On the referendum #31: Project Maven, procurement, lollapalooza results & nuclear/AGI safety

On the referendum #31: Project Maven, procurement, lollapalooza results & nuclear/AGI safety

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

‘[R]ational systems exhibit universal drives towards self-protection, resource acquisition, replication and efficiency. Those drives will lead to anti-social and dangerous behaviour if not explicitly countered. The current computing infrastructure would be very vulnerable to unconstrained systems with these drives.’ Omohundro.

‘For progress there is no cure…’ von Neumann

This blog sketches a few recent developments connecting AI and issues around ‘systems management’ and government procurement.

The biggest problem for governments with new technologies is that the limiting factor on applying new technologies is not the technology but management and operational ideas which are extremely hard to change fast. This has been proved repeatedly: eg. the tank in the 1920s-30s or the development of ‘precision strike’ in the 1970s. These problems are directly relevant to the application of AI by militaries and intelligence services. The Pentagon’s recent crash program, Project Maven, discussed below, was an attempt to grapple with these issues.

‘The good news is that Project Maven has delivered a game-changing AI capability… The bad news is that Project Maven’s success is clear proof that existing AI technology is ready to revolutionize many national security missions… The project’s success was enabled by its organizational structure.

This blog sketches some connections between:

  • Project Maven.
  • The example of ‘precision strike’ in the 1970s, Marshal Ogarkov and Andy Marshall, implications for now — ‘anti-access / area denial’ (A2/AD), ‘Air-Sea Battle’ etc.
  • Development of ‘precision strike’ to lethal autonomous cheap drone swarms hunting humans cowering underground.
  • Adding AI to already broken nuclear systems and doctrines, hacking the NSA etc — mix coke, Milla Jovovich and some alpha engineers and you get…?
  • A few thoughts on ‘systems management’ and procurement, lessons from the Manhattan Project etc.
  • The Chinese attitude to ‘systems management’ and Qian Xuesen, combined with AI, mass surveillance, ‘social credit’ etc.
  • A few recent miscellaneous episodes such as an interesting DARPA demo on ‘self-aware’ robots.
  • Charts on Moore’s Law: what scale would a ‘Manhattan Project for AGI’ be?
  • AGI safety — the alignment problem, the dangers of science as a ‘blind search algorithm’, closed vs open security architectures etc.

A theme of this blog since before the referendum campaign has been that thinking about organisational structure/dynamics can bring what Warren Buffett calls ‘lollapalooza’ results. What seems to be very esoteric and disconnected from ‘practical politics’ (studying things like the management of the Manhattan Project and Apollo) turns out to be extraordinarily practical (gives you models for creating super-productive processes).

Part of the reason lollapalooza results are possible is that almost nobody near the apex of power believes the paragraph above is true and they actively fight to stop people learning from extreme successes so there is gold lying on the ground waiting to be picked up for trivial costs. Nudging reality down an alternative branch of history in summer 2016 only cost ~£106 so the ‘return on investment’ if you think about altered GDP, technology, hundreds of millions of lives over decades and so on was truly lollapalooza. Politics is not like the stock market where you need to be an extreme outlier like Buffett/Munger to find such inefficiencies and results consistently. The stock market is an exploitable market where being right means you get rich and you help the overall system error-correct which makes it harder to be right (the mechanism pushes prices close to random,  they’re not quite random but few can exploit the non-randomness). Politics/government is not like this. Billionaires who want to influence politics could get better ‘returns on investment’ than from early stage Amazon.

This blog is not directly about Brexit at all but if you are thinking — how could we escape this nightmare and turn government institutions from hopeless to high performance and what should we focus on to replace the vision of ‘influencing the EU’ that has been blown up by Brexit? — it will be of interest. Lessons that have been lying around for over half a century could have pushed the Brexit negotiations in a completely different direction and still could do but require an extremely different ‘model of effective action’ to dominant models in Westminster.

*

Project Maven: new organisational approaches for rapid deployment of AI to war / hybrid-war

The quotes below are from a piece in The Bulletin of Atomic Scientists about a recent AI project by the Pentagon. The most interesting aspect is not the technical details but the management approach and implications for Pentagon-style bureaucraties.

Project Maven is a crash Defense Department program that was designed to deliver AI technologiesto an active combat theater within six months from when the project received funding… Technologies developed through Project Maven have already been successfully deployed in the fight against ISIS. Despite their rapid development and deployment, these technologies are getting strong praise from their military intelligence users. For the US national security community, Project Maven’s frankly incredible success foreshadows enormous opportunities ahead — as well as enormous organizational, ethical, and strategic challenges.

‘In late April, Robert Work — then the deputy secretary of the Defense Department — wrote a memo establishing the Algorithmic Warfare Cross-Functional Team, also known as Project Maven. The team had only six members to start with, but its small size belied the significance of its charter… Project Maven is the first time the Defense Department has sought to deploy deep learning and neural networks, at the level of state-of-the-art commercial AI, in department operations in a combat theater…

‘Every day, US spy planes and satellites collect more raw data than the Defense Department could analyze even if its whole workforce spent their entire lives on it. As its AI beachhead, the department chose Project Maven, which focuses on analysis of full-motion video data from tactical aerial drone platforms… These drone platforms and their full-motion video sensors play a major role in the conflict against ISIS across the globe. The tactical and medium-altitude video sensors of the Scan Eagle, MQ-1C, and MQ-9 produce imagery that more or less resembles what you see on Google Earth. A single drone with these sensors produces many terabytes of data every day. Before AI was incorporated into analysis of this data, it took a team of analysts working 24 hours a day to exploit only a fraction of one drone’s sensor data.

‘The Defense Department spent tens of billions of dollars developing and fielding these sensors and platforms, and the capabilities they offer are remarkable. Whenever a roadside bomb detonates in Iraq, the analysts can simply rewind the video feed to watch who planted it there, when they planted it, where they came from, and where they went. Unfortunately, most of the imagery analysis involves tedious work—people look at screens to count cars, individuals, or activities, and then type their counts into a PowerPoint presentation or Excel spreadsheet. Worse, most of the sensor data just disappears — it’s never looked at — even though the department has been hiring analysts as fast as it can for years… Plenty of higher-value analysis work will be available for these service members and contractors once low-level counting activity is fully automated.

‘The six founding members of Project Maven, though they were assigned to run an AI project, were not experts in AI or even computer science. Rather, their first task was building partnerships, both with AI experts in industry and academia and with the Defense Department’s communities of drone sensor analysts… AI experts and organizations who are interested in helping the US national security mission often find that the department’s contracting procedures are so slow, costly, and painful that they just don’t want to bother. Project Maven’s team — with the help of Defense Information Unit Experimental, an organization set up to accelerate the department’s adoption of commercial technologies — managed to attract the support of some of the top talent in the AI field (the vast majority of which lies outside the traditional defense contracting base). Figuring out how to effectively engage the tech sector on a project basis is itself a remarkable achievement…

‘Before Maven, nobody in the department had a clue how to properly buy, field, and implement AI. A traditional defense acquisition process lasts multiple years, with separate organizations defining the functions that acquisitions must perform, or handling technology development, production, or operational deployment. Each of these organizations must complete its activities before results are handed off to the next organization. When it comes to digital technologies, this approach often results in systems that perform poorly and are obsolete even before they are fielded.

Project Maven has taken a different approach, one modeled after project management techniques in the commercial tech sector: Product prototypes and underlying infrastructure are developed iteratively, and tested by the user community on an ongoing basis. Developers can tailor their solutions to end-user needs, and end users can prepare their organizations to make rapid and effective use of AI capabilities. Key activities in AI system development — labeling data, developing AI-computational infrastructure, developing and integrating neural net algorithms, and receiving user feedback — are all run iteratively and in parallel…

‘In Maven’s case, humans had to individually label more than 150,000 images in order to establish the first training data sets; the group hopes to have 1 million images in the training data set by the end of January. Such large training data sets are needed for ensuring robust performance across the huge diversity of possible operating conditions, including different altitudes, density of tracked objects, image resolution, view angles, and so on. Throughout the Defense Department, every AI successor to Project Maven will need a strategy for acquiring and labeling a large training data set…

‘From their users, Maven’s developers found out quickly when they were headed down the wrong track — and could correct course. Only this approach could have provided a high-quality, field-ready capability in the six months between the start of the project’s funding and the operational use of its output. In early December, just over six months from the start of the project, Maven’s first algorithms were fielded to defense intelligence analysts to support real drone missions in the fight against ISIS.

‘The good news is that Project Maven has delivered a game-changing AI capability… The bad news is that Project Maven’s success is clear proof that existing AI technology is ready to revolutionize many national security missions

The project’s success was enabled by its organizational structure: a small, operationally focused, cross-functional team that was empowered to develop external partnerships, leverage existing infrastructure and platforms, and engage with user communities iteratively during development. AI needs to be woven throughout the fabric of the Defense Department, and many existing department institutions will have to adopt project management structures similar to Maven’s if they are to run effective AI acquisition programs. Moreover, the department must develop concepts of operations to effectively use AI capabilities—and train its military officers and warfighters in effective use of these capabilities…

‘Already the satellite imagery analysis community is working on its own version of Project Maven. Next up will be migrating drone imagery analysis beyond the campaign to defeat ISIS and into other segments of the Defense Department that use drone imagery platforms. After that, Project Maven copycats will likely be established for other types of sensor platforms and intelligence data, including analysis of radar, signals intelligence, and even digital document analysis… In October 2016, Michael Rogers (head of both the agency and US Cyber Command) said “Artificial Intelligence and machine learning … [are] foundational to the future of cybersecurity. … It is not the if, it’s only the when to me.”

‘The US national security community is right to pursue greater utilization of AI capabilities. The global security landscape — in which both Russia and China are racing to adapt AI for espionage and warfare — essentially demands this. Both Robert Work and former Google CEO Eric Schmidt have said that leadership in AI technology is critical to the future of economic and military power and that continued US leadership is far from guaranteed. Still, the Defense Department must explore this new technological landscape with a clear understanding of the risks involved…

‘The stakes are relatively low when AI is merely counting the number of cars filmed by a drone camera, but drone surveillance data can also be used to determine whether an individual is directly engaging in hostilities and is thereby potentially subject to direct attack. As AI systems become more capable and are deployed across more applications, they will engender ever more difficult ethical and legal dilemmas.

‘US military and intelligence agencies will have to develop effective technological and organizational safeguards to ensure that Washington’s military use of AI is consistent with national values. They will have to do so in a way that retains the trust of elected officials, the American people, and Washington’s allies. The arms-race aspect of artificial intelligence certainly doesn’t make this task any easier…

‘The Defense Department must develop and field AI systems that are reliably safe when the stakes are life and death — and when adversaries are constantly seeking to find or create vulnerabilities in these systems.

‘Moreover, the department must develop a national security strategy that focuses on establishing US advantages even though, in the current global security environment, the ability to implement advanced AI algorithms diffuses quickly. When the department and its contractors developed stealth and precision-guided weapons technology in the 1970s, they laid the foundation for a monopoly, nearly four decades long, on technologies that essentially guaranteed victory in any non-nuclear war. By contrast, today’s best AI tech comes from commercial and academic communities that make much of their research freely available online. In any event, these communities are far removed from the Defense Department’s traditional technology circles. For now at least, the best AI research is still emerging from the United States and allied countries, but China’s national AI strategy, released in July, poses a credible challenge to US technology leadership.’

Full article here: https://thebulletin.org/project-maven-brings-ai-fight-against-isis11374

Project Maven shows recurring lessons from history. Speed and adaptability are crucial to success in conflict and can be helped by new technologies. So is the capacity for new operational ideas about using new technologies. These ideas depend on unusual people. Bureaucracies naturally slow things down (for some good but mostly bad reasons), crush new ideas, and exclude unusual people in order to defend established interests. The limiting factor for the Pentagon in deploying advanced technology to conflict in a useful time period was not new technical ideas — overcoming its own bureaucracy was harder than overcoming enemy action. This is absolutely normal in conflict (e.g it was true of the 2016 referendum where dealing with internal problems was at least an order of magnitude harder and more costly than dealing with Cameron).

As Colonel Boyd used to shout to military audiences, ‘People, ideas, machines — in that order!’

*

DARPA, ‘precision strike’, the ‘Revolution in Military Affairs’ and bureaucracies

The Project Maven experience is similar to the famous example of the tank. Everybody could see tanks were possible from the end of World War I but over 20 years Britain and France were hampered by their own bureaucracies in thinking about the operational implications and how to use them most effectively. Some in Britain and France did point out the possibilities but the possibilities were not absorbed into official planning. Powerful bureaucratic interests reinforced the normal sort of blindness to new possibilities. Innovative thinking  flourished, relatively, in Germany where people like Guderian and von Manstein could see the possibilities for a very big increase in speed turning into a huge nonlinear advantage — possibilities applied to the ‘von Manstein plan’ that shocked the world in 1940. This was partly because the destruction of German forces after 1918 meant everything had to be built from scratch and this connects to another lesson about successful innovation: in the military, as in business, it is more likely if a new entity is given the job, as with the Manhattan Project to develop nuclear weapons. The consequences were devastating for the world in 1940 but, lucky for us, the nature of the Nazi regime meant that it made very similar errors itself, e.g regarding the importance of air power in general and long range bombers in particular. (This history is obviously very complex but this crude summary is roughly right about the main point)

There was a similar story with the technological developments mainly sparked by DARPA in the 1970s including stealth (developed in a classified program by the legendary ‘Skunk Works’, tested at ‘Area 51’), global positioning system (GPS), ‘precision strike’ long-range conventional weapons, drones, advanced wide-area sensors, computerised command and control (C2), and new intelligence, reconnaissance and surveillance capabilities (ISR). The hope was that together these capabilities could automate the location and destruction of long-range targets and greatly improve simultaneously the precision, destructiveness, and speed of operations. 

The approach became known in America as ‘deep-strike architectures’ (DSA) and in the Soviet Union as ‘reconnaissance-strike complexes’ (RUK). The Soviet Marshal Ogarkov realised that these developments, based on America’s superior ability to develop micro-electronics and computers, constituted what he called a ‘Military-Technical Revolution’ (MTR) and was an existential threat to the Soviet Union. He wrote about them from the late 1970s. (The KGB successfully stole much of the technology but the Soviet system still could not compete.) His writings were analysed in America particularly by Andy Marshall at the Pentagon’s Office of Net Assessment (ONA) and others. ONA’s analyses of what they started calling the Revolution in Military Affairs (RMA) in turn affected Pentagon decisions. In 1991 the Gulf War demonstrated some of these technologies just as the Soviet Union was imploding. In 1992 the ONA wrote a very influential report (The Military-Technical Revolution) which, unusually, they made public (almost all ONA documents remain classified). 

The ~1978 Assault Breaker concept

Screenshot 2019-03-01 16.06.35

Soviet depiction of Assault Breaker (Sergeyev, ‘Reconnaissance-Strike Complexes,’ Red Star, 1985)

Screenshot 2019-03-01 16.07.48

In many ways Marshal Ogarkov thought more deeply about how to develop the Pentagon’s own technologies than the Pentagon did, hampered by the normal problems that the operationalising of new ideas threatened established bureaucratic interests, including the Pentagon’s procurement system. These problems have continued. It is hard to overstate the scale of waste and corruption in the Pentagon’s horrific procurement system (see below).

China has studied this episode intensely. It has integrated lessons into their ‘anti-access / area denial’ (A2/AD) efforts to limit American power projection in East Asia. America’s response to A2/AD is the ‘Air-Sea Battle’ concept. As Marshal Ogarkov predicted in the 1970s the ‘revolution’ has evolved into opposing ‘reconnaissance-strike complexes’ facing each other with each side striving to deploy near-nuclear force using extremely precise conventional weapons from far away, all increasingly complicated by possibilities for cyberwar to destroy the infrastructure on which all this depends and information operations to alter the enemy population’s perception (very Sun Tzu!).

Graphic: Operational risks of conventional US approach vs A2/AD (CSBA, 2016)

Screenshot 2019-03-01 16.12.17

The penetration of the CIA by the KGB, the failure of the CIA to provide good predictions, the general American failure to understand the Soviet economy, doctrine and so on despite many billions spent over decades, the attempts by the Office of Net Assessment to correct institutional failings, the bureaucratic rivalries and so on — all this is a fascinating subject and one can see why China studies it so closely.

*

From experimental drones in the 1970s to drone swarms deployed via iPhone 

The next step for reconnaissance-strike is the application of advanced robotics and artificial intelligence which could bring further order(s) of magnitude performance improvements, cost reductions, and increases in tempo. This is central to the US-China military contest. It will also affect everyone else as much of the technology becomes available to Third World states and small terrorist groups.

I wrote in 2004 about the farce of the UK aircraft carrier procurement story (and many others have warned similarly). Regardless of elections, the farce has continued to squander billions of pounds, enriching some of the worst corporate looters and corrupting public life via the revolving door of officials/lobbyists. Scrutiny by our MPs has been contemptible. They have built platforms that already cannot be sent to a serious war against a serious enemy. A teenager will be able to deploy a drone from their smartphone to sink one of these multi-billion dollar platforms. Such a teenager could already take out the stage of a Downing Street photo op with a little imagination and initiative, as I wrote about years ago

The drone industry is no longer dependent on its DARPA roots and is no longer tied to the economics of the Pentagon’s research budgets and procurement timetables. It is driven by the economics of the extremely rapidly developing smartphone market including Moore’s Law, plummeting costs for sensors and so on. Further, there are great advantages of autonomy including avoiding jamming counter-measures. Kalashnikov has just unveiled its drone version of the AK-47: a cheap anonymous suicide drone that flies to the target and blows itself up — it’s so cheap you don’t care. So you have a combination of exponentially increasing capabilities, exponentially falling costs, greater reliability, greater lethality, greater autonomy, and anonymity (if you’re careful and buy them through cut-outs etc). Then with a bit of added sophistication you add AI face recognition etc. Then you add an increasing capacity to organise many of these units at scale in a swarm, all running off your iPhone — and consider how effective swarming tactics were for people like Alexander the Great.

This is why one of the world’s leading AI researchers, Stuart Russell (professor of computer science at Berkeley) has made this warning:

‘The capabilities of autonomous weapons will be limited more by the laws of physics — for example, by constraints on range, speed and payload — than by any deficiencies in the AI systems that control them. For instance, as flying robots become smaller, their manoeuvrability increases and their ability to be targeted decreases… Despite the limits imposed by physics, one can expect platforms deployed in the millions, the agility and lethality of which will leave humans utterly defenceless

‘A very, very small quadcopter, one inch in diameter can carry a one- or two-gram shaped charge. You can order them from a drone manufacturer in China. You can program the code to say: “Here are thousands of photographs of the kinds of things I want to target.” A one-gram shaped charge can punch a hole in nine millimeters of steel, so presumably you can also punch a hole in someone’s head. You can fit about three million of those in a semi-tractor-trailer. You can drive up I-95 with three trucks and have 10 million weapons attacking New York City. They don’t have to be very effective, only 5 or 10% of them have to find the target.

‘There will be manufacturers producing millions of these weapons that people will be able to buy just like you can buy guns now, except millions of guns don’t matter unless you have a million soldiers. You need only three guys to write the program and launch them. So you can just imagine that in many parts of the world humans will be hunted. They will be cowering underground in shelters and devising techniques so that they don’t get detected. This is the ever-present cloud of lethal autonomous weapons… There are really no technological breakthroughs that are required. Every one of the component technologies is available in some form commercially… It’s really a matter of just how much resources are invested in it.’

There is some talk in London of ‘what if there is an AI arms race’ but there is already an AI/automation arms race between companies and between countries — it’s just that Europe is barely relevant to the cutting edge of it. Europe wants to be a world player but it has totally failed to generate anything approaching what is happening in coastal America and China. Brussels spends its time on posturing, publishing documents about ‘AI and trust’, whining, spreading fake news about fake news (while ignoring experts like Duncan Watts), trying to damage Silicon Valley companies rather than considering how to nourish European entities with real capabilities, and imposing bad regulation like GDPR (that ironically was intended to harm Google/Facebook but actually helped them in some ways because Brussels doesn’t understand them).

Britain had a valuable asset, Deep Mind, and let Google buy it for trivial money without the powers-that-be in Whitehall understanding its significance — it is relevant but it is not under British control. Britain has other valuable assets — for example, it is a potential strategic asset to have the AI centre, financial centre, and political centre all in London, IF politicians cared and wanted to nourish AI research and companies. Very obviously, right now we have a MP/official class that is unfit to do this even if they had the vaguest idea what to do, which almost none do (there is a flash of hope on genomics/AI).

Unlike during the Cold War when the Soviet Union could not compete in critical industries such as semi-conductors and consumer electronics, China can compete, is competing, and in some areas is already ahead.

The automation arms race is already hitting all sorts of low skilled jobs from baristas to factory cleaning, some of which will be largely eliminated much more quickly than economists and politicians expect. Many agricultural jobs are being rapidly eliminated as are jobs in fields like mining and drilling. Look at a modern mine and you will see driverless trucks on the ground and drones overhead. The implications for millions who make a living from driving is now well known. (This also has obvious implications for the wisdom of allowing millions of un-skilled immigrants and one of the oddities of Silicon Valley is that people there simultaneously argue a) politicians are clueless about the impact of automation on unskilled people and b) politicians should allow millions more unskilled immigrants into the country — an example of how technical people are not always as rational about politics as they think they are.)

This automation arms race will affect different countries at different speeds depending on their exposure to fields that are ripe for disruption sooner or later. If countries cannot tax those companies that lead in AI, they will have narrower options. They may even be forced into a sort of colony status. Those who think this is an exaggeration should look at China’s recent deals in Africa where countries are handing over vast amounts of data to China on extremely unfavourable terms. Huge server farms in China are processing facial recognition data on millions of Africans who have no idea their personal data has been handed over. The western media focuses on Facebook with almost no coverage of these issues.

In the extreme case, a significant lead in AI for country X could lead to a self-reinforcing cycle in which it increasingly dominates economically, scientifically, and militarily and perhaps cannot be caught as Ian Hogarth has argued and to which Putin recently alluded.

China’s investment in AI — more data = better product = more users = more revenue  = better talent + more data in a beautiful flywheel…

China has about x3 number of internet users than America but the gap in internet and mobile usage is much larger. ‘In China, people use their mobile phones to pay for goods 50 times more often than Americans. Food delivery volume in China is 10 times more than that of the United States. And shared bicycle usage is 300 times that of the US. This proliferation of data — with more people generating far more information than any other country – is the fuel for improving China’s AI’ (report).

Screenshot 2018-08-03 16.53.14

Screenshot 2018-08-03 17.02.34

Screenshot 2018-08-03 16.57.10

China’s AI policy priority is clear. The ‘Next Generation Artificial Intelligence Development Plan‘ announced in July 2017 states that China should catch America by 2020 and be the global leader by 2030.  Xi Jinping emphasises this repeatedly.

Screenshot 2018-08-03 17.05.15

 

*

Some implications for entangling AI with WMD — take a Milla Jovovich lookalike then add some alpha engineers…

It is important to consider nuclear safety when thinking about AI safety.

The missile silos for US nuclear weapons have repeatedly been shown to be terrifyingly insecure. Sometimes incidents are just bog standard unchecked incompetence: e.g nuclear weapons are accidentally loaded onto a plane which is then left unattended on an insecure airfield. Coke, great unconventional hookers and a bit of imagination get you into nuclear facilities, just as they get you into pretty much anywhere.

Cyber security is also awful. For example, in a major  2013 study the Pentagon’s Defense Science Board concluded that the military’s systems were vulnerable to cyberattacks, the government was ‘not prepared to defend against this threat’, and a successful cyberattack could cause military commanders to lose ‘trust in the information and ability to control U.S. systems and forces [including nuclear]’ (cf. this report). Since then, the NSA itself has had its deepest secrets hacked by an unidentified actor (possibly/probably AI-enabled) in a breach much more serious but infinitely less famous than Snowden (and resembles a chapter in the best recent techno-thriller, Daemon).

This matches research just published in the Bulletin of Atomic Scientists on the most secure (Level 3/enhanced and Level 4) bio-labs. It is now clear that laboratories conducting research on viruses that could cause a global pandemic are extremely dangerous. I am not aware of any mainstream media in Britain reporting this (story here).

Further, the systems for coping with nuclear crises have failed repeatedly. They are extremely vulnerable to false alarms, malicious attacks or even freaks like, famously, a bear (yes, a bear) triggering false alarms. We have repeatedly escaped accidental nuclear war because of flukes such as odd individuals not passing on ‘launch’ warnings or simply refusing to act. The US National Security Adviser has sat at the end of his bed looking at his sleeping wife ‘knowing’ she won’t wake up while pondering his advice to the President on a counterattack that will destroy half the world, only to be told minutes later the launch warning was the product of a catastrophic error. These problems have not been dealt with. We don’t know how bad this problem is: many details are classified and many incidents are totally unreported.

Further, the end of the Cold War gave many politicians and policy people in the West the completely false idea that established ideas about deterrence had been vindicated but they have not been vindicated (cf. Payne’s Fallacies of Cold War deterrence and The Great American Gamble). Senior decision-makers are confident that their very dangerous ideas are ‘rational’

US and Russian nukes remain on ‘launch on warning’ — i.e a hair trigger — so the vulnerabilities could recur any time. Threats to use them are explicitly contemplated over crises such as Taiwan and Kashmir. Nuclear weapons have proliferated and are very likely to proliferate further. There are now thousands of people, including North Korean and Pakistani scientists, who understand the technology. And there is a large network of scientists involved in the classified Soviet bio-weapon programme that was largely unknown to western intelligence services before the end of the Cold War and has dispersed across the world.

These are all dangers already known to experts. But now we are throwing at these wobbling systems and flawed/overconfident thinking the development of AI/ML capabilities. This will exacerbate all these problems and make crises even faster, more confusing and more dangerous.

Yes, you’re right to ask ‘why don’t I read about this stuff in the mainstream media?’. There is very little media coverage of reports on things like nuclear safety and pretty much nobody with real power pays any attention to all this. If those at the apex of power don’t take nuclear safety seriously, why would you think they are on top of anything? Markets and science have done wondrous things but they cannot by themselves fix such crazy incentive problems with government institutions.

*

Government procurement — ‘the horror, the horror’

The problem of ‘rational procurement’ is incredibly hard to solve and even during existential conflicts problems with incentives recur. If state agencies, out of  fear of what opponents might be doing, create organisations that escape most normal bureaucratic constraints, then AI will escalate in importance to the military and intelligence services even more rapidly than it already is. It is possible that China will build organisations to deploy AI to war/pseudo-war/hybrid-war faster and better than America.

In January 2017 I wrote about systems engineering and systems management — an approach for delivering extremely complex and technically challenging projects. (It was already clear the Brexit negotiations were botched, that Heywood, Hammond et al had effectively destroyed any sort of serious negotiating position, and I suggested Westminster/Whitehall had to learn from successful management of complex projects to avert what would otherwise be a debacle.) These ideas were born with the Manhattan Project to build the first nuclear bomb, the ICBMs project in the 1950s, and the Apollo program in the 1960s which put man on the moon. These projects combined a) some of the most astonishing intellects the world has seen of which a subset were also brilliant at navigating government (e.g von Neumann) and b) phenomenally successful practical managers: e.g General Groves on Manhattan Project, Bernard Schriever on ICBMs and George Mueller on Apollo.

The story we are told about the Manhattan Project focuses almost exclusively on the extraordinary collection of physicists and mathematicians at Los Alamos but they were a relatively small part of the whole story which involved an engineer building an unprecedented operation at multiple sites across America in secret and with extraordinary speed while many doubted the project was possible —  then coordinating multiple projects, integrating distributed expertise and delivering a functioning bomb.

If you read Groves’ fascinating book, Now It Can Be Told, and read a recent biography of him, in many important ways you will acquire what is effectively cutting-edge knowledge today about making huge endeavours work — ‘cutting-edge’ because almost nobody has learned from this (see below). If you are one of the many MPs aspiring to be not just Prime Minister but a Prime Minister who gets important things done, there are very few books that would repay careful study as much as Groves’. If you do then you could avoid joining the list of Major, Blair, Brown, Cameron and May who bungle around for a few years before being spat out to write very similar accounts about how they struggled to ‘find the levers of power’, couldn’t get officials to do what they want, and never understood how to get things done.

Screenshot 2019-02-22 13.13.41

Systems management is generally relevant to the question: how best to manage very big complex projects? It was relevant to the referendum (Victoria Woodcock was Vote Leave’s George Mueller). It is relevant to the Brexit negotiations and the appalling management process between May/Hammond/Heywood/Robbins et al, which has been a case study in how not to manage a complex project (Parliament also deserves much blame for never scrutinising this process). It is relevant to China’s internal development and the US-China geopolitical struggle. It is relevant to questions like ‘how to avoid nuclear war’ and ‘how would you build a Manhattan Project for safe AGI?’. It is relevant to how you could develop a high performance team in Downing Street that could end the current farce. The same issues and lessons crop up in every account of a Presidency and the role of the Chief of Staff. If you want to change Whitehall from 1) ‘failure is normal’ to 2) ‘align incentives with predictive accuracy, operational excellence and high performance’, then systems management provides an extremely valuable anti-checklist for Whitehall.

Given vital principles were established more than half a century ago that were proved to do things much faster and more effectively than usual, it would be natural to assume that these lessons became integrated in training and practice both in the worlds of management and politics/government. This did not happen. In fact, these lessons have been ‘unlearned’.

General Groves was pushed out of the Pentagon (‘too difficult’). The ICBM project, conducted in extreme panic post-Sputnik, had to re-create an organisation outside the Pentagon and re-learn Groves’ lessons a decade later. NASA was a mess until Mueller took over and imported the lessons from Manhattan and ICBMs. After Apollo’s success in 1969, Mueller left and NASA reverted to being a ‘normal’ organisation and forgot his successful approach. (The plans Mueller left for developing a manned lunar base, space commercialisation, and man on Mars by the end of the 1980s were also tragically abandoned.)

While Mueller was putting man on the moon, MacNamara’s ‘Whizz Kids’ in the Pentagon, who took America into the Vietnam War, were dismantling the successful approach to systems management claiming that it was ‘wasteful’ and they could do it ‘more efficiently’. Their approach was a disaster and not just regarding Vietnam. The combination of certain definitions of ‘efficiency’ and new legal processes ensured that procurement was routinely over-budget, over-schedule, over-promising, and generated more and more scandals. Regardless of failure the MacNamara approach metastasised across the Pentagon. Incentives are so disastrously misaligned that almost every attempt at reform makes these problems worse and lawyers and lobbyists get richer. Of course, if lawmakers knew how the Manhattan Project and Apollo were done — the lack of ‘legal process’, things happening with a mere handshake instead of years of reviews enriching lawyers! — they would be stunned.

Successes since the 1960s have often been freaks (e.g the F-16, Boyd’s brainchild) or ‘black’ projects (e.g stealth) and often conducted in SkunkWorks-style operations outside normal laws. It is striking that US classified special forces, JSOC (equivalent to SAS/SBS etc), routinely use a special process to procure technologies outside the normal law to avoid the delays. This connects to George Mueller saying late in life that Apollo would be impossible with the current legal/procurement system and it could only be done as a ‘black’ program. 

The lessons of success have been so widely ‘unlearned’ throughout the government system that when Obama tried to roll out ObamaCare, it blew up. When they investigated, the answer was: we didn’t use systems management so the parts didn’t connect and we never tested this properly. Remember: Obama had the support of the vast majority of Silicon Valley expertise but this did not avert disaster. All anyone had to do was read Groves’ book and call Sam Altman or Patrick Collison and they could have provided the expertise to do it properly but none of Obama’s staff or responsible officials did.

The UK is the same. MPs constantly repeat the absurd SW1 mantra that ‘there’s no money’ while handing out a quarter of a TRILLION pounds every year on procurement and contracting. I engaged with this many times in the Department for Education 2010-14. The Whitehall procurement system is embedded in the dominant framework of EU law (the EU law is bad but UK officials have made it worse). It is complex, slow and wasteful. It hugely favours large established companies with powerful political connections — true corporate looters. The likes of Carillion and lawyers love it because they gain from the complexity, delays, and waste. It is horrific for SMEs to navigate and few can afford even to try to participate. The officials in charge of multi-billion processes are mostly mediocre, often appalling. In the MoD corruption adds to the problems.

Because of mangled incentives and reinforcing culture, the senior civil service does not care about this and does not try to improve. Total failure is totally irrelevant to the senior civil service and is absolutely no reason to change behaviour even if it means thousands of people killed and many billions wasted. Occasionally incidents like Carillion blow up and the same stories are written and the same quotes given — ‘unbelievable’, ‘scandal’, ‘incompetence’, ‘heads will roll’. Nothing changes. The closed and dysfunctional Whitehall system fights to stay closed and dysfunctional. The media caravan soon rolls on. ‘Reform’ in response to botches and scandals almost inevitably makes things even slower and more expensive — even more focus on process rather than outcomes, with the real focus being ‘we can claim to have acted properly because of our Potemkin process’. Nobody is incentivised to care about high performance and error-correction. The MPs ignore it all. Select Committees issue press releases about ‘incompetence’ but never expose the likes of Heywood to persistent investigation to figure out what has really happened and why. Nobody cares.

This culture has been encouraged by the most senior leaders. The recent Cabinet Secretary Jeremy Heywood assured us all that the civil service could easily cope with Brexit and  the civil service would handle Brexit fine and ‘definitely on digital, project management we’ve got nothing to learn from the private sector’. His predecessor, O’Donnell, made similar asinine comments. The fact that Heywood could make such a laughable claim after years of presiding over expensive debacle after expensive debacle and be universally praised by Insiders tells you all you need to know about ‘the blind leading the blind’ in Westminster. Heywood was a brilliant courtier-fixer but he didn’t care about management and operational excellence. Whitehall now incentivises the promotion of courtier-fixers, not great managers like Groves and Mueller. Management, like science, is regarded contemptuously as something for the lower orders to think about, not the ‘strategists’ at the top.

Long-term leadership from the likes of O’Donnell and Heywood is why officials know that practically nobody is ever held accountable regardless of the scale of failure. Being in charge of massive screwups is no barrier to promotion. Operational excellence is no requirement for promotion. You will often see the official in charge of some debacle walking to the tube at 4pm (‘compressed hours’ old boy) while the debacle is live on TV (I know because I saw this regularly in the DfE). The senior civil service now operates like a protected caste to preserve its power and privileges regardless of who the ignorant plebs vote for.

You can see how crazy the incentives are when you consider elections. If you look back at recent British elections the difference in the spending plans between the two sides has been a tiny fraction of the £250 billion p/a procurement and contracting budget — yet nobody ever really talks about this budget, it is the great unmentionable subject in Westminster! There’s the odd slogan about ‘let’s cut waste’ but the public rightly ignores this and assumes both sides will do nothing about it out of a mix of ignorance, incompetence and flawed incentives so big powerful companies continue to loot the taxpayer. Look at both parties now just letting the HS2 debacle grow and grow with the budget out of control, the schedule out of control, officials briefing ludicrously that the ‘high speed’ rail will be SLOWED DOWN to reduce costs and so on, all while an army of privileged looters, lobbyists, and lawyers hoover up taxpayer cash. 

And now, when Brexit means the entire legal basis for procurement is changing, do these MPs, ministers and officials finally examine it and see how they could improve? No of course not! The top priority for Heywood et al viz Brexit and procurement has been to get hapless ministers to lock Britain into the same nightmare system even after we leave the EU — nothing must disrupt the gravy train! There’s been a lot of talk about £350 million per week for the NHS since the referendum. I could find this in days and in ways that would have strong public support. But nobody is even trying to do this and if some minister took a serious interest, they would soon find all sorts of things going wrong for them until the PermSec has a quiet word and the natural order is restored…

To put the failures of politicians and official in context, it is fascinating that most of the commercial world also ignores the crucial lessons from Groves et al! Most commercial megaprojects are over-schedule, over-budget, and over-promise. The data shows that there has been little improvement over decades. (Cf. What You Should Know About Megaprojects, and Why, Flyvbjerg). And look at this  2019 article in Harvard Business Review which, remarkably, argues that managers in modern MBA programmes are taught NOT TO VALUE OPERATIONAL EXCELLENCE! ‘Operational effectiveness — doing the same thing as other companies but doing it exceptionally well — is not a path to sustainable advantage in the competitive universe’, elite managers are taught. The authors have looked at a company data and concluded that, shock horror, operational excellence turns out to be vital after all! They conclude:

‘[T]he management community may have badly underestimated the benefits of core management practices [and] it’s unwise to teach future leaders that strategic decision making and basic management processes are unrelated.’ [!]

The study of management, like politics, is not a field with genuine expertise. Like other social sciences there is widespread ‘cargo cult science’, fads and charlatans drowning out core lessons. This makes it easier to understand the failure of politicians: when elite business schools now teach students NOT to value operational excellence, when supposed management gurus like MacNamara actually push things in a worse direction, then it is less surprising people like Cameron and Heywood don’t know know which way to turn. Imagine the normal politician or senior official in Washington or London. They have almost no exposure to genuinely brilliant managers or very well run organisations. Their exposure is overwhelmingly to ‘normal’ CEOs of public companies and normal bureaucracies. As the most successful investors in world history, Buffett and Munger, have pointed out for over 50 years, many of these corporate CEOs, the supposedly ‘serious people’, don’t know what they are doing and have terrible incentives.

But surely if someone recently created something unarguably massively world-changing,  like inventing the internet and personal computing, then everyone would pay attention, right? WRONG! I wrote this (2018) about the extraordinary ARPA-PARC episode, which created much of the ecosystem for interactive personal computing and the internet and provided a model for how to conduct high-risk-high-payoff technology research.

There is almost no research funded on ARPA-PARC principles worldwide. ARPA was deliberately made less like what it was like when it created the internet. The man most responsible for PARC’s success, Robert Taylor, was fired and the most effective team in the history of computing research was disbanded. XEROX notoriously could not overcome its internal incentive problems and let Steve Jobs and Bill Gates develop the ideas. Although politicians love giving speeches about ‘innovation’ and launching projects for PR, governments subsequently almost completely ignored the lessons of how to create superproductive processes and there are almost zero examples of the ARPA-PARC approach in the world today (an interesting partial exception is Janelia). Whitehall, as a subset of its general vandalism towards science, has successfully resisted all attempts at learning from ARPA for decades and this has been helped by the attitude of leading scientists themselves whose incentives push them toward supporting objectively bad funding models. In science as well as politics, incentives can be destructive and stop learning. As Alan Kay, one of the crucial PARC researchers, wrote:

‘The most interesting thing has been the contrast between appreciation/exploitation of the inventions/contributions versus the almost complete lack of curiosity and interest in the processes that produced them… [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”.’

Or as George Mueller said later in life about the institutional imperative and project failures:

‘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.

So, on one hand, radical improvements in non-military spheres would be a wonderful free lunch. We simply apply old lessons, scale them up with technology and there are massive savings for free.

But wouldn’t it be ironic if we don’t do this — instead, we keep our dysfunctional systems for non-military spheres and carry on the waste, failure and corruption but we channel the Cold War and, in the atmosphere of an arms race, America and China apply the lessons from Groves, Schreiver and Mueller but to military AI procurement?!

Not everybody has unlearned the lessons from Groves and Mueller…

*

China: a culture of learning from systems management

‘All stable processes we shall predict. All unstable processes we shall control.’ von Neumann.

In Science there was an interesting article on Qian Xuesen, the godfather of China’s nuclear and space programs which also had a profound affect on ideas about government. Qian studied in California at Caltech where he worked with the Hungarian mathematician Theodore von Kármán who co-founded the Jet Propulsion Laboratory (JPL) which worked on rockets after 1945.

In the West, systems engineering’s heyday has long passed. But in China, the discipline is deeply integrated into national planning. The city of Wuhan is preparing to host in August the International Conference on Control Science and Systems Engineering, which focuses on topics such as autonomous transportation and the “control analysis of social and human systems.” Systems engineers have had a hand in projects as diverse as hydropower dam construction and China’s social credit system, a vast effort aimed at using big data to track citizens’ behavior. Systems theory “doesn’t just solve natural sciences problems, social science problems, and engineering technology problems,” explains Xue Huifeng, director of the China Aerospace Laboratory of Social System Engineering (CALSSE) and president of the China Academy of Aerospace Systems Science and Engineering in Beijing. “It also solves governance problems.”

The field has resonated with Chinese President Xi Jinping, who in 2013 said that “comprehensively deepening reform is a complex systems engineering problem.” So important is the discipline to the Chinese Communist Party that cadres in its Central Party School in Beijing are required to study it. By applying systems engineering to challenges such as maintaining social stability, the Chinese government aims to “not just understand reality or predict reality, but to control reality,” says Rogier Creemers, a scholar of Chinese law at the Leiden University Institute for Area Studies in the Netherlands…

‘In a building flanked by military guards, systems scientists from CALSSE sit around a large conference table, explaining to Science the complex diagrams behind their studies on controlling systems. The researchers have helped model resource management and other processes in smart cities powered by artificial intelligence. Xue, who oversees a project named for Qian at CALSSE, traces his work back to the U.S.-educated scientist. “You should not forget your original starting point,” he says…

‘The Chinese government claims to have wired hundreds of cities with sensors that collect data on topics including city service usage and crime. At the opening ceremony of China’s 19th Party Congress last fall, Xi said smart cities were part of a “deep integration of the internet, big data, and artificial intelligence with the real economy.”… Xue and colleagues, for example, are working on how smart cities can manage water resources. In Guangdong province, the researchers are evaluating how to develop a standardized approach for monitoring water use that might be extended to other smart cities.

‘But Xue says that smart cities are as much about preserving societal stability as streamlining transportation flows and mitigating air pollution. Samantha Hoffman, a consultant with the International Institute for Strategic Studies in London, says the program is tied to long-standing efforts to build a digital surveillance infrastructure and is “specifically there for social control reasons” (Science, 9 February, p. 628). The smart cities initiative builds on 1990s systems engineering projects — the “golden” projects — aimed at dividing cities into geographic grids for monitoring, she adds.

‘Layered onto the smart cities project is another systems engineering effort: China’s social credit system. In 2014, the country’s State Council outlined a plan to compile data on individuals, government officials, and companies into a nationwide tracking system by 2020. The goal is to shape behavior by using a mixture of carrots and sticks. In some citywide and commercial pilot projects already underway, individuals can be dinged for transgressions such as spreading rumors online. People who receive poor marks in the national system may eventually be barred from travel and denied access to social services, according to government documents…

‘Government documents refer to the social credit system as a “social systems engineering project.” Details about which systems engineers consulted on the project are scant. But one theory that may have proved useful is Qian’s “open complex giant system,” Zhu says. A quarter-century ago, Qian proposed that society is a system comprising millions of subsystems: individual persons, in human parlance. Maintaining control in such a system is challenging because people have diverse backgrounds, hold a broad spectrum of opinions, and communicate using a variety of media, he wrote in 1993 in the Journal of Systems Engineering and Electronics. His answer sounds like an early road map for the social credit system: to use then-embryonic tools such as artificial intelligence to collect and synthesize reams of data. According to published papers, China’s hard systems scientists also use approaches derived from Qian’s work to monitor public opinion and gauge crowd behavior

‘Hard systems engineering worked well for rocket science, but not for more complex social problems, Gu says: “We realized we needed to change our approach.” He felt strongly that any methods used in China had to be grounded in Chinese culture.

‘The duo came up with what it called the WSR approach: It integrated wuli, an investigation of facts and future scenarios; shili, the mathematical and conceptual models used to organize systems; and renli. Though influenced by U.K. systems thinking, the approach was decidedly eastern, its precepts inspired by the emphasis on social relationships in Chinese culture. Instead of shunning mathematical approaches, WSR tried to integrate them with softer inquiries, such as taking stock of what groups a project would benefit or harm. WSR has since been used to calculate wait times for large events in China and to determine how China’s universities perform, among other projects…

‘Zhu … recently wrote that systems science in China is “under a rationalistic grip, with the ‘scientific’ leg long and the democratic leg short.” Zhu says he has no doubt that systems scientists can make projects such as the social credit system more effective. However, he cautions, “Systems approaches should not be just a convenient tool in the expert’s hands for realizing the party’s wills. They should be a powerful weapon in people’s hands for building a fair, just, prosperous society.”’

In Open Complex Giant System (1993), Qian Xuesen compares the study of physics, where large complex systems can be studied using the phenomenally successful tools of  statistical mechanics, and the study of society which has no such methods. He describes an overall approach in which fields spanning physical sciences, study of the mind, medicine, geoscience and so on must be integrated in a sort of uber-field he calls ‘social systems engineering‘.

‘Studies and practices have clearly proved that the only feasible and effective way to treat an open complex giant system is a metasynthesis from the qualitative to the quantitative, i.e. the meta—synthetic engineering method. This method has been extracted, generalized and abstracted from practical studies…’

This involves integrating: scientific theories, data, quantitative models, qualitative practical expert experience into ‘models built from empirical data and reference material, with hundreds and thousands of parameters’ then simulated.

This is quantitative knowledge arising from qualitative understanding. Thus metasynthesis from qualitative to quantitative approach is to unite organically the expert group, data, all sorts of information, and the computer technology, and to unite scien- tific theory of various disciplines and human experience and knowledge.’

He gives some examples and gives this diagram as a high level summary:

Screenshot 2019-02-22 17.31.33

So, China is combining:

  • A massive ~$150 billion data science/AI investment program with the goal of global leadership in the science/technology and economic dominance.
  • A massive investment program in associated science/technology such as quantum information/computing.
  • A massive domestic surveillance program combining AI, facial recognition, genetic identification, the ‘social credit system’ and so on.
  • A massive anti-access/area denial military program aimed at America/Taiwan.
  • A massive technology espionage program that, for example, successfully stole the software codes for the F-35.
  • A massive innovation ecosystem that rivals Silicon Valley and may eclipse it (cf. this fascinating documentary on Shenzhen).
  • The use of proven systems management techniques for integrating principles of effective action to predict and manage complex systems at large scale.

America led the development of AI technologies and has the huge assets of its universities, a tradition (weakening) of welcoming scientists (since they opened Princeton to Einstein, von Neumann and Gödel in the 1930s), and the ecosystem of places like Silicon Valley.

It is plausible that China could find a way within 15 years to find some nonlinear asymmetries that provide an edge while, channeling Marshal Ogarkov, it outthinks the Pentagon in management and operations.

*

A few interesting recent straws in the AI/robotics wind

I blogged recently about Judea Pearl. He is one of the most important scholars in the field of causal reasoning. He wrote a short paper about the limits of state-of-the-art AI systems using ‘deep learning’ neural networks — such as the AlphaGo system which recently conquered the game of GO — and how these systems could be improved. Humans can interrogate stored representations of their environment with counter-factual questions: how to instantiate this in machines? (Also economists, NB. Pearl’s statement that ‘I can hardly name a handful (<6) of economists who can answer even one causal question posed in ucla.in/2mhxKdO‘.)

In an interview he said this about self-aware robots:

‘If a machine does not have a model of reality, you cannot expect the machine to behave intelligently in that reality. The first step, one that will take place in maybe 10 years, is that conceptual models of reality will be programmed by humans. The next step will be that machines will postulate such models on their own and will verify and refine them based on empirical evidence. That is what happened to science; we started with a geocentric model, with circles and epicycles, and ended up with a heliocentric model with its ellipses.

We’re going to have robots with free will, absolutely. We have to understand how to program them and what we gain out of it. For some reason, evolution has found this sensation of free will to be computationally desirable… Evidently, it serves some computational function.

‘I think the first evidence will be if robots start communicating with each other counterfactually, like “You should have done better.” If a team of robots playing soccer starts to communicate in this language, then we’ll know that they have a sensation of free will. “You should have passed me the ball — I was waiting for you and you didn’t!” “You should have” means you could have controlled whatever urges made you do what you did, and you didn’t.

[When will robots be evil?] When it appears that the robot follows the advice of some software components and not others, when the robot ignores the advice of other components that are maintaining norms of behavior that have been programmed into them or are expected to be there on the basis of past learning. And the robot stops following them.’

A DARPA project recently published this on self-aware robots.

‘A robot that learns what it is, from scratch, with zero prior knowledge of physics, geometry, or motor dynamics. Initially the robot does not know if it is a spider, a snake, an arm—it has no clue what its shape is. After a brief period of “babbling,” and within about a day of intensive computing, their robot creates a self-simulation. The robot can then use that self-simulator internally to contemplate and adapt to different situations, handling new tasks as well as detecting and repairing damage in its own body

‘Initially, the robot moved randomly and collected approximately one thousand trajectories, each comprising one hundred points. The robot then used deep learning, a modern machine learning technique, to create a self-model. The first self-models were quite inaccurate, and the robot did not know what it was, or how its joints were connected. But after less than 35 hours of training, the self-model became consistent with the physical robot to within about four centimeters…

‘Lipson … notes that self-imaging is key to enabling robots to move away from the confinements of so-called “narrow-AI” towards more general abilities. “This is perhaps what a newborn child does in its crib, as it learns what it is,” he says. “We conjecture that this advantage may have also been the evolutionary origin of self-awareness in humans. While our robot’s ability to imagine itself is still crude compared to humans, we believe that this ability is on the path to machine self-awareness.”

‘Lipson believes that robotics and AI may offer a fresh window into the age-old puzzle of consciousness. “Philosophers, psychologists, and cognitive scientists have been pondering the nature self-awareness for millennia, but have made relatively little progress,” he observes. “We still cloak our lack of understanding with subjective terms like ‘canvas of reality,’ but robots now force us to translate these vague notions into concrete algorithms and mechanisms.”

‘Lipson and Kwiatkowski are aware of the ethical implications. “Self-awareness will lead to more resilient and adaptive systems, but also implies some loss of control,” they warn. “It’s a powerful technology, but it should be handled with care.”’

Robot paper HERE.

Press release HERE.

Recently, OpenAI, one of the world leaders in AI founded by Sam Altman and Elon Musk, announced:

‘… a large-scale unsupervised language model which generates coherent paragraphs of text, achieves state-of-the-art performance on many language modeling benchmarks, and performs rudimentary reading comprehension, machine translation, question answering, and summarization — all without task-specific training… The model is chameleon-like — it adapts to the style and content of the conditioning text. This allows the user to generate realistic and coherent continuations about a topic of their choosing… Our model is capable of generating samples from a variety of prompts that feel close to human quality and show coherence over a page or more of text… These samples have substantial policy implications: large language models are becoming increasingly easy to steer towards scalable, customized, coherent text generation, which in turn could be used in a number of beneficial as well as malicious ways.’ (bold added).

Screenshot 2019-02-15 11.48.37

OpenAI has not released the full model yet because they take safety issues seriously. Cf. this for a discussion of some safety issues and links. As the author says re some of the complaints about OpenAI not releasing the full model, when you find normal cyber security flaws you do not publish the problem immediately — that is a ‘zero day attack’ and we should not ‘promote a norm that zero-day threats are OK in AI.’ Quite. It’s also interesting that it would probably only take ~$100,000 for a resourceful individual to re-create the full model quite quickly.

A few weeks ago, Deep Mind showed that their approach to beating human champions at GO can also beat the world’s best players at StarCraft, a game of IMperfect information which is much closer to real life human competitions than perfect information games like chess and GO. OpenAI has shown something similar with a similar game, DOTA.

 

*

Moore’s Law: what if a country spends 1-10% GDP pushing such curves?

The march of Moore’s Law is entangled in many predictions. It is true that in some ways Moore’s Law has flattened out recently…

Screenshot 2018-03-12 11.55.21

… BUT specialised chips developed for machine learning and other adaptations have actually kept it going. This chart shows how it actually started long before Moore and has been remarkably steady for ~120 years (NVIDIA in the top right is specialised for deep learning)…

Screenshot 2018-03-12 11.56.15

NB. This is a logarithmic scale so makes progress seem much less dramatic than the ~20 orders of magnitude it represents.

  • Since Von Neumann and Turing led the development of the modern computer in the 1940s, the price of computation has got ~x10 cheaper every five years (so x100 per decade), so over ~75 years that’s a factor of about a thousand trillion (1015).
  • The industry seems confident the graph above will continue roughly as it has for at least another decade, though not because of continued transistor doubling rates which has reached such a tiny nanometer scale that quantum effects will soon interfere with engineering. This means ~100-fold improvement before 2030 and combined with the ecosystem of entrepreneurs/VC/science investment etc this will bring many major disruptions even without significant progress with general intelligence.
  • Dominant companies like Apple, Amazon, Google, Baidu, Alibaba etc (NB. no big EU players) have extremely strong incentives to keep this trend going given the impact of mobile computing / the cloud etc on their revenues.
  • Computers will be ~10,000 times more powerful than today for the same price if this chart holds for another 20 years and ~1 million times more powerful for the same price than today if it holds for another 30 years. Today’s multi-billion dollar supercomputer performance would be available for ~$1,000, just as the supercomputer power of a few decades ago is now available in your smartphone.

But there is another dimension to this trend. Look at this graph below. It shows the total amount of compute, in petaflop/s-days, that was used to train some selected AI projects using neural networks / deep learning.

‘Since 2012, the amount of compute used in the largest AI training runs has been increasing exponentially with a 3.5 month-doubling time (by comparison, Moore’s Law had an 18-month doubling period). Since 2012, this metric has grown by more than 300,000x (an 18-month doubling period would yield only a 12x increase)… The chart shows the total amount of compute, in petaflop/s-days, that was used to train selected results that are relatively well known, used a lot of compute for their time, and gave enough information to estimate the compute used. A petaflop/s-day (pfs-day) consists of performing 1015neural net operations per second for one day, or a total of about 1020operations. ‘ (Cf. OpenAI blog.)

Screenshot 2018-05-19 17.04.04

The AlphaZero project in the top right is the recent Deep Mind project in which an AI system (a successor to the original AlphaGo that first beat human GO champions) zoomed by centuries of human knowledge on GO and chess in about one day of training.

Many dramatic breakthroughs in machine learning, particularly using neural networks (NNs), are open source. They are scaling up very fast. They will be networked together into ‘networks of networks’ and will become x10, x100, x1,000 more powerful. These NNs will keep demonstrating better than human performance in relatively narrowly defined tasks (like winning games) but these narrow definitions will widen unpredictably.

OpenAI’s blog showing the above graph concludes:

‘Overall, given the data above, the precedent for exponential trends in computing, work on ML specific hardware, and the economic incentives at play, we think it’d be a mistake to be confident this trend won’t continue in the short term. Past trends are not sufficient to predict how long the trend will continue into the future, or what will happen while it continues. But even the reasonable potential for rapid increases in capabilities means it is critical to start addressing both safety and malicious use of AI today. Foresight is essential to responsible policymaking and responsible technological development, and we must get out ahead of these trends rather than belatedly reacting to them.’ (Bold added)

This recent analysis of the extremely rapid growth of deep learning systems tries to estimate how long this rapid growth can continue and what interesting milestones may fall. It considers 1) the rate of growth of cost, 2) the cost of current experiments, and 3) the maximum amount that can be spent on an experiment in the future. Its rough answers are:

  1. ‘The cost of the largest experiments is increasing by an order of magnitude every 1.1 – 1.4 years.
  2. ‘The largest current experiment, AlphaGo Zero, probably cost about $10M.’
  3. On the basis of the Manhattan Project costing ~1% of GDP, that gives ~$200 billion for one AI experiment. Given the growth rate, we could expect a $200B experiment in 5-6 years.
  4. ‘There is a range of estimates for how many floating point operations per second are required to simulate a human brain for one second. Those collected by AI Impacts have a median of 1018 FLOPS (corresponding roughly to a whole-brain simulation using Hodgkin-Huxley neurons)’. [NB. many experts think 1018 is off by orders of magnitude and it could easily be x1,000 or more higher.]
  5. ‘So for the shortest estimates … we have already reached enough compute to pass the human-childhood milestone. For the median estimate, and the Hodgkin-Huxley estimates, we will have reached the milestone within 3.5 years.’
  6. We will not reach the bigger estimates (~1025FLOPS) within the 10 year window.
  7. ‘The AI-Compute trend is an extraordinarily fast trend that economic forces (absent large increases in GDP) cannot sustain beyond 3.5-10 more years. Yet the trend is also fast enough that if it is sustained for even a few years from now, it will sweep past some compute milestones that could plausibly correspond to the requirements for AGI, including the amount of compute required to simulate a human brain thinking for eighteen years, using Hodgkin Huxley neurons.’

I can’t comment on the technical aspects of this but one political/historical point. I think this analysis is wrong about the Manhattan Project (MP). His argument is the MP represents a reasonable upper-bound for what America might spend. But the MP was not constrained by money — it was mainly constrained by theoretical and engineering challenges, constraints of non-financial resources and so on. Having studied General Groves’ book (who ran the MP), he does not say money was a problem — in fact, one of the extraordinary aspects of the story is the extreme (to today’s eyes) measures he took to ensure money was not a problem. If more than 1% GDP had been needed, he’d have got it (until the intelligence came in from Europe that the Nazi programme was not threatening).

This is an important analogy. America and China are investing very heavily in AI but nobody knows — are there places at the edge of ‘breakthroughs with relatively narrow applications’ where suddenly you push ‘a bit’ and you get lollapalooza results with general intelligence? What if someone thinks — if I ‘only’ need to add some hardware and I can muster, say, 100 billion dollars to buy it, maybe I could take over the world? What if they’re right?

I think it is therefore more plausible to use the US defence budget at the height of the Cold War as a ‘reasonable estimate’ for what America might spend if they feel they are in an existential struggle. Washington knows that China is putting vast resources into AI research. If it starts taking over from Deep Mind and OpenAI as the place where the edge-of-the-art is discovered, then it WILL soon be seen as an existential struggle and there would rapidly be political pressures for a 1950s/1960s style ‘extreme’ response. So a reasonable upper bound might be at least 5-8 times bigger than 1% of GDP.

Further, unlike the nuclear race, an AGI race carries implications of not just ‘destroy global civilisation and most people’ but ‘potentially destroys ABSOLUTELY EVERYTHING not just on earth but, given time and the speed of light, everywhere’ — i.e potentially all molecules re-assembled in the pursuit of some malign energy-information optimisation process. Once people realise just how bad AGI could go if the alignment problem is not solved (see below), would it not be reasonable to assume that even more money than ~8% GDP will be found if/when this becomes a near-term fear of politicians?

Some in Silicon Valley who already have many billions at their disposal are already calculating numbers for these budgets. Surely people in Chinese intelligence are doodling the same as they listen to the week’s audio of Larry talking to Demis…?

*

General intelligence and safety

‘[R]ational systems exhibit universal drives towards self-protection, resource acquisition, replication and efficiency. Those drives will lead to anti-social and dangerous behaviour if not explicitly countered. The current computing infrastructure would be very vulnerable to unconstrained systems with these drives.’ Omohundro.

Shane Legg, co-founder and chief scientist of Deep Mind, said publicly a few years ago that there is a 50% probability that we will achieve human level AI by 2028, a 90% probability by 2050, and ‘I think human extinction will probably occur‘. Given Deep Mind’s progress since he said this it is surely unlikely he thinks the odds now are lower than 50% by 2028. Some at the leading edge of the field agree.

‘I think that within a few years we’ll be able to build an NN-based [neural network] AI (an NNAI) that incrementally learns to become at least as smart as a little animal, curiously and creatively learning to plan, reason and decompose a wide variety of problems into quickly solvable sub-problems. Once animal-level AI has been achieved, the move towards human-level AI may be small: it took billions of years to evolve smart animals, but only a few millions of years on top of that to evolve humans. Technological evolution is much faster than biological evolution, because dead ends are weeded out much more quickly. Once we have animal-level AI, a few years or decades later we may have human-level AI, with truly limitless applications. Every business will change and all of civilisation will change…

In 2050 there will be trillions of self-replicating robot factories on the asteroid belt. A few million years later, AI will colonise the galaxy. Humans are not going to play a big role there, but that’s ok. We should be proud of being part of a grand process that transcends humankind.’ Schmidhuber, one of the pioneers of ML, 2016.

Others have said they believe that estimates of AGI within 15-30 years are unlikely to be right. Two of the smartest people I’ve ever spoken to are physicists who understand the technical details and know the key researchers and think that dozens of Nobel Prize scale ideas will probably be needed before AGI happens and it is more likely that the current wave of enthusiasm with machine learning/neural networks will repeat previous cycles in science (e.g with quantum computing 20 years ago) — great enthusiasm, the feeling that all barriers are quickly falling, then an increasingly obvious plateau, spreading disillusion, a search for new ideas, then a revival of hope and so on. They would bet more on a 50-80 year than a 20 year scale.

Of top people I have spoken to and/or followed their predictions, it’s clear that there is a consensus that mainstream economic analysis (which is the foundation of politicians’ and media discussion) seriously underestimates the scale and speed of social/economic/military/political disruption that narrow AI/automation will soon cause. But predictions on AGI are unsurprisingly all over the place.

Chart: predictions on AGI timelines (When Will AI Exceed Human Performance? Evidence from AI Experts)

Screenshot 2019-02-28 10.00.31

Screenshot 2019-02-28 10.22.40

Many argue there even if Moore’s Law continues for 30 years (millionfold performance improvement) this may mean nothing significant for general intelligence, even if narrow AI transforms the world in many ways. Some experts think that estimates of the human brain’s computational capacity widely believed in the computer science world are actually orders of magnitude wrong. We still don’t know much about basics of the brain such as how long-term memories are formed. Maybe the brain’s processes will be much more resistant to understanding than ‘optimists’ assume.

But maybe relatively few big new ideas are needed to create world-changing capabilities. ‘Just’ applying great engineering and more resources to existing ideas allowed Deep Mind to blow past human performance metrics. I obviously cannot judge competing expert views but from a political perspective we know for sure that there is inherent uncertainty about how we discover new knowledge and this means we are bound to be surprised in all sorts of ways. We know that even brilliant researchers working right at the edge of progress are often clueless about what will happen quite soon and cannot reliably judge ‘is it less than 1% or more like 20% probability?’ questions. For example:

‘In 1901, two years before helping build the first heavier-than-air flyer, Wilbur Wright told his brother that powered flight was fifty years away. In 1939, three years before he personally oversaw the first critical chain reaction in a pile of uranium bricks, Enrico Fermi voiced 90% confidence that it was impossible to use uranium to sustain a fission chain reaction.’ (Yudkowsky)

Fermi’s experience suggests we should be extremely careful and put more resources into thinking very hard about how to minimise risks viz both narrow and general AI.

Those right at the edge of genetic engineering, such as George Church and Kevin Esvelt, are pushing for their field to be forcibly opened up to make it safer. As they argue, the current scientific approach and incentive system is essentially a ‘blind search algorithm’ in which small teams work in secret without being able to predict the consequences of their work and cannot be warned by those who do understand. A blind search algorithm is a bad approach for things like bioweapons that can destroy billions of lives and it is what we now have. The same argument applies to AGI.

We also know that political people and governments are slow to cope with major technological disruptions. Just look at TV. It’s been dominating politics since the 1950s. It is roughly 70 years old. Many politicians still do not understand it well. The UK state and political parties are in many ways much less sophisticated in its use of TV than groups like Hezbollah. This is even more true of social media. Also look at how unfounded conspiracy theories about fake news and social media viz the referendum and Trump have gripped much of the ‘educated’ class that thinks they see through fake news that fools the uneducated! Journalists are awarded THE ORWELL AWARD(!) for spreading fake news about fake news (and it’s not ‘lies’, they actually believe what they say)! (My experience is it’s much easier to fool people about politics if they have a degree than if they don’t because those with a degree tend to spend so much more energy fooling themselves.) This is not encouraging particularly if one considers that politicians are directly incentivised to understand technologies like TV and internet polling for their own short-term interests yet most don’t.

From cars to planes it has taken time for us to work out how to adapt to new things that can kill us. Given that 1) conventional research is ‘a blind search algorithm’, 2) our politicians are behind the curve on 70 year-old technologies and 3) there is little prospect of this changing without huge changes to conventional models of politics, we must ask another question about secrecy v openness and centralised vs decentralised architectures.

One of the leaders of the 3D printing / FabLab revolution wrote this comparing the closed v open models of security:

‘The history of the Internet has shown that security through obscurity doesn’t work. Systems that have kept their inner workings a secret in the name of security have consistently proved more vulnerable than those that have allowed themselves to be examined — and challenged — by outsiders. The open protocols and programs used to protect Internet communications are the result of ongoing development and testing by a large expert community. Another historical lesson is that people, not technology, are the most common weakness when it comes to security. No matter how secure a system is, someone who has access to it can always be corrupted, wittingly or otherwise. Centralized control introduces a point of vulnerability that is not present in a distributed system.’ (Bold added)

As we saw above, the centralised approach has been a disaster for nuclear weapons and we survived by fluke. Overall the history of nuclear security is surely a very relevant and bad signal for AI safety. I would bet a lot that Deep Mind et al are all hacked and spied on by China and Russia (at least) so I think it’s safest to plan on the assumption that dangerous breakthroughs will leak almost instantly and could be applied by the sort of people who spy for intel agencies. So it is natural to ask, should we take an open/decentralised approach towards possible AGI?

(Tangential thought experiment: if you were in charge of an organisation like the KGB, why would you not hack hedge funds like Renaissance Technologies and use the information for your own ‘black’ hedge fund and thus dodge the need for arguments over funding (a ‘virtuous’ circle of espionage, free money, resources for more effective R&D and espionage plus it minimises the need for irritating interactions with politicians)? How hard would it be to detect such activity IF done with intelligent modesty? Given someone can hack the NSA without their identity being revealed, why would they not be hacking Renaissance and Deep Mind, with a bit of help from a Milla Jovovich lookalike whose reading a book on n-dimensional string theory at the bar when that exhausted physics PhD with the access codes staggers in to relax?)

This seems to collide with another big problem — the alignment problem.

Stuart Russell, one of the world’s leading researchers, is one of those who has been very forceful about the fundamental importance of this: how do we GUARANTEE that entities more intelligent than us are aligned with humanity’s interests?

‘One [view] is: It’ll never happen, which is like saying we are driving towards the cliff but we’re bound to run out of gas before we get there. And that doesn’t seem like a good way to manage the affairs of the human race. And the other [view] is: Not to worry — we will just build robots that collaborate with us and we’ll be in human-robot teams. Which begs the question: If your robot doesn’t agree with your objectives, how do you form a team with it?’ .

Eliezer Yudkowsky, one of the few working on the alignment problem, described the difficulty:

‘How do you encode the goal functions of an A.I. such that it has an Off switch and it wants there to be an Off switch and it won’t try to eliminate the Off switch and it will let you press the Off switch, but it won’t jump ahead and press the Off switch itself? And if it self-modifies, will it self-modify in such a way as to keep the Off switch? We’re trying to work on that. It’s not easy… When you’re building something smarter than you, you have to get it right on the first try.

So, we know centralised systems are very vulnerable and decentralised systems have advantages, but with AGI we also have to fear that we have no room for the trial-and-error of decentralised internet style security architectures — ‘you have to get it right on the first try’. Are we snookered?! And of course there is no guarantee it is even possible to solve the alignment problem. When you hear people in this field describing ideas about ‘abstracting human ethics and encoding them’ one wonders if solving the alignment problem might prove even harder than AGI — maybe only an AGI could solve it…

Given the media debate is dominated by endless pictures of the Terminator and politicians are what they are, researchers are, understandably, extremely worried about what might happen if the political-media system makes a sudden transition from complacency to panic. After all, consider the global reaction if reputable scientists suddenly announced they have discovered plausible signals that super-intelligent aliens will arrive on earth within 30 years: even when softened by caveats, such a warning would obviously transform our culture (in many ways positively!). As Peter Thiel has said, creating true AGI is a close equivalent to the ‘super-intelligent aliens arriving on earth’ scenario and the most important questions are not economic but political, and in particular: are they friendly and can we stop them eliminating us by design, bad luck, or indifference?

Further, in my experience extremely smart technical people are often naive about politics. They greatly over-estimate the abilities of prime ministers and presidents. They greatly under-estimate the incentive problems and the degree of focus that is required to get ANYTHING done in politics. They greatly exaggerate the potential for ‘rational argument’ to change minds and wrongly assume somewhere at the top of power ‘there must be’ a group of really smart people working on very dangerous problems who have real clout. Further, everybody thinks they understand ‘communication’ but almost nobody does. We can see from recent events that even the very best engineering companies like Facebook and Google can not just make huge mistakes with the political/communication world but not learn (Facebook hiring Clegg was a sign of deep ignorance inside Facebook about their true problems). So it’s hard to be optimistic about the technical people educating the political people even assuming the technical people make progress with safety.

Hypothesis: 1) minimising nuclear/bio/AI risks and the potential for disastrous climate change requires a few very big things to change roughly simultaneously (‘normal’ political action will not be enough) and 2) this will require a weird alliance between a) technical people, b) political ‘renegades’, c) the public to ‘surround’ political Insiders locked into existing incentives:

  1. Different ‘models for effective action’ among powerful people, which will only happen if either (A) some freak individual/group pops up, probably in a crisis environment or (B) somehow incentives are hacked. (A) can’t be relied on and (B) is very hard.
  2. A new institution with global reach that can win global trust and support is needed. The UN is worse than useless for these purposes.
  3. Public opinion will have to be mobilised to overcome the resistance of political Insiders, for example, regarding the potential for technology to bring very large gains ‘to me’ and simultaneously avert extreme dangers. This connects to the very widespread view that a) the existing economic model is extremely unfair and b) this model is sustained by a loose alliance of political elites and corporate looters who get richer by screwing the rest of us.

I have an idea about a specific project, mixing engineering/economics/psychology/politics, that might do this and will blog on it separately.

I suspect almost any idea that could do 1-3 will seem at least weird but without big changes, we are simply waiting for the law of averages to do its thing. We may have decades for AGI and climate change but we could collide with the WMD law of averages tomorrow so, impractical as this sounds, it seems to me people have to try new things and risk failure and ridicule.

Please leave comments/corrections below…

Further reading

An excellent essay by Ian Hogarth, AI nationalism, which covers some of the same ground but is written by someone with deep connections to the field whereas I am extremely non-expert but interested.

AI safety is one of those subjects that is taken extremely seriously by a tiny number of people that has almost zero overlap with the policy/government world. If interested, then follow @ESYudkowsky. Cf. Intelligence Explosion Microeconomics, Yudkowsky.

Drones go to work, Chris Anderson (one of the pioneers of commercial drones). This explains the economics and how drones are transforming industries.

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

Autonomous technology and the greater human good. Omohundro. ‘Military and economic pressures are driving the rapid development of autonomous systems. We show that these systems are likely to behave in anti-social and harmful ways unless they are very carefully designed. Designers will be motivated to create systems that act approximately rationally and rational systems exhibit universal drives towards self-protection, resource acquisition, replication and efficiency. Those drives will lead to anti-social and dangerous behaviour if not explicitly countered. The current computing infrastructure would be very vulnerable to unconstrained systems with these drives. We describe the use of formal methods to create provably safe but limited autonomous systems. We then discuss harmful systems and how to stop them. We conclude with a description of the ‘Safe-AI Scaffolding Strategy’ for creating powerful safe systems with a high confidence of safety at each stage of development.’ I strongly recommend reading this paper if interested in this blog.

Can intelligence explode? Hutter.

Read this 1955 essay by von Neumann ‘Can we survive technology?. VN was involved in the Manhattan Project, inventing computer science, game theory and much more. This essay explored the essential problem that the scale and speed of technological change have suddenly blown past political institutions. ‘For progress there is no cure…’

The recent Science piece on Qian Xuesen and systems management is HERE.

Qian Xuesen – Open Complex Giant System, 1993.

I wrote this (2018) about the extraordinary ARPA-PARC episode, which created much of the ecosystem for interactive personal computing and the internet and provided a model for how to conduct high-risk-high-payoff technology research.

I wrote this Jan 2017 on  systems management, von Neumann, Apollo, Mueller etc. It provides a checklist for how to improve Whitehall systematically and deliver complex projects like Brexit.

The Hollow Men (2014) that summarised the main problems of Westminster and Whitehall.

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

My review of Allison’s book on the US-China contest and some thoughts on how Bismarck would see it.

On ‘Expertise’ from fighting and physics to economics, politics and government.

I blogged a few links to AI papers HERE.

On the referendum #24F: Another central claim of the Observer/Channel 4 conspiracy blows up

Yesterday I posted Facebook’s evidence showing that the central allegation of the Observer/Channel 4 conspiracy theory — that Vote Leave used the infamous data obtained by Cambridge Analytica — was provably false.

Today, the Spectator blows up other claims.

The most striking bit of ‘evidence’ the Observer produced recently was a video which they claimed showed the ‘destruction of evidence’ and a ‘coverup’. At the time I said that it did not show who or what the Observer claimed. (I won’t post it to avoid spreading fake news.)

The Spectator carries a statement from Vote Leave directors sent to the Electoral Commission proving that the Observer claims are entirely false:

‘This statement concerns a serious allegation against Ms Victoria Woodcock recently made by Shahmir Sanni et al, which we have reviewed urgently and needed to respond to more immediately, alleging what was variously described as data deletion on, or removal of access permissions from, Vote leave’s ‘BeLeave’ folder on March 17th 2017. We are now in a position to respond on this matter following a forensic review of Vote Leave’s Google Drive.

Ms Woodcock did not on that date access, delete, amend, or change permissions for any data or files on the BeLeave folder, as alleged by the so-called whistle-blowers and as is purported to be shown in the GIF published by The Observer. Allegations that claim she did are false and are based on misconceptions and misunderstandings of how Google Drive works.

‘Prior to March 17th 2017 Ms Woodcock was the Data Controller for Vote Leave and, in preparation for closedown, the majority of documents on its drive had been incorporated by her into a super-folder in her name. As a next step in the closedown process, it was decided that Ms Woodcock should hand over her responsibilities on March 17th 2017, and accordingly, on that date, her access to the Vote Leave Gdrive was removed. Later that day, continuing the closedown, at the direction of the Board and as a part of a standard data protection exercise, permissions were removed from folders across the Gdrive (of which the BeLeave folder was a part) for a group of high-level users (this group included, but was by no means merely, Ms Woodcock and the other two individuals shown in the GIF).

‘Ms Woodcock’s name appears as the user making the changes because she had been the super-administrator and data controller, so the “Victoria Woodcock” account was a convenient one to use, to achieve best visibility across the G Drive; the changes were in fact made by an authorised Vote Leave administrator, using her account, at a time when Ms Woodcock had had her access removed so would therefore not even have known that this activity was taking place. Independent IT consultants have verified that no BeLeave files were deleted from the folder. Permissions were removed, not by Victoria Woodcock; from folders across the drive, not just the BeLeave folder; and for a wider group than the three individuals shown in the Observer’s GIF.

‘These allegations against Ms Woodcock are therefore groundless.’ (Emphasis added)

In short, VW was removed from access to the drive before the video was taken, the video does NOT show her, it shows a different person to the Observer’s claim doing something completely different to the Observer’s claim, and nothing was deleted. (I was removed from access to this system long before 17 March 2017.)

Everything the Observer/C4 claimed about this GIF/video was wrong. No responsible media organisation should repeat the libellous allegations from Observer/C4. 

*

2 other interesting snippets re the EU today.

1/ The GDPR legislation is horrific. One of the many advantages of Brexit is we will soon be able to bin such idiotic laws. We will be able to navigate between America’s poor protection of privacy and the EU’s hostility to technology and entrepreneurs. It doesn’t matter that this Government will sign up to a shockingly bad deal that purports to keep us in such stuff. The deal will be binned. With Brexit, it is the long-term that counts most — not what ministers like DD say and sign.

Hacks should ask around big companies for lunatic documents circulating to staff giving them directions on how to behave under GDPR to see what I mean. From baby photos to sickies, hacks will have a field day.

Also note that Whitehall is happy to spend huge amounts of time and effort passing GDPR and associated bullshit while stalling on preparations to make the UK a ‘third country’ under EU law and claiming to ministers that preparations to leave the EU are ‘illegal’ (and requiring they make written notices to Permanent Secretaries and other classic moves of the normal bureaucratic chess match). This Government is so comical that we will soon leave the EU without preparing to leave the EU AND we will not even prepare to leave the EU after we have already left because officials continue to argue such preparations are illegal 2019-2020 and DD has already conceded the argument. (Officials use various devices including our supposed obligations under A50.) If I had wanted to create a story to demonstrate my long-running claims about Whitehall, I could hardly have bettered this.

Whitehall spends much more time implementing new EU law than preparing to get out of EU law and almost all Ministers have so little grip of their departments, and have so little support from May (herself an avatar for Heywood and Robbins), that they meekly acquiesce. The Cabinet even now has never insisted on a single discussion with responsible officials over preparations — a dereliction of duty that will be seen by history as similar to the failure of the pre-World War I Cabinet to have discussions about UK military commitments to France.

2/ During the campaign VL warned that the ECJ would use the Charter of Fundamental Rights (NB. NOT the ECHR/HRA) to interfere with UK intelligence services and police. Cameron and Osborne claimed this was ‘lies’ even though it was perfectly obvious this would happen to anybody reading ECJ cases.

An example of what we warned about is HERE. Today’s judgment undermines the Investigatory Powers Act 2016 using the Charter of Fundamental Rights. The Conservatives used to claim that these powers were vital for national security and fighting crime.

The ECJ will soon decide in the Privacy International on further aspects of the Five Eyes Agreement. During the campaign, Cameron, Osborne, Grieve and their collaborators claimed EU law would have no effect on the Five Eyes agreement. An example of the repeated dishonesty by Grieve on this subject is HERE. Grieve claimed that Gove’s statements during the referendum were wrong, ‘unfounded and indeed untenable’ and so on. It is Grieve’s claims on the Charter that are factually and legally wrong and ‘untenable’ in the light of the actual law and actual ECJ decisions. Grieve’s repeated bullshit on this issue should be called out by broadcast interviews. The treatment of him as an impartial expert is absurd. He is no expert and he is repeatedly dishonest on the subject.

Today’s judgment will be one of many if we remain ‘aligned’ to EU law and the Charter. Are MPs going to win the argument that we will leave the EU but leave the ECJ in charge of our response to terrorism? Not long-term. (And this is why we will win a referendum on the ECHR too.)

Of course, every single bit of advice that VL gave pre-referendum about what to do has been ignored by the Conservative government and MPs generally, from how to handle A50 to the need for investment in the NHS to this issue.

VL said that there should be ‘notwithstanding ECA1972’ legislation to remove the ECJ from any interference with the intelligence services, which would be strongly supported by Leave and Remain voters. Instead, the Government will accept this judgment and do nothing about it despite their previous promises. Ministers will, as usual, be easily bamboozled by officials waving ‘legal advice’ at them, just as Heywood bamboozled them into their catastrophic decisions on A50 by waving ‘legal advice’ at them.

This is just one small example of how extremely rubbish this government is and why it is vital that there are radical changes as soon as Brexit happens next March. This government, Parliament, and Whitehall generally are not remotely able to cope with the hard reboot of Brexit. Vote Leave warned them they could not do Brexit with the normal dysfunctional management processes of Whitehall. They ignored this advice and have collapsed into repeated and inescapable shambles.

Many in SW1 think that willpower can bounce them from the actual branch of reality we are on to a neighbouring branch of the multiverse where they can escape the referendum, just as many Brexit supporters think willpower can bounce them into a branch of the multiverse where we can escape all the disastrous effects of the May government. Both are wrong. ‘Reality cannot be fooled’ indefinitely. A hard rain is coming for SW1…

On the referendum #21: Branching histories of the 2016 referendum and ‘the frogs before the storm’

‘Politics is gambling for high stakes with other people’s money… Politics is a job that can be compared with navigation in uncharted waters. One has no idea how the weather or the currents will be or what storms one is in for. In politics, there is the added fact that one is largely dependent on the decisions of others, decisions on which one was counting and which then do not materialise; one’s actions are never completely one’s own. And if the friends on whose support one is relying change their minds, which is something that one cannot vouch for, the whole plan miscarries… One’s enemies one can count on – but one’s friends!’ Bismarck.

‘The most important thing is not to fool yourself and you are the easiest person to fool.’ Feynman. 

‘He lies like an eyewitness.’ Russian proverb.

In January 2014 I left the Department for Education and spent the next 18 months away from politics. A few days after the 2015 election I wrote a blog about Michael Gove’s new job touching on the referendum. When I wrote it I assumed I would carry on studying and would not be involved in it. About ten days later I was asked by an assortment of MPs, rich businessmen, and campaigners including Matthew Elliott to help put together an organisation that could fight the referendum. I was very reluctant and prevaricated but ended up agreeing. I left my happy life away from SW1 and spent eight weeks biking around London persuading people to take what was likely to be a car crash career decision – to quit their jobs and join a low probability proposition: hacking the political system to win a referendum against almost every force with power and money in politics. In September we had an office, in October ‘Vote Leave’ went public, in April we were designated the official campaign, 10 weeks later we won.

Why and how? The first draft of history was written in the days and weeks after the 23 June and the second draft has appeared over the past few weeks in the form of a handful of books. There is no competition between them. Shipman’s is by far the best and he is the only one to have spoken to key people. I will review it soon. One of his few errors is to give me the credit for things that were done by others, often people in their twenties like Oliver Lewis, Jonny Suart, and Cleo Watson who, unknown outside the office, made extreme efforts and ran rings around supposed ‘experts’. His book has encouraged people to exaggerate greatly my importance.

I have been urged by some of those who worked on the campaign to write about it. I have avoided it, and interviews, for a few reasons (though I had to write one blog to explain that with the formal closing of VL we had made the first online canvassing software that really works in the UK freely available HERE). For months I couldn’t face it. The idea of writing about the referendum made me feel sick. It still does but a bit less.

For about a year I worked on this project every day often for 18 hours and sometimes awake almost constantly. Most of the ‘debate’ was moronic as political debate always is. Many hours of life I’m never getting back were spent dealing with abysmal infighting among dysfunctional egomaniacs while trying to build a ~£10 million startup in 10 months when very few powerful people thought the probability of victory was worth the risk of helping us. (Two rare heroes who put up a lot of their own money and supported the team were Peter Cruddas and Stuart Wheeler.) Many of those involved regarded their TV appearances as by far the most important aspect of the campaign. Many regarded Vote Leave as ‘the real enemy’.

It is hard to explain the depth of TV derangement that gobbles up SW1 souls. Much of politics involves very similar tragi-comic scenes re-created by some of the basic atoms of human nature – fear, self-interest and vanity. The years, characters, and contexts change, the atoms shuffle, but the stories are the same year after year, century after century. Delusions and vanity dominate ‘rationality’ and ‘public service’. Progress, when it comes, is driven by the error-correcting institutions of science and markets when political institutions limit the damage done by decision makers at the apex of centralised hierarchies. It rarely comes from those people, and, when it does, it is usually accidental or incidental to their motives.

Discussions about things like ‘why did X win/lose?’ are structured to be misleading and I could not face trying to untangle everything. There are strong psychological pressures that lead people to create post facto stories that seem to add up to ‘I always said X and X happened.’ Even if people do not think this at the start they rapidly construct psychologically appealing stories that overwrite memories. Many involved with this extraordinary episode feel the need to justify themselves and this means a lot of rewriting of history. I also kept no diary so I have no clear source for what I really thought other than some notes here and there. I already know from talking to people that my lousy memory has conflated episodes, tried to impose patterns that did not actually exist and so on – all the usual psychological issues. To counter all this in detail would require going through big databases of emails, printouts of appointment diaries, notebooks and so on, and even then I would rarely be able to reconstruct reliably what I thought. Life’s too short.

I’ve learned over the years that ‘rational discussion’ accomplishes almost nothing in politics, particularly with people better educated than average. Most educated people are not set up to listen or change their minds about politics, however sensible they are in other fields. But I have also learned that when you say or write something, although it has roughly zero effect on powerful/prestigious people or the immediate course of any ‘debate’, you are throwing seeds into a wind and are often happily surprised. A few years ago I wrote something that was almost entirely ignored in SW1 but someone at Harvard I’d never met read it. This ended up having a decisive effect on the referendum.

A warning. Politics is not a field which meets the two basic criteria for true expertise (see below). An effect of this is that arguments made by people who win are taken too seriously. People in my position often see victory as confirmation of ideas they had before victory but people often win for reasons they never understand or even despite their own efforts. Cameron’s win in 2015 was like this – he fooled himself about some of the reasons why he’d won and this error contributed to his errors on the referendum. Maybe Leave won regardless of or even despite my ideas. Maybe I’m fooling myself like  Cameron. Some of my arguments below have as good an empirical support as is possible in politics (i.e. not very good objectively) but most of them do not even have that. Also, it is clear that almost nobody agrees with me about some of my general ideas. It is more likely that I am wrong than 99% of people who work in this field professionally. Still, cognitive diversity is inherently good for political analysis so I’ll say what I think and others will judge if there’s anything to learn.

Apologies for the length but I didn’t have time to make it shorter. The next ones will be short.

*

Reality has branching histories, not ‘a big why’

Much political analysis revolves around competing simple stories based on one big factor such that, in retrospect, ‘it was always clear that immigration would trump economic interest / Cameron’s negotiation was never going to be enough / there is an unstoppable populist tide’, and so on. Alternatives are quickly thought to have been impossible (even if X argued the exact opposite repeatedly). The big event must have had an equally big single cause. Confirmation bias kicks in and evidence seeming to suggest that what actually happened would happen looms larger. People who are quite wrong quickly persuade themselves they were ‘mostly right’ and ‘had a strong feeling’ unlike, of course, the blind fools around them. Soon our actual history seems like the only way things could have played out. Brexit had to happen. Trump had to win.

You see these dynamics all the time in historical accounts. History tends to present the 1866 war between Prussia and Austria as almost inevitable but historians spend much less time on why Bismarck pulled back from war in 1865 and how he might have done the same in 1866 (actually he prepared the ground so he could do this and he kept the option open until the last minute). The same is true about 1870. When some generals tried to bounce him into a quick preventive war against Russia in the late 1880s he squashed them flat warning against tying the probability of a Great Power war to ‘the passions of sheep stealers’ in the Balkans (a lesson even more important today than then). If he had wanted a war, students would now be writing essays on why the Russo-German War of 1888 was ‘inevitable’. Many portray the war that broke out in August 1914 as ‘inevitable’ but many decisions in the preceding month could have derailed it, just as decisions derailed general war in previous Balkan crises. Few realise how lucky we were to avoid nuclear war during the Cuban Missile crisis (cf. Vasili Arkhipov) and other terrifying near-miss nuclear wars. The whole 20th Century history of two world wars and a nuclear Cold War might have been avoided if one of the assassination attempts on Bismarck had succeeded. If Cohen-Blind’s aim had been very slightly different in May 1866 when he fired five bullets at Bismarck, then the German states would certainly have evolved in a different way and it is quite plausible that there would have been no unified German army with its fearsome General Staff, no World War I, no Lenin and Hitler, and so on. The branching histories are forgotten and the actual branch taken, often because of some relatively trivial event casting a huge shadow (perhaps as small as a half-second delay by Cohen-Blind), seems overwhelmingly probable. This ought to, but does not, make us apply extreme intelligent focus to those areas that can go catastrophically wrong, like accidental nuclear war, to try to narrow the range of possible histories but instead most people in politics spend almost all their time on trivia.

We evolved to make sense of this nonlinear and unpredictable world with stories. These stories are often very powerful. On one hand the work of Kahneman et al on ‘irrationality’ has given an exaggerated impression. The fact that we did not evolve to think as natural Bayesians does not make us as ‘irrational’ as some argue. We evolved to avoid disasters where the probability of disaster X happening was unknowable but the outcome was fatal. Rationality is more than ‘Bayesian updating’. On the other hand our stories do often obscure the branching histories of reality and they remain the primary way in which history is told. The mathematical models that illuminate complex reality in the physical sciences do not help us much with history yet. Only recently has reliable data science begun to play an important role in politics.

Andrew Marr wrote recently about the referendum with a classic post facto ‘big event must be caused by one big factor’ story:

‘Connected to this is the big “why?”. I don’t think we voted to leave the EU because of clever tacticians or not-quite-clever-enough pollsters, or even because Johnson decided that one of his columns was better than another. I think we voted to leave because so many British people had been left behind economically and culturally for so long, and were furious about it; and because, from the 2008 financial crisis onwards, they had accumulated so much contempt for the political elites. In these circumstances any referendum narrows down to a single question: “Are you happy with the way things are?” The answer was “no”.’ Andrew Marr, October 2016.

‘The big why?’ is psychologically appealing but it is a mistake. In general terms it is the wrong way to look at history and it is specifically wrong about the referendum. If it were accurate we would have won by much more than we did given millions who were not ‘happy with the way things are’ and would like to be out of the EU reluctantly voted IN out of fear. Such stories oversimplify and limit thinking about the much richer reality of branching histories.

*

Branching histories in 2016: three powerful forces, many possible campaigns

Sometimes the outcome of a vote is clear before a campaign starts such that it is reasonable to say ‘the campaign didn’t matter’ other than in the negative sense that, provided it avoids huge disasters, the twists and turns, the exact messages and adverts, thousands of decisions taken and so on very likely had no impact on the binary outcome. For example, Reagan’s re-election campaign in 1984 or Blair’s re-election campaign in 2001 were campaigns like this. Both won by so much and were clearly predicted by very large and historically very unusual poll leads well in advance. It is not plausible to say that the weeks of campaigning affected who won. At most the campaigns affected the scale of victory.

The referendum was not like this. Throughout the second half of 2015 and the beginning of 2016 the averages of polls – the only sensible way to look at polls – showed clear IN leads. All polls showed significant shifts towards Leave in the last five weeks (then a shift towards Remain at the end but this was at least partly because London-based pollsters changed their methodology thinking that they were making them more accurate – they fooled themselves). Polls tracking deeper attitudes that had been consistent for years suddenly changed in the last few months in ways that were significant given the close outcome. Recent claims that the polls ‘really’ showed Leave ahead all the time should be taken with very large pinches of salt given their dodgy statistical claims, charlatan authors like Matt Goodwin (who treats data dishonestly), and the inherent impossibility of discovering the truth of such a question.

One example from our private ICM polls (I will post the data tables for all these): Vote Leave asked people to choose between these options regularly to probe attitudes to the EU that are more informative than just the referendum question. The 11 point gain for ‘strong out’ is much bigger than the margin of error, is supported by other data, and is clearly significant.

screenshot-2016-11-27-13-58-44

The cold reality of the referendum is no clear story, no ‘one big causal factor’, and no inevitability – it was ‘men going at it blind’. The result was an emergent property of many individual actions playing out amid a combination of three big forces (see below). Many of these actions were profoundly nonlinear and interdependent and the result that we actually witnessed was very close. If about 600,000 people – just over 1% of registered voters – had decided differently, IN would have won. This is a small enough margin that it could easily have happened if quite a few specific events and decisions had turned out differently. If just one person had behaved differently the dominant story now would be ‘the economy was always going to trump a revolt against the elites, the status quo and “the economy stupid” always win’ – which is what the overwhelming majority of pundits said before 23 June and in some cases had drafted for their columns after the vote.

For example, if Michael Gove had stayed out of the campaign then Vote Leave would almost certainly have either collapsed (which it nearly did anyway) or been forced into fighting the campaign on a losing message like ‘Go Global’, a firm favourite for many years among a subset of MPs and Farage’s inner circle (Leave.EU adopted this as its first slogan) and a total loser with the public. (Therefore another counterfactual: why did Cameron and Osborne not try very hard to get a clear commitment from Gove that all he would do is issue a statement but would carry on with his day job and would not campaign? I hope he would have refused but it was worth a shot and they didn’t try very hard.)

Without Boris, Farage would have been a much more prominent face on TV during the crucial final weeks, probably the most prominent face. (We had to use Boris as leverage with the BBC to keep Farage off and even then they nearly screwed us as ITV did.) It is extremely plausible that this would have lost us over 600,000 vital middle class votes.

Without Victoria Woodcock, an absolutely phenomenal manager and by far the single most important person in the management of Vote Leave (and who would have been running Downing Street now but for the Gove-Boris debacle – more branching histories), we would not have been able to build anything like the structure we did and this could easily have cost us the winning margin of votes.

Anybody who says ‘I always knew X would win’ is fooling themselves. What actually happened was one of many branching histories and in many other branches of this network – branches that almost happened and still seem almost real to me – we lost.

Problems with Vote Leave

This is not a claim that ‘we won because of the Vote Leave campaign’. Our campaign failed to do much that we should have done. There were powerful connections between:

  • infighting over who appeared on broadcast and strategy,
  • the lack of resources (many kept clear because of the infighting and many used infighting as an excuse to keep clear of something they thought was doomed),
  • the extreme difficulty of finding a governance system that could work,
  • four crucial posts held by the wrong people (including the disastrous John Mills as first Chairman),
  • the fundamental structure of how the media works (see below),
  • the extreme difficulty of getting prominent people to say on TV what research showed was necessary to win, and
  • the lack of anything resembling a well-organised mass movement.

Despite many years to prepare, the eurosceptic community had built remarkably little to prepare for the battle. On the ground were many small ineffective and often warring little groups and essentially no serious machinery (though Business for Britain had begun to build a business network). All this had to be built almost entirely from scratch in an environment in which many of those in charge of the small groups were sure we would lose, were less interested in winning than they were in ‘preserving our group’s identity Dominic’, and were keen to get their hands on cash being handed out by Leave.EU on condition that they not contribute to the campaign with Vote Leave. At various points UKIP HQ sent out emails to UKIP activists telling them not to work with Vote Leave and some senior activists were told by Farage’s gang that they would lose their UKIP jobs if they helped our ground campaign (luckily most of those out on the ground ignored these instructions but they were disruptive).

The office implemented the winning message in ~125 million leaflets and nearly a billion targeted digital adverts regardless of all complaints. We recruited more active volunteers (~12,000) in 10 months than UKIP in 25 years (~7,000 according to Farage). Our GOTV effort targeted crucial voters identified by traditional polling, a new type of experimental polling, the ground campaign, and the social media campaign, all overseen by the data science team. But until the last 4-5 weeks we had a big problem getting those going on TV to give the same message. The office could only do so much. If Boris, Gove, and Gisela had not supported us and picked up the baseball bat marked ‘Turkey/NHS/£350 million’ with five weeks to go, then 650,000 votes might have been lost. In the awful weekly campaign committee meetings, there were constant complaints and arguments for variations on ‘Go Global’ (until all the polls swung our way and people remembered ‘I’ve always said stick with 350 million’.) The Big Three knocked this back despite great pressure.

Some people had spent a quarter of a century talking about things that appealed to about 10% of the population and they would not pay attention to what millions of normal people actually knew and thought (‘I’ve spent years trying to ignore the NHS in elections Dominic and I’m not going to change now’ said many like Peter Bone). Media planning was extremely hard. Paul Stephenson’s media team of half a dozen, massively outnumbered by hundreds of officials, did a fantastic job but we could have done so much more if more MPs had been more determined and more supportive.

It should be remembered that the net effect of Conservative MPs was strongly supportive of IN. We won despite the net effort of Conservative Party MPs, not because of them, though the support from a small fraction was vital. Although Leave voters were more enthusiastic and determined than Remain voters, Cameron and Osborne were more focused on winning than most Leave MPs were. (Almost all Labour MPs seemed to be in a parallel universe until they got intelligence from their constituencies about postal votes after which they panicked ineffectually.)

Most of the MPs we dealt with were not highly motivated to win and lacked extreme focus, even those who had been boring everybody about this for decades. They sort of wanted to win but they had other priorities. They were very happy having dinner parties and gossiping. They were very happy coming to meetings with people they thought were important. This wasted enormous amounts of time as we had to create a string of Potemkin committees for people to attend while the core team actually did the campaign, then reinvent them as people became convinced that there were other secret meetings that they were being excluded from. They were very happy to be on the Today Programme. But they didn’t want to win that much. Not enough to work weekends. Not enough to stop having all their usual skiing holidays and winter beach holidays. Not enough to get out on the streets day after day.  Not enough to miss a great shooting weekend. Not enough, most of them, to risk annoying a Prime Minister who they thought would still control their next job after 23 June.

This lack of motivation is connected to another important psychology – the willingness to fail conventionally. Most people in politics are, whether they know it or not, much more comfortable with failing conventionally than risking the social stigma of behaving unconventionally. They did not mind losing so much as being embarrassed, as standing out from the crowd. (The same phenomenon explains why the vast majority of active fund management destroys wealth and nobody learns from this fact repeated every year.)

Our core campaign team were not like this. They sacrificed weekends, holidays, and family events. They worked like dogs week in week out for little money often treated with appalling rudeness by people calling from their beach loungers (Boris, Gisela and Gove were three notable exceptions and all three were liked by junior staff partly because of their good, therefore rare, manners). We were happy to risk looking stupid to win. We knew that almost nobody in SW1 understood or agreed with what we were doing. We also knew we had more chance of winning if we did not explain a lot of it – most importantly the entire digital and data science element which (combined with the ground campaign and GOTV) gave us a chance to exploit strong network effects  (and which we hid from the Board and MPs, see HERE).

Example… We were urged by everyone to hire a big advertising agency and do traditional posters. ‘When can we discuss our posters?’ I was asked constantly by people who would then try to explain to me their creative ideas (‘we need another Labour Isn’t Working, Dominic, I’ve got an idea for a picture of the globe and arrows…’). One of the few reliable things we know about advertising amid the all-pervasive charlatanry is that, unsurprisingly, adverts are more effective the closer to the decision moment they hit the brain. Instead of spending a fortune on an expensive agency (with 15% going to them out of ‘controlled expenditure’) and putting up posters to be ‘part of the national conversation’ weeks or months before the vote, we decided to 1) hire extremely smart physicists to consider everything from first principles, 2) put almost all our money into digital (~98%), 3) hold the vast majority of our budget back and drop it all right at the end with money spent on those adverts that experiments had shown were most effective (internal code name ‘Waterloo’). When things are digital you can be more empirical and control the timing. The world of advertising agencies and PR companies were sure we had screwed up because they did not see what we were doing. (Tim Bell told everybody we were doomed because we hadn’t hired one of his companies.) This points to another important issue – it is actually hard even for very competent and determined people to track digital communication accurately, and it is important that the political media is not set up to do this. There was not a single report anywhere (and very little curiosity) on how the official Leave campaign spent 98% of its marketing budget. There was a lot of coverage of a few tactical posters.

There were some MP heroes.

Example… Steve Baker often disagreed with me, sometimes very strongly, but he was a rare person in the campaign – an honest man. Not only did Steve win some important Parliamentary battles he also played a vital role during the attempted coup of 25 January. If he had thrown in his lot with the coup, it might have proved fatal. Instead he spoke honestly about the situation. We did not agree and we were both under pressure from a set of people who thought that ‘if they [HQ/MPs] control the campaign we will lose, we [HQ/MPs] must control it’. We came to an agreement that we both stuck to. With five weeks to go, there was an attempt to revive the coup by a couple of VL Board members working with players from the January coup like Malcolm Pearson. The demand was to replace the Big Three (Boris, Gisela, Gove) and the core campaign team with Farage, and replace £350 million / NHS with ‘go global’ trade babble. This didn’t get past the usual weekend boozy chats partly because of Steve Baker telling them he thought it a mad plan. This also shows how volatile the situation was right until the end and how few prominent eurosceptics even then understood that a) the £350 million / NHS argument was necessary to win and b) their ‘go global’ message was a total loser.

Other MPs also made significant personal sacrifices – backbenchers like Anne Marie Trevelyan and Graham Stringer, and ministers like George Eustice and Dominic Raab.

Rough balance of forces

The IN side started with huge structural advantages.

  1. IN started in 2015 well ahead in the polls and had the advantage of having the status quo on its side which is intrinsically easier to explain than change is, as lots of historical data around the world shows. Usually the ‘change’ campaign has to start considerably ahead in order to win as it loses support as the campaign goes on. This argument was even stronger with something so much bigger and more complex like the EU. VL had to persuade millions of people to risk a profound change. Those on the IN side made this point repeatedly for many months. They were right then. After 23 June many of them say the exact opposite – it’s so complex to explain all the wonders of the EU, they say, and so easy to argue for change. This is laughable.
  2. IN had the government at its heart including the Downing Street machine, the Cabinet Office, and Government departments and agencies all of which added up to thousands of people including hundreds of press officers. Cameron and Heywood also instructed Permanent Secretaries not to share EU material with Secretaries of State supporting Vote Leave in order that they did not have access to new information about all the ways in which EU law affected policy. (In general Whitehall has made great efforts to hide the scope of EU control. It also preserves, Potemkin-style, old processes like circulating Cabinet papers ‘for approval’ where the only acceptable response is ‘approve’ – it is not actually legally possible not to ‘approve’ but still the papers are sent round via the absurd red box system daily.) VL had a few dozen effective people and no access to the official machine other than some leaks. We had a research team of about five. MPs proved largely useless in helping this team.
  3. IN controlled one side of the renegotiation and its timing. VL was at the mercy of events and could not get any ministers supporting us until the process ended.
  4. IN controlled the timing of the referendum. VL had to plan resources on the basis of many scenarios.
  5. IN controlled the Cabinet and junior ministers – bribes for support and threats to deter. They had the chance to set the terms for how ministers engaged in the campaign (though they partly blew this). VL had to meet ministers in secret, could guarantee them no jobs, and (as was pointed out to me by many) could not dodge the basic truth that purely from a personal career perspective it was usually better to support the PM.
  6. IN controlled the governing party and the Parliamentary timetable and procedures. VL had to work with a small number of MPs many of whom had spent many years in constant opposition to their own leadership and were unused to any sort of discipline or collective action.
  7. IN set the legal rules. VL faced a huge imbalance in how these worked. For example, Cameron even during the official campaign could do huge events at places like the British Museum and the IN campaign did not have to account for such events as part of their £7 million. Meanwhile VL was told by the Electoral Commission that if people we did not even know put up huge signs that appeared on TV we might get billed for them. There were many other consequences of the imbalance. E.g. the Government’s legal timetable meant we had to commit before the official start of the campaign to a load of activity that would occur after the official start of the campaign without knowing if we would be the official campaign and therefore legally entitled to spend this money. We therefore had to choose between either a) not do various things, be sure we would not break the law, and lower the chances of winning or b) do the right thing for the campaign and riski being judged to have broken the law. Obviously we did (b) though we had to hide this choice from some of those on our Board as this was exactly the sort of thing some of them were very weak about.
  8. IN had access to huge resources – financial, personnel etc. IN had the support of almost every entity with power in Britain, Europe, and the world from the senior civil service to the CBI to the big investment banks, to Obama and the world bureaucracy (G20, UN, IMF etc).  Very few senior people were prepared to risk supporting us. Those who did mostly did so in a small way and on their own terms without getting involved in our campaign. While IN could send out name after name to deliver their message, we could depend on very few names who would deliver our message. The Government machine, the Commission, and the Cabinet Office were effective in scaring off prominent people from supporting us; many of them told us (some embarrassed) about the phone calls they’d had and their ‘duty to shareholders’ and so on. Advanced media planning was almost impossible and we had to shuffle things around at short notice constantly. IN had millions more than us before the campaign ever started and used this money for direct voter communication. We could not afford this. We sent out one 10 million voter mailing to people identified by the physicists just before the spending limits started and we could only do this by tricking some of those on our Board about the numbers. (I was also  helped by Peter Cruddas saying, ‘Don’t worry about the fundraising situation, don’t listen to everybody panicking, just do whatever it takes to do the campaign, if the money doesn’t come I guarantee I’ll put in whatever you need’. I knew I could trust him. This gave us vital flexibility and also meant we could ignore some of those on the Board who were more focused on whether they may be liable for a bill post-23/6 than they were on winning.)
  9. IN had the support of most journalists and senior management in the main broadcasters. The broadcasters let the Government set the agenda on TV for almost the entire campaign, apart from ten crucial days after the immigration numbers on 26 May. VL had the support of some powerful papers but we were overwhelmed on TV news. (Two broadcast journalists who were conspicuous by their unusual professionalism and determination to act fairly despite the behaviour of some of their management were Laura K and Allegra Stratton.)
  10. IN started with legal access to vast amounts of electoral data from at least three political parties, unofficial / illegal access to vast amounts of data from things like CCHQ data and the Crosby/Messina models built during the campaign, and vast amounts of commercial data. (CCHQ laughably claimed that there were ‘Chinese walls’ that prevented any abuse of Party data.) VL had none of these things. We could not even afford to buy standard commercial datasets (though the physicists found ingenious ways around this). We had no way even to acquire the electoral roll until the official process allowed us in early 2016, after which we had to wait a couple of months for LAs to fulfil their legal obligations to provide us with the data (which they did patchily and often late).
  11. IN had a great boost to its fortunes in the form of a network linking Nigel Farage, Aaron Banks, assorted peers (e.g. Malcolm Pearson), MPs (e.g. Bill Cash), businessmen (e.g. Richard Smith), and a handful of Vote Leave Board members (including the one-time Chairman John Mills) and some staff foisted on us (one of whom won the title of the most repellent person I’ve met in politics – Nigel Griffiths, an ex-MP who some female staff refused to be in the same room with). Farage put off millions of (middle class in particular) voters who wanted to leave the EU but who were very clear in market research that a major obstacle to voting Leave was ‘I don’t want to vote for Farage, I’m not like that’. He also put off many prominent business people from supporting us. Over and over they would say ‘I agree with you the EU is a disaster and we should get out but I just cannot be on the same side as a guy who makes comments about people with HIV’.

On 25 January 2016 a network of these characters launched a coup. But for the actions of Stephen Parkinson, Paul Stephenson,  and Victoria Woodcock (supported by most but not all of the office) it would have succeeded. This would have given control of the official campaign to the Farage crowd. They ran with vapid slogans like ‘Be in the know’. Ironically for a group of people who claim to be anti-SW1 they rehashed the classic losing SW1 eurosceptic trope for 25 years – ‘Go Global’ – showing how little they understood the electorate and mass communication. They rejected the connection between immigration, £350 million and the NHS, which was absolutely vital, as the IN side has said after 23 June (see below). They published dumb offensive videos. They talked about privatising the NHS. They built little grassroots organisation and their claims about social media were (and remain ) laughable. Farage himself admitted after 23 June that they did not have the organisation to run the campaign if they had won designation: ‘quite what we would have done if we had got it I’m not really sure!’, which sums them up (Shipman, Location 4,150). The media would have covered this gang’s official campaign as a version of their own book – a bunch of childish dodgy boozers on an ego trip.

Before the 2015 election Farage said to me at Stuart Wheeler’s that he knew he could not be the leading face of the campaign – ‘I’m one of the generals but I can’t lead the army’ he said, to my relief. When I next saw him in the summer, I was amazed at how his tune had changed, his obsession with the debates, and his pessimism. One can only understand some of the behaviour from those around Farage if you realise that much of their operation was about positioning Farage for what they assumed would be defeat.

One of the biggest problems during the campaign and biggest misconceptions after concerns this issue. Those who argued ‘we need one campaign’ were wrong. Those who argue now ‘we would have won by more if there’d been one campaign’ are wrong. One campaign would have meant total bedlam and 60-40 defeat.

If MPs had had extreme focus on winning then they would not have used Farage as leverage against us viz official designation and therefore much of the infighting could have been avoided as Farage would have done a sensible deal with us early, realising much earlier that we would not compromise over him running the campaign under any circumstances. By encouraging Farage to think that he could get a much more prominent position, people like Bill Cash nearly destroyed everything.

Given all these huge advantages, if their campaign had been of equal effectiveness to Vote Leave then, all else remaining equal, Cameron would almost certainly (>95% likely) have won.

Why did all these forces not add up to overwhelming and devastating firepower? If you want to understand the combination of things that gives us largely dysfunctional government and therefore undermined the IN campaign – a mix of selecting and promoting the wrong people, wrong education and training, bad incentives, anti-adaptive institutions and so on – then read this in which I explain in detail why Whitehall does not and cannot work properly.

The approximate truth

The closest approximation to the truth that we can get is that Leave won because of a combination of 1) three big, powerful forces with global impact: the immigration crisis, the financial crisis, and the euro crisis which created conditions in which the referendum could be competitive; 2) Vote Leave implemented some unrecognised simplicities in its operations that focused attention more effectively than the other side on a simple and psychologically compelling story, thus taking advantage of those three big forces; and 3) Cameron and Osborne operated with a flawed model of what constitutes effective political action and had bad judgement about key people (particularly his chief of staff and director of communications) therefore they made critical errors. Even if (1) and (2) had played out the same, I think that if that duo had made one of a few crucial decisions differently they would very likely have won.

When I started to research opinion in 2014-15 and compared it to my experience of the euro campaign (1999-2002), it was clear three forces had changed opinion on the EU.

1) The immigration crisis. 15 years of immigration and, recently, a few years of the migration crisis from the East and Africa, dramatically portrayed on TV and social media, had a big effect. In 2000, focus groups were already unhappy with immigration but did not regard it as a problem caused by the EU. By 2015, the EU was blamed substantially for the immigration/asylum crisis and this was entangled with years of news stories about ‘European courts’ limiting action against terrorists and criminals. Actually often these stories concerned the Strasbourg court of the ECHR (not the ECJ) though, ironically, the EU’s adoption of its Charter of Fundamental Rights meant that many issues concerning the ECHR became relevant to the EU debate, something that almost nobody in SW1 realised and we tried and largely failed to explain (one of the very few who did understand this was Boris’s wife, an accomplished lawyer, who I discussed this with in autumn 2015).

2) The 2008 financial crisis. This undermined confidence in Government, politicians, big business, banks, and almost any entity thought to be speaking for those with power and money. Contra many pundits, Miliband was right that the centre of gravity has swung against free markets. Even among the world of Thatcherite small businesses and entrepreneurs opinion is deeply hostile to the way in which banks and  public company executive pay work. Over and over again outside London people would rant about how they had not/barely recovered from this recession ‘while the politicians and bankers and businessmen in London all keep raking in the money and us mugs on PAYE are paying for the bailouts, now they’re saying we’ve just got to put up with the EU being crap or else we’ll be unemployed, I don’t buy it, they’ve been wrong about everything else…’ All those amazed at why so little attention was paid to ‘the experts’ did not, and still do not, appreciate that these ‘experts’ are seen by most people of all political views as having botched financial regulation, made a load of rubbish predictions, then forced everybody else outside London to pay for the mess while they got richer and dodged responsibility. They are right. This is exactly what happened.

Many Tory MPs and ‘free market’ pundits / think tankers are living in a fantasy world in which they want hostility to big business to end even though everybody can see that those who failed largely escaped responsibility and have even gone back to doing the same things. (I’ve argued since 2001 for big changes on executive pay to almost zero effect. SW1 is full of people who think they’re ‘defending markets’ but are actually defending the opposite – corporate looting. In the 1930s Britain put people in jail because of what happened in the 1920s. We should have done the same after 2008.)

3) The euro crisis. Britain joined the EEC because it was a basket case in the 1970s and ‘Europe’ was seen as a modernising force that could help us recover and improve the economy and living standards. As the euro crisis hit, millions saw Greece in chaos, even flames, for month after month. This undermined confidence in the EU as a modern successful force – ‘it’s so bad even Germany’s in trouble now because of the euro’, ‘not even Germany can afford to sort this out’, people would say.

Together these three big forces undermined confidence in the EU project as a modern force for progress that brings prosperity and solves problems and pushed it into about 30-35% of the population (younger, richer, better educated) which increasingly saw the EU in terms of ‘are you racist / supporter of Farage?’ This feeling was central in 1975. It diminished gradually but was still partly there 1999-2002 when I was doing focus groups on the euro. (It is why I had so many arguments at the time with eurosceptics explaining to them that if we accepted Blair’s framing of the euro debate as IN/OUT of the EU, we would lose. Our two slogans were therefore ‘Europe yes, euro no’ and ‘Keep the pound, keep control’.)

Second, they undermined confidence in those in charge. There had been strong anti-Westminster feelings growing for over a decade. In 2004 with James Frayne and my uncle I set up the campaign to fight the referendum on the North East Regional Assembly as a training exercise for an EU referendum (then envisaged after Blair’s 2005 victory). We came from behind and won 80-20 (not a misprint) despite having almost no money, no support, and the entire North East establishment against us because we exploited this feeling (‘politicians talk, we pay‘ was our slogan). SW1 ignored the result. It did not appreciate the scale of this growing force even after the financial crash and expenses scandal. Normal electoral politics and the structural grip of established political parties fooled insiders about the extent of support for people like Cameron. Cameron won negatively – because he was not Brown or Miliband. There was very little positive feeling for him. They fought the referendum with him and Osborne at the front as if they were fighting Brown or Miliband and asking people to make a choice: this is not how most people saw it.

These three big forces and the failure of the parties to cope, combined with the daily resentment of paying taxes for the bill of the 2008 Crash, meant that in a vote like 2016 where people did not have to vote to stop Brown or Miliband ‘stealing my money’, millions who were unpersuaded by Cameron/Osborne felt free to vote positively for something (‘take back control’) and against a duo they disliked, distrusted, and saw as representative of politicians’ failure over many years.

These three big forces had global impact and had much more effect on people who pay a normal amount of attention to politics than every speech, article, pamphlet and ‘campaign’ about the EU over 15 years, the sum total of which had almost no discernible effect.

Those who think I am exaggerating the relative lack of influence of conscious SW1 activity could consider another example – the Gove education reforms 2010-14 (which I was closely involved with). These reforms were one of the most prominent stories of the 2010-15 Government with thousands of stories and broadcast discussions. I researched public attitudes to these reforms after I resigned from government in January 2014 (contrary to widespread belief the Cameron operation spent very little time and resources before 2014 on researching public opinion, they were focused on the media rather than the public). Approximately nothing of our arguments  – including the years of speeches by Blair too – had got through to the public.The entire SW1 media debate had approximately no impact on public opinion. People had some idea of some changes if they had kids in school but knew almost nothing of the arguments. Consider how much more motivated people were to learn about this than they were about the EU. (Part of the reason is that the language that Cameron and SW1 generally used was about ‘choice, competition’ and so on. I was almost totally unsuccessful in persuading people to talk about the issue in a different way which is one of the reasons I spent so little time on communication and almost all my time on management in the DfE. Gove knew the problem but also knew that there was no chance of getting Cameron to do things differently.)

This is relevant to the immigration argument in particular. Many pundits who described themselves as ‘modernisers’ wrote columns over the years arguing that immigration was an issue because Cameron was making foolish promises about it and the media therefore paid more attention to it. This is wrong. Cameron’s foolish promises certainly made his situation worse but it is wrong to think that public interest in an issue is proportional to the attention paid by politicians and newspapers in SW1. The public only pays attention to a tiny subset of issues that politicians and the media bang on about. It is largely impossible to predict which things will catch fire and which will not, though process stories and ‘scandals’ almost always have zero effect and insiders repeatedly get this wrong. Long before there was any prominent media discussion of ‘the Australian points system’ you could hear it being discussed in focus group after focus group to an extent that was very surprising to me and was very surprising to every single person I discussed it with, including Farage (who adopted the policy because of focus groups, the causal chain was not – Farage talks >> focus groups respond).

Making these three forces even more powerful was the nature of the reaction from those in charge in the EU and Britain – a general failure not only to grip the problems but even to show that they understood what the problems were. There was clearly no sensible movement for reform of the EU. As it lurched from crisis to crisis, its only response was ‘the EU needs more power’ (this is, of course, the founding logic of the Monnet-Delors system). The British Government clearly had no sensible plan for dealing with the EU’s crises and dysfunction. Worse, their responses were often obviously rubbish, such as the ‘tens of thousands’ immigration promise that people could see had no chance of being met yet politicians just kept repeating it. People naturally concluded – these guys in London don’t grasp the seriousness of the problems, they haven’t a clue what to do, and are treating us like idiots. Cameron’s renegotiation did not change this view. The Government therefore entered the campaign in a very different state to Wilson in 1975.

These three forces meant that by summer 2015 only about a third of the electorate positively wanted to be inside the EU. Another third strongly wanted to leave and were not worried about the economy. Another fifth had roughly the view that – the EU is rubbish, I’d like to be outside, but I’m worried about the short-term effects on jobs and living standards so maybe I’ll vote IN (see the ICM table above). Further, our research showed that the strong Leave third was significantly more enthusiastic about the referendum than the strong Remain third and the swing fifth, and therefore more likely to vote.

Vote Leave exploited these forces

I will go into this in much more detail and I will ignore all management/operational issues here.

Our story rested on five simple foundations that came from listening very hard to what people really knew, thought, and said:

1. ‘Let’s take back control’. The overall theme. When I researched opinion on the euro the best slogan we could come up with was ‘keep control’. I therefore played with variations of this. A lot of people have given me a lot of credit for coming up with it but all I really did was listen. (NB. ‘back’ plays into a strong evolved instinct – we hate losing things, especially control.)

2. ‘The official bill of EU membership is £350 million per week – let’s spend our money on our priorities like the NHS instead.’ (Sometimes we said ‘we send the EU £350m’ to provoke people into argument. This worked much better than I thought it would. There is no single definitive figure because there are different sets of official figures but the Treasury gross figure is slightly more than £350m of which we get back roughly half, though some of this is spent in absurd ways like subsidies for very rich landowners to do stupid things.)

Pundits and MPs kept saying ‘why isn’t Leave arguing about the economy and living standards’. They did not realise that for millions of people, £350m/NHS was about the economy and living standards – that’s why it was so effective. It was clearly the most effective argument not only with the crucial swing fifth but with almost every demographic. Even with UKIP voters it was level-pegging with immigration. Would we have won without immigration? No. Would we have won without £350m/NHS? All our research and the close result strongly suggests No. Would we have won by spending our time talking about trade and the Single Market? No way (see below).

NB. Unlike most of those on our side the IN campaign realised the effectiveness of this, as Cooper, Coetze and others said after 23 June. E.g. ‘The power of their £350 million a week can’t be overstated.’ Andrew Cooper, director of strategy for the IN campaign.

Some people now claim this was cynical and we never intended to spend more on the NHS. Wrong. Boris and Gove were agreed and determined to do exactly this. On the morning of 24 June they both came into HQ. In the tiny ‘operations room’ amid beer cans, champagne bottles, and general bedlam I said to Boris – on day one of being PM you should immediately announce the extra £100 million per week for the NHS [the specific pledge we’d made] is starting today and more will be coming – you should start off by being unusual, a political who actually delivers what they promise. ‘Absolutely. ABSOLUTELY. We MUST do this, no question, we’ll park our tanks EVERYWHERE’ he said. Gove strongly agreed. If they had not blown up this would have happened. The opposite impression was created because many Tories who did not like us talking about the NHS reverted to type within seconds of victory and immediately distanced themselves from it and the winning campaign. Unlike Gove and Boris they did not learn from the campaign, they did not listen to the public. Until people trust that the NHS is a financial priority for Tories, they will have no moral authority to discuss management issues. This obvious fact is psychologically hard to absorb because of the strength of gang feelings in politics.

(There are already myths about some of these events. The press conference of 24 June is now written up as the two of them ‘terrified of what they had done’ but this is completely wrong. They were subdued partly because they were genuinely sad about Cameron and partly because they did not want to be seen as dancing on his grave. Some of the media created the psychologically compelling story that they were regretful / frightened about victory but this was not at all their mood in HQ on the morning of 24 June. Boris came in punching the air like Maradona after a great goal, hugging staff and clearly euphoric. It is completely wrong to portray him as regretful.)

3. ‘Vote Leave to take back control of immigration policy. If we stay there will be more new countries like Turkey joining and you won’t get a vote. Cameron says he wants to “pave the road” from Turkey to here. That’s dangerous. If we leave we can have democratic control and a system like Australia’s. It’s safer to take back control.’

I was surprised at what a shock it was to IN when we hit them with Turkey. By the time this happened they were in an almost impossible position. I wanted them to announce a veto. It would not have been believed and would have had the opposite effect – people would have taken the danger of Turkey joining more seriously. If your life depended on winning for IN, the answer is clear: they should have said long before the campaign started as part of the renegotiation process that they would veto any accession.

4. ‘The euro is a nightmare, the EU is failing, unemployment is a disaster, their debts and pensions are a disaster, if we stay YOU will be paying the bills. It’s safer to take back control and have a new relationship based on free trade and friendly cooperation instead of the European Court being in charge of everything…’ (This is not an official text, just a summary of the notion off the top of my head.)

5. Anti-Establishment. E.g. We aligned our campaign with those who were furious with executive pay / corporate looting (about 99% of the country). We aligned ourselves with the public who had been let down by the system.

Mandelson regarded this as ‘sheer nerve, sheer chutzpah’. It was obvious. The hard thing was sticking to it despite the sensibilities of many of our own supporters. One of the most effective TV performances of the campaign was the day Boris hit the theme of corporate looting in a market square. No10 were rightly panicked and in response pushed out Heseltine a few hours later to make a very personal attack on Boris. This made sense tactically but was a strategic error. All such personal attacks helped persuade Boris to up the ante. This was vital with a month to go when the immigration figures came out. Rudd and others argue that Cameron should have attacked Boris and others more. Wrong. They should have played it Zen publicly and had a much better black ops team.

Cameron/Osborne mistakes

I’ll go into this separately but just to give a few examples…

1. Cameron never had to offer the referendum in the first place. His sudden U-turn was a classic example of how his Downing Street operation lurched without serious thought in response to media pressure, not because of junior people but because of Cameron himself and his terrible choice of two main advisers (Llewellyn and Oliver). This happened many times and I wrote about all the damage it caused on other issues after I left government (HERE). This was the biggest example. It was a product of a deeper error – a combination of his failure of party management (misleading them about the best way to handle the party) and failure to understand how swing voters really think and therefore the dangers of a vote (see below).

2. If Cameron/Osborne had had a top notch person like David Plouffe running their campaign and they did as they were told then they would have won (>95% confidence), all else being equal. They were warned many times by their closest friends about Oliver and Llewellyn, including by Gove, but would not listen.

3. Their renegotiation was flawed from the start and badly undermined their central message. They compounded their errors in 2015 by accepting the pathetic deal in 2016.  If they had walked away in February then Vote Leave would quickly have imploded and the flying monkeys would have taken over the campaign.

4. They made themselves too prominent in the campaign and were too crude. Lacking a feel for psychology they gradually undermined their own message. Oliver thought Obama’s ‘back of the queue’ was brilliant. It was counterproductive. They thought ratcheting up the warnings to DEFCON 1 was effective. It was counterproductive.

5. They doubled down on ‘tens of thousands’. They thought they would lose credibility if they didn’t. The opposite was true. They should have dropped this in 2015 – for example, in an exclusive to the Independent on a Saturday in early August 2015 – and gone into the campaign without it. Every time they defended it they were helping us.

6. They suckered themselves into over-prioritising their coalition versus message. Blair’s campaign against us in the North East did the same. When you do this you lose focus and clarity which is usually fatal. The error was perhaps most visible the day Cameron unveiled an absurd poster that effectively listed all the ‘serious people’ on their side and – creative genius! – a blank page for us. A total waste of valuable time. The fact of being the Government meant the broadcasters let them lead the news almost all the time but they often wasted it like this. (I would bet that that ad was never put in focus groups or if it was the results were ignored.)

7. One of my basic criticisms of Cameron/Osborne from the start was the way they steered by pundit. During the 2015 election Crosby partly corrected this and they partly learned the lesson. But left to their own devices in the referendum when under pressure they defaulted to their instincts at a crucial moment. The reaction to the dreadful murder was an example of how the media and SW1 can live effectively in a parallel universe. Somehow they convinced themselves that this event might undo over a decade of growing hostility for those in power. They therefore tried to push the theme that actually MPs are great, ‘they are in it for good reasons’ and so on. The media led themselves into a dead end and No10, defaulting to their instincts of steering by pundit, followed. As soon as I saw Osborne and Matt Hancock wasting their time tweeting broken multicoloured hearts and encouraging #weloveourMP, I knew they had screwed their own OODA loop. We knew from focus groups (conducted by the brilliant Henry de Zoete who also played a crucial role in coordinating the digital and data science teams) that opinion outside London was extremely different to that of MPs and those in charge of most news. We went straight back to what we knew were the winning messages leaving Hancock and co to tweet broken hearts.

BUT BUT… Roland Rudd and others have attacked them for their basic strategy of focus on the economy and argue there should have been ‘a positive campaign for the EU’. WRONG. Cameron and Osborne were right about this big call. There was not enough time or money to change basic attitudes. As the campaign developed and there were signs of pressure from Rudd and others I crossed my fingers and hoped they would shift strategy. No10 were right to ignore him.

I suspect that in general big mistakes cause defeat much more often than excellent moves cause victory. There are some theoretical reasons to suspect this is true from recent statistical analysis of human and computer decisions in chess. Two results are particularly interesting. 1) The very best computers seem to make moves that preserve  the widest possible choices in the future, just as the most effective person in politics for whom we have good sources, Bismarck, operated always on the principle of ‘keep two irons in the fire’. (We tried to mimic this by adopting a message that we thought had the highest probability of  winning in the largest number of plausible branching futures, hence £350m/NHS.) 2) Even great humans are distinguishable from great computers by their propensity to make clear tactical errors occasionally amid the fog of war. This is significant enough that it wipes out the advantage of going first – i.e. it being ‘your move’ is seen as a plus but in fact it is a minus for humans because of the probability of a significant error, while for computers this effect is absent. (See Human and Computer Preferences at Chess, 2014. It would be very interesting to know if these results are supported by the recent success of Deep Mind with computer GO.)

Summary of the false dichotomy

False: ‘Leave won because of the campaign.’ E.g. Without 15 years of out of control immigration, our message of ‘take back control’ would not have had enough traction. Campaigns can ride big waves but they almost never make them.

False: ‘Leave won because of a big event [immigration, 2008 crash etc], the campaign was irrelevant.’ E.g. If the campaign had not deployed £350 million and the NHS (which almost nobody on our side liked), we would not have neutralised/overwhelmed Project Fear.

True: ‘Leave won because 1) three big forces created conditions in which the contest was competitive, AND 2) Vote Leave exploited the situation imperfectly but effectively, AND 3) Cameron/Osborne made big mistakes. If just one of these had been different, it is very likely IN would have won.’

Overall, the now-mocked conventional wisdom that ‘the status quo almost always wins in referendums like this’ obviously has a lot of truth to it and it only proved false this time because of a combination of events that was improbable.

*

A ‘miracle’ to get 48%? Beaten by lies? Corbyn the AWOL saviour?

Since losing many inside the IN campaign now talk dejectedly as if they could never have won and tell rationalising fairy tales. They are wrong. They almost did win. Some have latched onto the idea that they were overwhelmed by an epic, global force of ‘right-wing populism’. Mandelson defends himself by saying  48% looks ‘like a miracle’ given the populist tide. Most have latched onto the idea that their ‘complex truth’ was overwhelmed by ‘simple lies’ and they are happy with their comforting ‘post-truth’ sobriquet – a delusion that leaves them very vulnerable to being shocked again. Many have even argued that they lost because they could not persuade Corbyn to make more speeches.

These stories are psychologically preferable to the idea that their own errors caused defeat (just as it is for some of those in Hilary’s campaign) but should not be taken seriously.

The least plausible claim is that Corbyn sabotaged what was otherwise a winning campaign. This is argued mainly by the same people (including Mandelson) who in a party context also argue that Corbyn is a joke who nobody takes seriously. The idea that more speeches by Corbyn would have persuaded vital swing voters has no good evidence. These people wanted to ‘take back control’. Corbyn’s message was – there should be not just more immigration but no limits on it. There are not many branching histories in which this is a winner.

This ‘epic global force’ of ‘populism’ was thought by the same people before 23 June to be puny in comparison with the force of the combined Establishment hammering a message of economic fear in support of the status quo. Having underestimated certain trends in public opinion the same people are now exaggerating them (see below).

This is connected to ‘complexity’. Month after month they argued (including to us in private discussions) that they would win largely because they had the advantage of the status quo – an advantage proved in votes around the world over many years. They were right. That was a big advantage. It is much simpler to argue for the status quo than for a very complex change – that is exactly why most ‘change’ referendums lose, just as they briefed the media. Now they say ‘The EU is very complex, it requires a lot of information to explain it’ (Craig Oliver). Their claim that actually they had the ‘complex’ argument to make against our ‘simple lies’ is laughable for exactly the reasons they gave themselves before they came unstuck.

Connected to this idea is that the great rationalists Cameron and Osborne – they of Project Fear and their comic ’emergency budget’ and in 2015 the pictures of Salmond picking pockets designed successfully to persuade the English that the Scots would steal their money – were undone by a great surge of ’emotion’. Osborne is taking this delusion so far he is writing a book titled ludicrously ‘The age of unreason’. When you lose and you blame it on millions of people being overtaken by ‘unreason’ – after previously winning by exploiting nationalist hostility – it’s a sure sign that you are the one not reasoning straight and able to face your errors. For the likes of Osborne it is ‘irrational’ to reject the views of people like him. For most of us, people like Osborne are not experts to be trusted – they are charlatans not to be taken seriously.

Many of those who blame defeat on ‘lies’, including Cameron, Osborne, and Clegg themselves told flat-out lies. One example will do. Cameron and Osborne claimed repeatedly on TV, almost always unchallenged, that their new deal meant ‘after six months if you haven’t got a job you have to leave’. This is not an argument over the fairness of using a gross/net figure, like ‘£350 million’, or even a properly bogus figure like the Treasury’s £4,000 per household figure. It is a different category of claim – a flat out 100% lie. (For more details see HERE.) How much time did TodayNewsnight, and the Guardian spend explaining to people that the PM and Chancellor were lying through their teeth? Approximately none. Why? Because very few of those complaining about lies really are cross about ‘lies’ – they are cross they lost and they are not so interested in discussing a lie that undermines the pro-EU campaign’s attempt to neutralise fear of immigration.

Further, many of the same people spent the entire campaign saying ‘Vote Leave has admitted a Leave vote means leaving the Single Market, this is what will happen make no mistake…’ and now say ‘the Single Market was not an issue, Vote Leave never had a policy on it and there is no mandate for leaving it’. Cameron, Osborne, Mandelson, Campbell and Clegg spent much of the last 20 years lying through their teeth to further their own interests and prestige. Now they whine about ‘lies’. They deserved worse than they got – and reasonable Remain-ers deserved better leadership.

Fools and knaves

Many of those who worked on the IN side are now wrongly attacked as fools by pundits who would have praised them as geniuses had they won, while many on the OUT campaign are wrongly praised.

Example… ‘If Remain wins Cameron ought to be hailed as the genius strategist of western democratic politics’ (Rentoul). Pundits who wrongly hailed Cameron as a genius after the 2015 election now wrongly describe him as a bumbling oaf. He was neither – he was the best of a bad bunch picked pseudo-randomly in a broken system and out of his depth. 600,000 votes either way does not make one set of people geniuses and another set of people morons. Geniuses in politics are rarer than in maths and physics and nobody involved in the referendum on either side is remotely close to one. Some of those who worked on the IN side were much more able than many on the winning side. It does not make sense to label people on the IN side idiots because of errors made by Cameron, Osborne, Llewellyn, and Oliver.

Example: many have said to me ‘you were so clever to hold back on immigration until the start of purdah’. Wrong. It is true that we did not do much on immigration before the 10 week official campaign. That is because, as I wrote in 2014, we did not need to. It was far more important to plant other seeds and recruit support that would have been put off if we had focused early on immigration. Immigration was a baseball bat that just needed picking up at the right time and in the right way. The right time was before purdah and we set in motion during January-April a series of things like the free referendum address with the right message but we could not persuade many prominent people to do what was needed until after 26 May. The right way was via the NHS (unifying) – not ‘we want our country back’ of Farage (divisive). The timing was not ‘a brilliant move’ by me, it was a combination of good luck and seizing a tactical chance to persuade people of something I’d failed to do for weeks, but such things get rewritten as such if you win.

It is also foolish to see the conflict in terms of who is ‘nicer’ and ‘nastier’. I don’t think the people on our side are nicer. There are lovely and loathsome people, liars and charlatans on both sides.

Many OUT-ers talk as if we were destined to win. Wrong. The IRA used to say ‘you have to get lucky every time but we only have to get lucky once’. For Leave to win a string of events had to happen many of which were independently improbable or 50-50 and therefore the combination was very improbable. The result was certainly not an inevitable outcome of ‘the great British public simply voting for democracy’ as some romantics delude themselves.

*

Oblonsky and the frogs before the thunderstorm: fashion, delusions of the educated, and the Single Market

‘I feel that, in some ways, this was a conflict between good forces in society and bad forces. I feel that the bad forces on 23 June won a very significant victory.’ Matthew Parris.

Matt Ridley: Matthew, you’re not saying that 17 million people are, deep down, racists? 

Matthew Parris: Yes. (Spectator, December 2016)

Why is almost all political analysis and discussion so depressing and fruitless? I think much has to do with the delusions of better educated people. It is easier to spread memes in SW1, N1, and among Guardian readers than in Easington Colliery.

Generally the better educated are more prone to irrational political opinions and political hysteria than the worse educated far from power. Why? In the field of political opinion they are more driven by fashion, a gang mentality, and the desire to pose about moral and political questions all of which exacerbate cognitive biases, encourage groupthink, and reduce accuracy. Those on average incomes are less likely to express political views to send signals; political views are much less important for signalling to one’s immediate in-group when you are on 20k a year. The former tend to see such questions in more general and abstract terms, and are more insulated from immediate worries about money. The latter tend to see such questions in more concrete and specific terms and ask ‘how does this affect me?’. The former live amid the emotional waves that ripple around powerful and tightly linked self-reinforcing networks. These waves rarely permeate the barrier around insiders and touch others.

These factors are deepened by the fact that almost all of those whose job it is to explain politics and campaigns have never been responsible for a complex organisation in general or a campaign in particular, so they are unsuited to understand how politics ripples out from decisions at the centre through dysfunctional bureaucracies to the ground. They almost always exaggerate the extent to which important decisions have been considered carefully by people who know what they are talking about. (The worse educated are actually often helped by their lack of education towards the truth.) They constantly discuss complex systems as though errors can be eradicated instead of asking how quickly errors are adapted to and learned from. This perspective biases them in favour of existing centralised systems that fail continually and against innovations with decentralised systems. They understand little about the challenges faced by small businesses and the lower middle classes.

The more closely involved people are in the media and politics the more they are driven by fashion and the feeling, rarely acknowledged and almost always rationalised, that ‘this is my gang’. Look at all those in SW1 who tweet attacks on Dacre to each other then retweet the praise from their friends, then look at those who attack them. Look at Robert Peston tweeting pictures of the London Eye and Habermas quotes on election night and his opponents ranting about ‘elites’. Both sides are just like football team fans defending their in-group and attacking their out-group enemies. The more they think of themselves as original the more likely they are to be conformist – and conformist within very narrow parameters.  We all fool ourselves but the more educated are particularly overconfident that they are not fooling themselves. They back their gang then fool themselves that they have reached their views by sensible, intelligent, reasoning.

This makes them particularly vulnerable to ‘influence operations’. It also makes them vulnerable to repeated errors about what the sort of people who ignore politics other than for a few weeks before voting time are thinking. It creates something of a paradox: it is almost impossible to get a good feel of public opinion, or of ‘the winning strategy’, by listening to those whose job it is to speculate about it. However often this happens, the lesson is never learned. It is very hard to see how it could change as it is so entangled with our evolved nature.

There is a wonderful passage in Anna Karenina that sums this up, much better than any ‘political scientist’ has done:

Oblonsky never chose his tendencies and opinions any more than he chose the style of his hat or coat. He always wore those which happened to be in fashion. Moving in a certain circle where a desire for some form of mental activity was part of maturity, he was obliged to hold views in the same way he was obliged to wear a hat. If he had a reason for preferring Liberalism to the Conservatism of many in his set, it was not that he considered the liberal outlook more rational but because it corresponded better with his mode of life… The Liberal Party said that marriage was an obsolete tradition which ought to be reformed, and indeed family life gave Oblonsky very little pleasure, forcing him to tell lies and dissemble, which was quite contrary to his nature. The Liberal Party said, or rather assumed, that religion was only a curb on the illiterate, and indeed Oblonsky could not stand through even the shortest church service without aching feet, or understand the point of all that dreadful high-flown talk about the other world when life in this world was really rather pleasant… Liberalism had become a habit with Oblonsky and he enjoyed his newspaper, as he did his after-dinner cigar, for the slight haze it produced in his brain.’

Towards the end of the novel, there is a discussion about the then big issue of Turkish atrocities and the rise of pan-Slavism. The old prince replies to the intellectuals who are talking rubbish about ‘the national feeling’ that they think is ‘sweeping the country’:

‘Yes, all the papers say the same thing. That’s true. So much the same that they are just like frogs before a storm! You can’t hear anything for their croaking.’

Many will reply, ‘Oblonsky is a dilettante, not a serious character, you can’t compare him with people like Robert Peston’. Oblonsky isn’t a dummy, he’s brighter than many of the posh duffers in his club. And also consider Anna’s husband, Karenin – a terrifying reminder that the ‘serious characters’ in politics are really no better than Oblonsky regarding fashion. In politics, just about all of us are some combination of Oblonsky and Karenin. If you think you aren’t, you’re probably fooling yourself. If you’re on TV a lot, you’re almost definitely fooling yourself.

There are many examples of how real Oblonskys, who control practically all important cultural institutions, think. They believed things about Stalin’s regime so outlandish that it is hard to appreciate now. They were more in favour of Britain joining the euro, not because they understood ‘the complexities’ better but because they were suckered into thinking about it as a moral test – are you on the side of the ‘baddies’ or the goodies’? As the BBC Europe editor said to me back then, in similar terms to Matthew Parris about the 2016 referendum, ‘the thing is Dominic, we like foreigners and cappuccinos and we hate racists’. Polls show that better educated people are less likely to have accurate views about the science of evolution and genetics (their desire to send moral signals suckers them into believing fairy tales).

The conformity of the educated is in some ways a good thing – most obviously, a basic consensus about things like not killing one’s domestic opponents that is extremely unusual historically. But it has many bad effects too. There is a collective lack of imagination which makes the system very susceptible to disastrous shocks. They share a narrow set of ideas about how the world works which mistakes their own view as the only possible sensible approach. They are aways writing about how ‘shocking’ things are to them – things that never were as low probability events as they imagine.  They can’t imagine something like Stalin deliberately creating a famine or deliberately murdering millions. They tell themselves that Hitler will be ‘more sensible in power’ and ‘engagement’ is the right path. Western liberals (like Clinton and many pro-euro campaigners) and conservatives (like Bush) talked of relations with Putin as if he is a normal western politician rather than an ex-KGB mafia overlord with views very far from western liberals. They tell each other ‘I can’t imagine President Trump, it just can’t happen’. Many conservatives are now telling themselves that they should not take Trump too literally but that too is a failure of imagination – his character is clear to those unblinded by gang mentality and he will govern in character.

The referendum was a great example of this. Large numbers of people better educated than average – the sort of people who work as producers at the BBC – talked about their vote like this:

‘Farage is racist, he hates gay people and made that comment about foreigners with HIV, he wants to turn the clock back and pull the drawbridge up, I’m not like that, my friends aren’t like that, I am on the other side to people like that, I am tolerant and modern, I will vote IN.’

All over the country sentiments almost identical to this were expressed in large numbers. The idea that millions of graduates voted because they ‘studied the issues’ is laughable to anybody who spent time measuring opinion honestly. Almost none of these people know more about what a Customs Union is than a bricky in Darlington. They did not vote on the basis of thinking hard about the dynamics of EMU or about how Brussels will cope with issues like gene drives. Millions thought – there’s two gangs and I know which one I’m in. Another subset of the better educated feared the short-term economic disruption of a Leave vote would cost them money. They also did not vote on the basis of deep consideration of the issues.

The modern day Oblonsky reads an op-ed about how ‘the CBI warns of the dangers of leaving the Single Market’ and ‘the dangers of racist extremists’ and, having no idea of what ‘the Single Market’ is, jabbers away at their dinner party about how concerned they are about leaving ‘the Single Market’, and a warm haze of knowing one is on the ‘good’ side of the argument envelops the brain.

When it comes to the central issues of the nature of the EU’s trading relationships and what a UK-EU relationship might look like outside the EU, we are dealing with a particularly strong example of this phenomenon. Not only do the Oblonskys not know what they are talking about, neither do almost any of the supposed experts and specialists.

Lots of people said to me ‘when are you going to set out the details of the UK-EU trade relationship if you win?’ What would have been the point of that?! Approximately nobody knows anything about the important details of how the EU works including the MPs who have spent years talking about it and the journalists who cover it – indeed, often those who talk about it most are the most ignorant (and most overconfident). This is still true six months after the vote – imagine how much more true it was in the six months before the vote.

I am not aware of a single MP or political journalist who understands the Single Market – its history, its nature, its dynamics, its legal system, the complex interactions between law, economics, business, history and so on. Cameron, Osborne and Clegg certainly don’t. Neither does Bill Cash. Neither does any head of the CBI. Neither do Jon Snow, Robert Peston, Evan Davis or John Humphreys so they do a rubbish job of exposing politicians’ ignorance.

The number of people who do is tiny. In our campaign there were two – Oliver Lewis and Richard Howell – who understood a large fraction of it and the common misconceptions. They constantly had to explain to MPs, MEPs, and journalists why their ideas were misunderstandings. Maybe there is a business/economics journalist somewhere who really understands it. There are certainly some exceptional lawyers who understand narrow aspects extremely well, though few of these also understand the political and business dimensions. I have spoken to many very successful business people and never met one on either side who understands the Single Market in depth. In the entire campaign I am not aware of a single programme on TV that even tried to delve into these issues seriously (Newsnight was particularly bad, combining smugness and vapidity such as dropping Evan Davis by helicopter on an offshore platform to babble about ‘sovereignty’ trying to make the Leave side look like a bunch of weirdo cranks). British elites handed over power to the Monnet-Delors project with barely one-in-a-thousand understanding in detail why, what it involved, and its likely evolution (and that  one-in-a-thousand almost all concluded that the public could not be trusted to know the truth – I’ll explore another time the ideas of this tiny group).

Further, it was clear that Cameron/Osborne intended to run a campaign based on hysterical warnings and bogus arguments/figures while ignoring the big questions about how the EU works and its trajectory. No10 tried to turn the whole complex issue into a question about whether the economy would grow a little bit slower over the next few years – a trivial issue relative to the significance of the overall question. They are not a duo who have ever engaged the public on a serious matter in a serious way. Their brains don’t work like that. They formed early habits of looking at everything through a very narrow prism of SW1 conventional political wisdom. Given this, the way the media works, how outnumbered we were among the influential broadcast media, and the way in which the media (inevitably to some extent) takes its lead from No10, why would I have tried to run a campaign based on educating normal people to a far higher level than the professionals and ‘experts’ who were fighting and covering the campaign? It would have been impossible to get even two sensible MPs to explain the same complex argument about such things on TV without cocking it up – it was hard enough to get people to say ‘let’s spend our money on our priorities’ without days of arguing. (If the vote had happened in 2017 and we’d had all that time to build sensibly more could have been done.)

We did try to get the media to focus on deeper questions of how the EU is run, its problems, its evolution and so on. We knew from the research that the more coverage of the EU, its powers, its record, its plans and so on the better for us. We had little to fear from serious policy discussion and much to gain. But we largely failed. (A big speech from Gove was turned by the Financial Times – yes, the FT that bemoaned the ‘low quality debate’ – into a story about whether he had ‘gaffed’ by mentioning Albania, though in plastering ‘Albania’ all over the place the FT accidentally helped us.) No10 calls up the BBC and says, ‘we’ve got a business letter tomorrow with dozens of household companies warning of Armageddon.’ If we published something worthy on the Eurozone’s debt and demographic nightmare, the structural problems of the Eurozone and implications of the Five Presidents’ Report, how far did this get? ‘Sounds boring. Who’s fronting it? Got any new names? Any chance of Boris putting the boot into Dave and George?’, is the first question from the BBC TV producer who has no interest in ‘the arguments’.

It was not in our power to change basics of how the media works. We therefore  twisted them to our advantage to hack the system.

Hack the medium, hack the message: ‘the alternative government’ launches Sunday 29 May

The media is obsessed with process and the snakes and ladders of careers. Many hacks said to me words to the effect: ‘I don’t care about the issues, I care about whether Cameron will still be PM at the end of the year.’ We could not match No10 in the golden currency of ‘names’. But we could give the media an even more valuable currency – a leadership story. When Boris and Gove decided to go for it after 29 May  immigration numbers, we launched the story of ‘the alternative government’.

The media were understandably obsessed with this story so we served it up to them in such a way that they also had to cover our message. For 10 days, we dominated the news with a set of stories on the Australian points system, VAT on fuel, Turkey, the NHS and so on all based on ‘it’s safer to take back control’. Broadcasters lapped it up – even ITV News which barely pretended to be impartial was useful.

What did the public hear? They heard that prominent Conservatives, particularly Boris and Gove, did not trust Cameron’s promises or warnings and that there was an alternative path – we could ‘take back control’, have ‘an Australian style immigration system’, and ‘spend our money on our priorities like the NHS’. In an environment in which the central arguments concerning trade and the economy were incomprehensible to the ‘experts’ themselves and the history and dynamics of the EU either unknown to or suppressed by broadcasters, people chose between two simple stories. Vote Leave’s was more psychologically compelling, given the three powerful forces at work and No10’s errors.

(NB. Whoever leaked the Hilary email story was probably doing something similar. This played into the media obsession with scandal and process such that they spent a ridiculous amount of time on it despite probably 80% of them wanting Hilary to win. It shows how powerfully the media is in the grip of dynamics they rarely reflect on themselves. Putin’s communication maestro, Surkov, uses these sorts of tricks all the time. Cf. Peter Pomerantsev’s great book, a must read for any MP before they pontificate on Putin’s mafia government.)

*

The political media and how to improve it

High prestige pundits and editors yield great power over the stories told (and have far more power over politicians like Cameron, unfortunately, than they realise) but the field is not based on real expertise. Fields dominated by real expertise are distinguished by two features: 1) there is enough informational structure in the environment such that reliable predictions are possible despite complexity and 2) there is effective feedback so learning is possible.

Neither condition applies generally to politics or the political media. In the most rigorous studies done, it has been shown that in general political experts are little better than the proverbial dart throwing chimp and that those most confident in their big picture views and are most often on TV  – people like Robert Peston, Jon Snow, and Evan Davis – are the least accurate political ‘experts’ (cf. HERE).

We know that cognitive diversity is vital for political accuracy yet almost all political institutions and the media – including the dominant people at Newsnight, the Economist, the FT, and Parliament – are actually remarkably homogenous, as discussed above, and they herd around very similar ideas about how the world works. Scientists and entrepreneurs in particular are almost totally excluded from political influence.

There is no structure to hold them to account either internally or externally so, like anyone when not forced to be rigorous, they fool themselves. It is normal to write month after month that the IN campaign cannot lose because of XYZ then just as confidently and authoritatively explain why IN lost without any intermediate step of identifying and explaining errors.

Despite the rise of social media most people get most of their news from TV. TV coverage of politics rarely illuminates much because there is no clear way to decide who is right about anything. The format makes it almost impossible for any useful discussion to happen. Interviewers, politicians, and pundits talk past each other with no clarity about assumptions. Questions are vague, often meaningless, posed by interviewers who rarely have more than a thin bluffer’s understanding of any policy issue and the same is usually true of those answering; the more famous the interviewer, the less likely it is they know anything about, say, education policy and like David Cameron they are bluffing. (As soon as a story is deemed ‘political’ it is taken out of the hands of specialists (who are very rarely actually specialists anyway) and given to ‘political’ hacks who have no idea of the policy.) Most of those professionally involved are much more interested in the ‘horse race’ political dimension than the policy. They obsess on process and scandal but most people have no interest in the process or ‘scandals’  because they assume ‘they’re all dodgy in some way’. Nobody tries to make predictions that can be checked and the shows don’t take what is said seriously enough to catalogue it. Simplistic stories compete so political analysis is dominated by endless false dichotomies.

Those making the shows do not understand how people learn so the dead format recycles grim clichés like Evan Davis saying ‘… economy down the plug hole’, while filming an actual plug hole, or Nick Robinson saying ‘… will the economy take off’ standing in front of a plane actually taking off (both of these have happened). Every night the News contains reports that are a mix of incomprehensible, facile, and boring to millions while also usually at best simplistic and often just wrong when it comes to policy / issues. The possibilities of the medium are largely ignored.

Insiders think of the masses as being irrational in paying so little attention to political debate. I think they are rational. If you want to understand politics you should read serious things and invest time and effort in researching public opinion. You should particularly make an effort to invert your point of view and consider opinions very different to your own. Time spent watching/listening to shows like Newsnight and Today is not just wasted – it is actively distorting reality and making you less informed. I often meet people who are cleverer than those in politics and successful but they have deluded views about politics because they pay too much attention to political analysis. Overall, unless you are professionally involved in politics you will be better off if you stop >95% active reading of political analysis. You will miss occasional worthwhile things but the effort of sorting them is not worth it. If something is genuinely very good / unusual and you have avoided isolating yourself in an echo chamber that insulates you from opinions very different to your own then someone reliable will send it to you. Even if you are professionally involved in politics I would do roughly the same. Extreme focus on important things you can control will repay far far more than time spent reading speculation about things you can’t control.

I read very little punditry during the campaign – just enough to preserve a sense of the gaps between the ‘croaking frogs’ and the real world. If I’d had less infighting to deal with I’d have read even less as I could have been less concerned about tracking certain things. In my entire time in the DfE (three years) I never listened to Today once (I listened to a handful of interviews on the web). I focused on managing priorities and saying ‘No, stop, that’s a waste of time’ every day.

This situation is particularly ironic because the media industry is in a panic about the internet, falling ad revenues and profits, the collapse of print journalism and so on.

A better way…

There is a better way.

Example 1. Shows should require precise quantitative predictions about well-formed questions as Superforecasters do. Newspapers should do the same when interviewing people. The next step is using this process to push people towards admitting conditional errors like ‘if I am proved wrong about X by date Y then I will admit I was wrong to claim Z’. If political shows pushed their guests to do this and kept track of the predictions it could have a big positive effect. (Next time they come on you can flash up their record on a screen so the public can see how often they are right.) It is vital to change incentives so people are encouraged to admit errors and learn instead of fooling themselves constantly. For those who refuse it would be easy to develop a protocol that categorises their vague comments and puts numbers on them. This will push them to ‘correct the record’.

Example 2. Rip up the format for political shows and base broadcasts on a) an empirical assessment of what people actually know and b) the science of how people really learn and how best to communicate. Instead of the tedious low-information interviews, imagine what could be done if one had a mix of artists, scientists, and policy specialists trying really hard to use the possibilities of film to explain things, then used cutting edge data science to test how effective they were as part of a learning cycle driving higher quality. A news broadcast now contains much less  information content and much higher noise than reading. The only way to improve this is experimenting with formats in a scientific way. Doing this would force those making the news to think more about policy and the audience would be much more engaged. People are interested in policy and ‘how X will affect me, my family, and my community’. It would also obviously require a lot of changes in the media but this is coming anyway because existing business models are blowing up.

Example 3. Pay for this partly by firing most of your political commentators like Dan Hodges. Broadcasters, fire 90% of your political correspondents. They are a waste of money. Hire a much smaller number of much better people with radically different skills and backgrounds and a different focus. (By doing #1 you will soon see who is more/less accurate so you’ll have a good benchmark.)

Together these changes could improve the quality by a factor of x10 or more. The same principle of focusing on precise quantitative predictions about well-formed questions could also be used to improve policy making and management of bureaucracies by developing clusters of well-formed questions that ‘surround’ a vaguer big question that is not so susceptible to measurement.

For example, break down ‘will Britain leaving the EU be a success?‘ into dozens of simpler more precise questions that can be quantified and which together give a useful part of an overall answer. This process could be put on a prediction platform for little money and dramatically improve the quality of decisions. The Superforecasters new consultancy could do this pretty simply with little help and not much money.

As usual in systems that are failing, the youngest people understand the problems and possibilities best while the most senior / prestigious figures can’t think of anything to do other than get overpaid for what they’ve always done. If you run a big media company, you should replace the expensive old schoolers like Jon Snow with  younger, cheaper, and brighter new schoolers with an extreme focus on the public, not SW1.

*

An example of a simple, powerful media story that is wrong and contributed to forecasting errors on Brexit – ‘the centre ground’

One of the most misleading stories in politics is the story of ‘the centre ground’. In this story people’s views are distributed on an X-axis with ‘extreme left’ at one end, ‘extreme right’ at the other end, and ‘the centre ground’ in the middle. People in ‘the centre’ are ‘moderate’. ‘Extremists’ are always ‘lurching’ while ‘sensible moderates’ are urged to ‘occupy the centre’.

This story is one of the dominant features of political discussion and the basis for endless interviews, columns, and attempts at political ‘strategy’. The story is deeply flawed and where it is not trivially true it is deeply misleading.

Swing voters who decide elections – both those who swing between Conservative/Labour and those who swing between IN/OUT – do not think like this. They support much tougher policies on violent crime than most Tory MPs AND much higher taxes on the rich than Blair, Brown, and Miliband. They support much tougher anti-terrorism laws than most Tory MPs AND they support much tougher action on white collar criminals and executive pay than Blair, Brown, and Miliband.

One of the key delusions that ‘the centre ground’ caused in SW1 concerned immigration. Most people convinced themselves that ‘swing voters’ must have a ‘moderate’ and ‘centre ground’ view between Farage and Corbyn. Wrong. About 80% of the country including almost all swing voters agreed with UKIP that immigration was out of control and something like an Australian points system was a good idea. This was true across party lines.

This was brought home to me very starkly one day. I was conducting focus groups of Conservative voters. I talked with them about immigration for 20 minutes (all focus groups now start with immigration and tend to revert to it within two minutes unless you stop them). We then moved onto the economy. After two minutes of listening I was puzzled and said – who did you vote for? Labour they all said. An admin error by the company meant that I had been talking to core Labour voters, not core Tory voters.  On the subject of immigration, these working class / lower middle class people were practically indistinguishable from all the Tories and UKIP people I had been talking to.

The media tried to categorise Vote Leave as ‘right wing’ while Tory MPs and Farage’s gang were screaming at me about our championing the NHS and our attacks on the indefensible pay of FTSE CEOs. SW1 did not understand our appeal but the crucial voters did because they do not think as the ‘experts’ think they think. We tried to speak to a majority in the country. Cameron and Osborne have never won even 40%. They approached it as they did previous battles but this greatly limited their appeal. Most UKIP and Tory voters (rather than MPs/insiders) agreed with us on the NHS and executive pay while also agreeing with us on the need to take back control of immigration policy from a system that has obviously failed. Our campaign was neither Left nor Right in the eyes of the crucial audience.

The media made a similar mistake with Trump. Trump did lots of things wrong and the post facto re-branding of his campaign as ‘brilliant’ is very silly. BUT he had a national message the core of which appealed to a big majority and which defied categorisation as Left/Right. Again the media do not realise this – they label it, like Vote Leave, as ‘populist right’ (abetted by some charlatan academics). But the reason why it is successful is exactly because it is not a simple right-wing message.

It doesn’t occur to SW1 and the media that outside London their general outlook is seen as extreme. Have an immigration policy that guarantees free movement rights even for murderers, so we cannot deport them or keep them locked up after they are released? Extreme. Have open doors to the EU and don’t build the infrastructure needed? Extreme. Take violent thugs who kick women down stairs on CCTV, there is no doubt about their identity, and either don’t send them to jail or they’re out in a few months? Extreme. Have a set of policies that stops you dealing with the likes of ‘the guy with the hook’ for over a decade while still giving benefits to his family? Extreme. Ignore warnings about the dangers of financial derivatives, including from the most successful investor in the history of the world, and just keep pocketing the taxes from the banks and spending your time on trivia rather than possible disasters? Extreme. Make us – living on average wages without all your lucky advantages – pay for your bailouts while you keep getting raises and bonuses? Extreme and stupid – and contemptible.

These views are held across educational lines, across party lines, and across class lines. Cameron, Blair, and Evan Davis agree about lots of these things and tell people constantly why they are wrong to think differently but to millions they are the extremists.

(This is not a post facto rationalisation. I wrote about the centre ground and the EU in 2014 HERE.)

*

Why I got involved and my role

Winning the referendum against Cameron was not the way I wanted things to happen. I thought the chances of winning a referendum against a PM on the other side, with all the possibilities for him to mobilise the system behind IN, were low. Many prominent Eurosceptics (not all) lobbied for it out of a combination of self-promotion and not knowing how to solve the real problem – what should the UK-EU relationship be? The referendum was very useful for many Out-ers: it provided a much simpler political focus than figuring out a complex positive agenda, removed the need for difficult thinking and action, and gave people a chance to pose on the side of ‘democracy’. I thought it foolish to push for a referendum while simultaneously not building a serious movement to win it. (I had tried to start building such a movement in 2004 after the euro battle was clearly won but could not persuade crucial people so decided to drop the issue for a while.) Romantic long shots are rarely wise in politics particularly if there is a better path.

I thought it wiser and safer to wait for Cameron to go then try to capture the Tory leadership and change the UK-EU relationship from Downing Street with someone who actually wanted to solve the problems (Cameron’s best friends would not claim that he wanted to spend his time trying to solve these deep problems, he wanted not to think about the EU and got into an existential battle he never wanted). If you are going to have a referendum, then have it when controlling the institutions and when you can set the agenda. A British PM could invite the EU to evolve such as to include a) those in the euro, Single Market and ‘free movement’, and b) those outside some or all of those three but with free trade and friendly cooperation between all. The chances are low that there would have been support for fundamental change but then a divorce could have happened after a serious clarifying debate which would have occurred ~2018-25, including the Eurozone countries figuring out what they would do. This would have been a much better way to proceed than the charade of Cameron’s ‘renegotiation’. Either Europe would have embraced a new and more open architecture (unlikely) or the Government would have won a Leave referendum with 60-70% and prompted a lot of clarifying thought across Europe.

I also thought it foolish of Cameron to cave into the pressure and promise a referendum in 2013. So did Gove and Osborne both of whom told Cameron not to do it. He mistakenly thought it would take the wind out of UKIP’s sails and did not understand why it would actually boost UKIP and Farage. (This was not hard to foresee and I suspect part of the problem was that Cameron did not appreciate that him promising a referendum would be thought by most as just a typical pre-election lie.) The idea that there was an irresistible force for a referendum is pushed by Farage’s and Cameron’s supporters. They are both wrong. The country supported one but without any passion outside the small fraction who had long been passionate about it. Most Tory MPs did not want it. Most Tory donors thought the timing was wrong and wanted a focus on stopping Miliband who they feared. Those MPs who did want it could mostly have been bought off or distracted in other ways – a mix of some policy, gongs, bribes, and so on in the usual fashion. Putting a date on the vote was particularly mistaken – it would have been far better to leave it open-ended ‘in the next Parliament’.

Once the election happened there was a sudden panic among OUT-ers. UKIP was an organisational disaster. There was no national campaign prepared. There were many tiny groups who often hated each other more than they wanted to win and were conditioned to expect failure and defeat. There was an abundance of people who thought that the campaign was quite simple – put me on TV, they thought, and the nation will appreciate my natural leadership. There was practically nothing of what was actually needed. Many quickly flipped into panic mode assuming the vote was unwinnable.

Having opposed the push for a referendum, I was faced with an uncomfortable choice in May 2015. Either keep out of politics, refuse to help, and then feel miserable about the tragicomic campaign, or re-engage with people I did not want to work with, feel miserable about the tragicomic campaign, and in almost every way make my life worse. In many ways irrationally, I chose the latter. My thinking was something like this: the chance of changing the whole political system (more profoundly than in a normal election) comes along very rarely, the chaos of the eurosceptics and the complacency of Cameron creates a very slim bridge to seize control and do it, a small chance of very high impact is worth the gamble. About a month or so later my wife was pregnant. If the timing had been slightly different I might well have stayed retired.

Why do it?

I thought that Leaving would improve the probability of 1) Britain contributing positively to the world and 2) minimising dangers. I thought it would:

  1. minimise Britain’s exposure to the problems caused by the EU;
  2. improve the probability that others in Europe would change course before more big crises hit, e.g. by limiting free movement which is the biggest threat to continued free trade;
  3. require and therefore hopefully spark big changes in the fundamental wiring of UK government including an extremely strong intelligent focus on making Britain the best place in the world for science and education;
  4. improve the probability of building new institutions for international cooperation to minimise the probability of disasters.

The foundation problem with the EU was best summarised by the brilliant physicist David Deutsch, the man who extended Alan Turing’s 1936 paper on computation into the realm of quantum mechanics. Deutsch said:

‘The EU is incompatible with Britain’s more advanced political culture. I’m voting Leave… [E]rror correction is the basic issue, and I can’t foresee the EU improving much in this respect… [P]reserving the institutions of error correction is more important than any policy… Whether errors can be corrected without violence is not a “concern” but a condition for successfully addressing concerns.’

Healthy and effective systems like our immune system and the English common law allow constant and rapid error-correction. Unhealthy and ineffective systems like the EU and modern Whitehall departments block error-correction. They are extremely centralised and hierarchical therefore information processing is blocked and problems are not solved. In politics this often leads to disasters when more and more resources are devoted to reinforcing failure. NB. This most fundamental question played effectively no role in the debate.

This fundamental problem generates its other problems. It arises because of how Monnet and Delors created its institutions deliberately in opposition to the Anglo-American system they bitterly opposed. The Foreign Office romantic delusion of ‘influence’ was peddled by every PM since Thatcher. Every one left office having demonstrated how empty the hope is. True influence comes from demonstrating success – not sitting in meetings for forty years in an institution that is programmed on principles that guarantee worse error-correction than the evolved institutions  of the Anglo-American system.

I will go into the problems of the EU another time. I will just make one important point here.

I thought very strongly that 1) a return to 1930s protectionism would be disastrous, 2) the fastest route to this is continuing with no democratic control over immigration or  human rights policies for terrorists and other serious criminals, therefore 3) the best practical policy is to reduce (for a while) unskilled immigration and increase high skills immigration particularly those with very hard skills in maths, physics and computer science, 4) this requires getting out of the EU, 5) hopefully it will prod the rest of Europe to limit immigration and therefore limit the extremist forces that otherwise will try to rip down free trade.

One of our campaign’s biggest failures was to get even SW1 to think seriously about this, never mind millions of voters. Instead the false idea spread and is still dominant that if you are on the side of free trade, think controlled immigration generally a positive force, and want more international cooperation rather than a return to competing nation states then you must support the EU. I think this error is caused by the moral signalling and gang mentality described above.

What was my role?

My role mainly involved:

a) trying to suppress/divert/overcome internal coalition warfare to a level where about ten crucial people were protected enough to do their jobs,

b) building the team,

c) management,

d) taking a small number of important decisions about policy, message, money, and the machine,

e) providing clear focus and priorities, including the vital job that nobody likes of saying ‘no’ to hundreds of people (thus making (a) harder), and

f) dealing with big problems.

The media tends to suggest my role was mainly talking to them. This is wrong. The same happened with my role in the DfE. In both projects my main role was management. Serious management means extreme focus and this requires saying No an awful lot. Contrary to the media story, I dislike confrontation and rows like most people but I am very strongly motivated by doing things in a certain way and am not motivated by people in SW1 liking me. This is often confused with having a personality that likes fighting with people. One of the basic reasons so much in politics is mismanaged is that so often those responsible are more interested in social relations than in results and unlike in other more successful fields the incentives are not structured to control this instinct.

Many have written that I got involved with this because of ‘hate’ or ‘loathing’ for Cameron. Wrong. I do not hate Cameron. I do not respect him, which is different. I thought that he was in politics for bad reasons – essentially because he was someone who wanted ‘To Be’, not someone who wanted ‘To Do’ (see the Colonel Boyd speech) and his priority was himself and a small gang, not the public. I also thought Cameron was mostly (not all) bad at the job, despite having some of the  necessary temperamental characteristics, and was flattered by having Brown then Miliband as opponents. I didn’t object to him blocking me from Government in 2010. He was entitled not to hire someone who did not take him seriously and ignored the orders of his Chief of Staff.

I spent a few years of my life (1999-2002) trying to stop Blair on the euro before anyone had heard of Cameron. In 2004 I co-founded the campaign that won the referendum on the North East Regional Assembly 80-20 as a training exercise for a possible future EU referendum. My motivation was the issue itself – not personal antipathy for Cameron or anybody else. I’ve never been a party person. I’m not Tory, libertarian, ‘populist’ or anything else. I follow projects I think are worthwhile.

Farage’s motley crew claim that I did this campaign in order to lose it deliberately then get a job in No 10 with Cameron. It is pointless to discuss this theory though the fact that they understood so little about the political environment, and struggled to use Google, was an important fact.

I am not clever, I have a hopeless memory, and have almost no proper ‘circle of competence’. I made lots of mistakes in the campaign. I have had success in building and managing teams. This success has not relied on a single original insight of any kind. It comes from applying what Charlie Munger calls unrecognised simplicities of effective action that one can see implemented by successful people/organisations.

Effective because they work reliably, simple enough that even I could implement them, and ‘unrecognised’ because they are hiding in plain sight but are rarely stolen and used. I found 10-15 highly motivated people who knew what they were doing and largely left them to get on with it while stopping people who did not know what they were doing interfering with them, we worked out a psychologically compelling simple story, and we applied some simple management principles that I will write about another time. It is hard to overstate the relative importance in campaigns of message over resources. Our success is an extreme example given the huge imbalance in forces on either side. In many ways Trump’s victory has little resemblance to what we did but in this respect he is another example.

We also got lucky.

*

I will post a number of blogs of the referendum to try to answer some basic questions including:

What were the main political, operational, financial/budgetary, and data/digital lessons from the campaign?

What worked and did not work?

How confident can we be about these judgements?

There is a natural set of categories and I will post links to blogs below:

  • Some basic numbers that summarise important elements.
  • Strategy, message, polls.
  • Policy.
  • Data and digital.

On data science, digital marketing, canvassing software made available for download. (NB. There has been some confusion about this blog. The VICS system is a web-based canvassing tool, the first proper one that works in the UK – it was one component of our overall data science approach and should not be equated with it. It is not a data science tool – it provided data to the data science team.)

  • The ground campaign.
  • The media.
  • Internal politics and the infighting.
  • Dynamics that affect ‘what next’.
  • The rules: how could they be improved to make future votes serve the public better?

Please leave comments and corrections below. I am happy to approve hostile comments if they have substance and will moderate comments to avoid putting sensible people off reading them.

On the Referendum #7: Transparency for our Potemkin government – Memo to ministers and spads thinking about how you could help the NO campaign

There have been many attempts to quantify the extent to which EU law (primary, secondary, Regulations, Directives, ECJ judgements etc) really determines what happens in the UK. It is inherently hard to come to an agreed answer given the combination of a) the sheer scale and complexity of EU law’s entanglement with domestic law over decades including things like domestic court interpretations of ECJ judgements, b) different definitions of regulation and the units of measurement, c) the desire of the civil service to obscure the issue, and so on.

You – ministers and spads – can contribute something valuable to this debate in a way that will help the NO campaign at a crucial time.

For those not in government reading this… One of the basic mechanisms of government is the ‘Cabinet write round’ system. This involves Secretaries of State being given lots of documents every night in their box from other departments. The SoS is supposed to read these documents and tick the relevant box on the attached form signalling assent, disagreement, comments etc. (When I find a copy of one in my papers I’ll post a photo.)

For entirely domestic things, this process can lead to disagreement and negotiation. An interesting aspect of our membership of the EU is that a large fraction of the documents concerning future law and administrative action come from the EU. For reasons that are opaque, the civil service continues with the write round system. It is, of course, a Potemkin system as ministers do not have a real power to oppose anything – the document in question will become law regardless of how the minister fills in the chitty. Still, the chitties are sent around so everybody can pretend they are in charge. This is a depressing process for some ministers but perhaps the Cabinet Office regards it as a Pavlovian exercise – ministers become habituated to simply tick everything without engaging their brains or ethics.

When occasionally a SoS refuses, the first step is the Private Office asks whether a mistake has been made. No? Are you sure minister? Off the chitty goes to the Cabinet Office (‘very courageous minister’). Step 2 is that the Cabinet Office emails to say – ‘Was your SoS drunk again last night, he seems to have rejected the EU Directive on XXX, better go and tell him to withdraw his objection pronto or Jeremy [Heywood] will be on it.’ This is normally enough to get SoS scuttling to retract his objection. Stage 3 is unusual – it involves the SoS not giving in at Stage 2. What happens then is that the SoS is informed by the private office that Ed Llewellyn has said that the Prime Minister agrees with Jeremy and insists on measure X. This flattens practically all objections. I have witnessed the very unusual Stage 4 – the SoS sends back a message asking for a meeting with Jeremy. Jeremy arrived. ‘This is EU law so there is no basis for us to object.’ Gove: ‘Why do we get sent these stupid forms to fill out then if we can’t stop these awful things, this is going to waste hundreds of millions of pounds for nothing?’ Jeremy [a chuckle]: ‘Haha, yes, so I’ll inform the Prime Minister that you agree after all, we will mention to European officials that ministers have grave concerns, I’m sure Oliver will look at it further, goodbye Michael.’ Game Over: ‘All your base belong to us’, as the old video game said…

The fairy tale that Britain still has Cabinet government involves maintaining this Potemkin process.

I have asked No10 spads a few times over the years what proportion of things they see come from the EU. The estimates have been 50-60%. When I was in the DfE, I would occasionally do surveys of Gove’s box, going through every single paper in it, to see the proportion of EU stuff. I would estimate the same – typically about half, though sometimes much more (though obviously volume does not equate to importance).

With the referendum coming, this will be an important question. The usual surveys will not answer the question. So what could you do?

From now, start collecting stats on a daily basis of the proportion of EU stuff in the box. Spad, create a GoogleDoc – obviously do not use the official system – so that the minister can simply fill in the box on the grid for each day. For the minister (or you) to jot ‘x%’ each day and fill in a GoogleDoc grid will take no more than a few seconds per day, less than a minute per week, less than an hour in a year. (If they are technically hopeless just get him to jot a figure and you fill it in.)  Also, you could take a few photos of some of the boxes (‘This one was 80% EU stuff’) and save them to Dropbox for future use. Keep copies of the 1% most stupid, irrational, and wasteful things. Add ECHR/HRA stuff too – that is all relevant particularly given Cameron is going to do nothing at all about the Charter of Fundamental Rights (NB. this is the EU thing, not the ECHR). No officials will know you are doing it. Neither Heywood nor Llewellyn will be able to know you are doing it.

After Cameron returns from the EU proclaiming triumph and some of you resign, you will then have a record of contemporaneously collected stats on the real importance of EU affairs in Government. You will be able to publish this. It will be recorded over a year or so and therefore have hundreds of data points. Much more than other surveys on this question, people will take it seriously – particularly when you explain it at a press conference holding up some photos and copies of the most stupid documents. It will be impossible for No10 to rebut it effectively. They will not be able to publish documents that could refute it. Heywood will give a statement saying that your claims are wrong but nobody will believe him.

This is a simple thing that could have a significant impact at the right time. You all know how much EU stuff is hidden by Whitehall and how much effort goes into pretending that ministers decide things that were really decided by some lobbyist in a Brussels hotel years ago. You know these Kafka-esque bureaucratic processes, redolent of the dying days of the Austro-Hungarian Empire, that characterise modern Whitehall. DO SOMETHING ABOUT IT!

Lots of you now won’t know whether you are going to resign but you can do this without anybody knowing so you have something useful if you do decide to resign; if you don’t you can delete it all, no harm done. Boris, we know you read this blog, you could do the same thing in the Mayor’s office and surely there will be some committees Cameron puts you on shortly to try to keep you quiet…

Please suggest ideas about how to improve this process,

Dominic