“The 2016 election cycle will be remembered for many things, but for those who work in politics, it may be best remembered as the year that political data reached maturity.”
— Andrew Therriault

In 2004, 22-year old Andrew Therriault wanted to get a Democrat president elected. “I got in my car and drove to Ohio,” he told me. “I walked around college campuses, signing students up to vote and all that.” But John Kerry did not get elected, and Therriault “wound up after the election, back at my parents’ house, broke, depressed and unemployed”. Talking to strangers about politics, he concluded, was not his forte.

Ten years later, Therriault found a job that did feel right: Director of Data Science for the DNC, the organising body of the US Democratic Party. By 2016, he was editing the O’Reilly publication Data And Democracy, from which the above quote is taken, describing in detail exactly how data can be used for political campaigning.

With hindsight, the 2016 election cycle was probably not best remembered as “the year that political data reached maturity”. It was, however, the year that many journalists and researchers discovered for the first time how data-driven, personalised political campaigning works, when they sought explanations for the political shocks of Britain voting Leave and the US voting Trump.

But Barack Obama had already won two Presidential elections using those techniques with increasing sophistication. A Guardian article in 2012 spared a single paragraph to privacy concerns between gushing about “a vast digital data operation that for the first time combines a unified database on millions of Americans with the power of Facebook to target individual voters to a degree never achieved before”, and the way, probably unconsciously, “the individual volunteer will be injecting all the information they store publicly on their Facebook page — home location, date of birth, interests and, crucially, network of friends — directly into the central Obama database”.

Despite the furore over Cambridge Analytica, and the supposed (but implausible) psychological manipulation of 2016 voters via social media, digital political campaigning remains the norm. Hardly surprising, in a world where we increasingly turn to personalised channels, and especially social media, for our news and commentary. How else could campaigners reach potential voters?

In the recent UK local and mayoral elections, the Labour Party spent over half a million pounds (£570,160) on advertising via Meta alone — that’s Facebook and Instagram. The Conservatives lagged behind with a mere £336,668, perhaps because their adverts tended to target pensioners who spend less time on social media: one with the headline “Tax the Codgers” was shown only to over 55s. Thanks to WhoTargetsMe you can now dig into the broad strategies of digital political campaigners, though not the details of how specific ads are targeted.

I asked Andrew Therriault how many data points he would have on a typical voter — dozens? Hundreds? Thousands? “Thousands,” he says, as if that’s obvious. But to think of that as a list of thousands of things he’d know about an individual voter would be misleading. Yes, it would include voter registration details — name, address, age — and perhaps answers previously given to surveys. But to that basic voter database Therriault’s team would add census data for the neighbourhood. Political campaigns also regularly use commercial data from data brokers such as Experian and Acxiom, collected from our interactions with businesses. Even though it’s hard to match data from those sources to individual voters, they can be used to build models, so what they do know about you can be used to predict things that they don’t know.

More from this author

We need to calm down about screen time

By Timandra Harkness

Being able to profile individual voters helps a campaigner decide who to target, and tailor the message to them. There are two kinds of people worth targeting: those who might be persuaded to support your side, and those who need a push to get out and vote. “Persuade or mobilise,” as Andrew Therriault puts it.

Like other kinds of personalised advertising, political campaigns aim to show the most effective message to the right person, at the right time. And, also like other adverts, digital media allow the advertiser to monitor which ads are the most effective for this kind of person, by running trials of different variants to see which gets the best response. It is, in short, a huge feat of digital technology. It would be a mistake, however, to see the technology itself as driving this political change. Rather, the change from mass movements to “The Personal Is Political” predates the capacity to deliver a tailored message to an individual on a device that accompanies them everywhere.

The era of mass politics is not very old. In 1848, “the springtime of the peoples”, a huge Chartist demonstration in London tried to deliver a petition calling for universal male suffrage, secret ballots, and other reforms to parliament. According to the Illustrated London News, it took a cart drawn by four farm horses to carry the signatures in “five huge bales or bundles”. The government fortified all the bridges in London to prevent the assembled masses reaching Westminster, and went on to arrest the leaders and suppress the movement.

The road to every citizen getting a vote would be long and hard. In the US, race was the bar to democratic equality: in the UK, it was property, which excluded some men until 1918, and a third of women until 1928. In France, where all men did get the vote in 1848, women would wait until 1944.

But the ability to vote a government in — or out — was not the only form of mass politics. By 1874 around 10%  of the adult population of Britain were members of TUC-affiliated unions (mostly men). Organisations pressing for equal rights for women, or for non-whites in America, also demanded equality beyond the vote. However, by the time Martin Luther King made his “I Have A Dream” speech in 1963, the character of those mass movements was changing. Even as King called on the US Government to make good the “promissory note” to all Americans, black and white, guaranteeing “the unalienable rights of life, liberty and the pursuit of happiness”, that universalist vision was fragmenting and turning inwards.

Tom Hayden, later Senator Hayden, was in the immense crowd that heard King that day. After being inspired by Jack Kerouac’s On The Road to hitchhike across America, Hayden became a Civil Rights activist and spent some time in jail. There he drafted what would become the 1962 Port Huron Statement, a manifesto for Students for a Democratic Society:

“We oppose the depersonalisation that reduces human beings to the status of things,” it declares. “The goal of man and society should be human independence; a concern not with image or popularity but with finding a meaning in life that is personally authentic.” Politics, he said, should be a “means of finding meaning in personal life”, imbuing it with a therapeutic sensibility. The Port Huron Statement captures a moment at the beginning of a new political age. It talks of “loneliness, estrangement, isolation”, and “the felt powerlessness of ordinary people.” It matters not only that people lack power, but that they feel powerless.

The distinction between politics and personal feelings was fraying, and as the Sixties drew to a close, a slogan emerged to capture this new kind of politics. “The Personal is Political” is the title of a short 1969 article written by Carol Hanisch, an activist in the Mississippi Civil Rights Movement, and a founder of New York Radical Women, the group that protested at the Miss America contest in 1968. They were all women who were active in other movements — Civil Rights, Anti-War or the New Left — who wanted to discuss politics without men present. After the meeting ended, they kept arguing into the small hours at a nearby restaurant over apple pie a la mode.

Suggested reading

Big Tech has stolen our children

By Matt Feeney

I asked Hanisch what she thought of the shift towards seeing the world in terms of identity. “It cannot be a positive change,” she told me. “We lose all sense of a need for unity to take on the source(s) of our oppression. Instead it invites us to escape the real world by “living in our heads”. It allows us to believe that if we can change ourselves and/or just a few people, everything will be fine, which is, of course, nonsense. We’re talking society-wide oppression here, not just individual attitudes. You can’t ‘identify’ yourself out of oppression, though over the centuries people sure have tried!”

Today, we take disagreement very personally. Not only issues that are entangled with our everyday lives, our feelings or how we see ourselves: even geopolitical issues or which political party we vote for are taken much more personally. People are less open to social connection with those who disagree with them politically. Survey after survey has shown that, increasingly, we are more likely to see those who disagree with us politically as closed-minded, selfish, hypocritical, immoral or lazy, and less likely to call them intelligent or honest. Negative feeling towards the other side, politically, has been steadily increasing since the Eighties.

I asked Andrew Therriault if he thought politics in general had become more personal.  “Yeah, from both the voters and the candidate side it’s become more individualised, in terms of even the basic facts we’re working with,” he says. “This idea that your political reality can be whatever you want it to be, and whatever someone else wants to tell you it should be. And there does not seem to be that sort of common grounding we once had when things were less personalised.”

Politics, in other words, is turning into a form of self-expression. It is no longer an arena to which we all bring our conflicting visions of how the world should be, and try to persuade others to join us in working towards ours — or even, in which we might listen to other people’s ideas and change our minds.

***

This article includes material drawn from Technology Is Not The Problem, published by HarperCollins.

That a private American company should gain such a foothold in the NHS would normally cause concern among privacy and health service advocates. And that it could be Palantir — part-funded in its early days by the CIA and with Western armies among its clients — is causing high anxiety. “Is Palantir really the kind of company we want at the very heart of the National Health Service?” asks Cori Crider, director of Foxglove, a campaign group dedicated to challenging the excesses of tech giants. “This is a company who, at the start of the pandemic, had no track record of working with healthcare staff. They’re not a healthcare company. They weren’t a health data company. They were essentially a tech company who supported spies, police, the military and border forces.”

Palantir, in partnership with Accenture, is in the running with Quantexa, a British company partnered with IBM, and Oracle Cerner to build the “Federated Data Platform” (FDP), described by the NHS as a system “which will enable NHS organisations to bring together operational data – currently stored in separate systems – to support staff to access the information they need in one safe and secure environment”. NHS England says there are currently no plans to include GP records in the FDP, but privacy campaigners fear that that could change with mission creep, and if it did, Palantir could gain access to them. The winner of the FDP contract was supposed to have been named in September, an announcement then delayed to the middle of this month, and now said to be imminent — but few in the NHS believe it will be anyone but Palantir.

The NHS — budget this year £182 billion — has the largest repository of health data in the world. It services around a million GP appointments a day, and over a quarter of a million hospital appointments. According to consultants Ernst and Young, the information gleaned from NHS activities could be worth £9.6 billion a year. With the advent of AI, mining this treasure trove of data could mean the discovery of better medicines and treatments, improved patient care and vast savings. But in the wrong hands, it could see the NHS being used as a cash-cow, and it could mean a loss of privacy, the end of patient trust, and the slow death of the NHS as we know it.

So, how did Palantir get in the running — and why are they being hotly tipped as a shoo-in for the FDP contract, worth £480m over seven years? The company, named after the crystal “seeing stones” in J.R.R Tolkien’s Lord of the Rings, was founded by Thiel, a Stanford law graduate, in 2003 with the intention in his words of adapting PayPal’s anti-fraud software in order to “reduce terrorism while preserving civil liberties” (he had sold PayPal to eBay for $1.5 billion in October 2002). Before the 9/11 attack on New York, the US intelligence agencies had been blind-sided; Thiel was convinced his seeing stones would give them better vision in the future.

An early investor in Palantir was the CIA, through a venture capital arm named In-Q-Tel. The first of four iterations of Thiel’s data systems — Palantir Gotham — is used by the US National Security Agency, the FBI, the Western military, police, US immigration services and fraud investigators. It is capable of pulling together vast amounts of data and making connections that could solve problems insoluble to mere humans. It is currently being used by the Ukrainian army in its war with Russia.

Suggested reading

Peter Thiel on the dangers of progress

By Mary Harrington

There have been reports — neither confirmed nor denied by Palantir — that its software helped to locate Osama bin Laden. Former CIA director George Tenet once said he wished the agency had had Gotham before 9/11, the suggestion being that it might have pulled together disparate dots into one large picture of terror plot planning. Another iteration, called Metropolis, is a finance analytics platform widely used by hedge funds and banks. In March 2020, at the onset of the pandemic, executives from major tech companies, including Palantir, Google, Meta and Amazon, were asked by the UK government whether they could help with the planning and implementation of its responses to Covid-19. Palantir immediately stepped in, offering the use of another of its systems, Foundry… for just £1. Its offer was accepted by beleaguered health officials.

Palantir advisers were embedded in the NHS while Foundry was used in planning to effectively distribute medicines, ventilators and vaccines, monitor staff levels and intensive care bed availability — creating the kind of joined-up health response that would not have been previously possible with the UK’s fragmented network of trusts and care providers. Once a vaccine became available, it was involved in its rollout. By the end of the pandemic, Palantir’s £1 punt had paid off.

“A £1 contract seemed odd at the time, but it obviously gave them a foot in the door,” says Dr David Wrigley, digital lead on the BMA’s GP Committee, which has expressed concerns over the use of patients’ data on the FDP. “They will have learned about NHS systems and contracts, seen how the system operates and how best to bid for other contracts going forward.” Palantir’s next move to curry favour with the UK government was to offer its services again, for six months, to help with the logistics of housing Ukrainian refugees following the Russian invasion. But this time it charged nothing whatsoever. The company says it is proud of the role that it played at a time of great suffering for Ukrainian families. But this move did its relationship with the UK, and its long-term income, no harm at all.

Since then, Palantir has been awarded several other NHS contracts worth £60 million, with none going out to tender. Last month, it was awarded £5.5million to continue running Homes for Ukraine over the next 12 months for the Department for Levelling Up, Housing and Communities (DLUHC) — again with no tendering process. The probity of this deal was questioned by the National Audit Office (NAO) in a report published last week. It pointed out that: “In February 2023, the Government’s Chief Commercial Officer, wrote to Palantir noting his concern about the practice of offering services to public sector customers for a zero or nominal cost to gain a commercial foothold, contrary to the principles of public procurement which usually require open competition.” Asked about this criticism, Palantir said there was: “nothing unusual in a business offering a prospective customer the opportunity to trial before purchase. Indeed the Government’s own guidance states that ‘where a service is being outsourced for the first time, a pilot should be run’ and that ‘if you’re buying a product, consider asking for a demonstration or trial on a smaller scale’.”

More from this author

The forgotten victims of online child abuse

By Steve Boggan

But a more polemical vision of how Thiel sees British people’s relationship with the NHS was revealed in January when he made some unguarded comments during a Q&A session at the Oxford Union debating society. He said they were suffering from “Stockholm syndrome”, held ideologically captive by their healthcare system, and that the NHS needed “market mechanisms” to improve, adding that: “In theory, you just rip the whole thing from the ground and start over… In practice, you have to somehow make it all backwards compatible in all these ridiculous British ways.” He said British people needed to stop thinking the NHS was “the most wonderful thing in the world” and accept that it is an “iatrogenic” institution. An iatrogenic illness is a malady caused by the treatment itself. “Highways create traffic jams, welfare creates poverty, schools make people dumb, and the NHS makes people sick,” he said.

Palantir quickly distanced itself from his comments. Joanna Peller, the company’s UK health lead, said: “Peter Thiel made these comments as a private individual and our CEO [Alex Karp, at Palantir since 2004] has made it clear he firmly disagrees with them and that he wishes ‘we had a health care system in the US that served the poor and underserved as well as I perceive the British system does’.” Few people in the NHS would argue that its countrywide IT systems couldn’t be improved, but many billions of pounds have been wasted in previous unsuccessful attempts to merge them. At the moment, there are 42 administered by “Integrated Care Systems” comprising hospital trusts, local authorities and care providers.

The first attempt at digitising and uniting records held by GPs and hospital trusts was launched in 2002 and called the National Programme for IT. It began with a budget of £6.2 billion but was mired in technical difficulties and a lack of trust among health staff. It was scrapped in 2011 after racking up £12 billion in costs. One of the private contractors who pulled out of the scheme before it was junked was Accenture, Palantir’s current bidding partner. In 2021 the government launched the General Practice Data for Planning and Research system, designed to enable records to be used for more than simple personal care. Advances in technology meant that the data in them could be used for the same purposes envisaged for the FDP — better care, medicines and healthcare planning.

However, patients were advised that where their data was to be shared for anything beyond care — for research or planning — they had the right to withhold their data under an NHS “National Data Opt-Out”, and they did in their droves. “More than a million people opted out because there was a lack of trust in the system,” says Foxglove’s Crider. “And each time someone opts out, the data set gets smaller and so its usefulness for research is diminished.” Sam Smith, policy lead at medConfidential, which campaigns for patient privacy, says he believes even more people will want to opt-out if Palantir wins the FDP contract. But there is confusion over whether they will be able to. Smith explains:

“The NHS England website says people will not be able to opt out, but this is contradictory because we all have the right to opt out of sharing our personal data if it is to be used for anything other than patient care, for purposes such as research and planning — but that is exactly what FDP is supposed to be used for… Last month, Sir Chris Whitty [chief medical officer for England] urged people not to opt out precisely because their information could help with research and planning. So what is it to be? Can people opt out, or can’t they?”

In a letter to the Health and Social Care Committee in August, Lord Markham, under-secretary of state for health, said the government was considering reforming the National Data Opt-Out. I asked NHS England whether these reforms would make it easier or harder for patients to opt out, but it did not answer the question. On the subject of Palantir and the FDP, a spokesperson said:

“The NHS is conducting a fair and transparent procurement process for a supplier of the federated data platform, in line with public contracts regulations, and this process has not yet concluded… Better use of data brings huge benefits for patients — a federated data platform will ensure more joined-up care and better use of resources, connecting existing NHS data in one secure environment to help speed up diagnosis and reduce waiting times and hospital stays.”

Palantir’s Peller emphasises that patients’ data is safe with the company. “Unlike many other technology companies, we’re not in the business of collecting, mining or selling data,” she says. “What we do is provide tools that help customers understand and organise the information that they hold, along with training and support in using those tools. We are legally defined as the ‘data processor’ while the customer is the ‘data controller’. Customers retain full ownership of their data. Palantir is granted no rights to such data and can only carry out that processing which it is instructed to do by the customer, while UK government customer data is only hosted in the UK.” She says Palantir is “now helping to reduce the care backlog and ensure patients receive vital treatment sooner in 36 NHS Trusts across England”. These are trusts invited (some say pressured) by NHS England to run trials of Foundry, and this is the reason Palantir is considered a shoo-in for the FDP contract — it is too embedded to get out.

However, in response to a written Parliamentary question in March, health minister Will Quince said nine of these had been “paused” while two had been suspended. A variety of reasons were later given, ranging from a lack of funds to administrative delays, and assurances were made in August that some of the paused pilots were up and running again. But at least two — Liverpool Heart & Chest Hospital, and Milton Keynes University Hospital — have indicated that the system wasn’t right for them. I asked NHS England how many of the 36 pilots were currently operational, but received no answer. This only confuses the picture, as Palantir insists they all are.

Suggested Reading

How Turbo-Wokism broke America

By David Samuels

Whichever company is announced as the provider of the FDP, its success will not only be judged on whether the NHS is finally blessed with the joined-up digital wizardry it so badly needs, but also on whether it manages to provide that and retain the trust of patients. “If a person goes to see their GP and holds back important or embarrassing information because they’re afraid it will be shared with a private company, then that is not only dangerous for the patient, but also very sad,” says the BMA’s Wrigley. “Because that could mean the end of the doctor-patient relationship as we know it.”

This policy divergence created an opportunity to examine the impact of mask mandates. Those areas that rescinded their mandates could function as control groups for evaluating the effect of the policy on cases and mortality. But neither public health experts nor any of the major media outlets took up this opportunity. The reason, most of them would likely have said if pressed on the subject, was there was nothing to learn: “the science” was settled.

For their part, critics of masking and other Non-Pharmaceutical Interventions (NPIs) typically relied on principled assertions of freedom. This gave them little reason to examine the evidentiary basis of these policies, since they would have rejected them on moral grounds even if they worked. Only a few sceptical observers drilled into the data that could be found on sites such as the New York Times — even if the paper’s own reporters made little of it. The most prolific of these was Ian Miller, who over the past two years has published copious data-driven commentary on the track record of various Covid public health interventions.

Miller arrives time and again at the same conclusion: that the ad hoc pandemic mitigation policies rolled out since 2020 have systematically failed to achieve goals. Miller has now compiled one subset of his graphs and commentaries into a book titled Unmasked: The Global Failure of Covid Mask Mandates which focuses solely on the most ubiquitous pandemic containment strategy deployed by governments worldwide: masks.

The endorsement of masking by medical bodies and public health authorities worldwide, Miller shows, entailed the abandonment of a longstanding view that masks were a useless and even harmful intervention. Over the previous decades, numerous randomised controlled trials had assessed the efficacy of masks in controlling the spread of respiratory viruses like influenza, and pandemic simulations had evaluated their potential.

A document published by the World Health Organisation in 2019 framed the results of these studies in no uncertain terms: “there was no evidence that face masks are effective in reducing transmission of laboratory-confirmed influenza”. It’s unsurprising, then, that when the CDC briefed reporters on the pandemic in February of 2020, masking was not even mentioned among the NPIs that might be deployed. The UK government, too, stated early in 2020 that there was no evidence to support masking.

More from this author

How vaccine mandates became a political weapon

By Geoff Shullenberger

After the CDC and other agencies revised their guidance in April 2020, Dr. Anthony Fauci, by then a staunch advocate of masks, claimed that he and other officials had discouraged the public from obtaining masks to ensure there were adequate supplies for health care workers. Ever since, promoters of masking have cited Fauci’s “noble lie” to account for the abrupt reversal of prior guidance. But as Miller notes, it was not just during the early months of the pandemic that officials said masks were ineffective. They had said so for years, and Fauci had advised against masks not just in public statements but in private emails in early 2020.

In light of the earlier consensus, Miller’s findings in Unmasked should not be surprising. As we might have predicted based on a plethora of trials and meta-analyses published prior to the pandemic, mask mandates have had little to no demonstrable impact on curbing the spread of the virus. Miller reaches this conclusion by comparing areas with mask mandates of longer and shorter duration with each other and with areas that never imposed mandates at all. The results, he shows, simply do not support the standard adage that “masks save lives”.

For example, Los Angeles County, the most populous county in the US, imposed one of the earliest mask mandates in the United States, just after the CDC released its new guidance in early April 2020. By May, the county was requiring masks outdoors as well as indoors and enjoying 96-97% compliance. (It also imposed an array of other strict mitigation measures.) But it had case and mortality rates well above the national average throughout 2020 and 2021. LA currently ranks 3rd among California counties in its Covid death rate, faring worse than numerous counties with mask mandates of shorter duration and lower levels of compliance.

It’s possible to quibble with Miller’s methodology. As he readily acknowledges, we don’t know precisely how LA and other heavily masked areas would have fared against Covid in the absence of mandates, and the fact that mask mandates have tended to be deployed alongside social distancing and other NPIs muddies the water. Miller’s comparisons of LA’s outcomes with those of neighbouring Orange County, which has had no mask mandate since last year but has had lower case and death rates, seem rather conclusive, but a sceptic might note confounding variables such as demographic differences, population density, and so on.

Suggested reading

Why the experts are losing

By Malcolm Kyeyune

Such criticisms would be more damaging to Miller’s case if the officials and experts who have embraced mask mandates were more modest in their own assertions. Not only did they jettison years of prior research pointing to the minimal efficacy of masking, they also made empirically indefensible claims that masks alone could end the pandemic.

For instance, Miller compiles the astonishing claims made by Dr. Robert Redfield, who directed the CDC until just a year ago. In September 2020, Redfield described masks as “the most powerful public health tool we have”, claimed they offered “more Covid protection than vaccines would”, and stated: “If we [wore masks] for 6, 8, 10, 12 weeks, we’d bring this pandemic under control.” While they may not rule out some small effects of masking, Miller’s charts expose claims such as Redfield’s as delusional propaganda.

Likewise, those sceptical of what Miller’s comparative graphs can teach us should be even more suspicious of the shoddy studies published by the CDC over the past two years. For example, one study claimed to show a decline in Covid cases in certain Arizona counties as a result of mask mandates. But it failed to note that cases declined at a similar rate over the same period in Arizona counties without a mandate. In the months following the completion of the study, moreover, cases rose well above prior levels in counties with mandates still in place. Such studies produced a mirage of success by limiting the period of study, conflating correlation and causation, and avoiding comparisons with non-mandate counties.

The most disastrous failing of the experts has been their lack of curiosity about the actual results of the policies they have staked their reputations on. Mask mandates have been, in Miller’s phrase, “a population-wide experiment”, but few within the US scientific and medical establishment have seemed interested in parsing the resulting data, leaving that task to outsiders like Miller. Astonishingly, there have been just two randomised controlled trials on masking published since the pandemic began. One found no significant effect at all, while the other found a small effect of 11% for surgical masks and no significant effect for cloth ones. The first was largely ignored or dismissed, while the second was optimistically glossed as proving that masks work.

Moreover, even the most bullish case for the technical efficacy of at least some higher-quality masks does not constitute a case for mask mandates, a distinction that most commentary elides. The only way to measure the efficacy of mandates is to look at their actual track record. This is what Miller has done, and the result, he argues, is clear: “mask mandates have demonstrated very little impact, if any, on case curves throughout the United States and in many other international locations.”

Suggested reading

The Left-wing case against vaccine mandates

By Thomas Fazi and Toby Green

Miller is justifiably derisive about the experts who have oversold dubious policies at every turn, but the ironic implication of his book is that much of the expert guidance from prior to 2020 has been vindicated. Before Covid appeared, scientists and officials advised time and again that masks would be ineffective at containing a pandemic respiratory virus, and the evidence Miller has compiled suggests they were correct.

Two years into the pandemic, the experts are now the last to acknowledge the accuracy of their earlier predictions. This raises the question of why they changed course and sacrificed their own credibility in the process. Miller confines himself to the data, and if there’s a limitation to his book, it’s that he does not offer any compelling explanation of why the expert class threw itself a policy it once regarded as worse than useless.

It is not difficult to see why mask mandates proved irresistible to politicians. Masks are the perfect form of hygiene theatre, conveying an intuitive sense of safety regardless of demonstrable efficacy at scale. They also offload responsibility for controlling the pandemic to ordinary people. The overcrowding of ICUs can be blamed on the bad behavior of “anti-maskers”, rather than on the allocation of resources by governments and hospital CEOs. When cases and deaths spike, it is the fault of the citizenry, not the leadership.

The scientific and medical establishment’s uncritical support of masks and other dubious policies is just the latest manifestation of its lack of independence from political imperatives. After several years of finding themselves at the receiving end of rhetorical assaults from rising Right-wing populists, the experts seized on the pandemic as an opportunity to reassert their own status and authority — and that of the liberal-technocratic politicians with whom they are largely aligned.

In the long run, however, the temporary political advantages they gained from this will be outweighed by the discrediting effect of their embrace of censorship, propaganda, and rule by decree. Making us all put masks on was the expert class’s mask-off moment — and what we can see now isn’t pretty.

The country has halved road deaths in a decade — a bigger fall than anywhere else in the world. In Oslo they went even better. There were no pedestrian or cyclist fatalities on the streets of the Norwegian capital, a city with a population bigger than Glasgow. There was, unfortunately, one traffic death after a driver smashed his car into a fence in June. But this was still a big improvement on the previous year when five motorists died — let alone going back to 1975, when traffic there took 41 lives.

Compare this with Britain, which has the world’s fourth safest roads despite slipping from first place in the global league table at start of this century. Glasgow saw nine fatalities in the most recent year for which data is available. Leeds, significantly smaller than Oslo, had 26 road deaths, including 15 pedestrians and one cyclist. In our own capital, preliminary figures indicate 73 pedestrians and six cyclists were killed, with the youngest confirmed casualty just three years old.

The reason for Norway’s success story is simple: good government. Politicians decided it was possible to eliminate road deaths — and thus a moral imperative to do everything possible to achieve this bold aim to make the nation safer for its citizens. There was nothing dramatic in reaching their goal. There was just the hard slog of slow improvement through systemic changes with lower speed limits, fewer parking spaces, safer cars, improved roads, smarter technology, better-funded traffic police and more cycling infrastructure. One key aspect was intense focus on safe ‘heart zones’ around primary schools.

Suggested reading

Test article

By Graeme Archer

They did not even rely on their own model. Instead they took the ‘Vision Zero’ concept pioneered 23 years ago in neighbouring Sweden, which decided that safety should not be traded for mobility. A law enshrined the idea that “it can never be ethically acceptable that people are killed or seriously injured when moving within the road transport system.” This was an amazing concept when you consider how numbed the modern world became to the primacy of cars on roads and over-riding need for speed in everyday lives. “It’s all about humans taking back the streets from cars,” said Raymond Johansen, Oslo’s mayor. “It is a win-win situation for our safety, our health, our quality of life and the environment.”

Vision Zero is being copied in other countries, including several British cities. But there are wider lessons in this triumph beyond the reduction of road carnage. For it serves as a timely reminder that politicians can make a huge difference — in this case, saving scores of lives. This is important to remember at a time when there is deep scepticism over Westminster, especially in light of the Brexit debacle, and when profound dissatisfaction with politics corrodes so many Western democracies. Talking to voters amid elections and protests in several countries including Britain, I have heard time and again the sense of despair over political leadership with the belief that politicians are distant, out of touch and uncaring. Often, people are right to be frustrated by political failure, although wrong to believe demagogic populists offer solutions to complex societal problems.

Norway’s achievement reminds us that leaders can be bold. But it also shows that improvements to collective well-being are often the result of slow, incremental change rather than the sort of headline-grabbing initiatives beloved by politicians. Just look at education, where standards have slowly improved highlighted by England’s steady rise in the international Pisa ranking for reading over the past decade. One strand of the success comes down to Labour starting to push phonics from 1997 under Tony Blair, then the Tories strengthening the policies, tweaking the curriculum and introducing new tests after returning to government in 2010.

The Norwegian effort shows a need for clear goals that stretch across departments.

Suggested reading

Could the Cummings nerd army fix broken Britain?

By Tom Chivers

This is something understood by Dominic Cummings, the combative driving force behind Boris Johnson’s government, in his determination to shake up Whitehall. He may be driven by despair over bureaucratic inertia and struggle to hide contempt for many politicians, yet he seems fuelled by recognition that government can be a force for good. This is a welcome contrast to the intuitive scepticism over the state that scars many conservatives on both sides of the Atlantic. It underlines why it makes sense to replace cabinet committees with focused decision-making bodies, another attempt to shatter the silos that disfigure too much of the public sector.

There are plenty more examples of such governmental success. The Institute for Government highlighted Ireland’s launch in 1997 of a ten-year plan to half numbers of ‘consistently poor’ citizens. Politicians and officials agreed the definition, analysed the causes, set targets, co-ordinated clear policy across departments, built consensus and learned from flaws that emerged. They achieved their aim five years ahead of schedule despite seeing four rival parties involved in government over this period. Another approach was taken by Singapore to empty its sprawling slums within a decade by setting up a specialist body with statutory powers to deliver a clear goal. Closer to home, it took New Labour’s ‘Rough Sleepers Unit’ only two years to slash the number of people sleeping on the streets by two-thirds with relentless focus on the issue and cutting across departments.

In his biography Tony Blair talks about setting up his delivery unit to “focus like a laser on an issue”, driven by “first-class data” to ensure he could “performance-manage it to solution”, which might all sound familiar to Cummings. Yet the past three Tory teams have all told me they were determined not to be enslaved by demands of ‘the grid’, the schedule of  announcements and events used to control the daily agenda, so it will be interesting to see if this one ends up feeding the voracious media with pathetic ‘initiatives’ and stunts to divert attention. The omens are not good. And will this government also end up frustrated by slow pace of change while buffeted by disruptive events? As Blair also said, political capital tends to have been whittled away by the time a prime minister has learned how to grapple with government and pull the levers of power.

by the same author

Why are we locking up patients for profit?

By Ian Birrell

It will take boldness to resolve some of Britain’s most intractable issues such as poor productivity, regional imbalances and a corrosive social care crisis. These get endlessly discussed but never resolved. And here lies another crucial lesson from those Norwegian roads. Politicians can act bravely, must focus on evidence and should brush aside the bleatings of lobby groups protecting patches. One business body in Oslo, after all, resisted the reforms designed to save lives with warnings that politicians were turning their city into a ghost town.

But to embed change, it is worth reaching across political divides to find consensus. Eight years ago I visited Texas to write about pioneering prison reform. Jerry Madden, a veteran Republican appointed to oversee criminal justice, told me he was asked not to build any more prisons as costs were rising in a state that locked up more people than any other place in the world. Yet he faced a conundrum: they expected 17,700 more inmates over the next six years. So he reached out to his Democrat rival. The two men accepted their differences on sentencing and drug issues, but agreed they both wanted to reduce the prison population. The result was innovative reform that invested in rehabilitation rather than incarceration, sparking a movement that has swept the nation, changed the debate and reduced prison populations.

Just imagine if such an approach was applied to social care in this country, a crisis that causes such pain to thousands of families while snarling up the health service. This catastrophe has many causes, from big cuts to local authority budgets through to staffing shortages intensified by Brexit. Johnson clearly does not have a plan, despite his claim when entering Downing Street, so why not do as he promised by seeking urgent cross-party consensus to deliver a service that is humane, effective and free at the point of use? Not with another independent inquiry, but with clear determination to sort out the mess.

After all, if Norway can stop children dying on its streets, surely Britain can save elderly and disabled citizens from living hell?

Small-c conservative working-class voters have migrated to the Conservative Party because of immigration and Brexit. On the other side, successful educated cosmopolitans opt for Labour or the Lib Dems.

As a number of academics, including me, have pointed out, apart from attitudes to the EU, nothing predicts whether someone voted Leave better than their views on immigration. This shows up clearly in UnHerd’s data.

For instance, as figure 1 reveals, 55% of the variation in immigration attitudes across constituencies can be explained by Brexit vote share, a powerful association. The relationship is especially strong toward the top left, with the strongest Remain seats far more pro-immigration than elsewhere.

Figure 1

Sources: UnHerd Britain / Brexit data from: “Areal interpolation and the UK’s referendum on EU membership”, Chris Hanretty, Journal Of Elections, Public Opinion And Parties, Online Early Access, https://dx.doi.org/10.1080/17457289.2017.1287081

This rising cultural cleavage made a powerful impact in the 2017 election despite the fact Corbyn’s far-Left image gave prosperous voters a strong incentive to vote against him. But what makes Hackney different from Boston on immigration attitudes? At the constituency level, this is about demographics, of which the key ones are, in order of importance: the proportion with degrees, the level of ethnic diversity and average age. The share with a degree varies from under 5% in Clacton to over 50% in Battersea. Immigration attitudes largely track this measure, as figure 1 shows. The reason higher education is so important is not because it boosts earnings, but because it is a measure of psychological openness and a liberal worldview, both of which incline curious minds toward the pursuit of higher education.

Figure 2

Sources: UnHerd Britain / Norris, ‘British General Election Constituency Results’

When it comes to explaining Brexit, Trump or any other western national populist vote, it’s breathtaking how resilient the narrative of the ‘left behind’ is. This suggests that those who feel they have lost out during the economic modernisation of the British economy are striking back at the elites. Yet, as I mentioned after the Brexit vote, economic factors do not tell us much about who voted Leave or Remain. As figure 3 clearly shows, unemployment level — as with other purely material factors — can’t explain immigration attitudes. The flat regression line and wide spread of constituencies around it tells us there just isn’t a story here.

Figure 3

Sources: UnHerd Britain / Norris, ‘British General Election Constituency Results’

Beyond education, what social indicators underpin immigration attitudes? An important — but not dominant — factor is ethnicity. Many are fond of pointing out that places with few immigrants seem to be the ones most concerned about immigration. Yet this statement completely sidesteps the fact that a high-immigrant area will have lots of immigrants in it! Since immigrants and those who are from minority communities tend, all else being equal, to be more pro-immigration than White Britons, highly diverse constituencies are also more open to immigration, something evident in figure 4.

Even so, there is considerable variation around the trend line. Constituencies with a more educated population tend to lie above the line, and those with fewer degrees below it. In addition, there is an important current of immigration restrictionism among ethnic minorities, especially those who are UK-born. This is more pronounced in constituencies with a higher Sikh and Hindu component to their minority population since these groups are more restrictionist than other minorities, and less so in places with a higher Afro-Caribbean share, as they are relatively liberal on immigration compared to other non-White British groups.

Figure 4

Sources: UnHerd Britain / Norris, ‘British General Election Constituency Results’. NB: excludes constituencies with less than 20% non-White British share (a majority of seats)

The share of the population aged 20-24 is also an important correlate of immigration attitudes, with younger constituencies more ‘open’ and older ones more ‘closed’ in immigration outlook. Yet this matters considerably less than education or ethnicity. The share of passport holders, however, is an indicator, even when we account for its relationship with education, age and ethnicity. Figure 5 shows the importance of this association.

Figure 5

Sources: UnHerd Britain / Norris, ‘British General Election Constituency Results’

Education, ethnicity, share without a passport and age account for nearly 80% of the variation in immigration opinion across constituencies. This means almost all of the differences in immigration attitudes over place have to do with these demographics. But is there any role here for the properties of place — whether local tradition, regional economy or cultural atmosphere?

The answer is a qualified yes. Even with controls for education, ethnicity, age and share without a passport, London emerges as somewhat more open, and the East of England and South East somewhat less so. We should bear in mind, however, that regional differences are small. Apart from London (at 46% agreement), the share agreeing with open immigration fluctuates in a tight band from 32% in Wales to 36% in Scotland. When you account for education, ethnicity and age composition, London remains only 3 points more pro-immigration than the supposedly restrictionist East of England. Moreover, some of this may be due to social indicators we haven’t included in the model, so the geographic effect may be weaker still.

Geography is important, but it’s also vital to appreciate its limits, which too many pundits fail to do after perusing an election map. A map of SNP vote by British region would suggest that living in Scotland explains 100% of the variation in SNP voting among people. It doesn’t. Mapping state income and the Democratic vote in America would show that richer states back the Democrats. Yet richer individuals aren’t more likely to vote Democratic than poorer people. In other words, you can’t simply read individual-level relationships off aggregate correlations. More than that, most of the variation (90% or more) in attitudes lies within constituencies, not between them.

Suggested reading

How bigoted is Brexit?

By tom shivers

Consider university education, the most important cross-constituency predictor of immigration attitudes in Britain. When we drill into the survey data on individuals (as distinct from constituencies) — using the British Election Study — we find that education is much less important. That is, once we take psychological factors such as the death penalty into account, having a degree makes only a modest difference to immigration attitudes.

Figure 6, based on a sample of around 20,000 individuals, shows that a person’s views on the death penalty matters vastly more than whether they have a degree in predicting whether they want less immigration. Almost 9 in 10 strong death penalty supporters back restriction against barely 3 in 10 strong opponents of capital punishment.

Next to this 60-point chasm, the roughly 5-point gap between those with and without degrees pales into insignificance. While graduates are less likely (41%) than non-grads (62%) to back the death penalty, this is beside the point. The massive difference between pro- and anti-death penalty supporters holds as strongly within the population of university graduates as it does among non-graduates.

Figure 6

Source: British Election Study Internet Panel Study wave 9, 2016

As I argue in my book Whiteshift, immigration attitudes are the fulcrum around which the politics of western societies are realigning. This is because those whose psychological make-up inclines them to see difference as disorder and change as loss are voting for parties that promise to slow immigration. This isn’t generally about competition for jobs or services, which is a safer narrative to voice, but instead mainly about majority-ethnic and what I term ‘ethno-traditional’ national identities. Conservative voters feel that these are being unsettled by the rapid ethnic shifts sweeping across western countries.

The Brexit vote was primarily driven by the immigration question, though the focus on the economic risks of Brexit is presently overshadowing the subterranean force of this issue. Should Johnson ink a deal while failing to reduce migration levels, we should expect this debate to return — intruding as sharply into British politics as it did in the runup to 23 June, 2016.

Whether people agree or disagree with that statement opens a window onto one of the issues that politics doesn’t talk about enough, yet which matters a great deal.

The politics of transgenderism are under-explored for several reasons. Politicians’ timidity in the face of activists’ ferocity is one big factor. But there’s another, revealed in today’s UnHerd Britain data — that it’s a fiendishly complex issue that doesn’t fit easily into the narratives and categories that structure so much of British political debate. Or at least, political debate at Westminster.

Even a quick glance at the results at the UnHerd Britain results shows how enthusiasm for, and scepticism about, teenagers being able to change their gender at will are not distributed along conventional party lines.

Yes, the places where there is greatest support for that opening statement lean to Labour: Bethnal Green and Bow, Islington South and Finsbury, Jeremy Corbyn’s Islington North. Here, well over half the respondents are happy for children to “self-identify” as male, female or something else. Fewer than one in five people express any doubts.

But look at the other end of the spectrum and you’ll find that some of the places where people are most sceptical are also solid Labour. In Doncaster North and Wansbeck in Northumberland, well over a third of people are not persuaded that children should choose their gender.

Tory seats split in a similar way. Chelsea and Fulham and the Cities of London and Westminster are near the top of the “agree” list, yet true blue places such as Buckingham and South West Surrey are among the most sceptical.

This probably helps to explain why neither party has found the trans issue a comfortable one to accommodate. Even though both Labour and Tory leaderships (at least, Theresa May’s) have in recent years enthusiastically embraced calls for self-ID, both parties have also faced significant internal resistance.

Among the most organised sceptics of trans orthodoxy have been women from the trade union movement, traditional left-wingers who set up the Woman’s Place UK campaign. They have been surprised to find themselves fighting on the same side as social conservatives; Tory David TC Davies has become an unlikely hero to some lifelong left-wing feminists because of his willingness to air their concerns about this issue in Parliament.

What, if anything, unites the places that share similar perspectives of transgender issues?

A superficial analysis would conclude that this is just part of the “anywheres vs somewheres” debate that is en vogue in parts of Westminster. That’s the same story that dominates a lot of Brexit analysis, where Remainers are all urban sophisticates and Leavers all nativists in bleak post-industrial backwaters. Yet just as Phillip Cowley and Rob Ford brilliantly showed on these pages recently, a closer look at the transgender polling shows this issue can’t be neatly slotted into the “rich liberal cities — poor conservative towns” template either.

Yes, scepticism about that gender question is strong in some places that really have been “left behind” — Clacton in Essex and Boston and Skegness in Lincolnshire are, among other things, physically isolated, as much by dismal transport connections as by geography.

But what about Wolverhampton South West, barely 20 miles from Birmingham, and the Selly Oak seat where support for gender change is substantially higher? Wansbeck can not easily be written off as a north-eastern throwback. The seat is centred on the relatively affluent market town of Morpeth, home to better-off Newcastle commuters. (I went to the town’s comprehensive school; it was hardly Brideshead Revisited, but nor was it a Geordie version of Deliverance.)

Meanwhile, South West Surrey and Bracknell are safely inside the (large, affluent) London commuter belt. Bracknell is barely 40 miles from central London; the train takes little more than an hour. These are not the “left behind” towns of metropolitan media imagination, but they have significant doubts about a change in social mores that some people living just a few miles ago might consider unquestionable.

For the most enthusiastic supporters of transgenderism, there can be “no debate” — accepting the fluidity of self-defined gender is a given, something that posterity will judge was both inevitable and enlightened.

Yet it’s striking that in only in seven of all the constituencies did a majority (more than 50%) agree with that opening statement; everywhere else, the majority either questioned a staple of trans orthodoxy, or expressed no view. Some in the “Neither” column will be wary of expressing a view, but most will not yet have engaged with the issue in depth.

“No debate”? This poll shows that the debate about gender in 21st Century Britain is not just unresolved. It has yet to even begin in earnest.

Back then, it was still relatively straightforward: survey a group of people, usually between 1,000 and 2,000, ‘weight’ the answers to make them more representative of the whole UK population, and interpret the results. The methodological arguments that raged back then among the pointy-heads —  telephone polls vs online polls; how to deal with past vote recall and turnout — have either now been settled or are at best secondary issues.

Polling companies were still, only half a decade ago, headed by larger-than-life public figures, more likely to have had a journalistic background than a statistical one, who regularly went on TV to “read the runes” for a grateful nation. For them, polling, like politics, was an art not a science.

Suggested reading

The royalty loyalty of the working classes

By Paul Embery

Today, the question everyone asks during an election is still the same one — who is going to win? — but the world of opinion polling looks very different. We have, of course, all been burned by a succession of difficult elections with unexpected results that various polling companies failed to predict and as a result the media and public are rightly much more circumspect. We want to know more, before believing anyone’s predictions.

‘Data scientists’ are gradually replacing those grand ‘pollsters’ who used to offer confident insights with a neat turn of phrase. These new number nerds are likely to be under 30 and might well know nothing about politics. Instead of simple surveys and uniform national swings, they are using complex statistical models. For people whose bread and butter relies on being seen as a savvy political ‘expert’, all this amounts to an existential threat.

The most powerful data modelling technique in politics at the moment, is something called MRP. It stands for “multilevel regression with post-stratification” —  not exactly catchy — and the number of people who fully understand it in the UK can be counted on two hands.

But, roughly speaking, you conduct a huge survey, normally 10,000 or more so that you have sufficient numbers to look at small subsets of people, and you then analyse what the most predictive characteristics are behind the question you’re interested in, for example whether race, education or income is more predictive of how people are going to vote (that’s the ‘multilevel regression’ part).

Then, once you have identified the most predictive characteristics, you use what you know about the people in each constituency, and in combination with any local effects you observe in your sample, you can estimate the outcome in every constituency in the land (that’s the ‘post stratification’).

Suggested reading

Call yourself a Monarchist?

By James Kanagasooriam

With this method, instead of crude national swings, you create something close to a simulated version of the whole electorate, in miniature, in all its complex glory.

The defining moment for MRP arrived during the general election of 2017. The conventional opinion polls had showed a dramatic narrowing, with most still showing Theresa May’s Conservatives significantly ahead on election day. The Westminster bubble was expecting her to build on, not lose, David Cameron’s majority.

But as well as their conventional poll, YouGov had also run an MRP model, and, instead of a majority this one pointed to something very different: a hung parliament, with the Tories losing ‘safe’ seats such as Kensington, Canterbury and Ipswich. Two weeks before polling day they gave it to the Times, who put it on the front page with the headline: Shock Poll Predicts Tory Losses.

Almost nobody believed it, but it turned out to be astonishingly accurate.

In the polling world, the pattern of the turbulent past few elections has been one of repeatedly trying to correct for the mistakes of the previous election  (and sometimes overcorrecting in the process). Because YouGov’s model turned out to be so accurate, right down to Canterbury and Kensington, this time everybody is waiting for the YouGov MRP to drop. It is expected next week, and if it shows a large Tory majority, I would expect the pound to surge against the dollar within seconds.

Suggested reading

Does Scotland resent the Royals?

By Euan McColm

At UnHerd, though, we’re interested in the things everyone else is ignoring, and so we’ll leave to YouGov the unenviable task of forecasting the main result.

Instead, we’re doing something slightly different. We’ve teamed up with an impressive startup called Focaldata that specialises in MRP — and we’re doing what we believe is a first. Using the same technique that predicted the result last time, we’re mapping how the 632 constituencies of Britain differ on six big issues of the moment that aren’t being much talked about in the election campaign: attitudes to the Royal Family, gender, immigration, religion, free speech and tax. We’ve called it UnHerd Britain.

The first set of results, out today, reveal how different parts of the country feel about the Royal Family. The results are striking. Prince Andrew may be all over the news, and the new season of The Crown may be keeping people busy across the country, but nobody gets to vote on the Royal Family.

For the first time, today, we can see what would happen if they did — down to individual constituencies. For Buckingham Palace, it’s a mixed bag of news. Overall, their position is still secure: only 2 of 632 constituencies have more people disagreeing than agreeing with the phrase, “I am a strong supporter of the continued reign of the Royal Family.”

But all is not rosy. Scotland, while still tilting towards support of the royals in all 59 seats, is dramatically less in favour than the rest of the UK; the large cities, including London, are similarly much less enthusiastic. The groups that are growing fastest in our society — university graduates and ethnic minorities — are the groups least excited about the Royals.

It may sometimes feel like Brexit is the cause of all our divisions, but as UnHerd Britain will reveal in full Technicolor over the next few weeks, the Royal Family reign over a deeply divided kingdom across many more issues than that.

That’s because any statistic that is interesting enough to have become a news story, he claims, has probably passed through “so many filters that encourage distortion and selection”, that “the very fact that I am hearing a claim based on statistics is reason to disbelieve it”.

I have a similar rule of thumb, which I’ve said here before: if a statistic is interesting, it’s probably not true.

The trouble is that we can’t do without them. There are simply too many people and too many things in the world to understand what’s going on without simplifying them and aggregating them into statistics; we can’t think about every individual NHS patient, we need to think in terms of average survival rates or waiting times. We have to treat people like numbers if we want to do the best for those people, so we need to learn to be better at interpreting those numbers.

Suggested reading

Why flawed science should concern us all

By Tom Chivers

But statistics can be misleading — in some fascinating ways. In his book, Spiegelhalter, a Cambridge statistician and former president of the Royal Statistical Society, points some out; Simpson’s paradox, for instance. In 1996, the admissions data for five STEM subjects at Cambridge showed that men were more likely to be admitted: 24% of men’s applications were accepted, compared to 23% of women’s.

Is this evidence of bias? Well: possibly. But not, interestingly, of discrimination against women. In four of the five subjects – economics, engineering, computer science, medicine, and veterinary medicine – women were actually more likely to be accepted (in one, medicine, they were equal). Sometimes, that difference was quite stark – 16% of women compared to 12% of men were accepted onto the veterinary course; 32% of women against 26% of men into engineering.

How is this possible? Because medicine and veterinary medicine were much more popular, so your chances of being accepted were much lower; and because women were much more likely to apply for medical or veterinary degrees. By choosing the harder-to-access courses, women were less likely to be accepted overall, even though, individually, they were more likely to be accepted than a man applying for the same course. Similar findings have been seen elsewhere.

Another example is the “prosecutor’s fallacy”. Imagine a suspect is in court. If they’re innocent, there is only a one in a billion chance that their DNA would match the DNA found at the crime scene, and yet it does. Does this mean that there is only a one in a billion chance that they’re innocent? No; far from it. Nonetheless this has led to real women being wrongly convicted of child murder, after several of their children died of sudden infant death syndrome.

Suggested reading

Don't fight dirty with statistics

By Tom Chivers

An analogous problem is cancer testing. Say your new cancer test is 95% accurate; it will answer the question “do you have cancer” correctly 95% of the time. If you test positive, do you have a 95% chance of having cancer?

Not at all! Imagine that 1% of people have cancer. You run your test on 10,000 people. Of the 100 people who do have cancer, it rightly tells 95 of them that they do have cancer. Of the 9,900 people who don’t have cancer, it rightly tells 9,405 that they don’t have cancer. But this means that it has wrongly told 495 people that they do have cancer. If you test positive for cancer, your chances of having cancer are not 95%, but 95 in 590, or about 16%. The DNA-test situation is exactly the same: your test is “accurate” 999,999,999 times out of a billion, but will still regularly be wrong, because the thing it is looking for (the specific guilty person) is very rare.

These are just a couple of the reasons why even straightforward-sounding statistics can be misleading. They’re not even the big reasons – they’re just a couple that I thought you might not have heard of.

Things like statistical bias, small sample sizes, or multiple analyses of the same data will have more of an impact, as well as the Groucho Principle that the most dramatic findings are both more likely to be well-publicised and more likely to be false. But they’re illustrative of the fact that you need to be very careful of almost any statistical story you read.

Suggested reading

If you don't 'get' statistics it's just as bad as not 'getting' Shakespeare

By Graeme Archer

At this point, I should admit that I have a dirty little secret, which is that I am not very good at maths. I mean, I’m fine. I’m OK. I got an A at GCSE, in 1997. But I’m not good. This is a problem, because I write an awful lot about statistics. Spiegelhalter tries to make the case that you don’t need to be very good at maths to be good at statistics, and he sort of succeeds – quite a lot of it you can grasp via the concepts, or through visualisations. But still, as an indifferent mathematician, I often found it hard to follow him.

As a non-mathematician, I have a few shortcuts for working out whether a statistic is worth believing, which seem to have done all right for me so far. One, which Spiegelhalter stresses, is that often the best statistical analysis you can do is simply visualising the data. There was a bit of a recent kerfuffle about suicides among girls and young women going up 83% since 2012; but simply looking at the ONS chart showed that the numbers were small, the data was noisy, and the only way you got the 83% figure was by choosing the lowest year on record. (It’s an old trick.)

Another useful hack is to ask “is this a big number?” Lots of statistics are presented as bald numbers: “X people have died”. But out of how many? If 25 people have died of cancer out of a population of 50, that’s awful; if 25 people have died of cancer out of a population of 50 million, that’s a statistical miracle. That’s one reason why I’m always wary of stories that say, for instance, “10 or more black transgender women have been killed this year” without telling me how many black transgender women there actually are, and whether that is more or less than we should expect.

Similarly, if a story tells you that something has “gone up by 25%”, or that something “increases your risk by 90%”, without telling you what it’s increased it from and to – if it gives you the “relative risk” but not the “absolute risk”, in technical terms – that’s an indicator that it may be worth ignoring. If something doubles your risk from one in 100 million to one in 50 million, you probably don’t care.

A third is asking “can I be sure what’s causing this?” Spiegelhalter dedicates a chapter to how you establish causality. It is amazingly hard to do this in the absence of large-scale randomised controlled trials. Austin Bradford Hill, one of the scientists – with Richard Doll – who first established that smoking causes lung cancer, suggested ways of doing so; looking for plausible mechanisms, making sure that the supposed effect follows the supposed cause, seeing if there’s a “dose response” – that is, the more of the cause there is, the more of the effect you see – and, importantly, making sure that the effect is sufficiently large that it couldn’t be a coincidence or some statistical confounding.

If you’re looking to see if something (say, a rise in teen suicides) is caused by something else (say, social media) then it’s worth looking to see if any rise in teen suicides follows the growth in social media, and whether each doubling in social media use leads to a consistent increase in suicide rate.

Suggested reading

Do we really have a 'suicidal generation'?

By Tom Chivers

But also, it’s important to see whether the suicide rate tends to jump around a lot anyway. If it does, then perhaps there’s nothing to explain; it just happened to be low one year and high another. (Spiegelhalter has a useful trick for seeing whether you’d expect to see the variation you see, such as using the mean averages to perform a “Poisson distribution”, but just looking at a graph and seeing whether it bounces around a lot is a good start.)

The statistician George Box said: “All models are wrong, but some are useful.” When you’re reducing millions of people and events to a few numbers, you’re building a model of the world, and that model must be simpler than the world, so it will be wrong in some way. You’re always making assumptions and drawing arbitrary lines.

For instance, we talk about the “murder rate” in Britain; statisticians treat it like an arbitrary event that happens at random to people, like decaying subatomic particles, with a 0.0008% or so chance of it happening to any given person per year. But that’s not how it works. Each murder is the result of real people making real decisions, not some lightning bolt out of a clear blue sky; it just happens that in a large enough population, non-random events can be treated as though they were random.

We can’t do it any other way. We can’t look at every person, and you can’t possibly predict every individual event, whether it’s the outcomes of heart operations or teen suicides or murders. You have to aggregate them, round off the corners, make simplifying assumptions; we have to treat people like numbers. You’ll get it wrong a lot of the time, and you have to be immensely cautious about any conclusions you draw.

There is one simple step, though, which will help a lot of the time. It’s a bit nihilistic, perhaps, but it’s powerful. And that is the Groucho principle that we discussed at the beginning: if the statistic makes a headline, it’s probably wrong.

I see why victims of sexual assault are concerned. I sympathise with fears that their personal lives will be raked over by defence lawyers seeking to exonerate the accused. Our justice system allows character slurs to be aired in court to the extent that sometimes the victim feels as though they were the one on trial. Signing up to what has been described as a “digital strip search” makes victims feel like they are the criminals.

The police, for their part, say they are concerned about failed convictions, following a spate of collapsed trials when evidence uncovered from phones has not been properly shared with the defence. This move, presumably, is to emphasise the importance of early disclosure of evidence.

Suggested reading

Why are we playing games with our privacy?

By Peter Franklin

Now the police are responding to protests from victim support organisations by offering to meet stakeholders to discuss the new consent forms. In a letter designed to reassure them, the National Police Chiefs Council (NPCC) explains:

“When a crime is investigated, police will regularly seize devices of those accused. This is done using police powers and suspects are not required to consent. Coercive police powers are clearly not appropriate for use against complainants and access to their devices should be on the basis of specific, free and informed consent.” 

This does not reassure me, or quiet my alarm bells.

It only serves to confirm my fears that it’s not only victims who should be concerned about police taking powers to download information from our mobile phones in bulk. The truth is that any of us could find ourselves in a situation where the intimate details of our life are uploaded to a police computer. Any of us could be accused of a crime, or simply witness one. All of us should be asking – what are my rights in such a situation?

Just over a year ago, Privacy International – a charity that advocates for strong privacy protections in law and technology – released a report called “Digital Stop and Search; how the police can secretly download everything from your mobile phone”. After months of dogged investigations using the Freedom of Information Act they concluded that “in the UK, police are using highly intrusive technology to extract and store data from individual’s phones, on a questionable legal basis”.

It transpires that the police have been acquiring highly personal information this way for the past ten years. Some forces have downloaded data from tens of thousands of mobile phones – and with no real legal basis for doing so.

Suggested reading

Don't be afraid of the digital dustbin

By Sam Leith

The systems used – often “highly intrusive technology”– varies between forces, and not all have used it. But those that did were able to download far more information than even the phone’s owner could access. Deleted and hidden data – location, health, passwords – could be extracted. The contents of WhatsApp messages, which you might assume are private because they are encrypted, have been extracted and used in court. Then there’s all the easily accessible stuff – calendar, contacts, text messages, photographs – right there in the palm of the police officer’s hand. They could find out far more about you from the phone than they ever could from searching your house.

Unlike a physical search of your house, though, these data extraction procedures were carried out without a warrant and without a clearly defined legal framework. Any legal justifications that were offered by those forces that responded to Privacy International’s enquiries were often vague. Their defence relied heavily on PACE, the Police and Criminal Evidence Act of 1984, which Privacy International (PI) describes as “a piece of legislation written long before a phone became a device that could be used as a pocket surveillance tool”.

Following the publication of their report, Privacy International made a number of recommendations: guidance should be issued to the public about their rights; a warrant should be obtained before forensically examining anybody’s smartphone; and there should be independent oversight. They also invoked the principles of necessity and proportionality – and called on the Home Office to launch an independent review.

Suggested reading

The truth about Gen Z

By Eliza Filby

In Scotland, the report – combined with further investigations by The Sunday Herald – prompted parliamentary scrutiny. It’s the first time parliamentarians anywhere in the UK have called police to account in relation to key issues around data extraction technology. And as a result, Police Scotland, had to put their “cyber kiosk” roll out on hold.

The technology, which allows officers to circumvent passwords or other security measures to access data stored on mobile devices, was being used, according to the sub-committee on policing, by frontline officers “without any human rights, equality or community impact assessments, data protection or security assessments, and in the absence of any public information campaign”. No oversight. No recourse. No boundaries.

Following the publication of the PI report, the Information Commissioner’s Office is now conducting an investigation into “use of data extraction technology on the mobile phones of suspects, victims and witnesses”.

But rather than wait for the results of this enquiry, police went ahead with their consent form, asking victims to consent to a violation of their privacy and human rights or have their case dropped.

Perhaps they are hoping to address both the failures of the disclosure system and the absence of a formal basis for extracting mobile phone data with one form. If so, they’re misguided on both counts.

Suggested reading

How technology sidelines the citizen

By Carl Miller

Lack of data has not the been the problem behind those recent collapsed cases when the disclosure system failed.

One such high-profile rape case was dismissed, for example, after messages between the alleged victim and her friends were found to include positive references to the accused, Liam Allan, and to their ongoing relationship. Mr Allan had been under investigation for nearly two years and the police had gone through the accuser’s mobile phone data – including 57,000 text messages.

The data was in police files, but the human being responsible for going through it failed to find the key material. Since nothing of note was found, they refused requests to hand over the material to the defence, deeming it “not relevant”. The court case collapsed after three days after the data was finally handed over to the defence lawyers and messages from the victim asking for “casual sex” were found.

There’s no use having the information if you don’t know what you’re looking for – the data might as well be a ton of cuneiform tablets. The police don’t necessarily have the capacity or wherewithal to deal with it, and that’s why the disclosure system is flawed – not because they didn’t get their hands on it earlier.

Suggested reading

Prepare to be microchipped

By Peter Franklin

Furthermore, a consent form, signed under pressure, gives us no protection fromunre unregulated data extraction. And this applies to anyone – not only victims. We should all be be insisting on robust regulation and oversight of the “digital strip search”, because this potentially affects anybody who owns a mobile phone.

Naturally, we’re concerned about those companies who want to profile and target us, in order to sell us stuff. But news that being accused of a crime, witnessing a crime, or even being the victim of a crime – or even having been in touch with any of those people – could remove our fundamental right to privacy in the eyes of the law should worry us much more.

Police forces are currently under no obligation to tell people their rights, to inform them of what kind of information will be obtained, or to give them an inventory of what has been downloaded. There are also no limits on how long they can retain the data.

Suggested reading

If you want to get ahead, take a risk

By Timandra Harkness

Data from your mobile telephone has power. It could convict a criminal or exonerate an innocent person, whether you are the accuser, the accused, or a witness. Just as the home of a suspect can yield useful evidence, so too can an electronic device.

So while it may be proportionate and necessary to search such devices to solve or prevent crime, before investigators are allowed to enter our digital spaces, we should expect the same protections we are afforded when they search our homes.

Those legal protections not only guard us against incidental misuse of our data, such as security breaches or miscarriages of justice. They also draw a line around our basic right to privacy from the state.

Only a totalitarian regime should require every citizen to be transparent in the eyes of the law, simply to ease police investigations. The presumption of privacy should be the one to which police have to seek exceptions. Like the presumption of innocence in a defendant, it is a universal right because none of us knows when we might be the one to need it.

To re-read Neuromancer is to experience a series of shocks. The first shock is about the passage of time. William Gibson’s novel was first published 35 years ago this July; I read it a few years later. Since then, I’ve aged but the book hasn’t; neither story nor setting have dated at all – a testament to Gibson’s almost eerie prescience.

If you’ve heard of Neuromancer, it’s probably as the book that imagined – and named – “cyberspace”; the book that described the connected world a generation before we moved our culture, commerce and conversation online. Gibson described, “A consensual hallucination experienced daily by billions of legitimate operators, in every nation, by children being taught mathematical concepts” – at a time when few people had even a rudimentary computer in their home.

Case, his protagonist, is a hacker employed to steal secrets from corporations and governments, navigating a world made up of limitless data:

“Program a map to display frequency of data exchange, every thousand megabytes a single pixel on a very large screen. Manhattan and Atlanta burn solid white. Then they start to pulse, the rate of traffic threatening to overload your simulation. Your map is about to go nova. Cool it down. Up your scale. Each pixel a million megabytes. At a hundred million megabytes per second, you begin to make out certain blocks in midtown Manhattan…”

Again, remember that Neuromancer was published in 1984. While the projected numbers are out by several orders of magnitude, it is still a striking description of the big data world that would not come to pass for another three decades.

Yet re-reading the novel, I found that predicting the internet age and the dawn of big data was one of the less interesting feats Gibson pulls off. Neuromancer – and its sequels, Mona Lisa Overdrive and Burning Chrome – do so much more than that. It’s not really about technology, it’s about the way technology changes the way we live, the way we think and the way we feel when we inhabit two worlds – the real and the virtual. The real shock is how horribly familiar a world imagined in the early 1980s – by an author listening to Joy Division on his Sony Walkman – feels today.

Suggested Reading

Robots will kill us with kindness

By Peter Franklin

Sleepless, restless, rootless: Gibson’s world is all too recognisable to anyone who’s ever spent too much time online, ever stared too long at a glowing screen instead of doing something that actually matters. For some, the means to connect to that other world becomes an integral part of them: Neuromancer is full of people who plug enhancements and memories straight into their brain via sockets installed on their skulls. Gibson called them “microsofts” (the company of that name was nowhere near famous when he was writing).

Even though Gibson completely missed the potential for mobile telephones (landlines survive in Neuromancer, along with paper-based money: no-one gets everything right), his conception of people outsourcing their memory and perceptions to a small piece of tech – kept close to the body at all times – feels as familiar as the phone that’s always in my pocket.

He also captures the spirit of a world where technology, travel and global corporate branding have homogenised culture and experience. Case travels from Japan to the US to Turkey and onwards, but everywhere feels the same. His trip through Istanbul – in a driverless car that narrates the journey for passengers, of course – could be any big city.

Suggested Reading

Has a Chinese chancer won the gene-editing race?

By Tom Chivers

Celebrity too has gone global in the world of Neuromancer – in a way that feels strikingly contemporary. Tally Isham is a “simstim” celebrity – an icon whose fans experience her (subtly edited) life minute-by-minute in vivid technicolour. Long before the Kardashians were born or Snapchat was thought of, Gibson understood the limitless human hunger to see and feel the world through the lens of another, better, richer, prettier person. Just like today’s influencers, Isham doesn’t do anything meaningful – she just is.

To me, the most striking – and possibly unnerving – element of Neuromancer concerns the human body. A novel famous for the notion that human experience can be disconnected from flesh actually has a great deal to say – and predict – about human biology and how it can be changed.

In Case’s world, basic augmentation to change someone’s looks is routine: faces can be chosen like clothes off a rack, if you have the money. Muscles too: “joeboys” with enhanced physiques are the soldiers of crimelords, while omnipresent corporations – and wealthy individuals – mould and retain latter-day samurai, whose reflexes and skills are enhanced, and whose bodies sometimes have weapons installed. Meanwhile, Case’s sometime lover Molly has sunglass lenses grafted over her sockets, and retractable scalpel blades under her fingernails.

We’re not there yet, but the most advanced military forces of our day are paying growing attention to how the human body can be altered to make it more dangerous and more durable.

More interesting still is how, for those with the means, technology offers bodies that will last much, much longer than those of ordinary mortals. Neuromancer introduces Julius Dean, an importer and trafficker who is 135 years old, “his metabolism assiduously warped by a weekly fortune in serums and hormone…genetic surgeons reset the code of his DNA”. Another, even richer character is said to be 200 years old. In a world controlled by vastly wealthy corporations and their vastly wealthy owners, the 1% have realised that the ultimate treasure to amass is life itself.

If you’ve watched the Netflix series Altered Carbon, you’ve seen this idea taken to the next level, with the rich effectively achieving immortality. That show is based on a novel published in 2002, more than 20 years after Gibson began writing Neuromancer.

More from this author

Too posh to fail

By James Kirkup

William Gibson has said that Neuromancer is set in about 2035. What could happen in the next 16 years to take us closer to his cyberpunk future?

Today, if you talk to scientists, doctors and even insurers, you can glimpse a near future where the extension of current trends in public health, wealth and medical technology combine to widen existing gaps in life expectancy, to the point where the rich live significantly longer, healthier lives than everyone else. Last month, official statistics showed that the rich are living longer, while life expectancy for the poorest is falling.

Wealth inequality may well be the defining issue of today’s politics and economics. Health inequality may be tomorrow’s challenge. If my prediction is right, remember that William Gibson’s Neuromancer gave us an early warning, all those years ago in 1984.

But they were cheating. They picked 1998 deliberately since it was an outlier – an El Niño year much hotter than the years around it. If you were, on the other hand, to measure from 1997 or 1999, then there were lots of much hotter years on record; and the clear trend was that later years, on average, were hotter than earlier ones. It was a wobbly, noisy line, with some outliers, but the average temperature really was going up, and the only way you could hide that trend was by cherry-picking statistics.

I was thinking about this as I read the Sunday Times splash this week, which (using as-yet unavailable data from the Office for National Statistics) claimed that the “suicide rate among teenagers has nearly doubled in eight years”. It expressed concerns that we are raising “a suicidal generation”.

They said that the suicide rate among 15- to 19-year-olds in 2010 was just over three per 100,000. The ONS figures, due out in September, will (apparently) show that it is now over five per 100,000. Inevitably enough, the piece links the purported rise to the growth of social media since 2010.

Suggested reading

Why you can't rely on the news media to understand... science

By Tom Chivers

But this is – and I don’t want to get too technical here, but bear with me – absolute bollocks from top to bottom. It’s a masterclass in what scientists call “hypothesising after results are known”, or HARKing. If you have the data in front of you, then you can make it say almost anything you like.

First, it’s worth noting that very few teenagers kill themselves. The total number of suicide deaths among 15- to 19-year-olds in 2017 in England and Wales was 177, out of about 3.25 million. That means that small changes can look like big percentage swings. More important, though, the Sunday Times story did exactly what the climate deniers did. The year 2010 had the lowest rate of teen suicides of any year since at least 1981, when the ONS records begin. You could compare it with literally any other year and you’d see a rise.

Added to which, picking social media as your reason for is completely arbitrary. Social media did not start in 2010. The BBC TV series Sherlock, starring Benedict Cumberbatch and Martin Freeman, did, though. Maybe we should blame that.

You could, if you wanted to, use the same trick to tell the exact opposite story. Facebook was first released in 2004, when the suicide rate among 15- to 19-year-olds in England and Wales was 4.7. But after six years of social media being available, it had dropped to 3.1! It’s a life-saver, no?

Suggested reading

Why it's time to panic about kids and smartphones

By Peter Franklin

Or, you could use the trick to tell other, more political stories. For instance, one of the highest rates of suicide on record was 1998, shortly after the election of Tony Blair’s New Labour, when 6.1 out of every 100,000 teenagers killed themselves. But by the time they left office in 2010, it was – as we’ve seen! – an amazing 3.1; New Labour almost halved teen suicide! And with the election of three Tory-led governments in a row, it’s crept back up almost to where it was.

I have no idea if that’s true – and I should admit, I’ve done some shameless cherry-picking of my own, there, choosing 1998 instead of the year Blair actually swept to power, 1997, which was a less spectacular 5.5 per 100,000. You could tell a plausible-sounding story about cuts to mental health provision and social care and housing support and so on. But if you did so, you’d also have to explain why suicide in general, including among young people outside this very specific bracket, is on the decline.

The trouble is, for people who want to say that social media is causing an epidemic of suicide, is that there’s just no good evidence for it. For a start, as I just mentioned, there isn’t really an epidemic of suicide anyway. If you look at 10- to 29-year-olds, it’s gone from a consistent plateau of about 15 per 100,000 from the late 1980s to the early 2000s, down to about nine per 100,000 by about 2004, and that’s where it’s stayed, pretty much. There’s a similar story among the population as a whole.

Suggested reading

Why we should fear technophobia

By Tom Chivers

What’s more, when you look at the links between social media and mental wellbeing, they are deeply ambiguous and extremely weak – and the data doesn’t show causation, anyway. An enormous analysis of three major datasets by Oxford University scientists, published in Nature Human Behaviour last month, found a tiny, uncertain correlation between the two. But that correlation about the same as the correlation between well-being and regularly eating potatoes, or well-being and wearing glasses. We do not talk of potatoes as having “destroyed a generation”, or worry about bespectacled teens and suicide. The Human Behaviour study was also interesting because it showed how you could find scarier results – by doing exactly what the Sunday Times story did, in essence: by HARKing, getting your data first and then choosing the bits to display to make it look scary.

The sadness of it all is that this comes on the back of some extremely sensible advice being issued by the Royal College of Paediatrics and Child Health, and a surprisingly even-handed report by the House of Commons Science and Technology Committee. Both acknowledged the shortage of good evidence, and made common-sense suggestions for parents – in particular, rather than imposing hard limits, the RCPCH guidelines simply said that if you find that social media or screen time gets in the way of family time or things you want to do, then you should reduce it. It felt like a little turning point in the endless barrage of scaremongering. Hopefully it really was.

The climate deniers, by the way, appear to have largely stopped using the 1998 trick, after 2014, 2015 and 2016 were all hotter than 1998 and, in fact, each hotter than the last. The trick, in essence, stopped working. We don’t need extra reasons to hope that teen suicides go back down once more, but making it harder to scaremonger about social media would certainly be a bonus.

Steven Pinker
Author and Johnstone Family Professor of Psychology at Harvard

Our World in Data is a site that portrays our world with data rather than with headlines. The view is utterly different.

News is about what happens, like wars and epidemics, not what doesn’t happen, like peace and health. And it’s about what happens suddenly, whereas the world is often transformed gradually in increments that add up to momentous change.

The economist Max Roser has assembled an astonishing array of datasets about the world, with visualisation tools that make them easy to grasp, and concise commentaries that highlight the major trends and developments and explain why they happened. One comes away from the site with a radically different and far more accurate image of the world and times we live in.

Christina Sommers
Scholar, American Enterprise Institute

” The definition of insanity is doing the same thing over and over again and expecting a different result.” This quotation is ubiquitous and usually attributed to Albert Einstein. But there is no evidence he said it. According to the indispensable website Quote Investigator (QI), the closest match first appeared in a 1981 Narcotics Anonymous pamphlet. After that, it flouted around self-help circles for a few years. By 1990, newspapers began attributing it to Einstein. Now it’s an internet staple,

QI’s long list of fake quotations is a revelation. Edmund Burke never said, “The only thing necessary for the triumph of evil is that good men do nothing.” Still, the faux aphorism is a favorite on social media. Everyday, throngs of Twitter and Facebook users pass it along. Someone needs to direct them to Quote Investigator.

Peter Oborne
Journalist and author

I find Conflicts Forum, run by Alastair Crooke and Aisling Byrne, essential for understanding the emerging world order. Their subscription press service has a brilliant knack of pinpointing illuminating articles in unlikely places. In addition, Crooke provides a weekly essay which contains the best and clearest geopolitical analysis you will find anywhere. He has been proved right about events time and again.

Middle East Eye, a site I also write for that was founded three years ago, goes from strength to strength. It provides an illuminating new perspective on the Middle East, and its reporting is rigorous and impartial. MEE led the way on the rise of Mohammad Bin Salman, leading onto the Jamal Khashoggi tragedy.

Voice of Rohingya, a Twitter account (@VoiceRohingyaa) gives an appalling insight into the genocide against the Rohingya people. Finally, for British politics I love the Full Fact fact-checking service, while Hugo Dixon’s Infacts is consistently illuminating on Brexit.

Katharine Birbalsingh
Author and headmistress of Michaela Community School

When I want to read interesting viewpoints, I tend to turn to Quillette, the Australian online magazine and The Atlantic Monthly, the American online magazine.

I like Quillette because they are brave enough to question the usual dogma that weighs us all down. I may not agree with everything they say, but they certainly make me think. The Atlantic Monthly comes at news from the other side politically and from the other side of the world. I like the longer style essays that allow the author to delve deep into a subject. Again, I don’t always agree with the writers, but I enjoy the mental stimulation.

I also like watching The Rubin Report on YouTube, where David Rubin interviews someone for an hour about their area of expertise. He always chooses out-of-the-box thinkers and the interview style is conversational and inviting. I look forward to the holiday where I can really indulge in these three things!

Jim Waterson
Media Editor, The Guardian

I always used to read the print edition of York Evening Press after school and I still check its website first thing every morning. Despite relentless cuts by owner Newsquest, it’s a pretty good model of what a local newspaper should be, with reporters who know the community. If you’re a lucky enough to still have a decent local newspaper, then read it obsessively. You’re likely to pick up on trends and stories that London-based papers miss.

Transport shapes everyone’s lives, shapes where people work, and can make a massive different to quality of life. Also, trains are brilliant. London Reconnections is one of the best written most in-depth sites around. Their post explaining Uber’s business model in the capital is well-sourced, meticulously researched, and clearly written.

A culture magazine funded as a corporate project by a shaving business should be absolutely terrible. For some reason MEL Magazine isn’t and it consistently features weird, considered takes on modern online life. And they haven’t ever tried to sell me a razor.

Elizabeth Oldfield
Director, Theos

The podcast I reliably go to when I want to feel informed, inspired and better equipped be some use in the world is OnBeing.

Created by Krista Tippett, a former New York Times journalist and diplomat, it intelligently explores matters of faith, moral imagination and what a wise life looks like now. In our noisy information environment most outlets fight for attention by being louder, shorter, simpler and seemingly shallower.

Refreshingly, Krista has grown a faithful and engaged audience by interviewing the hidden influencers, the people too busy making the world a better place to spend time on self-promotion. Listening to it helps snap me out of imposed urgency and reminds me of the fundamental questions: what does it mean to be human, how do we want to live, and who will we be to each other?

Sarah Sands
Editor, BBC Radio 4’s Today

On Twitter, I follow mostly individuals, usually reporters, often the number two or three in the pecking order who are doing a lot of unseen work. For political websites, I’m not wildly unconventional. I go to the usuals: from Politico and the FT, to ConservativeHome, Slate, and The Hill to name a random a few.

Otherwise, I read National Geographic, and look at BBC Security Correspondent Frank Gardner’s Instagram for bird pictures and sunsets. I also like Eyob Derillo at the British Library on ancient illuminated manuscripts.

When it comes to podcasts, Brexitcast and Beyond Today from the BBC, as well as David Axelrod’s The Axe Files. Finally, Dead Ringers and anything to do with Tom Service on BBC Radio 3.

And UnHerd’s own Peter Franklin shares five tips on how to use Twitter without losing your mind

Ah, Twitter! Is there any social media site that’s more hated by its own users? We’re all addicted, of course – but here are some harm reduction tips for 2019.

Firstly, don’t argue with strangers. Obvious really, but all the same, just don’t.

Secondly, and further to the above, turn off your mentions – or at least restrict them to people you follow.

Thirdly, though Twitter is a truly horrible forum for debate, it’s a brilliant noticeboard. Try following fewer individuals (especially ranters, emoters, edgelords and news junkies) and follow more publications (especially those that publish and link to the work of thoughtful, insightful writers).”

Here are a couple of suggestions:

The New York Times is not exactly the most unheard-of publications – but for links to some of the impressive analytical journalism and infographics available in the world today, follow @upshotNYT and @NYTgraphics accounts.

On the other side of US political aisle, but very reasonably so, is the Institute for Family Studies. For links to eye-opening social research (and to brilliant writers and researchers like @lymanstoneky and @wilcoxNMP), follow the Inst. Family Studies account.

Breaking my rule about individual accounts, I can gratefully recommend Sunder Katwala’s – who manages to tweet about Brexit, immigration and other divisive issues without being divisive. I don’t always agree with him, but his tweets are a peaceful island in a roiling, angry, snarky ocean.

Fourthly, get a reader app like Pocket that allows you to quickly download articles from Twitter links.

Fifthly, once you’ve got something grown-up to read, get off Twitter.

Part One

New Year, New Media – part one

By Various

Though they provide different functions and services, this immense power resides in a shared business model: the extraction and analysis of data and the control of digital infrastructure – from mapping to cloud computing – that underpins all digital technology. And as a result of this control, they are also poised to dominate artificial intelligence markets, extending their power far into the future.

Further reading

How the tech lords privatised the public square

By Jamie Bartlett

Profiting from data generates a circular, expansive dynamic: the more data that is captured, the greater the potential for revenue. As such, the impulse to accumulate more and more dominates the strategies of the tech giants, and dictates how they develop and deploy their technology. By instinct, they seek to move into more markets, buying out more competitors, driven by their voracious appetite for more data and infrastructure. Amazon, for example, has filed for 5,860 patents, while Alphabet has acquired over 190 businesses since its foundation.

These companies will counter, accurately, that in the process they provide many benefits. Consumers get useful services at low to no immediate cost, with greater convenience through coordination, and with more personalisation. Businesses get access to new markets and experience lower transaction costs. At a societal level, their platforms increase access to information, and enable social connections and greater efficiency.

But this business model is driving platforms to behave in ways that create serious economic, social and political problems. By acquiring innovative start-ups and limiting access to their large datasets, platforms are almost certainly limiting the innovation potential of the economy.

Further reading

The inhuman politics of artificial intelligence

By Jon Cruddas

As the digital giants increasingly dominate their markets, their profits are soaring, and with relatively low levels of employment these labour-light digital firms are helping drive inequality. Their dominance in advertising markets has already eroded news media, while social network platforms have potentially been used to manipulate democratic proceedings. And as the power of the tech monopolies grows, these problems will grow with them.

In the face of such concentration, defenders of open and inclusive innovation must have a strategy for reining in the digital giants. So far, the development of the modern digital economy has largely been determined by powerful market actors, with little proactive response from government. Without action, the boundless ambition of the universal platforms threatens to undermine democracy, accelerate inequality and concentrate economic power.

The goal must be to move from conditions of ‘enclosure’, where data and infrastructure are dominated by a few tech giants, to an open and innovative digital ‘commons’, where they are organised as a collective resource. In the process, this can unlock a wave of more inclusive innovation and enterprise, supporting the revival of democracy.

Further reading

How technology sidelines the citizen

By Carl Miller

There are, of course, limits to what the UK can achieve with domestic levers ­– the digital economy is international. Nonetheless, we are not powerless. Much more can be done with bolder policymaking, and IPPR’s new report, The Digital Commonwealth, sets out an ambitious reform strategy to build a ‘digital commonwealth’.

First, we should strengthen competition law to foster innovation. The Competition and Markets Authority should be reformed so that it is better able to regulate tech companies. This should include a new requirement for major digital companies to open up their data upon entry to new markets in which they have a major advantage as a result of their existing data.

Second, where platform giants provide vital infrastructure services such as searching, matching, or communicating, under monopolistic conditions, we should regulate those activities and companies as public utilities. This should be done through a new Office for Digital Platforms, which should include powers to open up the data of the digital giants in cases and markets that are deemed to be in the public interest.

Further reading

To save democracy, we must disrupt it

By Carl Miller

Third, to better drive the curation and productive use of public data, a new public service corporation, Digital UK, should be created. This would establish a Digital Citizen Account and ‘public data stores’ through which useful insights from pooled, anonymised data could be shared. It could also oversee the creation of a national data portal, which would provide a single site linking to private- and public-sector data. Under appropriate privacy and security conditions this could provide an immense and shared resource for civil society, enterprise, and local communities to build better products and services.

Finally, we should learn from the cities at the cutting edge of digital governance, such as Barcelona, and get local authorities to set up Local Digital Commonwealth Strategies. These would aim to support the development and adoption of local platform services – from community decision-making apps to innovative businesses – leveraging public sector data sources and personal data from the opening-up of private sector data. This would ensure that value is retained and circulated among communities where the data is generated, to provide economic, social and environmental value.

Taken together, this is an ambitious agenda to rein in the power of the tech giants and foster inclusive innovation. The alternative – a status quo that will enable widening inequality and the growing concentration of economic and technological power – is the route to deepening democratic and social crisis. It is time we acted with an ambition fit for the machine age.

The most organised example of this is the founding of the Center for Humane Technology, whose mission is to “realign technology with humanity’s best interests”. But the most interesting is a clarion call from Facebook co-founder Chris Hughes to re-examine the basic business model that makes these companies tick. It’s this model that has rocketed both Facebook and Google past the banks and oil companies that used to dominate the ranking of the world’s corporate giants to the very top of the list. And it’s turned a handful of entrepreneurs such as Zuckerberg into the wealthiest tycoons on the planet.[1. Chris Hughes, ‘The wealth of our collective data should belong to all of us’, The Guardian, 27 April 2018]

And what’s that model? In one word, barter. It’s the revival of an approach to business that had almost died out in modern societies. When did you last offer the baker a chicken in exchange for a dozen bagels? Or pay the window cleaner in cabbages? Modern economies use money as the means of exchange.

There’s been little focus on the revival of the barter economy that lies at the heart of the business model of Facebook and similar companies. Since we don’t pay money to use their services, they seem to be “free”. But of course they are not. Each time we “like” something on Facebook, or check Google for the opening hours of a restaurant, we’re feeding tiny morsels of data into an enormous machine that has been brilliantly designed to turn that data into truly enormous quantities of cash.

We’re bartering our data. And while each “like” or search inquiry isn’t worth much, the aggregation of billions and billions of these data morsels means Facebook the company – despite all the recent controversies – is worth half a trillion dollars, and every year returns a profit to its investors of a staggering 30%. The company had income last year of $40,000,000,000, almost all of it from advertisements; after expenses, there was a profit of $15,000,000,000.[2. See Matt Egan, ‘Facebook and Amazon hit $500 billion milestone’, CNN Money, 27 July 2017; and Rakesh Sharma, , ‘How does Facebook make money?’, 25 April 2018]

The impact of this data-driven business model has been enormous, driving the growth of a handful of mega-companies. As I argued here on UnHerd, governments have entirely failed to apply traditional pro-competition policies (what in the US is called anti-trust) to prevent any single company from dominating the market.[3. The classic example: Luigi Zingales and Guy Rolnik, ‘A Way to Own Your Social-Media Data’, New York Times, 30 June 30 2017: “When Standard Oil, in the early 1900s, controlled 90 percent of oil refinery capacity in the United States, the federal government used its antitrust power to break it up into more than 30 smaller companies. A similar breakup was imposed 70 years later on AT&T.”]

Suggested reading

The cracks appearing in the digital economy's foundations

By David Z Morris

One reason for this is that the barter model is thoroughly confusing. Consumers have no idea how much they are “paying” for the service, and are far less aware of the fact that they are in the hands of a monopoly than they would be if there were only one company from which they could buy a car. Facebook looks a bit like a charity, and the idea that its ads are personalised for you can seem like it is doing you a favour. While the Cambridge Analytica scandal over how data was misused has tarnished this image (and hit the stock price), there’s no evidence it has made a long-term difference to how the company is seen and the loyalty of its more than two billion users.

As David Z. Morris pointed out recently, if competitors are to have any chance against these mammoths, governments will need to intervene. For example, Morris suggests, they could require that a portion of the data gathered by a company like Amazon be shared with its rivals. Or, as two University of Chicago professors recently argued in the New York Times, consumers could be given a property right to their own data.

Chris Hughes has a different idea, based on the idea that everyone should benefit. Writing in the The Guardian, this is how Zuckerberg’s Facebook co-founder captures the problem:

Nearly every moment of our lives, we’re producing data about ourselves that companies profit from. Our smartwatches know when we wake up, Alexa listens to our private conversations, our phones track where we go, Google knows what we email and search, Facebook knows what we share with friends, and our loyalty cards remember what we buy. We share all this data about ourselves because we like the services these companies provide, and business leaders tell us we must to make it possible for those services to be cheap or free…

Just like many other business leaders, Mark Zuckerberg describes this as a win-win – people stay in touch with friends and family more often through a free service, and businesses can more efficiently spend marketing dollars to fuel corporate profits, making Facebook free to all. But one party has benefited a lot more than anyone else: Facebook shareholders.

The question, therefore, is how to share that benefit more equally. One option is to clarify in law that we own our own data, and have a system of “data brokers” to sell it to the highest bidder. Another, which Hughes favours, is by taxing the use of data by these companies, and sharing the revenues among people whose data has been harvested. There’s a parallel, he suggests, in the US state of Alaska, where a portion of oil company taxes fund an annual payment of $1,500 to every resident.

Suggested reading

It's time to rein in the reckless monopolies

By Ann Pettifor

Hughes’ proposal is a creative solution, and before the companies start to complain about an additional tax burden they need to be reminded that neither Facebook nor Google invented the internet, or GPS (the basis of location services), or much else in the architecture that undergirds their business models and their enormous profitability. That technology was developed courtesy of the U.S. government.

What’s more, Google did not even invent the famous Google algorithm that enabled it to beat out early competitors in the search business. Its founders Larry Page and Sergei Brin did the work while they were students, funded by a US government grant.

It was public risk and public investment that got Silicon Valley and its megafauna going on their huge profits. Now the giants are hoovering up our data and turning it into cash, it’s time to take a long hard look at ways to re-structure our relationship with them.[4. On the public investment that underlies the digital companies, see Mariana Mazzucato’s book, The Entrepreneurial State. I discuss her argument briefly here: https://stage.unherd.com/briefings/big-techs-tipping-point/]

More on big tech...

How the big tech companies are eating free markets

By Charlotte Pickles

Short answer, your digital footprint – something each and every one of us has – is working against you. In 2014, the US Department for Transportation sanctioned airlines and travel agencies to collect personal user data – including nationality, marital status, address and travel history – to tailor fares, which it claimed allowed “more agile pricing and more personalised offerings”. A blunter term to describe it would be price discrimination – also known as “personalised pricing” – the practice of charging would-be digital customers differing prices based on their perceived willingness to cough up.

Of course, price discrimination has existed for decades outside of cyberspace in the form of early-bird specials for the elderly, or knock-off rates for students, but on the internet you shed the relative anonymity you have on the high street. You’re no longer part of a broad-brush vision of what a potential customer looks like, because online businesses already possess intimately detailed accounts of who you are, what you like and how much money they think you’re willing to spend with them. If it sounds ominous, that’s because it is.

These nebulous marketing practices build profiles of your habits and interests to better target adverts and search results for products, platforms or services directly at you. This data ranges from the basic – the kinds of websites you visit, how long you stay there, your IP address, your location and the type of device you’re using – to the extremely sensitive – health and financial info, your sexual orientation, age, income, parental status.

If that doesn’t sound sinister enough, the vast majority of this data is collected without your knowledge or consent, via cookies (text files which collect and store information including user names and browsing history), your IP address and tracking scripts which monitor your web browsing across sites, among other methods. In the case of the Thailand price disparity, the airline knows you’ve been looking regularly at flights between London and Southeast Asia, and therefore feels it can get away with charging you a higher amount. Every click, every new page opened, every refresh is probably being tracked.

I say probably, because there’s no definitive way of knowing. Businesses are incredibly secretive about how they track consumers across the web, which is partly down to wanting to keep their competition in the dark, but mostly down to fear of widespread customer backlash.The recent outrage over consultancy firm Cambridge Analytica’s alleged misuse of 50 million Facebook users’ personal information, taken without permission and reportedly used to create personalised adverts during the 2016 US Presidential election, demonstrates the potential pitfalls for companies who are less than forthcoming about their intentions for the data they collect. It’s little wonder consumers are losing faith in Big Tech.

Suggested reading

Who's protecting the vulnerable from predatory markets?

By Polly Mackenzie

The few concrete examples we have of internet businesses exercising digital price discrimination have been met with public anger, making the perpetrators even less likely to admit to dabbling with it in the future. In 2012, a Wall Street Journal investigation found that the website of stationer Staples displayed varying prices depending on the potential customer’s proximity to a competitor. Online travel company Orbitz was accused of directing Apple Mac-using visitors to higher-priced hotel rooms than visitors using PCs. The reasoning behind this, the WSJ claimed, was because Orbitz had determined Mac owners were inclined to spend 30% more per night on hotels.

Similarly, a 2014 study by Boston’s Northeastern University found Walmart and Home Depot were among numerous online retailers modifying the price of products in direct relation to consumers’ browsing history. Even Amazon, back in 2000, was caught charging returning customers higher prices for items than new visitors, which chief executive and founder Jeff Bezos attributed to “a random price test”, adding the company had “never tested and we never will test prices based on customer demographics”.

Some people might argue the moderation of prices based on a shoppers’ perceived ability to pay is a form of digital democracy, charging the rich to subsidise the poor. The rest of us see it as commercial extortion, fueled by covert tracking of information (which may not even be correct) to manipulate us into paying over the odds. What’s more, the sweeping reforms promised by the European Union’s forthcoming GDPR (General Data Protection Regulation) –  stating user consent to cookie tracking must be clearly given – are unlikely to force firms to explicitly tell us the kind of data they’re collecting on us, and what it’ll be used for. As tracking techniques advance, marketing strategies will likely be built around tracking our moods, creating a Big Brother-esque, price-fluctuating digital dystopia.

Just as the rise of fake news forced social media firms to address their own responsibilities in circulating misinformation, the mounting anger over the Cambridge Analytica scandal should force internet giants and retailers to be more transparent in their data collection and retention methods.The tide of public opinion is turning, and consumers are fed up of being treated like cash cows. While data is important, so is trust. Lose the latter, and you’ll find yourself without the former.

Suggested podcast

Test article

By Charlotte Pickles