experts has told me they think a large proportion of this problem could be solved in about a decade with a few tens of billions and something like an Apollo programme level of determination.

Not only is our destructive and disruptive power still getting bigger quickly – it is also getting cheaper and faster every year. The change in speed adds another dimension to the problem. In the period between the Archduke’s murder and the outbreak of World War I a month later it is striking how general failures of individuals and institutions were compounded by the way in which events moved much faster than the ‘mission critical’ institutions could cope with such that soon everyone was behind the pace, telegrams were read in the wrong order and so on. The crisis leading to World War I was about 30 days from the assassination to the start of general war – about 700 hours. The timescale for deciding what to do between receiving a warning of nuclear missile launch and deciding to launch yourself is less than half an hour and the President’s decision time is less than this, maybe just minutes. This is a speedup factor of at least 10³.

Economic crises already occur far faster than human brains can cope with. The financial system has made a transition from people shouting at each other to a a system dominated by high frequency ‘algorithmic trading’ (HFT), i.e. machine intelligence applied to robot trading with vast volumes traded on a global spatial scale and a microsecond (10^-6) temporal scale far beyond the monitoring, understanding, or control of regulators and politicians. There is even competition for computer trading bases in specific locations based on calculations of Special Relativity as the speed of light becomes a factor in minimising trade delays (cf. Relativistic statistical arbitrage, Wissner-Gross). ‘The Flash Crash’ of 9 May 2010 saw the Dow lose hundreds of points in minutes. Mini ‘flash crashes’ now blow up and die out faster than humans can notice. Given our institutions cannot cope with economic decisions made at ‘human speed’, a fortiori they cannot cope with decisions made at ‘robot speed’. There is scope for worse disasters than 2008 which would further damage the moral credibility of decentralised markets and provide huge chances for extremist political entrepreneurs to exploit. (* See endnote.)

What about the individuals and institutions that are supposed to cope with all this?

Our brains have not evolved much in thousands of years and are subject to all sorts of constraints including evolved heuristics that lead to misunderstanding, delusion, and violence particularly under pressure. There is a terrible mismatch between the sort of people that routinely dominate mission critical political institutions and the sort of people we need: high-ish IQ (we need more people >145 (+3SD) while almost everybody important is between 115-130 (+1 or 2SD)), a robust toolkit for not fooling yourself including quantitative problem-solving (almost totally absent at the apex of relevant institutions), determination, management skills, relevant experience, and ethics. While our ancestor chiefs at least had some intuitive feel for important variables like agriculture and cavalry 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 national institutions we have to deal with such crises are pretty similar to those that failed so spectacularly in summer 1914 yet they face crises moving at least ~10³ times faster and involving ~10⁶ times more destructive power able to kill ~10¹⁰ people. The international institutions developed post-1945 (UN, EU etc) contribute little to solving the biggest problems and in many ways make them worse. These institutions fail constantly and do not  – cannot – learn much.

If we keep having crises like we have experienced over the past century then this combination of problems pushes the probability of catastrophe towards ‘overwhelmingly likely’.

*

What Is To be Done? There’s plenty of room at the top

‘In a knowledge-rich world, progress does not lie in the direction of reading information faster, writing it faster, and storing more of it. Progress lies in the direction of extracting and exploiting the patterns of the world… And that progress will depend on … our ability to devise better and more powerful thinking programs for man and machine.’ Herbert Simon, Designing Organizations for an Information-rich World, 1969.

‘Fascinating that the same problems recur time after time, in almost every program, and that the management of the program, whether it happened to be government or industry, continues to avoid reality.’ George Mueller, pioneer of ‘systems engineering’ and ‘systems management’ and the man most responsible for the success of the 1969 moon landing.

Somehow the world has to make a series of extremely traumatic and dangerous transitions over the next 20 years. The main transition needed is:

Embed reliably the unrecognised simplicities of high performance teams (HPTs), including personnel selection and training, in ‘mission critical’ institutions while simultaneously developing a focused project that radically improves the prospects for international cooperation and new forms of political organisation beyond competing nation states.

Big progress on this problem would automatically and for free bring big progress on other big problems. It could improve (even save) billions of lives and save a quadrillion dollars (~$10¹⁵). If we avoid disasters then the error-correcting institutions of markets and science will, patchily, spread peace, prosperity, and learning. We will make big improvements with public services and other aspects of ‘normal’ government. We will have a healthier political culture in which representative institutions, markets serving the public (not looters), and international cooperation are stronger.

Can a big jump in performance – ‘better and more powerful thinking programs for man and machine’ – somehow be systematised?

Feynman once gave a talk titled ‘There’s plenty of room at the bottom’ about the huge performance improvements possible if we could learn to do engineering at the atomic scale – what is now called nanotechnology. There is also ‘plenty of room at the top’ of political structures for huge improvements in performance. As I explained recently, the victory of the Leave campaign owed more to the fundamental dysfunction of the British Establishment than it did to any brilliance from Vote Leave. Despite having the support of practically every force with power and money in the world (including the main broadcasters) and controlling the timing and legal regulation of the referendum, they blew it. This was good if you support Leave but just how easily the whole system could be taken down should be frightening for everybody .

Creating high performance teams is obviously hard but in what ways is it really hard? It is not hard in the same sense that some things are hard like discovering profound new mathematical knowledge. HPTs do not require profound new knowledge. We have been able to read the basic lessons in classics for over two thousand years. We can see relevant examples all around us of individuals and teams showing huge gains in effectiveness.

The real obstacle is not financial. The financial resources needed are remarkably low and the return on small investments could be incalculably vast. We could significantly improve the decisions of the most powerful 100 people in the UK or the world for less than a million dollars (~£10⁶) and a decade-long project on a scale of just ~£10⁷ could have dramatic effects.

The real obstacle is not a huge task of public persuasion – quite the opposite. A government that tried in a disciplined way to do this would attract huge public support. (I’ve polled some ideas and am confident about this.) Political parties are locked in a game that in trying to win in conventional ways leads to the public despising them. Ironically if a party (established or new) forgets this game and makes the public the target of extreme intelligent focus then it would not only make the world better but would trounce their opponents.

The real obstacle is not a need for breakthrough technologies though technology could help. As Colonel Boyd used to shout, ‘People, ideas, machines – in that order!’

The real obstacle is that although we can all learn and study HPTs it is extremely hard to put this learning to practical use and sustain it against all the forces of entropy that constantly operate to degrade high performance once the original people have gone. HPTs are episodic. They seem to come out of nowhere, shock people, then vanish with the rare individuals. People write about them and many talk about learning from them but in fact almost nobody ever learns from them – apart, perhaps, from those very rare people who did not need to learn – and nobody has found a method to embed this learning reliably and systematically in institutions that can maintain it. The Prussian General Staff remained operationally brilliant but in other ways went badly wrong after the death of the elder Moltke. When George Mueller left NASA it reverted to what it had been before he arrived – management chaos. All the best companies quickly go downhill after the departure of people like Bill Gates – even when such very able people have tried very very hard to avoid exactly this problem.

Charlie Munger, half of the most successful investment team in world history, has a great phrase he uses to explain their success that gets to the heart of this problem:

‘There isn’t one novel thought in all of how Berkshire [Hathaway] is run. It’s all about … exploiting unrecognized simplicities… It’s a community of like-minded people, and that makes most decisions into no-brainers. Warren [Buffett] and I aren’t prodigies. We can’t play chess blindfolded or be concert pianists. But the results are prodigious, because we have a temperamental advantage that more than compensates for a lack of IQ points.’

The simplicities that bring high performance in general, not just in investing, are largely unrecognised because they conflict with many evolved instincts and are therefore psychologically very hard to implement. The principles of the Buffett-Munger success are clear – they have even gone to great pains to explain them and what the rest of us should do – and the results are clear yet still almost nobody really listens to them and above average intelligence people instead constantly put their money into active fund management that is proved to destroy wealth every year!

Most people think they are already implementing these lessons and usually strongly reject the idea that they are not. This means that just explaining things is very unlikely to work:

‘I’d say the history that Charlie [Munger] and I have had of persuading decent, intelligent people who we thought were doing unintelligent things to change their course of action has been poor.’ Buffett.

Even more worrying, it is extremely hard to take over organisations that are not run right and make them excellent.

‘We really don’t believe in buying into organisations to change them.’ Buffett.

If people won’t listen to the world’s most successful investor in history on his own subject, and even he finds it too hard to take over failing businesses and turn them around, how likely is it that politicians and officials incentivised to keep things as they are will listen to ideas about how to do things better? How likely is it that a team can take over broken government institutions and make them dramatically better in a way that outlasts the people who do it? Bureaucracies are extraordinarily resistant to learning. Even after the debacles of 9/11 and the Iraq War, costing many lives and trillions of dollars, and even after the 2008 Crash, the security and financial bureaucracies in America and Europe are essentially the same and operate on the same principles.

Buffett’s success is partly due to his discipline in sticking within what he and Munger call their ‘circle of competence’. Within this circle they have proved the wisdom of avoiding trying to persuade people to change their minds and avoiding trying to fix broken institutions.

This option is not available in politics. The Enlightenment and the scientific revolution give us no choice but to try to persuade people and try to fix or replace broken institutions. In general ‘it is better to undertake revolution than undergo it’. How might we go about it? What can people who do not have any significant power inside the system do? What international projects are most likely to spark the sort of big changes in attitude we urgently need?

This is the first of a series. I will keep it separate from the series on the EU referendum though it is connected in the sense that I spent a year on the referendum in the belief that winning it was a necessary though not sufficient condition for Britain to play a part in improving the quality of government dramatically and improving the probability of avoiding the disasters that will happen if politics follows a normal path. I intended to implement some of these ideas in Downing Street if the Boris-Gove team had not blown up. The more I study this issue the more confident I am that dramatic improvements are possible and the more pessimistic I am that they will happen soon enough.

Please leave comments and corrections…

* A new transatlantic cable recently opened for financial trading. Its cost? £300 million. Its advantage? It shaves 2.6 milliseconds off the latency of financial trades. Innovative groups are discussing the application of military laser technology, unmanned drones circling the earth acting as routers, and even the use of neutrino communication (because neutrinos can go straight through the earth just as zillions pass through your body every second without colliding with its atoms) – cf. this recent survey in Nature.

Bureaucratic cancer and the sabotage of A Level reform

‘Bureaucracy is cancerous in head and limbs; only its belly is sound and the laws it excretes are the most straightforward shit in the world… With this bureaucracy including the judges on the bench we can have press laws written by angels and they cannot lift us from the swamp. With bad laws and good civil servants one can still govern, with bad civil servants the best laws cannot help.’ Otto von Bismarck, 1850.

‘I had the agreement in principle of my colleagues; I had the agreement in principle of the entire Landtag; and yet, although minister-president, I found myself absolutely unable to bring the matter one step further along. Agreement does not help me at all when passive resistance – from what direction in this complicated machine is impossible to learn – is conducted with such success that I am scarcely in a position after two to three years to answer even the most basic questions.’ Otto von Bismarck, 1878.

If the most effective political operator of the modern world frequently complained about the difficulty of enforcing policy against a hostile bureaucracy, we should not be surprised if similar problems recur over and over again.

Here is an interesting example of how education policy is made and how Whitehall works.

In 2012, we announced that the DfE would step back from controlling A Levels and give universities control. (Allegra Stratton ran the original story on Newsnight.) The main mechanism was ALCAB. It was a nightmare to set up partly because although subject experts very much wanted to be involved the administrators who control universities wanted to stay out of the controversy and said to us in the DfE ‘we don’t want to have to say publicly that A Level papers are bad’.

We forced ALCAB to be created. MG and I spent a lot of time in awful meetings forcing it through. Its main role was supposed to be an annual review of specific A Level papers so that professors XYZ could say ‘hopeless question in the Edexcel physics paper, it gets the definition of entropy wrong again, it fails to test XXX’ etc.

The DfE has closed this committee down. It emerged via this Times Higher Education story.

I pointed a few hacks to it. They have called the DfE press office and spads. Both of those entities were given a line from officials saying ‘ALCAB’s work is done, no story here’. (Cf. Forsyth’s blog here.)

This is a lie. The main role was an annual review process. This should have been conducted this year and 2016 in preparation for new A Levels in 2017. It was envisaged as a permanent role. Interestingly, the letters completely elide this main role out of existence and present ALCAB as having only a temporary role.

Now this annual review won’t happen.

This is almost a Jedi-level operation from DfE officials. The DfE hated giving away control, obviously, and hated ALCAB. The very point of the process – a sword of Damocles in the form of eminent professors saying ‘crap questions’ each year – was supposed to force the DfE, exam boards, and Ofqual to raise their game. You can imagine how popular this was. Now the situation will revert to the status quo – the DfE firmly in charge and those pesky professors who point out things like – specific papers do not test the maths skills in the specifications – are happily excluded, with no ‘unhelpful’ public scrutiny of standards.

I very much doubt that poor Nicky Morgan Nicky Morgan [*see end] realises what she has done. It was probably a letter buried deep in her box weeks ago that she had no reason to suspect meant she was being used to subvert reform and entrench Whitehall’s power. It is impossible for a new minister to spot all such things – you don’t know what you don’t know. We can also safely bet that No10 has not the faintest idea about what ALCAB is or what the annual review process was supposed to do.

This is how Whitehall closes down threats to its power. Although it is systemically incompetent viz policy and implementation, its real focus is on its own power, jobs, and money. To these, it pays careful attention and deploys its real skills.

It is possible that the hard struggle to improve A Levels and remove politicians’ and Whitehall’s grip of them is now substantially lost, without the MPs having a clue as to why and the details lost in a miasma of untraceable decisions and discussions.

Nicky Morgan and her spads should ask Rose (head of private office) and Wormald (Perm Sec) not just ‘how did this happen?’, but also ‘why were we and the press office given lies to tell the media?’ They would also be well advised to make clear that a repetition of this fancy footwork will mean someone fired. But of course this will have little effect. The officials are lining up their holidays and their own plans for the future, safe in the happy knowledge that whoever ‘wins’ the election, they will remain in charge. The MPs of all parties are largely content for this situation to continue. In the focus groups, swing voters will continue to say ‘they’re all the same’ with much more accuracy than they realise, but few in Westminster are really listening and even fewer know what is to be done…

I will blog a few reflections on No10’s ‘schools week’ tomorrow. NB. notice how, just as I wrote in The Hollow Men, this No10 ‘schools week’ is like all the others – two days of rubbish gimmicks, a self-inflicted cockup (‘real terms cuts to the budget’), followed by silence such that by Friday the 8 people who knew it was ‘schools week’ have themselves forgotten? Plus ca change…

Ps. If you want details on the devaluation of exams since 1988, and therefore why the annual review process was so important, read THIS.

---

 

UPDATE. Some have asked ‘how much confidence did you have in ALCAB doing a good job?’ Answer? Initially not much. They are all under huge pressure to say everything is fine. Initially for example, despite physics departments across the country  complaining about the removal of calculus from Physics A Level (complaints that practically none of them will repeat publicly because of fear of their VC office), it did not look like ALCAB would be much use and they rejected calls from various professors I know on this subject. There is massive political pressure to focus exclusively on the numbers taking an A Level rather than the quality  of the A Level.

But my hope was that by creating something that would be seen as the ‘voice of the university subject experts’, they would have to listen and adapt in order to maintain credibility and avoid embarrassing challenges. There are more and more enraged academics fed up of VC offices lying to the media and misrepresenting academics’ opinions. I thought that creating something would push the debate in increasingly sensible directions where the emphasis would be on the skills needed on arrival at university. Now, everything to do with A Levels is dominated by political not educational concerns about the numbers doing them and ‘access’. This has helped corrupt the exam system. If we had professors of physics, French, music etc every year publicly humiliating exam boards for errors, this would soon improve things from a low base and make it much harder for MPs and Whitehall to keep corrupting public exams.

[* I wrote ‘poor Nicky Morgan’ with the feeling – poor her, I know what it’s like to be pottering around in the DfE dealing with all sorts of problems before the horror of Question Time then someone walks in with a new bigger problem… But a few people email to say it sounds patronising which was not deliberate, hence deletion…]

Standards In English Schools Part I: The introduction of the National Curriculum and GCSEs

The Introduction to this series of blogs, HERE, sets out the background and goals.

There are many different senses in which people discuss ‘standards’. Sometimes they mean an overall judgement on the performance of the system as judged by an international test like PISA. Sometimes they mean judgements based on performance in official exams such as KS2 SATs (at 11) or GCSEs. Sometimes they mean the number of schools above or below a DfE ‘floor target’. Sometimes they mean the number of schools and/or pupils in Ofsted-defined categories. Sometimes people talk about ‘the quality of teachers’. Sometimes they mean ‘the standards required of pupils when they take certain exams’. Today, the media is asking ‘have Academies raised standards?’ because of the Select Committee Report (which, after a brief flick through, seems to have ignored most of the most interesting academic studies done on a randomised/pseudo-randomised basis).

This blog in the series is concerned mainly with the questions of – what has happened to the standards required of pupils when they take GCSEs and A Levels as a result of changes since the mid-1980s, and how do universities and learned societies judge the preparation of pupils for further studies. Have the exams got easier? Do universities and learned societies think pupils are well-prepared for further studies?

I will give a very short potted history of the introduction of GCSEs and the National Curriculum before examining the evidence of their effects. If you are not interested in the history, please skip to the Section B on Evidence. If you just want to see my Conclusions, scroll to the end for a short section.

I stress that my goal is not to argue for a return to the pre-1988 system of O Levels and A Levels. While it had some advantages over the existing system, it also had profound problems. I think that an unknown fraction of the cohort could experience far larger improvements in learning than we see now if they were introduced to different materials in different ways, rather than either contemporary exams or their predecessors, but I will come to this argument, and why I have this belief, in a later blog.

I have used the word ‘Department’ to represent the DES of the 1980s, the DfE of post-2010, and its different manifestations in between.

This is just a rough first stab at collecting things I’ve shoved in boxes, emails etc over the past few years. Please leave corrections and additions in Comments.

A. A very potted history

Joseph introduces GCSEs – ‘a right old mess’

The debate over the whole of education policy, and particularly the curriculum and exams, changed a lot after Callaghan’s Ruskin speech in 1976 and the Department’s Yellow Book. Before then, the main argument was simply about providing school places and the furore over selection. After 1976 the emphasis shifted to ‘standards’ and there was growing momentum behind a National Curriculum (NC) of some sort and reforms to the exam system.

Between 1979-85, the Department chivvied LAs on the curriculum but had little power and nothing significant changed. Joseph was too much of a free marketeer to support a NC so its proponents could not make progress.

Joseph was persuaded to replace O Levels with GCSEs. He thought that the outcome would be higher standards for all but he later complained that he had been hoodwinked by the bureaucratic process involving The Schools Examination Committee (SEC). He later complained:

‘I should have fought against flabbiness in general more than I did… I thought I did, but how do you reach into such a producer-oriented world? … “Stretching” was my favourite word; I judged that if you leant on that much else would follow. That’s what my officials encouraged me to imagine I was achieving… I said I’d only agree to unify the two examinations provided we established differentiation [which he defined as ‘you’re stretching the academic and you’re stretching the non-academic in appropriate ways’], and now I find that unconsciously I have allowed teacher assessment, to a greater extent than I assumed. My fault … my fault… it’s the job of ministers to see deeply… and therefore it’s flabby… You don’t find me defending either myself or the Conservative Party, but I reckon that we’ve all together made a right old mess of it. And it’s hurt most those who are most vulnerable.’ (Interview with Ball.)

I have not come across any other ministers or officials from this period so open about their errors.

The O Level survived under a different name as an international exam provided by Cambridge Assessment. It is still used abroad including in Singapore which regularly comes in the top three in all international tests. Cambridge Assessment also offers an ‘international GCSE’ that is, they say, tougher than the ‘old’ GCSE (i.e. the one in use now before it changes in 2015) but not as tough as the O Level. This international GCSE was used in some private schools pre-2010 along with ‘international GCSEs’ from other exam boards. From 2010, state schools could use iGCSEs. In 2014, the DfE announced that it would stop this again. I blogged on this decision HERE.

Entangled interests – Baker and the National Curriculum

In 1986, Thatcher replaced Joseph with Baker hoping, she admitted, that he would make up ‘in presentational flair what ever he lacked in attention to detail’. He did not. Nigel Lawson wrote of Baker that ‘not even his greatest friends would describe him as a profound thinker or a man with mastery of detail’. Baker’s own PPS said that at the morning meeting ‘the main issue was media handling’. Jenny Bacon, the official responsible for The National Curriculum 5-16 (1987), said that Baker liked memos ‘in “ball points” … some snappy things with headings. It wasn’t glorious continuous prose…[Ulrich, a powerful DES official] was appalled but Baker said “That’s just the kind of brief I want”.’

Between 1976 and 1986, concern had grown in Whitehall about the large number of awful schools and widespread bad teaching. Various intellectual arguments, ideology, political interests (personal and party), and bureaucratic interests aligned to create a National Curriculum. Thatcherites thought it would undermine what they thought of as the ‘loony left’, then much in the news. Baker thought it would bring him glory. The Department and HMI rightly thought it would increase their power. After foolishly announcing CTCs at Party Conference, thus poisoning their brand with politics from the start, Baker announced he would create a NC and a testing system at 7, 11, and 14.

The different centres of power disagreed on what form the NC would take. HMI lobbied against subjects and wanted a NC based on ‘areas of expertise’, not traditional subjects. Thatcher wanted a very limited core curriculum based on English, maths, and science. The Department wanted a NC that stretched across the whole curriculum. Baker agreed with the Department and dismissed Thatcher’s limited option as ‘Gradgrind’.

In order to con Thatcher into agreeing his scheme, Baker worked with officials to invent a fake distinction between ‘core’ and ‘foundation’ subjects. As Baker’s Permanent Secretary Hancock said, ‘We devised the notion of the core and the foundation subjects but if you examine the Act you will see that there is no difference between the two. This was a totally cynical and deliberate manoeuvre on Kenneth Baker’s part.’

The 1988 Act established two quangos to be what Baker called ‘the twin guardians of the curriculum’ – The National Curriculum Council (NCC), focused on the NC, and The Schools Examinations and Assessment Council (SEAC), focused on tests. Once the Act was passed, Baker’s junior minister Rumbold said that ‘Ken went out to lunch.’ Like many ministers, he did not understand the importance of the policy detail and the intricate issues of implementation. He allowed officials to control appointments to the two vital committees and various curriculum working groups. Even Baker’s own spad later said that Baker was conned into appointing ‘the very ones responsible for the failures we have been trying to put right’. Baker forlornly later admitted that ‘I thought you could produce a curriculum without bloodshed. Then people marched over mathematics. Great armies were assembled’, and he ‘never envisaged it would be as complex as it turned out to be’. Bacon, the official responsible for the NC, said that Baker ‘wasn’t interested in the nitty gritty’. Nicholas Tate (who was at the NCC and later headed the QCA) said that Baker was ‘affable but remote. He didn’t trouble his mind with attainment targets. He was resting on his laurels.’ Hancock, his Permanent Secretary, said that ‘after 1987 he became increasingly arrogant and impatient’. In 1989, Baker was moved to Party Chairman leaving behind chaos for his successor.

According to his colleagues, Baker was obsessed with the media, he did not try to understand (and did not have the training to understand) the policy issues in detail, and he confused the showmanship necessary to get a bill passed with serious management – he described himself as ‘a doer’ but the ‘doing’ in his mind consisted of legislation and spin. He did not even understand that there were strong disputes among teachers, subject bodies, and educationalists about the content of the NC – never mind what to do about these disputes. (Having watched the UTC programme from the DfE, the same traits were much in evidence thirty years later.)

Baker’s legacy 1989 – 1997: Shambles

Baker’s memoirs do not mention the report of The Task Group on Assessment (TGAT), chaired by Professor Paul Black, commissioned by Baker in 1987 to report on how the NC could be assessed. The plan was very complicated with ten levels of attainment having to be defined for each subject. Thatcher hated it and criticised Baker for accepting it. Meanwhile the Higginson Report had recommended replacing A Levels with some sort of IB type system. Bacon said that ‘the political trade-off was Higginson got ditched … and we got TGAT. In retrospect it may have been the wrong trade off.’

MacGregor could not get a grip of the complexity. He did not even hire a specialist policy adviser because, he said, ‘I didn’t feel I needed one.’ He blamed Baker for the chaos who, he said, ‘hadn’t spent enough time thinking about who was appointed to the bodies. He left it to officials and didn’t think through what he wanted the bodies to do. For the first year I was unable to replace anybody.’ The chairman of NCC described how they used ‘magic words to appease the right’ and get through what they wanted. The officials who controlled SEAC stopped the simplification that Thatcher wanted using the ‘legal advice’ card, claiming that the 1988 Act required testing of all attainment targets. (I had to deal with the same argument 25 years later.) MacGregor was trapped. He had an unworkable system and was under contradictory pressure from Thatcher to simplify everything and from Baker to maintain what he had promised.

Clarke bluffed and bullied his way through 18 months without solving the problems. His Permanent Secretary described the trick of getting Clarke to do what officials wanted: ‘The trick was to never box him into a corner… Show him where there was a door but never look at that door, and never let on you noticed when he walked through.’ Like MacGregor, Clarke blamed Baker for the shambles: ‘[Baker] had set up all these bloody specialist committees to guide the curriculum, he’d set up quango staff who as far as I could see had come out of the Inner London Education Authority the lot of them.’ Clarke solved none of the main problems with the tests, antagonised everybody, and replaced HMI with Ofsted.

After his surprise win, Major told the Tory Conference in 1992, ‘Yes it will mean another colossal row with the education establishment. I look forward to that.’ Patten soon imploded, the unions went for the jugular over the introduction of SATs, and by the end of 1993 Number Ten had backtracked on their bellicose spin and was in full retreat with a review by Dearing (published 1994). Suddenly, the legal advice that had supposedly prevented any simplification was rethought and officials told Dearing that the legal advice did allow simplification after all: ‘our advice is that the primary legislation allows a significant measure of flexibility’. (In my experience, one of the constants of Whitehall is that legal advice tends to shift according to what powerful officials want.) Dearing produced a classic Whitehall fudge that got everybody out of the immediate crisis but did not even try to deal with the fundamental problems, thus pushing the problems into the future.

The historian Robert Skidelsky, helping SEAC, told Patten ‘these tests will not run’ and he should change course but Patten shouted ‘That is defeatist talk.’ Skidelsky decided to work out a radically simpler model than the TGAT system with a small group in SEAC: ‘We pushed the model through committee and through the Council and sent it off to John Patten. We never received a reply. Six months after I resigned Emily Blatch approached me and said she had been looking for my paper on Assessment but no one seems to know where it is.’

Patten was finished. Gillian Shephard was put in to be friendly to the unions and quiet the chaos. Soon she and Major had also fallen out and the cycle of briefing and counter-briefing against Number Ten returned with permanent policy chaos. One of her senior officials, Clive Saville, concluded that ‘There was a great intellectual superficiality about Gillian Shephard and she was as intellectually dishonest as Shirley Williams. She was someone who wanted to be liked but wasn’t up to the job.’

A few thoughts on the process

The Government had introduced a new NC and test system and replaced O Levels with GCSEs. (They also introduced new vocational qualifications (NVQs) described by Professor Alan Smithers as a ‘disaster of epic proportions … utterly lightweight’.) The process was a disastrous bungle from start to finish.

Thatcher deserves considerable blame. She allowed Baker to go ahead with fundamental reforms without any agreed aims or a detailed roadmap. She knew, as did Lawson, that Baker could not cope with details yet appointed him on the basis of ‘presentational flair’ (media obsession is often confused with ‘presentational flair’).

The best book I have read by someone who has worked in Number Ten and seen why the Whitehall architecture is dysfunctional is John Hoskyns’ Just In Time. Extremely unusually for someone in a senior position in No10, Hoskyns both had an intellectual understanding of complex systems and was a successful manager. Inevitably, he was appalled at how the most important decisions were made and left Number Ten after failing to persuade Thatcher to tear up the civil service system. Since then, everybody in Number Ten has been struggling with the same issues. (If she had taken his advice history might have been extremely different – e.g. no ERM debacle.) His conclusion on Thatcher was:

‘The conclusion that I am coming to is that the way in which [Thatcher] herself operates, the way her fire is at present consumed, the lack of a methodical mode of working and the similar lack of orderly discussion and communication on key issues, means that our chance of implementing a carefully worked out strategy – both policy and communications – is very low indeed… Difficult problems are only solved – if they can be solved at all – by people who desperately want to solve them… I am convinced that the people and the organisation are wrong.’ (Emphasis added.)

Arguably the person who knowingly appoints someone like Baker is more to blame for the failings of Baker than Baker is himself. Major and the string of ministers that followed Baker were doomed. They were not unusually bad – they were representative examples of those at the apex of the political process. They did not know how to go about deciding aims, means, and operations. They were obsessed with media management and therefore continually botched the policy and implementation. They could not control their officials. They could not agree a plan and blamed each other. If they were the sort of people who could have got out of the mess, then they were the sort of people who would not have got into the mess in the first place.

Officials over-complicated everything and, like ministers, did not engage seriously with the core issue – what should pupils of different abilities be doing and how can we establish a process where we can collect reliable information. The process was dominated by the same attitude on all sides – how to impose a mentality already fixed.

It was also clearly affected by another element that has contemporary relevance – the constant churn of people. Just between summer 1989 and the end of 1992, there was: a new Permanent Secretary in May 1989, a new SoS in July 1989 (MacGregor), another new SoS in November 1990 (Clarke), a new PM and No10 team (Major), new heads for the NCC and SEAC in July 1991, then another new SoS in spring 1992 (Patten) and another new Permanent Secretary. Everybody blamed problems on predecessors and nobody could establish a consistent path.

Even its own Permanent Secretaries later attacked the DES. James Hamilton (1976-1983) was put into DES in June 1976 from the Cabinet Office to help with the Ruskin agenda and found a place where ‘when something was proposed someone would inevitably say, “Oh we tried that back in whenever and it didn’t work”…’. Geoffrey Holland (1992-3) admitted that, ‘It [DES] simply had no idea of how to get anything off the ground. It was lacking in any understanding or experience of actually making things happen.’

A central irony of the story shows how dysfunctional the system was. Thatcher never wanted a big NC and a complicated testing system but she got one. As some of her ideological opponents in the bureaucracy tried to simplify things when it was clear Baker’s original structure was a disaster, ministers were often fighting with them to preserve a complex system that could not work and which Thatcher had never wanted. This sums up the basic problem – a very disruptive process was embarked upon without the main players agreeing what the goal was.

Although the think tanks were much more influential in this period than they are now, Ferdinand Mount, head of Thatcher’s Policy Unit, made a telling point about their limitations: ‘Enthusiasts for reform at the IEA and the CPS were prodigal with committees and pamphlets but were much less helpful when it came to providing practical options for action. This made it difficult for the Policy Unit’s ideas to overcome the objections put forward by senior officials’. Thirty years later this remains true. Think tanks put out reports but they rarely provide a detailed roadmap that could help people navigate such reforms through the bureaucracy and few people in think tanks really understand how Whitehall works. This greatly limits their real influence. This is connected to a wider point. Few of those who comment prominently on education (or other) policy understand how Whitehall works, hence there is a huge gap between discussions of ideal policy and what is actually possible within a certain timeframe in the existing system, and commentators think that all sorts of things that happen do so because of ministers’ wishes, confusing public debate further.

I won’t go into the post-1997 story. There are various books that tell this whole story in detail. The National Curriculum remained but was altered; the test system remained but gradually narrowed from the original vision; there were some attempts at another major transformation (such as Tomlinson’s attempt to end A Levels, thwarted by Blair) but none took off; money poured into the school system and its accompanying bureaucracy at an unprecedented rate but, other than a large growth in the number and salaries of everybody, it remained unclear what if any progress was being made.

This bureaucracy spent a great deal of taxpayers’ money promoting concepts such as ‘learning styles’ and ‘multiple intelligences’ that have no proper scientific basis but which nevertheless were successfully blended with old ideas from Vygotsky and Piaget to dominate a great deal of teacher training. A lot of people in the education world got paid an awful lot of money (Hargreaves, Waters et al) but what happened to standards?

(The quotes above are taken mainly from Daniel Callaghan’s Conservative Party Education Policies 1976-1997.)

B. The cascading effects of GCSEs and the National Curriculum

Below I consider 1) the data on grade inflation in GCSEs and A Levels, 2) various studies from learned societies and others that throw light on the issue, 3) knock-on effects in universities.

1. Data on grade inflation in GCSEs and A Levels

We do not have an official benchmark against which to compare GCSE results. The picture is therefore necessarily hazy. As Coe has written, ‘we are limited by the fact that in England there has been no systematic, rigorous collection of high-quality data on attainment that could answer the question about systemic changes in standards.’ This is one of the reasons why in 2013 we, supported by Coe and others, pushed through (against considerable opposition including academics at the Institute of Education) a new ‘national reference test’ in English and maths at age 16, which I will return to in a later blog.

However, we can compare the improvement in GCSE results with a) results from international tests and b) consistent domestic tests uncontrolled by Whitehall.

The first two graphs below show the results of this comparison.

Chart 1: Comparison of English performance in international surveys versus GCSE scores 1995-2012 (Coe)

Chart 2: GCSE grades achieved by candidates with same maths & vocab scores each year 1996-2012 (Coe)

Professor Coe writes of Chart 1:

‘When GCSE was introduced in 1987 [I think he must mean 1988 as that was the first year of GCSEs or else he means ‘the year before GCSEs were first taken’], 26.4% of the cohort achieved five grade Cs or better. By 2012 the proportion had risen to 81.1%. This increase is equivalent to a standardised effect size of 1.63, 3 or 163 points on the PISA scale… If we limit the period to 1995 – 2011 [as in Chart 1 above] the rise (from 44% to 80% 5A*-C) is equivalent to 99 points on the PISA scale [as superimposed on Chart 1]… [T]he two sets of data [international and GCSEs] tell stories that are not remotely compatible. Even half the improvement that is entailed in the rise in GCSE performance would have lifted England from being an average performing OECD country to being comfortably the best in the world. To have doubled that rise in 16 years is just not believable…

‘The question, therefore, is not whether there has been grade inflation, but how much…’ [Emphasis added.] (Professor Robert Coe, ‘Improving education: a triumph of hope over experience‘, 18 June 2013, p. vi.)

Chart 2 plots the improving GCSE grades achieved by pupils scoring the same each year in a test of maths and vocabulary: pupils scoring the same on YELLIS get higher and higher GCSE grades as time passes. Coe concludes that although ‘it is not straightforward to interpret the rise in grades … as grade inflation’, the YELLIS data ‘does suggest that whatever improved grades may indicate, they do not correspond with improved performance in a fixed test of maths and vocabulary’ (Coe, ibid).

This YELLIS comparison suggests that in 2012 pupils received a grade higher in maths, history, and French GCSE, and almost a grade higher in English, than students of the same ability in 1996.

It is important to note that neither of Coe’s charts or measurements include the effects of either a) the initial switch from O Level to GCSE or b) what changed with GCSEs from 1988 – 1995. 

The next two charts show this earlier part of the story (both come from Education: Historical statistics, House of Commons, November 2012). NB. they have different end dates.

Chart 3: Proportion getting 5 O Levels / GCSEs at grade C or higher 1953/4 – 2008/9 

Chart 4: Proportion getting 1+ or 3+ passes at A Level 1953/4 – 1998/9

Chart 3 shows that the period 1988-95 saw an even sharper increase in GCSE scores than post-1995 so a GCSE/YELLIS style comparison that included the years 1988-1995 would make the picture even more dramatic.

Chart 4 shows a dramatic increase in A Level passes after the introduction of GCSEs. One interpretation of this graph, supported by the 1997-2010 Government and teaching unions, is that this increase reflected large real improvements in school standards.

There is GCSE data that those who believe this argument could cite. In 1988, 8% of GCSEs were awarded an ‘A’ in GCSE. In 2011, 23% of GCSEs were awarded an ‘A’ or ‘A*’ in GCSE. The DfE published data in 2013 which showed that the number of pupils with ten or more A* grades trebled 2002-12. This implies a very large increase in the numbers of those excelling at GCSE, which is consistent with a picture of a positive knock-on effect on improving A Level results.

However, we have already seen that the claims for GCSEs are ‘not believable’ in Coe’s words. It also seems prima facie very unlikely that a sudden large improvement in A Level results from 1990 could be the result of immediate improvements in learning driven by GCSEs. There is also evidence for A Levels similar to the GCSE/YELLIS comparison.

Chart 5: A level grades of candidates having the same TDA score (1988-2006)

Chart 5 plots A Level grades in different subjects against the international TDA test. As with GCSEs, this shows that pupils scoring the same in a non-government test got increasingly higher grades in A Levels. The change in maths is particularly dramatic from an ‘Unclassified’ mark in 1988 to a B/C in 2006.

What we know about GCSEs combined with this information makes it very hard to believe that the sudden dramatic increase in A Level performance since 1990 is because of real improvements and suggests another interpretation: these dramatic increases in A Level results reflected (mostly or entirely) A Levels being made significantly easier probably in order to compensate for GCSEs being much easier.

However, the data above can only tell part of the story. Logically, it is hard or impossible to distinguish between possible causes just from these sorts of comparisons. For example, perhaps someone might claim that A Level questions remained as challenging as before but grade boundaries moved – i.e. the exam papers were the same but the marking was easier. I think this is prima facie unlikely but the point is that logically the data above cannot distinguish between various possible dynamics.

Below is a collection of studies, reports, and comments from experts that I have accumulated over the past few years that throws light on which interpretation is more reasonable. Please add others in Comments.

(NB. David Spiegelhalter, a Professor of Statistics at Cambridge, has written about  problems with PISA’s use of statistics. These arguments are technical. To a non-specialist like me, he seems to make important points that PISA must answer to retain credibility and the fact that it has not (as of the last time I spoke to DS in summer 2014) is a blot on its copybook. However, I do not think they materially affect the discussion above. Other international tests conducted on different bases all tell roughly the same story. I will ask DS if he thinks his arguments do undermine the story above and post his reply if any.)

2. Studies 2007 – now 

NB1. Most of these studies are comparing changes over the past decade or so, not the period since the introduction of the NC and GCSEs in the 1980s.

NB2. I will reserve detailed discussion of the AS/A2/decoupling argument for a later blog as it fits better in the ‘post-2010 reforms’ section.

Learned societies. The Royal Society’s 2011 study of Science GCSEs: ‘the question types used provided insufficient opportunity for more able candidates … to demonstrate the extent of their scientific knowledge, understanding and skills. The question types restricted the range of responses that candidates could provide. There was little or no scope for them to demonstrate various aspects of the Assessment Objectives and grade descriptions… [T]he use of mathematics in science was examined in a very limited way.’ SCORE also published (2012) evidence on science GCSEs which reported ‘a wide variation in the amount of mathematics assessed across awarding organisations and confirmed that the use of mathematics within the context of science was examined in a very limited way. SCORE organisations felt that this was unacceptable.’

The 2012 SCORE report and Nuffield Report showed serious problems with the mathematical content of A Levels. SCORE was very critical:

‘For biology, chemistry and physics, it was felt there were underpinning areas of mathematics missing from the requirements and that their exclusion meant students were not adequately prepared for progression in that subject. For example, for physics many of the respondents highlighted the absence of calculus, differentiation and integration, in chemistry the absence of calculus and in biology, converting between different units… For biology, chemistry and physics, the analysis showed that the mathematical requirements that were assessed concentrated on a small number of areas (e.g. numerical manipulation) while many other areas were assessed in a limited way, or not at all… Survey respondents were asked to identify content areas from the mathematical requirements that should feature highly in assessments. In most cases, the biology, chemistry and physics respondents identified mathematical content areas that were hardly or not at all assessed by the awarding organisations.

‘[T]he inclusion of more in-depth problem solving would allow students to apply their knowledge and understanding in unstructured problems and would increase their fluency in mathematics within a science context.’

‘The current mathematical assessments in science A-levels do not accurately reflect the mathematical requirements of the sciences. The findings show that a large number of mathematical requirements listed in the biology, chemistry and physics specifications are assessed in a limited way or not at all within these papers. The mathematical requirements that are assessed are covered repeatedly and often at a lower level of difficulty than required for progression into higher education and employment. It has also highlighted a disparity between awarding organisations in their assessment of the use of mathematics within biology, chemistry and physics A-level. This is unacceptable and the examination system, regardless of the number of awarding organisations, must ensure the assessments provide an authentic representation of the subject and equip all students with the necessary skills to progress in the sciences.

‘This is likely to have an impact on the way that the subjects are taught and therefore on students’ ability to progress effectively to STEM higher education and employment.’ SCORE, 2012. Emphasis added.

The 2011 Institute of Physics report showed strong criticism from university academics of the state of physics and engineering undergraduates’ mathematical knowledge. Four-fifth of academics said that university courses had changed to deal with a lack of mathematical fluency and 92% said that a lack of mathematical fluency was a major obstacle.

‘The responses focused around mathematical content having to be diluted, or introduced more slowly, which subsequently impacts on both the depth of understanding of students, and the amount of material/topics that can be covered throughout the course…

‘Academics perceived a lack of crossover between mathematics and physics at A-level, which was felt to not only leave students unprepared for the amount of mathematics in physics, but also led to them not applying their mathematical knowledge to their learning of physics and engineering.’ IOP, 2011.

The 2011 Centre for Bioscience criticised Biology and Chemistry A Levels and preparation of pupils for bioscience degrees: ‘very many lack even the basics… [M]any students do not begin to attempt quantitative problems and this applies equally to those with A level maths as it does to those with C at GCSE. A lack of mathematics content in A level Biology means that students do not expect to encounter maths at undergraduate level. There needs to be a more significant mathematical component in A level biology and chemistry.’ The Royal Society of Chemistry report, The five decade challenge (2008), said there had been ‘catastrophic slippage in school science standards’ and that Government claims about improving GCSE scores were ‘an illusion’. (The Department said of the RSC report, ‘Standards in science have improved year on year thanks to 10 years of sustained investment and improvement in teaching and the education system – this is something we should celebrate, not criticise. Times have changed.’)

Ofqual, 2012. Ofqual’s Standards Review in 2012 found grade inflation in both GCSE and A-levels between 2001-03 and 2008-10: ‘Many of these reviews raise concerns about the maintenance of standards… In the GCSEs we reviewed (biology, chemistry and mathematics) we found that changes to the structure of the assessments, rather than changes to the content, reduced the demand of some qualifications.’

On A-levels, ‘In general we found that changes to the way the content was assessed had an impact on demand, in many cases reducing it. In two of the reviews (biology and chemistry) the specifications were the same for both years. We found that the demand in 2008 was lower than in 2003, usually because the structure of the assessments had changed. Often there were more short answer, structured questions’ (Ofqual, Standards Reviews – A Summary, 1 May 2012, found here).

Chief Executive of Ofqual, Glenys Stacey, has said: ‘If you look at the history, we have seen persistent grade inflation for these key qualifications for at least a decade… The grade inflation we have seen is virtually impossible to justify and it has done more than anything, in my view, to undermine confidence in the value of those qualifications’ (Sunday Telegraph, 28 April 2012).

The OECD’s International Survey of Adult Skills (October 2013). This assessed numeracy, literacy and computing skills of 16-24-year-olds. The tests were done over 2011/2012. England was 22nd out of 24 for literacy, 21st out of 24 for numeracy, and is 16th out of 20 for ‘problem solving in a technology-rich environment’.

PISA 2012. The normal school PISA tests taken in 2012 (reported 2013) showed no significant change between 2009-12. England was 21st for science, 23rd for reading, and 26th for mathematics. A 2011 OECD report concluded: ‘Official test scores and grades in England show systematically and significantly better performance than international and independent tests… [Official results] show significant increases in quality over time, while the measures based on cognitive tests not used for grading show declines or minimal improvements’ (OECD Economic Surveys: United Kingdom, 16 March 2011, p. 88-89). This interesting chart shows that in the PISA maths test the children of English professionals perform the same as children of Singapore cleaners (Do parents’ occupations have an impact on student performance?, PISA 2014).

Chart 6: Comparing pupil maths scores by parent occupation, UK (left) and Singapore (right) maths skills (PISA 2012)

TIMMS/PIRLS. The TIMMS/PIRLS tests (taken summer 2011, reported December 2012) told a similar story to PISA. England’s score in reading at age 10 increased since 2006 by a statistically significant amount. England’s score in science at age 10 decreased since 2007 by a statistically significant amount. England’s scores in science at age 14 and mathematics at ages 10 and 14 showed no statistically significant changes since 2007. (According to experts, the PISA maths test relies more on language comprehension than TIMMS which is supposedly why Finland scores higher in the former than the latter.)

National Numeracy (February 2012). Research showed that in 2011 only a fifth of the adult population had mathematical skills equivalent to a ‘C’ in GCSE, down a few percent from the last survey in 2003. About half of 16-65 year olds have at best the mathematical skills of an 11 year-old. A fifth of adults will struggle with understanding price labels on food and half ‘may not be able to check the pay and deductions on a wage slip.’

King’s College, 2009. A major study by academics from King’s College London and Durham University found that basic skills in maths have declined since the 1970s. In 2008, less than a fifth of 14 year-olds could write 11/10 as a decimal. In the early 1980s, only 22 per cent of pupils obtained a GCE O-level grade C or above in maths. In 2008, over 55 per cent gained a GCSE grade C or above in the subject (King’s College London/University of Durham, ‘Secondary students’ understanding of mathematics 30 years on‘, 5 September 2009, found here).

Chart 7: Performance on ICCAMS / CSMS Maths tests showing declines over time

Shayer et al (2007) found that performance in a test of basic scientific concepts fell significantly between 1976 and 2003. ‘[A]lthough both boys and girls have shown great drops in performance, the relative drop is greater for boys… It makes it difficult to believe in the validity of the year on year improvements reported nationally on Key Stage 3 NCTs in science and mathematics: if children are entering secondary from primary school less and less equipped with the necessary mental conditions for processing science and mathematics concepts it seems unlikely that the next 2.5 years KS3 teaching will have improved so much as more than to compensate for what students of today lack in comparison with 1976.’

Chart 8: Performance on tests of scientific concepts, 1976 – 2003 (Shayer)

Tymms (2007) reviewed assessment evidence in mathematics from children at the end of primary school between 1978 and 2004 and in reading between 1948 and 2004. The conclusion was that standards in both subjects ‘have remained fairly constant’.

Warner (2013) on physics. Professor Mark Warner (Cambridge University) produced a fascinating report (2013) on problems with GCSE and A Level Physics and compared the papers to old O Levels,  A Levels, ‘S’ Level papers, Oxbridge entry exams, international exams and so on. After reading it, there is no room for doubt. The standards demanded in GCSEs and A Levels have fallen very significantly.

‘[In modern papers] small steps are spelt out so that not more than one thing needs to be addressed before the candidate is set firmly on the right path again. Nearly all effort is spent injecting numbers into formulae that at most require GCSE-level rearrangements… All diagrams are provided… 1986 O-level … [is] certainly more difficult than the AS sample… 1988 A-level … [is] harder than most Cambridge entrance questions currently… 1983 Common Entrance [is] remarkably demanding for this age group, approaching the challenge of current AS… There is a staggering difference in the demands put on candidates… Exams [from the 1980s] much lower down the school system are in effect more difficult than exams given now in the penultimate years [i.e. AS].’

For example, the mechanics problems in GCSE Physics are substantially shallower than those in 1980s O Level, which examined concepts now in A Level. The removal of calculus from A Level physics badly undermined it. Calculus is tested in A Level Maths’ Mechanics I paper and Mechanics II and III test deeper material than Physics A Level. This is one of the reasons why Cambridge Physics department stopped requiring Physics A Level for entry and made clear that Further Maths A Level is acceptable instead (many say it is better preparation for university than physics A Level is).

Warner also makes the point that making Physics GCSE and A Level much easier did not even increase the number taking physics degrees, which has declined sharply since the mid-1980s. He concludes: ‘one could again aim for a school system to get a sizable fraction of pupils to manage exams of these [older] standards. Children are not intrinsically unable to attack such problems.’ (NB. The version of this report on the web is not the full version – I would urge those interested to email Professor Warner.)

Gowers (2012) on maths. Tim Gowers, Cambridge professor and Fields Medallist, described some problems with Maths A Level and concluded:

‘The general point here is of course that A-levels have got easier [emphasis added] and schools have a natural tendency to teach to the test. If just one of those were true, it would be far less of a problem. I would have nothing against an easy A-level if people who were clever enough were given a much deeper understanding than the exam strictly required (though as I’ve argued above, for many people teaching to the test is misguided even on its own terms, since they will do a lot better on the exam if they have not been confined to what’s on the test), and I would not be too against teaching to the test if the test was hard enough…

‘[S]ome exams, such as GCSE maths, are very very easy for some people, such as anybody who ends up reading mathematics at Cambridge (but not just those people by any means). I therefore think that the way to teach people in top sets at schools is not to work towards those exams but just to teach them maths at the pace they can manage.’

Durham University analysis gives data to quantify this conclusion. Pupils who would have received a U (unclassified) in Maths A-Level in 1988 received a B/C in 2006 – see above for Chart 5 showing this (CEM Centre Durham University, Changes in standards at GCSE and A-Level: Evidence from ALIS and YELLIS, April 2007). Further Maths A Level is supposedly the toughest A Level and probably it is but a) it is not the same as its 1980s ancestor and b) it now introduces pupils to material such as matrices that used to be taught in good prep schools.

I spent a lot of time 2007-14 talking to maths dons, including heads of departments, across England. The reason I quote Gowers is that I never heard anybody dispute his conclusion but he was almost the only one who would say it publicly. I heard essentially the same litany about A Level maths from everybody I spoke to: although there were differences of emphasis, nobody disputed these basic propositions. 1) The questions became much more structured so pupils are led up a scaffolding with less requirement for independent problem-solving. 2) The emphasis moved to memorising some basic techniques the choice of which is clearly signalled in the question. 3) The modular system a) encouraged a ‘memorise, regurgitate, forget’ mentality and b) undermined learning about how different topics connect across maths, both of which are bad preparation for further studies. (There are also some advantages to a modular system that I will return to.) 4) Many undergraduates, including even those in the top 5% at such prestigious universities as Imperial, therefore now struggle in their first year as they are not well-prepared by A Level for the sort of problems they are given in undergraduate study. (The maths department at Imperial became so sick of A Level’s failings that they recently sought and got approval to buy Oxford’s entrance exam for use in their admission system.)

I will not go into arguments about vocational qualifications here but note the conclusion of Alison Wolf whose 2011 report on this was not disputed by any of the three main parties:

‘The staple offer for between a quarter and a third of the post- 16 cohort is a diet of low-level vocational qualifications, most of which have little to no labour market value.’

3. Knock-on effects in universities

Serious lack of maths skills

There are many serious problems with maths skills. Part of the reason is that many universities do not even demand A Level maths. The result? As of about 2010-12, about 20% of Engineering undergraduates, about 40% of Chemistry and Economics undergraduates, and about 60-70% of Biology and Computer Science undergraduates did not have A Level Maths. Less than 10% of undergraduate bioscience degree courses demand A Level Maths therefore ‘problems with basic numeracy are evident and this reflects the fact that many students have grades less than A at GCSE Maths. These students are unlikely to be able to carry out many of the basic mathematical approaches, for example unable to manipulate scientific notation with negative powers so commonly used in biology’ (2011 Biosciences report). (I think that history undergraduates should be able to manipulate scientific notation with negative powers – this is one of the many things that should be standard for reasonably able people.)

The Royal Society estimated (Mathematical Needs, 2012) that about 300,000 per year need a post-GCSE Maths course but only ~100,000 do one. (This may change thanks to Core Maths starting in 2015, see later blog.) This House of Lords report (2012) on Higher Education in STEM subjects concluded: ‘We are concerned that … the level at which the subject [maths] is taught does not meet the requirements needed to study STEM subjects at undergraduate level… [W]e urge HEIs to introduce more demanding maths requirement for admissions into STEM courses as the lack, or low level, of maths requirements at entry acts as a disincentive for pupils to study maths and high level maths at A level.’ House of Lords Select Committee on Science and Technology, Higher Education in STEM subjects, 2012.

Further, though this subject is beyond the scope of this blog, it is also important that the maths PhD pipeline ‘which was already badly malfunctioning has been seriously damaged by EPSRC decisions’, including withdrawal of funding from non-statistics subjects which drew the ire of UK Fields Medallists, cf. Submission by the Council for the Mathematical Sciences to the House of Lords, 2011. The weaknesses in biology also feed into the bioscience pipeline: only six percent of bioscience academics think their graduates are well prepared for a masters in the fast-growing field of Computational Biology (p.8 of report).

Closing of language departments, decline of language skills

I have not found official stats for this but according to research done for the Guardian (with FOIs):

‘The number of universities offering degrees in the worst affected subject, German, has halved over the past 15 years. There are 40% fewer institutions where it is possible to study French on its own or with another language, while Italian is down 23% and Spanish is down 22%.’

As Katrin Kohl, professor of German at Jesus College (Oxford) has said, ‘The UK has in recent years been systematically squandering its already poor linguistic resources.’ Dawn Marley, senior lecturer in French at the University of Surrey, summarised problems across languages:

‘We regularly see high-achieving A-level students who have only a minimal knowledge of the country or countries where the language of study is spoken, or who have limited understanding of how the language works. Students often have little knowledge of key elements in a country’s history – such as the French Revolution, or the fact that France is a republic. They also continue to struggle with grammatical accuracy, and use English structures when writing in the language they are studying… The proposals for the revival of A-level are directly in line with what most, if not all, academics in language departments would see as essential.’ (Emphasis added.)

The same picture applies to classical languages. Already by 1994 the Oxford Classics department was removing texts such as Thucydides as compulsory elements in ‘Greats’ because they were deemed ‘too hard’. These changes continued and have made Classics a very different subject than it was before 1990. At Oxford, they introduced whole new courses (Mods B then Mods C) that do not require any prior study of the ancient languages themselves. The first year of Greats now involves remedial language courses.

I quote at length from a paper by John Davie, a Lecturer in Classics at Trinity College, Oxford, as his comments summarise the views of other senior classicists in Oxbridge and elsewhere who have been reluctant to speak out (In Pursuit of Excellence, Davie, 2013). Inevitably, the problems described are damaging the pipeline for masters, PhDs, and future scholarship.

‘Classics as an academic subject has lost much of its intellectual force in recent years. This is true not only of schools but also, inevitably, of universities, which are increasingly required to adapt to the lowering of standards…

‘In modernist courses…, there is (deliberately) no systematic learning of grammar or syntax, and emphasis is laid on fast reading of a dramatic continuous story in made-up Latin which gives scope for looking at aspects of ancient life. The principle of osmosis underlying this approach, whereby children will learn linguistic forms by constant exposure to them, aroused scepticism among many teachers and has been thoroughly discredited by experts in linguistics. Grammar and syntax learned in this piecemeal fashion give pupils no sense of structure and, crucially, deny them practice in logical analysis, a fundamental skill provided by Classics…

‘[W]e have, in GCSE, an exam that insults the intelligence… Recent changes to this exam have by general consent among teachers made the papers even easier.

‘In the AS exam currently taken at the end of the first year of A-level … students study two small passages of literature, which represent barely a third of an original text. They are asked questions so straightforward as to verge on the banal and the emphasis is on following a prescribed technique of answering, as at GCSE. Imagination and independent thought are simply squeezed out of this process as teachers practise exam-answering technique in accordance with the narrow criteria imposed on examiners.

‘The level of difficulty [in AS] is not substantially higher than that of GCSE, and yet this is the exam whose grades and marks are consulted by the universities when they are trying to determine the ability of candidates… Having learned the translation of these bite-sized chunks of literature with little awareness of their context or the wider picture (as at GCSE, it is increasingly the case that pupils are incapable of working out the Latin/Greek text for themselves, and so lean heavily on a supplied translation), they approach the university interview with little or no ability to think “outside the box”. Dons at Oxford and Cambridge regularly encounter a lack of independent thought and a tendency to fall back on generalisations that betray insufficient background reading or even basic curiosity about the subject. This need not be the case and is clearly the product of setting the bar too low for these young people at school…

‘At A2 … students read less than a third of a literary text they would formerly have read in its entirety.

‘There is the added problem that young teachers entering the profession are themselves products of the modernist approach and so not wholly in command of the classical languages themselves. As a result they welcome the fact that they are not required by the present system to give their pupils a thorough grounding in the language, embracing the less rigorous approach of modern course-books with some relief.

‘In the majority of British universities Classics in its traditional form has either disappeared altogether or has been replaced by a course which presents the literature, history and philosophy mainly (or entirely) in translation, i.e. less a degree course in Classics than in Classical Civilisation.

‘This situation has been forced upon university departments of Classics by the impoverished language skills of young people coming up from schools… It is not only the classical languages but English itself which has suffered in this way in the last few decades. Every university teacher of the classical languages knows that he cannot assume familiarity with the grammar and syntax of English itself, and that he will have to teach from scratch such concepts as an indirect object, punctuation or how a participle differs from a gerund…

‘Even at Oxford cuts have been made to the number of texts students are required to read and, in those texts that remain, not as many lines are prescribed for reading in the original Latin or Greek.

‘In the last ten years of teaching for Mods [at Oxford] I have been struck by how the first-year students who come my way at the start of the summer term appear to know less about the classical languages each year, an experience I know to be shared by dons at other colleges…

‘GCSE should be replaced by a modern version of the O-level that stretches pupils… This would make the present AS exam completely unsuitable, and either a more challenging set of papers should be devised, if the universities wish to continue with pre A-level interviewing, or there should be a return to an unexamined year of wide reading before the specialisation of the last year.

‘Although the present exam, A2, has more to recommend it than AS, it also would no longer be fit for purpose and would need strengthening. As part of both final years there should be regular practice in the writing of essays, a skill that has been largely lost in recent years because of the exam system and is (rightly) much missed by dons.’

This combination of problems explains why we funded a project with Professor Pelling, Regius Professor of Greek at Oxford, to fund teacher training and language enrichment courses for schools.

I will not go into other humanities subjects. I read Ancient & Modern History and have thoughts about it but I do not know of any good evidence similar to the reports quoted above by the likes of the Royal Society. I have spoken to many university teachers. Some, such as Professor Richard Evans (Cambridge) told me they think the standard of those who arrive as undergraduates is roughly the same as twenty years ago. Others at Oxbridge and elsewhere told me they think that essay writing skills have deteriorated because of changes to A Level (disputed by Evans and others) and that language skills among historians have deteriorated (undisputed by anyone I spoke to).

For example, the Cambridge Professor of Mediterranean History, David Abulafia, has contradicted Evans and, like classicists, pointed out the spread of remedial classes at Cambridge:

‘It’s a pity, then, that the director of admissions at Cambridge has proclaimed that the old system [pre-Gove reforms] is good and that AS-levels – a disaster in so many ways – are a good thing because somehow they promote access. I don’t know for whom he is speaking, but not for me as a professor in the same university…

‘[Gove] was quite right about the abolition of the time-wasting, badly devised and all too often incompetently marked AS Levels; these dreary exams have increasingly been used as the key to admissions to Cambridge, to the detriment of intellectually lively, quirky, candidates full of fizz and sparkle who actually have something to say for themselves…

‘Bogus educational theories have done so much to damage education in this country… The effects are visible even in a great university such as Cambridge, with a steady decline in standards of literacy, and with, in consequence, the provision in one college after another of ‘skills teaching’, so that students who no longer arrive knowing how to structure an essay or even read a book can receive appropriate ‘training’… Even students from top ranked schools seem to find it very difficult … to write essays coherently… In the sort of exams I am thinking of, essay writing comes much more to the fore and examiners would be making more subjective judgements about scripts. In an ideal world there would be double marking of scripts.’ Emphasis added.

Judging essay skills is a more nebulous task than judging the quality of mechanics questions. Also, there is less agreement among historians about the sort of things they want to see in school exams compared to mathematicians and physicists who largely (in my experience, I stress, which is limited) agree about the sorts of problems they want undergraduates to be able to solve and the skills they want them to have.

I will quote a Professor of English at Exeter University, Colin MacCabe, whose view of the decline of essay skills is representative of many comments I have heard, but I cannot say confidently that this view represents a consensus, despite his claim:

‘Nobody who teaches A-level or has anything to do with teaching first-year university students has any doubt that A Levels have been dumbed down… The writing of the essay has been the key intellectual form in undergraduate education for more than a century; excelling at A-level meant excelling in this form. All that went by the board when … David Blunkett, brought in AS-levels… A-levels … became two years of continuous assessment with students often taking their first module within three months of entering the sixth form. This huge increase in testing went together with a drastic change in assessment. Candidates were not now marked in relation to an overall view of their ability to mount and develop arguments, but in relation to their ability to demonstrate achievement against tightly defined assessment objectives… A-levels, once a test of general intellectual ability in relation to a particular subject, are now a tightly supervised procession through a series of targets. Assessment doesn’t come at the end of the course – it is the course… In English, students read many fewer books… Students now arrive at university without the knowledge or skills considered automatic in our day… One of the results of the changes at A-level is that the undergraduate degree is itself a much more targeted affair. Students lack of a general education mean that special subjects, dissertations etc are added to general courses which are themselves much more limited in their approach… One result of this is a grade inflation much more dramatic even than A-levels… [T]here is little place within a modern English university for students to develop the kind of intellectual independence and judgment, which has historically been the aim of the undergraduate degree.’ Observer, 22 August, 2004. (Emphasis added.)

If anybody knows of studies on history and other humanities please link in Comments below.

Oxbridge entrance

As political arguments increasingly focused on ‘participation’ and ‘access’, Oxford and Cambridge largely abandoned their own entrance exams in the 1990s. There were some oddities. Cambridge University dropped their maths test and were so worried by the results that they immediately asked for and were given special dispensation to reintroduce it and they have used one since (now known as the STEP paper, used by a few other universities). Other Cambridge departments who wanted to do the same were refused permission and some of them (including the physics department) now use interviews to test material they would like to test in a written exam. Oxford changed its mind and gradually reintroduced admission tests in some subjects. (E.g. It does not use STEP in maths but uses its own test which has more ‘applied’ maths.) Cambridge now uses AS Levels. Oxford does not (but does not like to explain why).

A Levels are largely useless for distinguishing between candidates in the top 2% of ability (i.e. two standard deviations above average). Oxbridge entry now involves a complex and incoherent set of procedures. Some departments use interviews to test skills that are i) either wholly or entirely untested by A Levels and ii) are not explicitly set out anywhere. For example, if you go to an interview for physics at Cambridge, they will ask you questions like ‘how many photons hit your eye per second from Alpha Centauri?’ – i.e. questions that you cannot cram for but from which much information can be gained by tutors watching how students grapple with the problem.

The fact that the real skills they want to test are asked about in interviews rather than in public exams is, in my opinion, not only bad for ‘standards’ but is also unfair. Rich schools with long connections to Oxbridge colleges have teachers who understand these interviews and know how to prepare pupils for them. They still teach the material tested in old exams and other materials such as Russian textbooks created decades ago. A comprehensive in east Durham that has never sent anybody to Oxbridge is very unlikely to have the same sort of expertise and is much more likely to operate on the very mistaken assumption that getting a pupil to three As is sufficient preparation for Oxbridge selection. Testing skills in open exams that everybody can see would be fairer.

I will return to this issue in a later blog but it is important to consider the oddities of this situation. Decades ago, open public standardised tests were seen as a way to overcome prejudice. For example, Ivy League universities like Harvard infamously biased their admissions system against Jews because a fair open process based on intellectual abilities, and ignoring things like lacrosse skills, would have put more Jews into Harvard than Harvard wanted. Similar bias is widespread now in order to keep the number of East Asians low. It is no coincidence that Caltech’s admissions policy is unusually based on academic ability and it has a far higher proportion of East Asians than the likes of Harvard.

Similar problems apply to Oxbridge. A consequence of making exams easier and removing Oxbridge admissions tests was to make the process more opaque and therefore biased against poorer families. The fascinating journey made by the intellectual Left on the issue of standardised tests is described in Steven Pinker’s recent influential essay on university admissions. I agree with him that a big part of the reason for the ‘madness’ is that the intelligentsia ‘has lost the ability to think straight about objective tests’. Half a century ago, the Left fought for standardised tests to overcome prejudice, now many on the Left oppose tests and argue for criteria that give the well-connected middle classes unfair advantages.

This combination of problems is one of the reasons why the Cambridge pure maths department and physics department worked with me to develop projects to redo 16-18 curricula, teacher training, and testing systems. Cambridge is even experimenting with a ‘correspondence Free School’ idea proposed by the mathematician Alexander Borovik (who attended one of the famous Russian maths schools). Powerful forces tried to stop these projects happening because they are, obviously, implicit condemnations of the existing system – condemnations that many would prefer had never seen the light of day. Similar projects in other departments at other universities were kiboshed for the same reason, as were other proposals for specialist maths schools as per the King’s project (which also would never have happened but for the determination of Alison Wolf and a handful of heroic officials in the DfE). I will return to this too.

C. Conclusions

Here are some tentative conclusions.

1. The political and bureaucratic process for the introduction of the GCSE and National Curriculum was a shambles. Those involved did not go through basic processes to agree aims. Implementation was awful. All elements of the system failed children. There are important lessons for those who want to reform the current system.
2. Given the weight of evidence above, it is hard to avoid the conclusion that GCSEs were made easier than O Levels and became easier still over time. This means that at least the top fifth are aimed aged 14 at lower standards than they would have been aimed at previously (not that O Levels were at all optimal). Many of them spend two years with low grade material and repeating boring drills, in order that the school can maximise its league table position, instead of delving deeper into subjects. Inflation seems to have stopped in the last two years, perhaps temporarily, but by the use of an Ofqual system known as ‘comparable outcomes’ which is barely understood by anybody in the school system or DfE.
3. A Levels, at least in maths, sciences, and languages, were quickly made easier after 1988 and not just by enough to keep pass marks stable but by enough to lead to large increases. Even A Level students are aimed at mundane tasks like ‘design a poster’ that are suitable for small children – not near-adults. (As I type this I am looking at an Edexcel textbook for Further Maths A Level which for some reason, Edexcel has chosen to decorate with the picture of a child in a ‘Robin’ masked outfit.)
4. The old ‘S’ level papers, designed to stretch the best A Level students, were abandoned which contributed to a decline of standards aimed for among the top 5%.
5. University degrees in some subjects therefore also had to become easier (e.g. classics) or longer (natural sciences) in order to avoid increases in failure rates. This happened in some subjects even in elite universities. Remedial courses spread, even in elite universities, to teach/improve skills that were previously expected on arrival (including Classics at Oxford and History at Cambridge). Not all of the problems are because of failures in schools or easier exams. Some are because universities themselves for political reasons will not make certain requirements of applicants. Even if the exam system were fixed, this would remain a big problem. On the other hand, while publicly speaking out for AS Levels, admissions officers also, very quietly, have been gradually introducing new, non-Government/Ofqual regulated, tests for admissions purposes. On this, it is more useful to watch what universities do than what they say.
6. These problems have cascaded right through the system and now affect the pipeline into senior university research positions in maths, sciences, and languages. For example, the lack of maths skills among biologists is hampering the development of synthetic biology and computational biology. It is very common now to have (private) discussions with scientists deploring the decline in English research universities. Just in the past few weeks I have had emails from an English physicist now at Harvard and a prominent English neuroscientist giving me details of these developments and how we are falling further behind American universities. As they say, however, nobody wants to speak out.
7. It is much easier to see what has happened at the top end of the ability curve, where effects show up in universities, than it is for median pupils. The media also  focuses on issues at the top end of the ability curve, A Levels, and the Russell Group.
8. Because politicians took control of the system and used results to justify their own policies, and because they control funding, debate over standards became thoroughly dishonest, starting with the Conservative government in the 1980s and continuing to now when academics are pressured not to speak out by administrators for fear of politicians’ responses. When governments are in control of the metrics according to which they are judged, there is likely to be dishonesty. If people – including unions, teachers, and officials – claim they deserve more money on the basis of metrics that are controlled by a small group of people operating an opaque process and controlling the regulator themselves, there is likely to be dishonesty.

An important caveat. It is possible that simultaneously a) 1-8 is true and b) the school system has improved in various ways. What do I mean?

This is a coherent (not necessarily right) conclusion from the story told above…

GCSEs are significantly easier than O Levels. Nevertheless, the switch to GCSEs also involved many comprehensives and secondary moderns dropping the old idea that maybe only a fifth of the cohort are ‘academic’ – the idea from Plato’s Republic of gold, silver, and bronze children, that influenced the 1944 Act. Instead, more schools began to focus more pupils on academic subjects. Even though the standards demanded were easier than in the pre-1988 exams, this new focus (combined with other things) at least led between 1988 and now to a) a reduction in the number of truly awful schools and b) more useful knowledge and skills at least for the bottom fifth of the cohort (in ability terms), and perhaps for more. Perhaps the education of median ability pupils stayed roughly the same (declining a bit in maths) hence the consistent picture in international tests, the King’s results comparing maths in 1978/2008, Shayer’s results and so on (above). Meanwhile the standards demanded by post-1988 A Levels clearly fell (at least in some vital subjects), as the changes in universities testify, and S Level papers vanished, so the top fifth of the cohort (and particularly the +2 standard deviation population, i.e. the top 2%) leave school in some subjects considerably worse educated than in the 1980s. (Given most scientific and technological breakthroughs come from among this top 2% this has a big knock-on effect.) Private schools felt incentivised to perform better than state schools on easier GCSEs and A Levels rather than pursue separate qualifications with all the accompanying problems. There remains no good scientific data on what children at different points on the ability curve are capable of achieving given excellent teaching so the discussion of ‘standards’ remains circular. Easier GCSEs and A Levels are consistent with some improvements for the bottom fifth, roughly stability for the median, significant decline for the top fifth, and fewer awful schools.

This is coherent. It fits the evidence sketched above.

But is it right?

In the next blog in this series I will consider issues of ‘ability’ and the circularity of the current debate on ‘standards’.

Questions?

If people accept the conclusions about GCSEs and A Levels (at least in maths, sciences, and languages, I stress again) how should this evidence be weighed against the very strong desire of many in the education system (and Parliament and Whitehall) to maintain a situation in which the vast majority of the cohort are aimed at GCSEs (or international equivalents that are not hugely different) and, for those deemed ‘academic’, A Levels?

Do the gains from this approach outweigh the losses for an unknown fraction of the ‘more able’?

Is there a way to improve gains for all points on the ability distribution?

I have been told that there is no grade inflation in music exams. Is this true? If YES, is this partly because they are not regulated by the state? Are there other factors? Has A Level Music got easier? If not why not?

What sort of approaches should be experimented with instead of the standard approaches seen in O Levels, GCSEs, and A Levels?

What can be learned from non-Government regulated tests such as Force Concepts Tests (physics), university admissions tests, STEP, IQ tests and so on?

What are the best sources on ‘S’ Level papers and what happened with Oxbridge entrance exams?

What other evidence is there? Where are analyses similar to Warner’s on physics for other subjects?

What evidence is there for university grade inflation which many tell me is now worse than GCSEs and A Levels?