Over the course of the past four months, Britain has been forced to grapple with a foundational question of moral philosophy and social science: what are rules, and how and why do they constrain us?
Policymakers have been drawn into fraught attempts to legislate for everyday social behaviour, from how many people we may meet in a park, to how close we may stand to someone in a pub. The document published in May, Our Plan to Rebuild: The UK Government’s Covid-19 Recovery Strategy, featured 53 uses of the word “should”. On 3 July, the Prime Minister implored those visiting the reopened pubs to “behave responsibly”. A form of moralised gossip has developed, focused upon rule breeches and loopholes being exploited around us.
This isn’t a standard political problem of jurisprudence or legislation, but entails a vast real-time experiment in moral psychology and behaviour change. The unveiling of a Covid-19 “alert system”, the Delphic command to “Stay Alert”, and a brand new Joint Biosecurity Centre suggest that we are entering a phase in which the ground-rules of everyday conduct will be constantly tweaked and recoded, in response to data. This cybernetic vision makes the question of how and why people stick to rules all the more pressing. It also renders the foundations of political authority precarious.
The immediate cause of this upsurge of normativity has, of course, been the nature of the coronavirus itself; its highly infectious and air-borne quality has thrown basic issues of public etiquette into question. And yet it is telling that there has been far more debate and controversy surrounding rules over the weeks that they have been progressively relaxed, than over the period when they were most strictly enforced.
Boris Johnson’s televised address on 10 May, which promised to reveal England’s path out of the lockdown, succeeded only in sowing confusion as to who exactly should be going to work and how exactly we should be socialising. The Dominic Cummings scandal that broke two weeks later generated mass fury that the government seemed unbeholden to its own rules.
Rules for the many: protesters hold a banner outside the home of Dominic Cummings. Credit: Peter Summers/Getty Images
The period of strict lockdown was tough to endure, but relatively easy to assent to. The slow relaxation of the lockdown has made life less stressful, but – greatly aided by Cummings – generated a new suspicion as to what precisely the rules are for, and who is actually obeying them. Photographs of crowded beaches and parks are shared on social media, encouraging a sense that others aren’t doing their bit. Research by the Oxford Reuters Institute reveals that trust in the government’s information about coronavirus plummeted over the course of May. Should a second spike require England to reintroduce restrictions, there must now be a serious danger that many people will simply decline to respect rules that seem neither enforceable by the state, nor respected by its most senior figures or many members of the public. Quite what would happen then is anyone’s guess.
The problem is that creating new laws is a relatively simple task, but creating new etiquettes and norms is anything but. While the state was deploying its legal sovereignty to impose universal restraints – to stay indoors, to close restaurants and pubs, to close schools – obedience was morally straightforward. But once the government becomes involved in handing out limited and complicated permissions, questions of fairness and conformity became far more heated. If the ongoing lockdown exit strategy is to involve constant tweaking of rules on a region-by-region basis (as witnessed in the local lockdown in Leicester), with schools being opened and closed as the infection rate fluctuates, we can expect resentments and suspicions to rise further. As rules lose their sense of universality and consistency, they come to feel arbitrary and unfair. The slow return of liberty is a more contentious process than its removal.
Rules appear very different when we are bound by them from how they appear to the outside observer. For followers of the 20th-century philosopher Ludwig Wittgenstein, we can only understand rules if we know how to follow them in practice. In The Concept of Law (1961), the legal scholar HLA Hart argued that laws cannot be grasped merely in terms of the behaviours they produce or the sanctions that enforce them, but also consist in a shared sense of obligation to uphold them.
That obligation is a social one: a rule or law that only applies to me is a nonsense. And if laws become purely a question of written instructions, punishment or predicted effects, they lose what makes them the law.
Similarly, the sociological tradition of ethnomethodology has sought to grasp the binding nature of informal norms, by studying interactions from the perspective of participants. In “breaching experiments” researchers went into everyday situations and deliberately disrupted micro-rules of social life (such as how close to stand to someone) to study the result. Students of the social psychologist Stanley Milgram conducted an experiment on the New York City subway in the early 1970s, in which they approached strangers and asked them to give up their seats without justification. The result was surprising: most did give up their seat, but the researchers found the experiments extraordinarily distressing to carry out.
Common to all these perspectives is a sense that human beings are fundamentally rule-bound creatures. Our capacity to understand one another (whether or not we’re speaking) is thanks to the customs we share. Etiquette, morality and eventually law itself are built upon this existential fact, that humanity and culture consist in conforming to ultimately inexpressible notions of how we ought to behave, and what happens if we don’t. By the same token, feelings of fairness and normal conduct penetrate all the way into our selves and society.
This illuminates something about the depth of public anger that broke over the Cummings scandal in May. Powerful people have exploited their positions for personal gain throughout history, while many political scandals that outrage commentators and opposition parties scarcely register with the public.
What was unusual about the Cummings furore was that it surrounded a rule that applied to an entire society, or so it seemed for a few weeks. A far costlier and more cynical breach, such as the 2012 Libor-fixing scandal, involved a norm to which most of us were ignorant or uninterested outsiders, and hence relatively unmoved by. Not so with the lockdown. The obvious analogy would be to the MPs’ expenses scandal, which also triggered mass public outrage, because it centred around a norm – the honest submission of receipts – that every member of society can understand, even if they don’t always obey it themselves.
Rules become something very different once they are viewed from the perspective of the outside observer. As we might find when watching a foreign sport for the first time, they become more like patterns or regularities. This is the perspective associated with behaviourism, the tradition of experimental psychology that attempts to circumvent questions about consciousness and meaning, and focus entirely on predicting how people will respond to different external stimuli. As far as the behaviourist is concerned, a rule is just one more way of seeking to alter someone’s behaviour, along with incentives and penalties. The aim in each case is to render behaviour more predictable, and therefore – ideally – more controllable, for whatever purpose.
The rights and wrongs of behaviourism have featured as a constant subplot in Britain’s response to the pandemic, partly due to the influence of the Behavioural Insights Team, or “nudge unit”, which has been a world-leader in developing behaviour change policies. What distinguishes “nudging” is its synthesis of economics with moral psychology, which produces a vision of individuals who are typically governed by incentives and self-interest, as economists presuppose, but sometimes governed by a sense of right and wrong, as moral psychology would assume. Thus, if I want you to give up smoking, it may be that the best way is to pay you to quit (the economic solution), but it may be that it would be more effective to convince you that most people have already quit (appealing to your sense of normality). The only way to know which is preferable to is to run a controlled trial.
Ever since the lockdown was first debated in March, policymakers have made the working assumption that people would not adhere adequately to the lockdown or not for very long (the precise source of this assumption is disputed, and the various behavioural psychologists on the Scientific Advisory Group for Emergencies (Sage) have distanced themselves from it). The implication was that our self-interest is simply too strong, and our respect for rules too weak. This now appears to have been a mistake (and a very costly one, as it resulted in the lockdown being delayed, with severe consequences for mortality levels), posing the question of whether British policy- making has been so drenched in economic orthodoxy over the decades as to find the idea of mass rule-following unimaginable.
To some extent, the challenge of coronavirus requires the use of behaviourists. The immediate challenge, after all, is not to establish some new social contract or vision of justice, but to alter behaviour until public safety is assured – to “Control the Virus”, as the government put it.
But even with a well-rounded theory of human psychology, which grants adequate space for moral instincts, this is a diminished basis for politics, which erodes the foundations of political authority. Once morality is conceived as a psychological tendency, it becomes just one more response to be triggered by some nudge or stimulus.
There is something nihilistic about a world-view that reduces morality to a type of predictable behavioural reaction. Rules lose their sense of rightness, and become instead a pure instrument to alter what someone is doing.
Nudging is the immediate precedent for this, but another would be the introduction of anti-social behaviour orders (Asbos) under New Labour, effectively a type of personalised law, that stipulated specific non-criminal behaviours that a person was not permitted to perform. The Asbo offered a glimpse of our pandemic future, with rules shattering into dozens of conditions, variations and person-by-person applications.
This is how rules morph into algorithms, ever-more complex sets of instructions, built around an “if/then” logic. By collecting more and more data, and running more and more behavioural tests, it should in principle become possible to steer behaviour in the desired direction (the term “cybernetics” derives from the Greek term kybernetes, meaning the steersman of a ship).
For a company like Facebook, which can scrape billions of data points on our relationships, interests and interactions, there is no need to understand what our friendships and cultural tastes mean to us, or what the term “friend” even means. It is sufficient to identify the specific, salient details that appear to influence how we behave, rendering us predictable. We are not rule-following creatures, but pattern-forming ones.
The government has stumbled into a sort of clumsy algorithmic mentality over the past couple of months, in its attempt to relax restrictions as quickly but as safely as possible. Ministers have been pressed on distinctions such as between “cleaners” and “window-cleaners” and the precise definition of a “household”. Rumours circulate of other ideas that have been in the mix, such as permitting each individual to select a “social bubble” of ten people outside their home.
There is a logic driving all this, but it is one only comprehensible to the data analyst and modeller, while seeming deeply weird to the rest of us.
The difference between the world as it appears to Facebook and the world as it presently appears to Whitehall is that, in the language of cybernetics, the former benefits from constant feedback. There is a “return path” constantly sending behavioural data back to Facebook to be analysed. The government would clearly like to establish a similar data analytics capacity to deal with coronavirus, if only it can find the technical competence. On the evidence of Leicester, where it seems nobody was aware of the surging rate of infections for several weeks, there is still a considerable way to go.
Nevertheless, the geeks that contributed to Vote Leave’s dubious success – data analyst Ben Warner, the AI contractor Faculty, and Cummings himself – are now applying their know-how to the pandemic. They are grabbing whatever data is available (social media, telecom bills, credit ratings) in an effort to map behaviour in real-time. The implications for privacy and the future opportunities for private contractors are grim.
To the algorithmic mind, there is no such thing as rule-breaking, only unpredicted behaviour. Thus, what most of us might view as a breach of a norm, Facebook views as a new data point to be learned from.
This also gets to the heart of the Cummings affair. What was notable in his rambling press conference in the Downing Street garden wasn’t just the lack of contrition, but also his peculiar, post-human sense of what rules actually are. He appeared to think that, with enough microcosmic empirical details of his behaviour, eventually his critics would be satisfied.
What he was either unable or unwilling to fathom was that, for the millions of people who had been stuck at home for weeks, driving the length of the country was not how the game of lockdown is played. Where most of us saw shared obligations, Cummings saw only lines of code and behaviours.
No doubt behavioural psychologists have much to offer policymaking in this crisis. The government’s legitimacy crisis isn’t the fault of nudgers such as David Halpern, but of how behaviourism meshes with Johnsonism. Johnson and his allies have frequently failed to understand what morality means to people, as if it is merely a habit that some of us happen to fall into.
Every time Johnson or his allies have been called to apologise for something (be it his past comments, shortage of PPE in hospitals, or the Cummings scandal), the answer has always been the same: “I’m sorry if you feel that this is wrong,” but never, “I’m sorry for this wrong.” It signals an eerie failure of understanding. As Wittgensteinians argue, a computer could analyse billions of situations in which the word “sorry” is used, and still never know what it means to be sorry. One sometimes gets the same impression listening to the PM seeking to draw a line under his government’s misdemeanours.
Johnson with a moral conscience wouldn’t be Johnson, and he certainly wouldn’t be prime minister. He entered No 10 on the promise to rip up the rules of public life, to “get Brexit done”. Cummings fancies himself as a kind of Nietzschean disrupter, unafraid to thwart convention. But these same qualities have come to appear downright dangerous to public safety.
Given the complexity of the coronavirus challenge, and the necessity of rewriting the rules of social life, the need for a sense of legality and social contract to undergird policy is all the greater. The intricacy of the lockdown regulations make it all the more important that a figure as prominent as Cummings exemplifies them in their spirit, rather than deconstructs them line by line.
The rules that have been imposed upon us these past few months are unlike anything we’ve experienced before: a mixture of law and algorithm, informal etiquette and formal code. The government can carry on imploring us to do the right thing, and not jump ship too soon. They can drag out motifs of national solidarity and protecting the NHS. But if people give up, and if the police are unable to constrain them, the question will arise of why people ever obey rules in the first place.
It then becomes all the graver that Britain, or at least England, is led by a leadership double-act that hasn’t the foggiest idea.
This article appears in the 08 Jul 2020 issue of the New Statesman, State of the nation