This is a draft article due to appear soon in the European Law Journal.
The rule of law is an ancient, global and malleable tenet of political morality. As a check on power, it requires equal subjection of everyone to the law, irrespective of wealth or status. Assessments of the health of the rule of law have tended to focus on the ability of the state to act arbitrarily, but private entities have in the past acquired power to rival the state and to rise, in effect, above the law. This article considers that the power and behaviour of today’s giant tech companies are throwing the rule of law out of kilter. They increasingly assume the trappings of the state but without the accountability. Meanwhile they seek to evade, dissuade or capture all attempts to regulate and enforce against the socio-economic and human rights damages that result from their business models. They profit from and perpetuate growing inequality within and between societies, which further undermines the principle of equality before the law. As policymakers grapple with new legislative techniques, they are unlikely to repair the consequences of extreme concentration of economic and political power without addressing underlying social injustices and over-deferential democratic institutions. Leviathan ultimately cannot co-exist with the rule of law – not because of its actions – but because of its power.
The rule of law is indispensable to any viable modern state. A society if it is to function predictably requires most if not all of its members to accept the supremacy of the prevailing rules. This ancient concept has waxed and waned, less itself a source of light than a reflector of more forceful imperatives, such as the constraint of the power of the state. Power is diffuse, however, and not confined to the machinery of government. There have been moments in the past where private entities have appeared to rival or even supplant the state in its ability to organise society. A debate is currently running about whether we are living through one such moment.
A few global digital monopolies seem immune to state control. The economic power and staggering levels of wealth accumulated by their leaders are as much a symptom as a cause of worsening inequality within societies. A privately-owned social media platform is able to amplify outrage so widely that it results in the formation of militias or the election of an autocrat. A monopoly search engine can threaten to withdraw its services because it objects to a draft law debated by the national parliament. Where the state is the guarantor of human rights and freedoms, and democratically accountable to the people for its actions, this is a perilous trend. It indicates that the normal levers of accountability are defective, and that people must instead depend on the voluntary benevolence of private entities. This may be considered the dark side of the rule of law: how a democratic state can expect to constrain through law a private power greater than itself, a power that even dons the external paraphernalia of the state in a bid for legitimacy.
In this article, I first explore the concept of the rule of law as a check on arbitrary power, how it can be distinguished from law itself, and how it relates to human rights, equality, democracy and justice. I will note that although the value of the rule of law is to constrain the powerful, power is not the sole preserve of the state. The application of the principle to non-state actors has tended to be overlooked until now, after two decades of the almost unimpeded rise to monopoly of a small number of multinational companies, based in the United States and China, that have exploited the opportunities of digitisation. Second, I highlight how the largest companies in the tech sector have wielded their extraordinary and growing power, and how concentrated power and harmful business models exacerbate societal inequalities. In particular, I will look at the how the biggest tech companies have begun to behave like surrogate states, the most brazen example being the ‘oversight board’ established by Facebook and described by its CEO as ‘almost like a Supreme Court’. Third, I will propose that these inequalities place the rule of law under pressure. Legal remedies for harmful corporate behaviour risk floundering through unequal access to justice, legalism and the tendency for these companies to assume, or to have conferred on them, the role of de facto regulator. Fourth, I will suggest how these ills might be remedied.
2. Fighting the Power
The Rule of law over time and space
The rule of law is universally venerated even as it is assailed. In February 2020, the Prime Minister of India proclaimed that ‘the rule of law is the foundation of societal values’, around the same time his government was abolishing Jammu and Kashmir’s constitutional guarantee of autonomy. In January 2021, the Chinese Communist Party (CCP) issued the ‘first special plan for advancing rule by law [by 2035] in the country since the founding of the People’s Republic’, applicable to all ‘exercise of power’ excepting the CCP leadership itself, and without any recognisable levers of democratic control. The European Union spent much of 2020 discussing how to tie the respect for the rule of law to disbursements of EU money to the avowedly illiberal regimes of Hungary and Poland. Globally, the rule of law is assessed to be in decline, especially in terms of substantive features (the ‘thick’ conception of the rule of law) rather than merely formal (or ‘thin’) rules and procedures.
As a tenet of political morality, the rule of law has become so heavily invested that it pays to revisit its origins and evolution. In other words, we have to think about power. Power is the ability of one person to coerce or persuade another person to do what he wants. A good law functions as a check on power. It may also act as a remedy for severe imbalances of power by investing weaker parties with the right and/or the means to defend their interests. Hobbes wrote that only an absolutist strong man could ensure that such good laws prevail; this Leviathan, however, could not himself be bound by the laws which he imposes on his subjects. Such authoritarian rule by law stands in contrast to the rule of law.
The rule of law implies that law, not any individual or group of individuals, governs the behaviour of human beings in a society. The notion has an ancient and global heritage, albeit owing its steady popularisation in particular to English legal philosophers from Henry de Bracton through Edward Coke to A. V. Dicey. One of the first Hindu Upanishads includes possibly the first definition of the rule of law: ‘Law is that which is the king of kings, nothing is superior to law.’ This is a fiction, of course, because laws are written, adopted, interpreted and enforced by humans. These legal iterations never end, and there is no original legal precedent. The concept of the rule of law thus evades definition or reduction. It is not the same as the law, nor is it rule by judges. Nevertheless, it is now a cornerstone of most liberal democratic constitutions. Article 6 of the United States Constitution places Congress under the obligation to support the Constitution; as Thomas Paine noted, only the law, and not even the democratically-elected representatives, should be king. Application for membership of the EU is conditional on respecting, and on commitment to promoting, the rule of law.
In the absence of a neat definition, the rule of law easily becomes a convenient vessel for all the things its advocates hold most dear, such as democracy, civil liberties and social justice. More problematically, in our age of ever-starker inequalities, the rule of law has been seen by some as a guarantee of respect for private property. Its essential value proposition is the protection of all people within a jurisdiction from the arbitrary exercise of power. Kofi Annan as Secretary General of the United Nations reported that it was part of governance: everyone is accountable to the law, which is enforced equally and independently adjudicated. Indeed, equality before the law seems a prerequisite for the rule of law, as has been the case in Europe, theoretically at least, since the French Declaration of the Rights of Man of 1789.
The Rule of law, democracy and human rights
The rule of law often appears as one constituent of a trinity of principles, alongside democracy and human rights. As with the Christian mystery of the Trinity, overlaps and tautology abound. The rule of law is a core value separate from respect for human rights, but also said to comprise these rights, like equality before the law and the right to judicial remedy. These liberal paraphernalia are not synonymous, but their relationship is intriguing. Human rights, like the right to privacy and freedom of expression, depend for their actuation on the rule of law so that those persons whose rights are infringed can demand redress, and the perpetrators of the infringement can be reprimanded and penalised. The rule of law is itself dependent on democracy, because enforcement can be arbitrary if not circumscribed by checks and balances, and by accountability to the general population. On the other hand, you could envisage a democratic system where, without human rights and/or rule of law, the will of the majority of electors prevails to the detriment of the rights and dignity of the minority. This was the state of things in classical Athens and the pre-emancipation United States. As Mireille Hildebrandt succinctly put it, ‘The rule of law establishes constitutional protection of citizens’ rights and liberties over and against their government, safeguarded by an independent judiciary that shares the authority of the state. This is called the paradox of the Rechtsstaat: the state gives its authority to those that judge citizens who contest the way the state uses its authority in a given case.’ Jurisprudence concerning the rule of law itself, at least in Europe, has tended to focus on ensuring the independence of the judiciary.
The Rule of law and justice
The rule of law bears close affinity to justice. Law is to be distinguished from justice. Justice is incalculable and infinite, whereas law is, to a degree, defined and predictable. Law can become an instrument of power and therefore of oppression or liberation. Through its generality and universality, law can serve justice to the weak and offended. As Derrida argued in ‘The Force of Law’, merely applying rules and conventions like a calculating machine would not necessarily be delivering justice. (This is what Dickens was thinking in Oliver Twist when he coined ‘the law is an ass’.) Justice, according to Derrida, requires analysis of the law and then decision – improvisation, a ‘madness’ – so that law ‘frames and facilitates reason and thoughtfulness in human affairs.’ Hence all laws, especially complex ones, and including codified fundamental rights, allow multiple avenues for interpretative legerdemain, for good or ill. The rule of law, equitably construed, can therefore be regarded as a means to ensuring justice.
The Rule of law and corporate power
This returns us to the question of power. In the modern age, ultimate power has typically, and certainly in theory, reposed in the state. The rule of law has stood as a theoretical counterpoint to and practical bulwark against the abuse of state power. Capitalism, in the meantime, has served up private agglomerations of wealth and patronage to rival or even dwarf the power of many sovereign states. One such entity, the East India Company, eventually wrested control of the government of Bengal in the 1750s. Edmund Burke railed against the Company’s colonialist greed and exploitation of the population. In his attempt to impeach Governor-General Warren Hastings, Burke turned to Natural Law. Repackaging the maxim eundem negotiatorem et dominum, he charged the East India Company with becoming ‘that thing which was supposed by the Roman law to be so unsuitable, the same power was a Trader, the same power was a Lord… a State in disguise of a Merchant, a great public office in disguise of a Countinghouse.’
Most theorists of the rule of law since Burke have not reckoned with private monopolies. However, the pioneers of modern competition law took aim deliberately at the concentrations of commercial power giving providers of essential services, like railroads and telecommunications, license to discriminate between individuals and competitors. John Sherman, defending his trailblazing antitrust bill before the United States Senate in 1890, famously declared that the country could no more accept ‘a king as a political power’ as a ‘king over the production, transportation, and sale of any of the necessities of life.’ After the Sherman Act, there ensued a debate that has never been resolved: was monopoly illegal and dangerous by sheer dint of its bigness, or was monopoly only illegal and dangerous where intent to unreasonably stifle competition had been proven. In one landmark case, U.S. v. Alcoa, a Federal appeal judge, citing a 1932 Supreme Court decision, ruled that ‘size carries with it an opportunity for abuse that is not to be ignored when the opportunity is proved to have been utilized in the past.’ Judge Learned Hand dismissed the defence that specific intent remained unproven – ‘for no monopolist monopolizes unconscious of what he is doing.’ Alcoa was hardly a corporate angel. With an almost complete monopoly over aluminium production in the US market, it had been colluding with Nazi German industry and was accused of refusing to expand capacity to prevent prices falling just when manufacture of warplanes became a national imperative.
The Nazis had declared cartels to be a virtue, hiving off to them hitherto public functions. By contrast, what became known as the ordo-liberal school regarded competition as essential for democracy, and the duties of the state to include preventing the creation and misuse of private economic power. This at the time fringe academic thinking found favour during the Allied occupation, one of whose priorities was breaking the monopolies that had enabled Hitler’s rise and had consolidated his control of the economy. Germany’s 1957 competition law, like Japan’s 1947 antimonopoly law, fulfilled one of the conditions of the return of sovereignty from the departing occupier. In parallel, similar provisions were incorporated into the treaty establishing the European Coal and Steel Community and its successor, the Treaty of Rome establishing the European Economic Community. Both closely resembled US law, albeit with more leniency towards concentration.
International human rights law largely developed in parallel to anti-monopoly and competition rules. Companies are not directly bound by international law, but by the laws of the jurisdiction in which they are based. Contracting states to the International Covenant on Civil and Political Rights and the European Convention on Human Rights are indeed obliged to secure to everyone within their jurisdiction human rights and freedoms, including where actions with an impact on those rights and freedoms take place outside their territory. More recently, the ‘Ruggie Principles’, endorsed by the UN Human Rights Council in 2011, address the power of multinational companies. The principles recommend that states act where businesses violate human rights and that businesses avoid ‘causing or contributing to adverse human rights impacts through their own activities, and address such impacts when they occur’. Businesses should ‘seek to prevent or mitigate adverse human rights impacts that are directly linked to their operations, products or services by their business relationships, even if they have not contributed to those impacts.’ The principles are non-binding and lack any enforcement mechanism. Liability under international criminal law only applies to individual corporate officers, though discussions on its extension are now ongoing which could see a company as a legal person tried domestically or before the International Criminal Court (to which neither China nor the United States is a party) for committing or assisting in the commission of a crime including human rights abuses.
The Rule of law and inequality
Neoliberal economics by the 1970s, incubated in the ‘Chicago School’ before taking root in politics and the judiciary, had begun to discredit any government intervention aimed at curbing the power or harmful behaviour of companies. It drove a deeper wedge between abstract legal notions and concrete socio-economic justice. Sherman in his great speech of 1890 had explicitly connected the lack of constraint on monopoly and the perpetuation and exacerbation of poverty. Among all ‘the problems that may disturb social order,’ he said, ‘none is more threatening than the inequality of condition, of wealth, and opportunity that has grown within a single generation out of the concentration of capital.’ A few years earlier, Yale professor Arthur Hadley, in his influential 1885 study Railroad Transportation, Its History and Laws, had remarked that most price discriminations ‘are in favor of the strong… As such they do great harm to the community by increasing inequalities of power.’ The danger monopolies posed to workers was also widely recognised at the time. Safeguards for unions and labour were inserted into the 1914 Clayton Act which closed loopholes in the Sherman Act that monopolists had been exploiting. The robustness of the rule of law was thus considered a function of the ability of the weakest to summon the law to their defence.
Inequality and poverty were irrelevant, however, to the new orthodoxy of leaving markets untrammelled, so long as ‘efficiency’, manifested principally by low prices to the consumer, was advanced. Richard Posner, doyen of the Chicago School with a senior judicial career spanning four decades, even argued that the ‘logic of the law might be economics’, and that ‘a second meaning of justice… is efficiency.’ This represented a radical deviation from the idea, gestated over the centuries, that overbearing power should be checked, and it gained general political traction in the United States and elsewhere. At the same time, it highlighted competing notions of the rule of law as a safeguard for respect for private property or as a force for justice and fairness.
A Turning point
The results of this deliberate deregulation, combined with globalized capital and concentration in digital markets, are to be seen now, early in the third decade of the 21st century. A small number of private American and Chinese companies now derive enormous profits from mediating human personal, commercial and political relations, aided by the recycling of talented personnel willing to trade loyalties between these companies and public administration. Public policy makers appeared mostly sanguine about this state of affairs at least until two events that happened to coincide at the turn of 2021: the ignominious climax of Donald Trump’s presidency, and the mysterious vanishing from public view of the founder of Ant Group and the third richest person in China, on the eve of the conglomerate’s Initial Public Offering. In very different ways, these developments have resurfaced a question lain relatively undisturbed since the wake of the Second World War: constitutions have evolved to constrain the power of the state through checks and balances, but what can constrain a private multinational company that threatens to become more powerful than the state?
3. The New Leviathans
On 6 January 2021, then-President Donald Trump incited his supporters, including domestic terrorist militias, to attack Congress while it was validating the states’ returns of electoral college votes after of the presidential election two months earlier. Within days, Trump was expelled from multiple privately-owned, US-based communications platforms, notably Twitter, Facebook and YouTube. Parler, a site founded by right-wing billionaires that refuses disclose its ownership and whose 4‑8 million user base included a significant chunk of right-wing extremists and conspiracy theorists, was effectively shut down when it was excluded from Android and iOS apps stores and from Amazon’s cloud.
Cue a fierce debate with a number of distinct strands. Strand one concerns ‘censorship’, and the rightness or wrongness of supposedly denying the First Amendment rights of the First Citizen and his devoted followers. Strand two concerns business models involving manipulation and deception, amplifying outrage and paranoia, facilitating the mobilization of extremist and violent groups, and the ubiquitous tracking of online behaviour, and all for the sake of keeping the ‘user’ engaged. A third strand of the debate concerns the legitimacy and sustainability of having a few private entities with the power to determine, through the design of proprietary algorithms and decisions in the boardroom, who gets a megaphone to a global audience and how content – whether ‘news’, search results, ‘relevant’ products or apps – should be ‘fed’ to whom.
This section is concerned with the third strand. I will suggest five ways that these companies have come to resemble Hobbes’ Leviathan, able effectively to remain above the law:
- By mimicking or pretending to exercise the rightful functions of the state;
- By purporting to apply the values native to their home jurisdiction when penetrating and operating in markets in the Global South;
- By evading, or lobbying to shape, regulation, and contesting every attempt to enforce it;
- By atomising consumers and workers to prevent collective action to contest abusive behaviour;
- By ensuring, through their ubiquity, that everyone is some way working for them.
Tribunes of the people
The promise of digitisation has been economic growth through exploitation of new opportunities to make certain tasks easier and quicker. Going digital has given sellers access to much wider markets, made information easier to gather and analyse, and allowed for personal contact anywhere and anytime, including with people you thought you would never hear from again. Now, with the dawning realisation that our way of life has set off environmental changes threatening the survival of not just other species but also our own future generations, digitisation bears the further burden of faint hopes of reversing those trends.
The profit-seeking companies that have thrived in this sector did ‘not enter or expand markets; they replace[d] (and rematerialize[d]) them.’ They have tried to conceal their purely economic motivation under various cloaks of Good Samaritanism – ‘building community’, ‘organizing the world’s information and making it universally accessible and useful’, ‘empowering businesses’ and so on. Since their respective markets tipped towards monopoly, this rhetoric has suffused broader PR strategies and, more recently, percolated into organisational structures. They do not conform to the theory of sovereign states; they have no territorial base nor Weberian monopoly on violence. On the other hand, they can comprise populations (users) managed via rules (protocols and proprietary algorithms) within sovereign territory (manicured environments), with carefully patrolled borders that resist interoperability and data portability. Facebook, whose CEO once said that the company was “in a lot of ways … more like a government than a traditional company,” now has a Vice President for Civil Rights. It has internal meetings to adjudicate whether to allow politicians to stoke racism and violence, and whether to tweak their algorithms to give ‘conservative’ postings equal amplification. The content of these deliberations, which will have profound impact on the public sphere, are only partially revealed by investigative journalism. It is still trying to launch a digital currency, and has an Oversight Board which Mark Zuckerberg had conceived as ‘almost’ a supreme court – almost in the sense, for instance, that the company is not bound to follow any of its ‘rulings’. The board will in selected cases interpret the ‘Community Standards’ as enforced by resident content moderators.
This positioning has been very successful. It has diverted media and policymakers attention away from questioning the legitimacy and sustainability of the companies’ great power, and instead garnered sympathy for their ‘great responsibility’ in making tough calls to preserve freedom of speech, avoid stifling innovation and support small businesses. An example of this ‘discursive capture’ is how the companies have managed, citing privacy, to resist pressure from law enforcement authorities for disclosure of information on their customers, and to present themselves as the honest innocents caught in the crossfire, like ‘trusted and neutral digital Switzerland[s].’. The emollient slogans and ‘community standards’ are in themselves unobjectionable; the problem lies in the evident disjunction with the brutal business imperative that actually drives decisions. The companies are able to publicly espouse human rights and democracy, apologise for what they claim were unforeseen consequences and promise to do better, and still shield themselves from any external audit of what they are doing.
The new reckoning is still poorly informed, however. Trump was not denied his freedom of opinion and expression; he had been deprived of a few digital privately-owned megaphones having exasperated their lawful owners with his bad behaviour. ‘Free speech does not mean free reach,’ as Renée Diresta put it, ‘There is no right to algorithmic amplification.’ The analogue presidential megaphone was still at his disposal, at least for another fortnight, though he seemed ill-inclined to use it, and he could have just launched his own digital one, but did not. This was unsurprising. Analogue media do not artificially amplify and microtarget white supremacist outrage. The market for these megaphones had become so concentrated that only a few were needed to radicalise thousands of citizens to attempt to overturn democracy. The deeper issue however was not ‘which types of freight can run on a railway’ but ‘who owns the track,’ and the responsibility for this predicament lay ultimately with the political class who allowed it to happen, not on the companies who had simply exploited the opportunities open to them. Handwringing over platforms’ acts of censorship thus plays to their preferred narrative of ‘great responsibility’; the more relevant question for the rule of law is how private entities were ever allowed to become the sentinels of the public sphere.
Around ten years ago national tech monopolies set forth to conquer the world. Whereas Silicon Valley adventurers sought to keep some distance from their government, their Chinese counterparts, were more inclined to flaunt their national origins: CCP interference in corporate affairs and ownership suggests a closer affinity to the East India Company that morphed from a commercial to an overtly military-colonial extension of the British state. In any case, there have been scant attempts to engage with, understand or respect the autonomy of the ‘data-rich’ populations they sought to ‘mine’ for value, particularly when preparations for floating on the stock market began in earnest. Colonialist echoes were unmistakable in the assumptions of powerful companies from the Global North that their technology could transform lives while masking their long term-profit motive. China through its own tech giants has sought to export its model of digitised surveillance and social coercion. Japan’s initiative for international ‘data flows with trust’ has so far floundered because countries in the Global South do not trust the more powerful countries and their industrial champions to respect the ‘policy space’ to make their own data governance decisions, notably on where the data should be stored.
The big platforms were no doubt tools for radicalising and mobilising the mob that stormed the US Capitol on January 6, and their earlier inaction was justified on grounds of freedom of speech enshrined in the US Constitution. But there was no First Amendment right in Myanmar compelling Facebook, as reported by the Office of the United Nations High Commissioner for Human Rights, to enter a ‘financial relationship with the security forces’, nor to enable the ‘spread and promotion of threats and the incitement to violence, hostility and discrimination’ on their social media and messaging services in spite of repeated warnings, before tens of thousands of Rohingyas were raped, tortured and murdered. Freedom of speech and of the press were, by contrast, guaranteed by the constitution of the Republic of the Philippines. Filipino journalist Maria Ressa presented Facebook – used by virtually every person with an internet connection in the country – with evidence of user accounts inventing and spreading disinformation in support of then presidential candidate Duterte, enabling his election and the death of thousands in his ‘war on drugs’. Genuine news outlets have since been sidelined and Ressa, along with other journalists on Facebook, has been smeared, harassed and subjected to death threats, culminating in a court convicting her of ‘cyber libel’ with the prospect of up to six years in jail. Research into YouTube suggests that the site has increasingly attracted publishers and consumers of extremist material and that there is a ‘radicalisation pipeline’ through recommendations and autoplay, but an accurate picture is obscured by the personalization of the service and the secrecy of the algorithm.
The one constant throughout is not staunch adherence to any free speech norms, nor opportunistic adaptation to local rules, nor even consistent application of company policies. Rather it is a massively profitable business model whose integrity is preserved until the pressure becomes overwhelming and some temporary concession is granted. The US and China have been able to exploit weak governance and the absence data of protection rules in the Global South to dominate the digital public sphere, while malicious actors, like Cambridge Analytica, have used populations as a low-risk testbed for manipulation techniques prior to deploying them in more lucrative markets. As Maya Steinitz states, MNCs in cross border contexts have ‘little incentive to act with the kind of care they would exercise if they were to internalize the costs of their management decisions… in the global context the risk is shifted not to an insurer…but rather to the world’s poorest.’ In the case of Silicon Valley’s merchant adventurers, wrote Nanjala Nyabola, their product’s ‘implicit central object’ was the Western, white male, and not ‘partly free countries like Kenya’. Meanwhile, their CEOs decline to give account before any parliament representing the people whose data they mine, other than US Congress: “It just doesn’t really make sense for me,” said Mark Zuckerberg, “to go to hearings in every single country that wants to have me show up and, frankly, doesn’t have jurisdiction to demand that.”
‘Senator, we sell ads’
The most highly valued companies in the world were founded at a moment when antitrust enforcement in the United States was being dismantled and, in China, before laws had had even been enacted. They achieved rapid growth through tactics such as running at a loss for years to achieve monopoly status, pursuing growth at all costs ignoring fake accounts, and a frenzy of mergers. Many of these acquisitions would have contravened pre-Chicago School norms precluding entities from competing on a platform that it owns. Others were self-avowed ‘land-grabs’ to kill off competitors or target tech startups with technology to spy on rivals. They can now, according to Julie Cohen, ‘leverage the logics of performative enclosure, productive appropriation, and expressive immunity’. The result has been a widespread hoodwinking of policymakers and legislators, encapsulated in Zuckerberg’s nonplussed answer to a question from a United States senator in 2018 on how Facebook were able to make money from a ‘free’ social media platform. I will explore in this section three typical techniques shoring up their political and economic power: evading enforcement, intense lobbying and atomisation of customers and workers.
Tech multinationals have an outsized bureaucratic capacity to evade or bend law to their advantage. First, they cloud the ‘definitional gateways’ that determine whether and to what extent their behaviour falls within the scope of the law. The lexicon of the digital economy helps them steer just beyond the reach of enforcement. People lured onto platforms to search for information, communicate, buy and sell are simply ‘users’ lucky enough to do so for ‘free’. Such was the narrative for years and only recently have consumer rights and competition law begun to reflect people’s entitlement to fair treatment online. The term ‘platform’ itself is defined by what is it not – not a broadcaster, not a content provider, not an employer – and what is not cannot be regulated.
Where they do fall within scope of regulation, their lawyers apply themselves to avoid the law biting. Sometimes legal provisions are so convoluted as to afford multiple escape hatches: a good example is Article 22 GDPR, which purports to outlaw seriously affecting and profiling people with ‘a decision based solely on automated processing’. But even when the law is clear, power prevails. Take, for instance, ‘consent’ to interference with one’s personal data processing, now minutely defined to the point of tautology in the General Data Protection Regulation (GDPR) as ‘freely-given, informed and specific’. Consent has of course always meant this, and to be interpreted in any other way would empty the concept of all meaning. Consent should be the signal of a human’s free agency when confronted by an unusual proposition. It should be never be taken as given; anyone requesting consent should be at least as prepared to be denied as to be given satisfaction. Yet this is not convenient to surveillance capitalism, which responds by turning consent into a mass commodity. In a travesty of the legislator’s intention of empowering the individual, the market now fizzes with tools for ‘collecting’ and ‘managing’ consent, getting people to sign on the dotted line and then making sure this can be proven in the event the conduct is contested. When digital freedom is reduced to an irritating popup, its easier to convince policymakers that consent is the enemy of growth and innovation.
Overall, it is important for unified monopolies that enforcement is fragmented. Concentrated digital markets have revealed the obvious synergies between competition and privacy; but big tech companies consistently argue, through what Cohen calls a unique ‘capacity for regulatory arbitrage’, for antitrust to be kept separate from the application of rules safeguarding other public goods, like privacy and democracy.
Big tech companies’ enormous expenditure on lobbying policymakers has accelerated. In 2020 Facebook and Amazon spent more (USD 19.68m and 17.86m respectively) than any other company in the United States, and in the EU Google and Microsoft more (EUR 5-6m each) than any other company according to EU transparency registers. Disclosed spending is only the visible tip of the iceberg: it does not include activities in other countries, hiring of law firms, sponsoring of academics and funding of think tanks.
An insight into the playbook for disruption of digital regulation agenda was gained in the form of a leaked confidential internal Google memo in October 2020. It included deploying friendly but apparently objective academics and think tanks to question planned new rules, sowing division within the European Commission, reframing the political narrative around costs to the economy and consumers, as well as more staple fayre, such as full-court pressing of each of the institutions and mobilising the United States government. The companies have actively promoted the libertarian view that regulation tends only make matters worse for smaller companies and consumers because only big companies can afford to comply, and costs of compliance have to be passed onto customers. In other words, regulatory intervention seems the only solution, yet in practice can only make things worse. The people hired to do the lobbying typically come from government institutions themselves, which facilitates access, as indicated by the hundreds of meetings with officials, and the social bonds between regulator and regulated.
To put this lobbying investment into perspective, the combined budget of the 27 national data protection authorities in EU is about €240m, and that of the US Federal Trade Commission for 2020 was $332m. The budget allocated by second largest philanthropic organisation in the world, the Open Society Foundations, to ‘information and digital rights’ was €0.91m in the EU, and €13.35m globally. Big tech companies command resources that dwarf not only those available to their competitors but also those of the bodies charged with regulating them, and of civil society organisations hoping to hold them to account.
Such an all-embracing influence machine cannot but reproduce within policymaking the ideology of the tech elites. This is reflected in ‘technosolutionism’, the notion that complex social problems can be fixed by technology rather than changes in human behaviour, and the conflation of technological progress with the private interests of the biggest companies. The greatest PR challenges for big tech in the 2010s were arguably the Snowden revelations and the Facebook-Cambridge Analytica data scandal. In both cases, however, big tech were able to frame the discussion to divert attention away from the business models which had created the conditions and opportunities for mass state surveillance and electoral manipulation. Democratic public policy debate is thereby distorted. With more vibrant competition, rivals would compete on the merits of their product; margins would be too fragile and not afford the luxury of surplus resources for lobbying for laws that suit a particular economic actor.
This formidable lobbying architecture notwithstanding, there are some laws are now in place that challenge big tech’s business models. A far greater weapon in their armoury is therefore litigation. Big tech aims to stymie every attempt at enforcement, either by settling behind closed doors or by pursuing all possible avenues in the courts at best to overturn or at least to delay. They are what Galanter long ago termed ‘repeat players’, the economic ‘haves’ whose interests are strategic and long-term rather than focused on the immediate risk of financial penalties. Although government in theory should be the ultimate repeat player, it is the powerful multinationals who have an in-built resource and incentive advantage over would-be enforcers. Powerful companies act solely in self-interest whereas the enforcing agency needs to weigh the political will for a prolonged dispute against the growing costs to the taxpayer the longer it drags on. In this way, Galanter reasoned, the ‘haves’ can determine the interpretation of the law, not only securing favourable rules but also privatising the debate so it takes place under cover of settlement.
One illustrative example is the Federal Trade Commission’s settlement with Facebook for repeated violations of a 2011 consent order. Such was the scale of Facebook’s neglect of the right to privacy that, according to the Washington Post, the Federal Trade Commission could have justified a fine of $7.5 trillion – more money than there is in circulation on the planet. Instead, eight years after the consent order, the Commissioners in a split 3-2 decision settled with Facebook for $5 billion – or $28 for each American affected. The majority justified themselves on grounds of the revealing hypothetical question, ‘Is the relief we would obtain through this settlement equal to or better than what we could reasonably obtain through litigation?’ They had calculated that Facebook would have successfully overturned any attempt to extract a sterner punishment for one of the most egregious privacy violations in history. Facebook’s share price rose when the settlement was announced.
Divide and conquer
Unions have not thrived in modern Silicon Valley. The flexibility to hire and fire while offering extravagant salaries in between suited both promiscuous young engineers and CEOs cultivating the myth of a flat democratic workplace. Attempts to unionise have resulted in summary dismissals; in response to the announcement of the launch of the Alphabet Workers Union, Google’s official response was that it would continue to negotiate with employees individually. Similarly they also aim to treat producers as individuals, algorithmically discriminating in the information, service and pricing offered to businesses that depend on their platforms.
The hundreds of millions scrolling and clicking within the walled gardens of the big platforms are similarly atomised. Everyone is uniquely tracked and microtargeted. Profiles created and maintained by automated processes are lawful under the GDPR so long as a human-in-the-loop can be adduced. Where companies deny processing any personal data, they can plead protection by means of trade secrets or intellectual property rules, reducing the individual’s rights ‘into an empty shell’. It also thwarts the public interest of understanding the impact of internal business decision making on groups of people and society generally.
The atomising instinct has an unlikely ally in modern human rights law. The conservative authors of the European Convention on Human Rights with its exclusion of social or collective rights believed they were establishing a bulwark against Bolshevism. The GDPR, one of the Convention’s most famous recent progenies, aims to safeguardthe interests of the individual whose data is processed by others; but it does not – it cannot – address the social, democratic and environmental externalities ensuing from the over-collection and misuse of data pertaining to the multitude. A recital muses on how ‘processing of personal data should be designed to serve mankind’ and that the right to data protection ‘must be considered in relation to its function in society’; but the law provides no immediate means for collective action. It opens a path for any individual to mandate an advocacy group to take up her cause, but only if her Member State so permits. In debates over state and federal level privacy laws in the United States, the inclusion of a private right to action is a major issue, as it would activate the potential for class actions against companies’ alleged violations, without having to wait for action by the Attorney General’s office.
Everyone now works for Amazon
Luigi Zingales of the University of Chicago often proposes a thought experiment to test the limits of antitrust in safeguarding democracy: ‘What if everyone worked for Amazon?’ Big tech’s populations of daily users surpass most countries in the world, and those using Facebook even outnumber the population of China. This is significant because the company’s actions, like the modification of algorithm or the expulsion of a misbehaving account holder, can have an impact broader and deeper than that of a state. Through Amazon’s Mechanical Turk and Google’s ubiquitous reCAPTCHA, individuals are already providing free labour to train the proprietary AI systems. Facebook employs on relatively low wages thousands of ‘content moderators’ to clear up the mess created by the company’s own business model, often with traumatic consequences for the staff concerned. Meanwhile, a snapshot of ongoing attempts to regulate and enforce compliance by the biggest tech companies can illustrate the current, costly impasse faced by public authorities around the world.
A Brussels conference on 8 December 2020 assembled the heads of several national competition agencies, plus the director-general for competition at the European Commission, and the only topic for discussion was what to do about Facebook, Google and Amazon. Each had a different story to tell, but with a recurring thread. The day after the discussion, the Federal Trade Commission and 48 state attorneys general filed separate complaints against Facebook alleging maintenance of illegal monopoly and anticompetitive conduct. This fell between two major filings against Google: in October, the US Department of Justice with (initially) 11 state attorneys filed allegations the company maintained unlawful monopolies in search and search advertising; in December, 10 states filed alleging exploitation of all sides of the adtech ecosystem to the detriment of publishers, advertisers and consumers, as well as collusion with Facebook to access communications over WhatsApp. All lawsuits call for structural remedies.
Meanwhile, the Bundeskartellamt’s decision in February 2019 that Facebook should cease combining personal data from different sources remained unimplemented, despite its being upheld by the Federal Court of Justice, because the company had lodged with the regional court an emergency appeal to suspend. Google’s acquisition of Fitbit was, according to reports around the time of the December conference, on the verge of being cleared by the European Commission, but the Australian Competition and Consumer Commission two weeks after the conference announced that it was not satisfied with the proposed remedies. On the regulatory front, the UK government had announced that the Competition and Markets Authority would include a Digital Markets Unit to enforce ‘a new code to govern the behaviour of platforms that currently dominate the market, such as Google and Facebook’; and the European Commission was about to unveil proposals governing large online platform now referred to as ‘gatekeepers’.
In the data protection arena, Facebook continued to refuse to amend business practices despite the CJEU judgments, in the two cases brought by Max Schrems, that the United States could not be deemed to provide adequate protections for personal data transferred from the EU. Facebook launched a PR drive in September at the same time as filing for a judicial review in the Irish high court to prevent the Data Protection Commissioner from proceeding with enforcement. Google was pursuing a challenge to a €600 000 fine imposed by the Belgian data protection authority for violations of the GDPR’s ‘right to be forgotten’.
The power of these few companies is such that the state now devotes to their control a sizeable and increasing proportion of its available resources. Whether in the form of writs submitted to the courts or draft bills tabled before parliaments, law is seen as the mechanism to bend private powerful actors to the will of the public democratic state. Behind each court filing and proposed regulation lies hours of hard toil of hundreds of officials and lawyers and clerical staff and senior managers. Moreover, politics and legislation depends on politicians, who in turn depend on sympathetic publishers. Almost all publishers are now ‘at best…. sharecroppers on Facebook’s massive industrial farm.’ Facebook could choose to adjust its algorithms to manipulate traffic, ad network readers to harm publishers. Politicians also depend on email, and recent evidence suggests Google’s gmail platform has arbitrarily promoted or disappeared candidates for public office. So long as such a possibility exists for a communication monopoly secretly to privilege or demote political speech, it is logical to assume a chilling effect on politicians.
In a sense, therefore, everyone is already working for or against Amazon and its ilk, and sometimes doing both at the same time.
4. Rule of law agonistes
The rule of law is relevant wherever power presents itself. The previous section discussed how the power of a handful of multinational corporations is rivalling that of sovereign states. In this section, I will suggest how, in three distinct ways, the power and behaviour of these companies undermines rule of law. First, they are complicit in a concentration of wealth that exacerbates inequality and in turn equality before the law. Second, they benefit from a tendency towards legalism in key areas of law that could otherwise restrain their activities, with the result of delays in and often denial of justice. Third, their increasing indispensability for certain public functions rightfully pertaining to the state, and the tendency towards laws that legitimise their power by giving them special responsibilities but without the means for enforcement, seem to allow them to act with impunity, as if they are above the law.
1) Unequal before the law
Equal access to justice is a target of the UN’s Sustainable Development Goals. The notion is understood by the World Justice Project to mean people’s ability to defend or enforce their rights, or to obtain a just resolution of their justiciable problems. The civil society organisation calculates that two-thirds of the world’s population, 5.1 billion people, cannot obtain justice for everyday problems, are excluded from the opportunities the law provides, or live in extreme conditions of injustice. Such areas of the Global South have proven very attractive to digital colonialists, monopoly is a proxy for this inequality, as described in the first Digital Economy Report from the United Nations Conference on Trade and Development. The report noted that by far the greater share of the wealth, intellectual property, and the seven most valuable companies (Microsoft, Apple, Amazon, Google, Facebook, Tencent and Alibaba), were concentrated in the United States and China, and that uneven digitisation has had the effect of exacerbating inequalities within and between countries.
Inequality concerns not only material possessions, but also social capital like available support networks, education and knowledge – what Pierre Bourdieu termed ‘cultural capital’. These assets determine a person’s ability to enforce her legal rights. The Covid-19 pandemic is widening inequalities: large companies have seen sharp rises in profits yet still laid off low wage workers; home schooling has resulted in children of poorer families falling further behind; global stock levels increased by trillions, but half of it has accrued to the richest 1%; tens of thousands of small business permanently closed. The richest men in the world, including the CEOs of Amazon and Facebook, have reaped a $400 billion dividend during the crisis.
Passing laws to control these powerful players seems easy when compared with the question of how the laws should be enforced. The effectiveness of a law in practice depends on relative power, not only procedure. As has been seen, the largest companies have more resources than both competitors and enforcers to attract the best legal talent. Their resources even exceed those available to most sovereign states. These resources are deployed to contest everything, from jurisdiction to definitions to balancing the interests of the vulnerable with deep-rooted commercial rights, while courts systems creak under excessive caseloads and resource constraints.
Even in more affluent nations, poverty and lack of education are at least as insurmountable a barrier as the absence of systems for rights protection that are susceptible to being instrumentalised for commercial interests and power consolidation. For most people affected by big tech business practices, however, the protection of law remains a dead letter. They are priced out of embarking on the ‘path to justice’ by several factors: the cost of legal fees; opportunity costs such as loss of personal time and foregone earnings while seeking out information and filling out forms; and less tangible costs, such as stress, mental health and privacy. Powerful commercial litigators may even view legal costs less as overheads or business risks and more as investments to yield future revenue in future. By contrast, private legal costs put most people off seeking justice at all, and the growing ‘social costs’ to the state of running courts and dispute mechanisms to adjudicate the claims fall to the taxpayer. A common strategy for the tech giants, as explained above, is delay – lucrative time for their contested business models to continue to generate revenue. Justice delayed adds to the cost of justice for private individuals, the agencies or organisations representing them, and for the state providing the justice infrastructure. Ironically but inevitably, tech-solutionism reappears offering fixes for problems in part caused by digital inequalities, such as in the form of access-to-justice mobile apps.
A comparison, though not possible here, of the efficacy of data protection rights and copyright could be salutary. Individuals and enforcers alike have consistently struggled to establish under the GDPR what data big companies have on their servers, whether their data practices are compliant, how to force them to change their practices and verify that they have done so. The major copyright holders, by contrast, have had the political and economic means to pursue platforms through the courts for years, with the result that both sides agree on rigorous self-regulation by the platforms. A typical victim of a privacy violation is in no position to bargain in this way.
2) Legalism over justice
Robert Bork is an unexpected villain in Homeland Elegies, Ayad Akhtar’s acclaimed novel-cum-autobiography. Akhtar chronicles through the eyes of a Pakistani immigrant family the decline of a vibrant America of small businesses into ‘corporate autocracy’ and a national mood ‘nasty, brutish, and nihilistic’. One passage of the book excoriates Bork, author of The Antitrust Paradox and President Reagan’s favourite judge, as ‘the Robespierre of the consumerist antitrust movement’ whose ‘notion that the collective good was determined solely by benefit to the consumer would prove to be the necessary lubricant in the world-historical shift to the form of free-market capitalism that has engulfed the planet.’ The consequences were felt particularly acutely by black-owned businesses.
Bork took the reductivism of the Chicago School out of academia and into the Supreme Court and the White House, so that economic efficiency – pegged to low prices and high output – became conveniently coterminous with the transfer of wealth to monopolies. It was a turn towards legalism, the ‘calculating machine’ that clings to the letter while disdaining the spirit of the law. It has enabled powerful companies to bamboozle enforcers and indefinitely defer justice. Big tech in particular has thrived in this environment, achieving dominant shares in markets by offering what appear to be free services. Consumers of these services are emasculated by take-it-or-leave-it (they almost always take) terms of service providing the companies with the legal latitude to do whatever they please with personal data and to shut out competitors. It transfers the onus onto individuals to protect their own rights and interests.
3) Above the law
Enforcement regimes cannot deliver accountability if the entities subject to enforcement can turn fines into a management business risk in company forecasts. The UK ICO in late 2020 fined Marriott £18.4 million for a 2014 breach affecting 339 million guest records – equivalent to 5 pence per affected customer or 0.6% of annual global turnover. The sanction had first been touted as £99m in 2019 (so around 25p per violation). As with the Federal Trade Commission settlement with Facebook on the 2011 consent order, a part of the deal was not to accept liability. Google confirmed its acquisition of Fitbit, disdaining the fact that the US Department of Justice had not even concluded its review of the merger, and that Australia’s competition authority had rejected it – the most the authority coulc now do is to impose a fine of around 400m USD. Law seems powerless to change the structure of incentives. Regulation that genuinely takes aim at the business models elicits threats to withdraw services altogether, as Google is currently attempting in Australia, the formula having succeeded in recent years in Spain and Germany.
Control over the digital environment and infrastructure gives the big tech companies power to enact business standards, and to govern the public sphere, consumers and workers. This usually leads policymakers to see them as a necessary part of the solution. Recent years have included attempts to make up for lost time with complex rules governing businesses in the digital economy. Some such rules, like the GDPR, apply equally across the board; others aim to create ‘asymmetric obligations’, where requirements become more onerous the more powerful the entity.
The big platforms are recruited to tackle copyright violations, terrorism and child abuse. Lucrative contracts are offered by law enforcement to deploy surveillance technology (from ‘smart’ door bells to social media trackers) developed for commercial purposes, which is always directed at the more vulnerable and disenfranchised – the poor, immigrants, people of colour. Microsoft’s ‘Death Star’ was indispensable to wiping out the malware on its ubiquitous operating systems in the wake of the SolarWinds hack in December 2020, as in past combating botnots and Russian election interference. In the early weeks of the pandemic when policy officials, entrepreneurs and coders were scrambling for digital solutions and had just began to coalesce around the idea of mobile contact tracing apps, Apple and Google jointly announced changes to the iOS and Android operating systems that would support one of the several competing protocols. The move was justified – creditably – on grounds of privacy and public interest, and probably spiked some of the more lunatic surveillance ideas in circulation, but its effect was to close off policy options to public health authorities. Some authorities complained, but were ultimately powerless, because governments had allowed the market for mobile operating systems to become a duopoly.
The delegation of certain democratic state functions to private actors is in principle wholly compatible with the rule of law. Accountability, however, cannot be delegated. Conferring additional responsibilities on powerful entities, whether through delegation or regulation, itself legitimises and entrenches their power in society, as if they had been granted a royal patent to guarantee security and facilitate communications and trade between everyone on the planet. The risk of such a symbiotic relationship is a complacent, overleveraged state ever less able or willing to enforce laws that are inconvenient to the monopolies it depends on. Reluctance to dismantle foreign monopolies grows for fear such action might choke the potential for more pliable domestic champions to achieve monopoly status.
The risk is therefore the replacement of the rule of law with the rule of men. The Digital Switzerlands that the tech giants may aspire to are in fact highly secretive corporate feudal states capable, in some instances, of summary dismissals of employees for attempting to unionise, or for exposing uncomfortable truths about treatment of people of colour. The inner workings of their business practices remain hidden despite their social consequences: non-disclosure agreements and non-compete clauses are the singular staple of Silicon Valley lawyers’ diet. Researchers and journalists are generally barred from interrogating why certain content is displayed to one individual but not to another, and privacy is cited to justify the opacity. They are even able to gag the public authorities charged with their supervision. The UK’s Information Commissioner was unable to answer an inquiry from a parliamentary committee as to whether a public commitment to an app audit, made by Zuckerberg himself before the US Congress in 2018, because it would have breached a ‘private agreement’ between regulator and regulated company.
Current trajectories are unlikely to see much change very soon. If enforcers continue to shirk litigation and permit mergers for fear of costly defeats, these companies can only get bigger, and the inequality of arms starker.
5. Binding Leviathan
Rosalyn Higgins, considering the rigid state/subject and object/individual dichotomy in international law, wrote that ‘We have erected an intellectual prison of our own choosing, and then declared it to be an unalterable constraint.’ The same might be said of the limitation of the rule of law to a constraint on ‘state’ power. Power is diffuse, but it is particularly concentrated at the moment in a few privately-owned entities based in China and the west coast of the United States who are, in the words of EU Commissioner Thierry Breton, ‘too big to care’. The sad reality is that some of the largest corporations in the world are now entangled – even if unintentionally – in the incitement of violence and racism and the abuse of human rights, potentially on a scale not seen since the Second World War. Their control over communications infrastructure gives them greater sway in shaping public opinion than even Big Oil, whose misinformation and lack of accountability bear a large share of responsibility for the climate emergency. Laws are required and are now being discussed, and nation states are looking for ways to return to the public purse a fairer chunk of big tech’s huge revenues. But laws and taxes alone cannot preserve the rule of law without other addressing the interlocking questions of power, compliance and justice through mechanisms to constrain digital corporate power. This section suggests three such mechanisms.
1) Private power and democracy
A potential tension within the concept of the rule of law has been magnified by the Borkian mindset between equality before the law and guaranteeing respect for private property. Ample evidence has been assembled that the private fortunes of the big tech companies has not always been fairly or lawfully amassed. Even without such evidence, the case remains for structural intervention to redress today’s serious imbalance of power between private monopoly and public democracy. But the objection is raised that breaking up monopolies operating according to toxic business models will simply result in several smaller companies with toxic business models. Were power evenly distributed, society could expect to tolerate and control objectionable business models; but experience suggests this is impossible when they are operated by entities powerful enough to evade, or shape to their own benefit, attempts to regulate and enforce. The argument that it is better to have a corporate monopoly than an overbearing state is similarly flawed, because a well-functioning democratic state, unlike the monopoly, is not a monolithic unity with a single operating imperative, and operates within constitutional checks and balances.
Authoritarian states solve the problem by maintaining a trapdoor through which corporate executives can be disappeared the moment they become a threat to the ruling regime. Only recently, China’s most famous CEO, Jack Ma, vanished at the time as the regime was issuing its ‘special plan’ for the rule of law, advancing comprehensive legislation governing cyberspace, updating its anti-monopoly rules and opening an antitrust investigation into Alibaba. It is a reminder that mafia-style abuses of the rule of law can and do co-exist alongside sophisticated governance frameworks.
Democracies need to find a just means of redressing the imbalance. If a company executive feels able, for instance, to decline an invitation even to answer questions before a committee of elected representatives, then that company should forfeit the right to trade in that market. They should not be able to treat elected representatives and government ministers as equals and be permitted to enter into secret settlements under cover of non-disclosure agreements. When a monopoly can threaten to withdraw a service because it objects to democratic deliberations on a new law, that signals that the balance of power has tipped dangerously towards private interests. The basic principle ought to be that no private entity should be so powerful that it can rival the authority of a democratic state.
2) Democratising the digital public sphere
The ideal public sphere, as conceived by Habermas, needs to be free from interference by both state and by private business, to allow for a critical consensus and public participation in the democratic process. Policymakers are now alive to the reality that the digital public sphere, once so promising, has been progressively privatised on their watch. The job of safeguarding freedom of expression, privacy and other human rights, and of the free exchange of ideas on which healthy democracy depends, cannot be consigned to the changing whims and tweaked algorithms of a private entity.
It has become a cliché of public policy that whatever is illegal online should be illegal online also. While this usually prefigures interventions to outlaw posting of various types of harmful content, it could also be a basis for outlawing the whole of surveillance capitalism. Online behavioural advertising and real time bidding are the online equivalent of stalking. ‘Smart speakers’ are the online equivalent of a nested spy, akin to the ‘Pair Up and Become Family’ policy in Xinjiang, where Uighurs are forced to host a Han Chinese Communist Party official in their homes. ‘Privacy,’ wrote Hildebrandt, ‘is also a public good that concerns a citizen’s freedom from unreasonable constraints on the construction of her identity. This freedom is a precondition for democracy and rule of law.’ Shoshana Zuboff calls this ‘sanctuary’, an inviolate space on her own or in the company of discreet companions, that ought to be the birth right of any individual sentient creature. danah boyd calls this control of one’s social situation. Being online may be like being in someone’s home, or in a public park. In neither offline scenario is a person presented with a choice between total anonymity or ‘consent’ to universal transparency. However, this is precisely the binary foisted on individuals once they are online.
Profit seeking is morally neutral, and companies are rightly obliged to deliver value to their shareholders. Nevertheless, as Woodrow Wilson said, in his ‘The New Freedom’ platform for the 1912 presidential elections, a modern company ‘cannot in any proper sense be said to base its rights and powers upon the principles of private property. Its powers are wholly derived from legislation. It possesses them for the convenience of business at the sufferance of the public.’ By this logic, it should not be acceptable for control over global communications channels to be based solely on the principles of private property.
Moreover, it should neither be acceptable for them to profit from practices that compromise core values of a democratic state, and not simply individual rights, through stalking people around the Internet, the amplification of inflammatory and conspiratorial content and the facilitating the mobilisation of armed militias. The comparison of monopoly intermediaries to weapons suppliers is unfair, because a weapon has one sole purpose, which is to cause harm. But the rule of law is not served by an incentive structure that rewards profiting from conspiracy and insurrection. At its simplest, it may require a simple prohibition on the sale of targeted advertising off the back of private individual behaviour online. Alternatively, platforms could be made liable for content they host if they profit from advertising. At the least, however, the company cannot be allowed to escape liability for negligence by denying knowledge and purpose; they should be required to explain their business practices and return revenue generated to society.
3) Bootstraps for the bootless
The rule of law cannot be abstracted from deep racial, gender and wealth inequalities that work against the principle of equality before the law. UNCTAD’s 2019 report on the global digital economy described how tech monopoly power is both a symptom and a cause of these inequalities. These inequalities are growing and expected to continue to grow for several years due to the pandemic. Martin Luther King in one of his last interviews said granting emancipation without an economic base was tantamount to ‘freedom to hunger’, and that it was absurd to tell a freed slave to pull himself up by his bootstraps when he had no boots to wear. More recently, the experience of post-Apartheid South Africa has demonstrated that constitutional human rights overlaid onto existing, deep-rooted social injustices cannot prevent rife corruption, violence and maladministration.
For the rule of law to endure, therefore, in the face of rapid digitisation often on terms determined by large multinational companies, progressive laws are not enough. Remedial laws need to be accompanied by effective social justice policies on education and health care. Emerging from a pandemic that has both exacerbated inequality and accelerated the pace of digitisation may provide an opportunity and the appetite for such policy interventions.
This article has advanced the thesis that private monopoly power is again threatening the rule of law, and I have offered only a tentative analysis.
The most valuable companies in the world have successfully exploited the opportunities of global digitisation to accumulate an unprecedented degree of political and economic power. The threat consists in the combination of this power with business models of surveillance and manipulation on a massive scale, resulting in societal and individual harm that is increasingly difficult to remedy due to unequal access to justice. I have argued that these companies are remarkable shape-shifters, grooming policymakers while they hammer enforcers and deplete resources in the justice system, contesting every attempt to hold them to account. They succeed in diverting political narratives away from their business models and power, and towards how the state can co-opt them to perform state functions, including alleviating problems arising from their own actions. This risks damaging faith in law and democracy.
The impact is global, and governments around the world are seeking to reassert the sovereignty of the state through new forms of regulation and antitrust enforcement and, in the case of authoritarian states, through intimidation and coercion. Big is not necessarily bad and often is a force for good. Success as such should not be penalized, and technology should be a force for progress. But when a private entity becomes so powerful that it can resist, or bend to its advantage, all attempts to curtail its most egregious behaviour, it is in effect above the law. If the rule of law is no more than the consistent application of existing rules while inequalities in power and wealth worsen, then it is likely only to perpetuate existing injustices, especially a post-pandemic world of rising temperatures. It would mean a prioritisation of order over justice. That, to return to the wisdom of King,  may be the biggest stumbling block on the way to dismantling unwieldy power concentrations in order to build back better.
 The views expressed in this article are entirely the personal views of the author and do not represent those of his current or previous employers.
 This is work in progress. Julie Cohen writes extensively and cogently on the nature of platforms and the adaptation of governance; see Cohen, J.E., Between Truth and Power, 2019 and Law for the Platform Economy, 51 U.C. Davis L. Rev. 133-2014 (2017). Mireille Hildebrandt has examined sovereignty, safety, freedom and human rights in cyberspace and the challenge to the rule of law posed by automated profiling and other techniques in intelligent networked environments; e.g. Hildebrandt, M., Profiling and the rule of law, IDIS 1, 2008, pp. 55–70, and Extraterritorial jurisdiction to enforce in cyberspace? Bodin, Schmitt, Grotius in cyberspace, Mireille Hildebrandt, University of Toronto Law Journal, 2013 63:2, pp. 196-224. Beth Stephens is an authority on the accountability of companies for complicity in human rights abuses; e.g. Stephens, B., The Amorality of Profit: Transnational Corporations and Human Rights, 20 Berkeley Journal International Law 45(2002). Douwe Korff wrote a useful paper for the Council of Europe in 2014 on ‘The Rule of law on the internet and in the wider digital world’ (https://rm.coe.int/16806da51c). Sally Hubbard, Lina Khan, Barry Lynn and Matt Stoller have exposed the nature of corporate power and reminded us of the purpose of antitrust. Shoshana Zuboff’s examination of ‘surveillance capitalism’ has brought political attention to intrusive business models. I am not aware of extensive scholarship on the implications of the power and behaviour of global digital monopolies for the rule of law.
 Press release (English) http://www.xinhuanet.com/english/2021-01/10/c_139656578.htm; text (Chinese) http://paper.people.com.cn/rmrb/html/2021-01/11/nw.D110000renmrb_20210111_1-01.htm
 The result is a ‘regulation on a general regime of conditionality for the protection of the Union budget’ applicable from 1 January 2021 that allowed the Commission to cut funding to a Member State’s beneficiaries if it can establish ‘that breaches of the principles of the rule of law in a Member State affect or seriously risk affecting the principles of sound financial management of the EU budget or the protection of the financial interests of the Union in a sufficiently direct way;’ Article 4 of text agreed by the European Parliament and Council https://data.consilium.europa.eu/doc/document/ST-9980-2020-INIT/en/pdf
 World Justice Project, rule of law Index 2020; https://worldjusticeproject.org/sites/default/files/documents/WJP-ROLI-2020-Online_0.pdf
 Hobbes, T., Leviathan, R. Tuck (ed.), 1991 , p. 184.
 Brihadaranyaka Upanishad, dated to 9th-6th century BCE and possibly the earliest Hindu scripture, states ‘Law [Dharma] is that which is the king of kings; nothing is superior to law. The law aided by the power of the king enables the weak to prevail over the strong;’ cited in Justice B.N. Srikrishna Pre-British Human Rights Jurisprudence, Lecture delivered at West Bengal National University of Juridical Sciences, 2010; http://docs.manupatra.in/newsline/articles/Upload/E073A7FF-96B6-481E-866B-16025B2BF2B7.pdf
 From Paine’s 1776 pamphlet, Common Sense; Addressed to the Inhabitants of America, on the Following Interesting Subjects.
 Article 49 Treaty on the European Union.
 The rule of law is not defined. In Europe it is generally considered correspondent with terms in other languages, eg état de Droit, Rechtstaat, praworządność etc.
 A clear thread runs from Clause 39 of Magna Carta – ‘No free man shall be seized or imprisoned, or stripped of his rights or possessions, or outlawed or exiled, or deprived of his standing in any way, nor will we proceed with force against him, or send others to do so, except by the lawful judgment of his equals or by the law of the land’ – through John Locke – ‘The Supream Power cannot take from any Man any part of his Property without his own consent, and any law that purports to do so is of no validity (The Second Treatise on Civil Government, 1689, §138) to the Fifth Amendment to the US Constitution – ‘no person shall be deprived of life, liberty, or property, without due process of law’.
 The rule of law and transitional justice in conflict and post-conflict societies: Report of the Secretary-General of the United Nations, 2004.
 Joseph Raz insisted that they be kept distinct: ‘If the rule of law is the rule of the good law [of democracy and human rights etc.] then to explain its nature is to propound a complete social philosophy. But if so the term lacks any useful function;’ ‘The rule of law and its Virtue’, The Law Quarterly Review,1977. Clustering is clearly favoured, however. ECHR Preamble recalls that the ‘common heritage of political traditions, ideals, freedom and the rule of law’ shared by governments of ‘like-minded’ European countries. The European Commission in 2014 said: ‘Respect for the rule of law is intrinsically linked to respect for democracy and for fundamental rights: there can be no democracy and respect for fundamental rights without respect for the rule of law and vice versa. Fundamental rights are effective only if they are justiciable. Democracy is protected if the fundamental role of the judiciary, including constitutional courts, can ensure freedom of expression, freedom of assembly and respect of the rules governing the political and electoral process’; A new EU Framework to strengthen the Rule of Law. Similarly in 2020: ‘The rule of law is enshrined in Article 2 of the Treaty on European Union as one of the common values for all Member States. Under the rule of law, all public powers always act within the constraints set out by law, in accordance with the values of democracy and fundamental rights, and under the control of independent and impartial courts. The rule of law includes principles such as legality, implying a transparent, accountable, democratic and pluralistic process for enacting laws; legal certainty; prohibiting the arbitrary exercise of executive power; effective judicial protection by independent and impartial courts, effective judicial review including respect for fundamental rights; separation of powers; and equality before the law;’ European Commission, 2020 Rule of Law Report: The rule of law situation in the European Union, 2020.
 Lynn argues that the work of building democracy and other rights could only begin after the arbitrary power of the king to grant licenses to monopolies was subjected to law and institutions; Liberty from All Masters, pp.90-95.
 Hildebrandt, ‘Profiling and the rule of law’, p.61.
 ‘Stripped of all technicalities, [the rule of law] means that government in all its actions is bound by rules fixed and announced beforehand—rules which make it possible to foresee with fair certainty how the authority will use its coercive powers in given circumstances and to plan one’s individual affairs on the basis of this;’ Hayek, F, The Road to Serfdom, Chicago, 1944, p.80.
 Derrida, J., Force of Law: The Mystical Foundation of Authority, in Deconstruction and the Possibility of Justice,edited by Cornell, D. and Rosenfeld, M.,1992, pp.24-29.
 Derrida, in an apparent misreading, attributes ‘the instant of a decision is madness’ to Kierkegaard. Robert Cover addressed how judges should rule on the basis of unjust or oppressive laws in the context of slavery in United States in the 1800s; Cover, R., Justice Accused: Slavery and the Judicial Process, 1975.
https://www.researchgate.net/publication/275498214_The_Rule_of_Law_against_the_Rule_of_Greed_Edmund_Burke_against_the_East_India_Company. Cf. St Augustine of Hippo: ‘Justice being taken away, then, what are kingdoms but great robberies? For what are robberies themselves, but little kingdoms?’ The City of God, Book IV, Chapter 4.
 United States v. Aluminum Company of America 148 F.2d 416 (2d Cir. 1945)
 Stoller, Goliath: The 100-year way between Monopoly Power and Democracy, 2019; Muller, C., The Aluminum Monopoly and the War. Political Science Quarterly, 60(1), 1945, pp. 14-43.
 Levy, H., Industrial Germany: A Study of Its Monopoly Organisations and Their Control by the State Levy, 1966, p. 159.
 Daniel A. Crane, Fascism and Monopoly, 118 MICH.L. REV.1315 (2020).
 Djelic., M-L. and Quack, S., Adaptation, Recombination and Reinforcement: The Story of Antitrust and Competition Law in Germany And Europe, Sciences Po publications, 2005.
 See, for example, Besson, S., The Extraterritoriality of the European Convention on Human Rights: Why Human Rights Depend on Jurisdiction and What Jurisdiction Amounts to, Leiden Journal of International Law, 2012, 25 (4), pp.857–884; Wilde, R., Human Rights Beyond Borders at the World Court: The Significance of the International Court of Justice’s Jurisprudence on the Extraterritorial Application of International Human Rights Law Treaties, Chinese Journal of International Law, Volume 12, Issue 4, December 2013, pp.639‑677.
 UN Human Rights Office of the High Commissioner, Guiding Principles on Business and Human Rights: Implementing the United Nations “Protect, Respect and Remedy” Framework, 2011; https://www.ohchr.org/documents/publications/guidingprinciplesbusinesshr_en.pdf
 Article 25 of the Rome Statute of the International Criminal Court. Domino, J., Crime as Cognitive Constraint: Facebook’s Role in Myanmar’s Incitement Landscape and the Promise of International Tort Liability (August 10, 2019). 52 Case W. Res. J. Int’l L. 143 (2020), pp173-8.
 Congressional Record: Proceedings and Debates of the Forty-Ninth Congress, Second Session, Volume 18, Part 1, 816; see discussion in Lynn, B., Liberty from All Masters, 2020.
 Posner, R., Economic analysis of law, 1986, p.25.
 Cohen, Law for the Platform Economy, p.135.
 Cohen, J. E., Cyberspace As/And Space, Georgetown Law Faculty Publications and Other Works, 807, 2007.
 Zuckerberg once wanted to sanction Trump. Then Facebook wrote rules that accommodated him, Washington Post, 29.06.2020.
 Pickard, V., The Return of the Nervous Liberals: Market Fundamentalism, Policy Failure, and Recurring Journalism Crises, The Communication Review, 18:2, 2015, pp. 82-97.
 https://blogs.microsoft.com/uploads/2017/03/Transcript-of-Brad-Smiths-Keynote-Address-at-the-RSA-Conference-2017.pdf; Eichensehr, K.E., Digital Switzerlands, University of Pennsylvania Law Review, Vol. 167, pp. 665-732.
Diresta, R., ‘Free Speech is not the same as free reach,’ Wired , 30.08.2018.
 Naughton, J, Guardian, 30.01.2021.
 Eichensehr, Digital Switzerlands, p. 681.
 Birhane, A., The Algorithmic Colonization of Africa, Real Life, 18.07.2019. On Facebook’s rebuffed attempts to offering a limited Internet in a number of countries with poor digital infrastructure in the Global South, conditional on providing their personal data by logging into their account, see LaFrance, A., Facebook and the New Colonialism, The Atlantic, 11.02.2016; and Thorat, D., Digital Infrastructures and Technoutopian Fantasies: The Colonial Roots of Technology Aid in the Global South, in Exploring Digital Humanities in India Pedagogies, Practices, and Institutional Possibilities, ed. Dodd, M. and Kalra, N, 2020.
 ‘If China could become a world power without a free Internet, why do African countries need a free internet?, one African leader is reported to have asked; Umejei. E, The imitation game: will China’s investments reshape Africa’s internet?, Power 3.0, 06.12.2018. Hawkins, A., ‘Beijing’s Big Brother tech needs African faces’, Foreign Policy, 24.07.2018.
 Greenleaf, Graham, G20 Makes Declaration of ‘Data Free Flow With Trust’: Support and Dissent, Privacy Laws & Business International Report, (2019) 160, pp. 18-19.
 Ribeiro, M.H., Ottoni, R., West, R., Almeida, V.A.F., and Meira, W., Auditing radicalization pathways on YouTube, Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency, 2020;
Munger, K., & Phillips, J. Right-Wing YouTube: A Supply and Demand Perspective, International Journal of Press/Politics, 2020.
 Facebook in particular had been to determined ‘to contort, hair-split and reimagine its rules to make sure nothing Trump posted would fall too far outside them;’ Oremus, W,, https://onezero.medium.com/facebook-chucked-its-own-rulebook-to-ban-trump-ecc036947f5d.
 Nyabola, N., Digital Democracy, Analogue Politics: How the Internet Era is Transforming Kenya, 2018; Klonick, K., The New Governors: The People, Rules, and Processes Governing Online Speech, 131 HARV. L. REV. 1598, 1603 (2018).
 Steinitz, M., The Case for an International Court of Civil Justice, 2019, pp79-80.
 Nyabola, Digital Democracy, p200-1.
 Duhigg., C., How Venture Capitalists Are Deforming Capitalism, The New Yorker, 23.11.2020; A ‘Dirty and Open Secret’: Can Social Media Curb Fake Followers?, Knowledge@Wharton 02.02.2018, https://knowledge.wharton.upenn.edu/article/twitter-and-the-bots/. Diana Moss calculated a total of 723 acquisitions by Google, Microsoft, Apple, Amazon and Facebook from 1987-2019; Moss, D.L., The Record of Weak U.S. Merger Enforcementin Big Tech, American Antitrust Institute, 08.07.2019. A similar trend is observed with Chinese big tech; Blown off course: China takes aim at its entrepreneurs, The Economist, 12.11.2020.
 Staff of H. Comm. n the Judiciary, 116th Cong., Investigation of competition of digital markets: majority staff report and recommendations, 39 (Comm. Print 2020), pp.149-160.
 Cohen, Law for the Platform Economy, p.39.
 For example, the Consumer Rights Directive (EU) 2019/216; https://ec.europa.eu/commission/presscorner/detail/en/IP_20_1187
 Edwards, L., and Veale, M., Slave to the Algorithm? Why a ‘Right to an Explanation’ Is Probably Not the Remedy You Are Looking For, 16 Duke Law & Technology Review 18, 2017, pp.18-84.
 Nissenbaum, Helen F., A Contextual Approach to Privacy Online, Daedalus 140 (4), Fall 2011, pp. 32-48.
 Cohen, Law for the Platform Economy, p.199.
 https://www.lepoint.fr/high-tech-internet/exclusif-comment-google-veut-faire-plier-bruxelles-28-10-2020-2398468_47.php; https://www.salon.com/2015/11/24/googles_insidious_shadow_lobbying_how_the_internet_giant_is_bankrolling_friendly_academics_and_skirting_federal_investigations/.
 Based on 2018 figures (excluding Cyprus, for whom figures were not available, and UK); Deloitte, Report on EU Data Protection Authorities, 2019.
 Federal Trade Commission, Agency Financial Report 2020.
 Galanter, M., Why the “Haves” Come out Ahead: Speculations on the Limits of Legal Change, Law & Society Review, 9(1), 1974, pp.97-8.
 The maximum civil penalty was $41 000 per violation; an estimated 181 million Americans were affected; https://www.washingtonpost.com/news/the-switch/wp/2018/04/09/how-big-could-facebooks-fine-theoretically-get-heres-a-hint-there-are-four-commas-and-counting/?noredirect=on
https://www.nytimes.com/2021/01/04/technology/google-employees-union.html?auth=login-email&login=email ; on the power and information disparities between big platforms and workers, see Calo, R., and Rosenblat, A., The Taking Economy: Uber, Information, and Power, Columbia Law Review, Vol. 117, 2017.
 Lynn, B., America Can Still Achieve Its Techno-Utopian Dream, Wired, 29.09.2020.
 Hildebrandt, Profiling and the rule of law, pp.12-13.
 Sedley, S., Be careful what you wish for, London Review of Books Vol. 40 No.16, 2018.
 Recital 4, Article 80 GDPR. This was confirmed by CJEU in Schrems v. Facebook Ireland Limited, judgment of 25 January 2018, C-498/16, EU:C:2018:37, in ruling that EU law did not permit a consumer to bring before her own national courts the claims of other consumers in her or another Member State. A new collective redress directive has been recently agreed that promises to remedy the situation; https://ec.europa.eu/commission/presscorner/detail/en/statement_20_1227 .
 https://scholarship.law.upenn.edu/cgi/viewcontent.cgi?article=9654&context=penn_law_review p686. ‘Misinformation dropped dramatically the week after Twitter banned Trump and some allies’; https://www.washingtonpost.com/technology/2021/01/16/misinformation-trump-twitter/
 ‘ Imagine if the financial markets are controlled by one monopoly company, say Goldman Sachs, and that company then owns the NYSE, which is the largest financial exchange, that then trades on that exchange to advantage itself, eliminate competition, and charge a monopoly tax on billions of daily transactions. That is the world of online display advertising today;’ https://www.texasattorneygeneral.gov/sites/default/files/images/admin/2020/Press/20201216%20COMPLAINT_REDACTED.pdf
 https://www.accc.gov.au/media-release/accc-rejects-google-behavioural-undertakings-for-fitbit-acquisition The merger was completed anyway on 14 January 2021.
 European Commission, Proposal for a regulation on contestable and fair markets in the digital sector (Digital Markets Act), COM/2020/842 final. The term gatekeepers or gateways has been in circulation for several years eg Lynskey, O., Regulating ‘Platform Power’, LSE Legal Studies Working Paper No. 1/2017; Acquisti A., Taylor C., Wagman L., The Economics of Privacy, 8 March 2016, Sloan Foundation Economics Research Paper No. 2580411, p. 3; European Data Protection Supervisor, Opinion 8/2016, Opinion on coherent enforcement of fundamental rights in the age of big data, 2016, p.8.
 Thompson, N. and Vogelstein, F., ‘Inside the two years that shook Facebook – and the world’, Wired, 2.12.2018.
 World Justice Project, Measuring the Justice Gap: People-Centered Assessment of Unmet Justice Needs Around the World, 2019.
 UNCTAD, Digital Economy Report, 2019 : Value Creation and Capture: Implications for Developing Countries.
 Bourdieu, P. The forms of capital, in Richardson, J., Handbook of Theory and Research for the Sociology of Education, 1986.
 The World in 2021:Covid-19 leaves a legacy of increased inequality, The Economist, 17.11.2020; ‘America’s biggest companies are flourishing during the pandemic and putting thousands of people out of work’, Washington Post, 16.12.2020; https://www.bruegel.org/2020/12/covid-19-has-widened-the-income-gap-in-europe/ ; https://www.theguardian.com/world/2021/jan/15/billionaires-net-worth-coronavirus-pandemic-jeff-bezos-elon-musk ; https://www.nytimes.com/2021/01/26/upshot/stocks-pandemic-inequality.html
 For example, see Semple, N., The Cost of Seeking Civil Justice in Canada, The Canadian Bar Review, 93(3), pp.639‑673, 2016.
 Gramatikov, M. A. (2009). A Framework for measuring the costs of paths to justice, The Journal Jurisprudence, 2(1), pp. 111-147.
 Equal Access to Justice: OECD Expert Roundtable Background Notes, 2015; https://www.oecd.org/gov/Equal-Access-Justice-Roundtable-background-note.pdf
 Cohen, Law for the Platform Economy, pp.172-184.
 Akhtar, A., Homeland Elegies, 2020, pp.266, 232-242.
 Bork, R., The Antitrust Paradox, 1978.
 Khan, L., Amazon’s Antitrust Paradox, Yale Law Journal Vol 126:3, 2017.
 At the Nuremberg Tribunal, company officers were prosecuted for crimes like the use of slave labour and supplying weapons for the Holocaust, and for aiding and abetting by publishing speeches and articles calling for the annihilation of Jews; Domino, Crime as Cognitive Constraint, pp. 178-9.
 In the EU, the Digital Services Act would require transparency of content moderation practices to prevent unjustified limitations on freedom of speech and public debate, address the vulnerabilities of their systems to intentional manipulation of their service, give people rights to seek redress for unfair take down of content, and target additional measures for very large online platforms with a user base greater than 10% of EU population (45 million). The Digital Markets Act would create special rules for ‘gatekeepers’ such as allowing access to data, more transparency for online advertising and prohibiting certain self-preferencing and lock-in of consumers.
 Eichensehr, Digital Switzerlands, p. 689; https://www.geekwire.com/2020/microsoft-unleashes-death-star-solarwinds-hackers-extraordinary-response-breach/
 https://techcrunch.com/2019/12/03/fired-google-workers-will-file-federal-complaint-alleging-the-company-wrongfully-terminated-them/ ; https://www.technologyreview.com/2020/12/04/1013294/google-ai-ethics-research-paper-forced-out-timnit-gebru/
 Higgins, R., Problems and Process: International Law and How We Use It, 1994, p.49.
 https://www.worldpoliticsreview.com/articles/29351/jack-ma-s-disappearance-and-the-dangers-of-doing-business-in-an-autocracy ; https://www.csis.org/chinas-emerging-cyber-governance-system ; https://www.forbes.com/sites/ywang/2020/12/24/china-launches-anti-monopoly-investigation-into-alibaba/; https://www.worldpoliticsreview.com/articles/29351/jack-ma-s-disappearance-and-the-dangers-of-doing-business-in-an-autocracy
 Habermas, J., The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society, 1989.
 Hildebrandt, Profiling and the rule of law, pp. 55–70.
 Zuboff, S., The Age of Surveillance Capitalism, 2019.
 boyd, d., It’s Complicated: The Social Lives of Networked Teens, 2014.
 On this, see Eli Pariser’s work on redesigning the digital public sphere while borrowing from the work of urbanist and sociologist William Whyte; https://reinreports.com/applying-the-lessons-of-public-places-to-workplaces/
 Shannon Raj Singh proposes to extend aiding and abetting liability to social media platforms; Move fast and break societies: the weaponisation of social media and options for accountability under international criminal law, Cambridge International Law Journal, 8, pp. 331-342, 2019. Domino suggests an international tort liability requiring party to compensate for harm caused by negligence or conduct; Crime as Cognitive Constraint, pp184-5
 These arguments are playing out in the United States in the context of section 230 of the Communications Decency Act and the European Commission’s proposal for a Digital Services Act.
 The passage – which was widely shared on social media on King’s birthday in January 2021 – is so powerful and eloquent that it deserves citing in full. ‘America freed the slaves in 1863 through the Emancipation Proclamation of Abraham Lincoln but gave the slaves no land or nothing in reality… to get started on. At the same time, America was giving away millions of acres of land in the west and the Midwest. Which meant there was a willingness to give the white peasants from Europe an economic base. And yet it refused to give its black peasants from Africa who came involuntarily, in chains, and had worked free for 244 years any kind of economic base. And so emancipation for the negro was really freedom to hunger. It was freedom to the winds and rains of heaven. It was freedom without food to eat or land to cultivate and therefore it was freedom and famine at the same time. And when white Americans tell the negro to lift himself by his own bootstraps, they don’t look over the legacy of slavery and segregation. Now I believe we ought to do all we can and seek to lift ourselves by our own bootstraps. But it is a cruel jest to say to the bootless man that he ought to lift himself by his own bootstraps.’; Martin Luther King interview with NBC, 1967.
 Letter from Birmingham Jail, 1963.