Big Tech Is Watching… and Being Watched
Published on March 22, 2019 on project syndicate
Roger McNamee, Zucked: Waking Up to the Facebook Catastrophe, Penguin Press, 2019.
Shoshana Zuboff, The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power, Public Affairs, 2019.
CAMBRIDGE – The maturation of the digital revolution has defined the first two decades of the twenty-first century. The central economic disasters of this period – the 2008 global financial crisis and consequent “Great Recession” – were driven by the proliferation of digital securities and the illusion that digital programs could measure and manage their attendant risks. And the social environment in which individuals consume information and execute transactions has been reconstructed according to the terms set by Amazon, Apple, Google, and Facebook.
Of these four Big Tech firms, Google and Facebook have pursued the most contentious business models. Those who search for information and connect with others on these platforms generate data that are fed into richly detailed user profiles, which are then monetized through the sale of targeted advertising. This heavy reliance on advertising is perhaps the least revolutionary aspect of the digital revolution. After all, the broadcast radio industry settled on the same model when it started running advertisements in 1922, much to the dismay of then-US Secretary of Commerce Herbert Hoover.
And yet, digital advertising is radically more efficient than its predecessors when it comes to targeting specified audiences, recording users’ responses and engagement, and leveraging the micro- and metadata they produce. It’s so efficient, in fact, that Alphabet (Google’s parent company) and Facebook have become two of the most valuable companies in the history of capitalism. But, of course, advertisers go where their audiences go, and people have flocked to the major digital platforms for the obvious benefits they provide.
For example, it is already hard to imagine conducting research or producing a work of scholarship without at least running a Google search for potential sources and archives. By the same token, it is easy to forget how much Facebook has reduced the hurdles to bringing likeminded people together, be it for public or private purposes. As Lara Putnam of the University of Pittsburgh and Theda Skocpol and Leah Gose of Harvard University have shown, private Facebook communities have become key vehicles for a nationwide mobilization of local “resistance” groups opposing US President Donald Trump – a movement that has far surpassed the scale of the Tea Party response to President Barack Obama.
The Techlash
Still, many have come to question whether the obvious benefits of digital technologies outweigh their costs in the form of lost privacy, social fragmentation, and new threats to democratic
political processes. This critical scrutiny is sure to be intensified given the horrific, real-time broadcast on Facebook of the massacre in the mosque in Christchurch, New Zealand.
The “techlash” now underway should not come as a surprise. A broad public response is to be expected in a political economy with institutional and regulatory channels for relief from the exercise of market power. Like the immediate re-regulation of financial institutions after the financial crisis, today’s techlash represents another step away from the market fundamentalism that dominated policymaking between the “Reagan Revolution” of the early 1980s and the end of George W. Bush’s presidency in 2008. Nothing new here: the emergence of dominant market forces has been met repeatedly with a concerted legal and political response. In the United States, this was true with respect to the continent-spanning railroads in the nineteenth century, the massive monopolies (“trusts”) of the fin de siècle era, and the giant corporations of the New Deal era.
If history is any guide, today’s techlash is likely still in its early stages. Legal and regulatory scrutiny of current business practices is intensifying, with the European Commission leading the way, ahead of progressive-leaning US states such as California. According to one industry newsletter, there are already 33 ongoing investigations in the US and Europe of Big Tech companies, from AT&T and Verizon to Uber and Airbnb, by way of Alphabet and Facebook.
The books under review represent another front in society’s response to Big Tech, though they could not be more different in terms of style and reader access. Zucked, by the venture capitalist Roger McNamee, offers a personal narrative of the author’s journey from investing early in Facebook and mentoring its founder, Mark Zuckerberg, to coming out as one of the company’s leading critics. By contrast, Surveillance Capitalism by Shoshana Zuboff, a professor emeritus at Harvard University, deploys a highly abstract analytical framework to argue that digital technologies have ushered in an altogether novel variant of capitalism itself.
“Move Fast and Break Things”
Both McNamee and Zuboff are on the same mission: to expose how the valued functions of “Search” (on Google) and “Connect” (on Facebook) turned into “Target” and “Manipulate” – as in behavioral manipulation to bring users’ actions into line with the goals of those who control the data (legitimately or otherwise). Beyond that, both books provide ample evidence of Google and Facebook’s controlling founders refusing to accept responsibility for what happens on their platforms (in Google’s case, this mostly concerns YouTube).
For example, liberated by Section 230 of the US Communications Decency Act of 1996, which immunized online platforms from legal liability for content provided by others, both companies have asserted the values of untrammeled free speech and avoided accountability for what their users share. As Zuboff puts it, they have “weaponized” the First Amendment of the US Constitution. When they are accused of complicity in the abuses that occur on their platforms, their standard response, McNamee writes, is to “deny, delay, deflect, dissemble,” eventually “admit and apologize,” and then repeat the cycle.
Zucked is the work of an insider witness with a deep understanding of the cultural context from which Facebook emerged. In a bibliographical essay at the end, McNamee reflects on the three eras of Silicon Valley, the last two of which he lived through personally. The first was the Apollo Era, when the US Department of Defense constructed the stage on which many an entrepreneur and venture capitalist would dance. Then came the Hippie Era, when computing was democratized by the personal computer and the advent of the Internet (itself a product of the first era).
But then came the Libertarian Era, exemplified by the “PayPal Mafia” (not least PayPal co-founder Peter Thiel, the first venture investor in Facebook) and others who shaped the process by which social media became the “killer app” of the digital age. It is not an accident that the Libertarian Era in Silicon Valley coincided with the neoliberal era of political economy. In the crucial early years of Google and Facebook, their founders were given a free pass – by stockholders and regulators alike – as they consolidated control of their companies and grew to dominate markets they themselves had created.
Respect Betrayed
Facebook’s stunning growth in scale and profitability is not what turned McNamee’s admiration into staunch opposition. Rather, it was the company’s move from “Target” to “Manipulate.” The step-by-step account of his journey delivers its message in the most effective way possible: as a personal story of betrayal and response.
McNamee is primarily concerned with the social and political consequences of Facebook’s business model, which is geared toward feeding and reinforcing the revealed preferences of its users. Hence, in 2017, he joined forces with Tristan Harris, an erstwhile “design ethicist” at Google, to launch the Center for Humane Technology. A former student of the Stanford University behaviorist B. J. Fogg, Harris is closely acquainted with the techniques Google and Facebook have deployed to influence behavior, often going to extremes far beyond what Fogg himself anticipated.
For example, McNamee recounts how Harris introduced him to the process by which digital platforms shape individual preferences into “filter bubbles” that are far more powerful than those implied by, say, watching only Fox News. Online, users participate actively – yet unwittingly – in refining the algorithms that drive their informational isolation, rendering the process far more efficient than in other media domains. Worse, this process itself is not isolated from the outside world, and can easily be manipulated by the likes of Cambridge Analytica and the Russian Internet Research Agency. And, as we’ve learned over the past year, part of Facebook’s push to maximize revenue entailed granting such third parties near-open access to users’ data.
McNamee did not originally set out to issue a public warning about Facebook. In fact, he raised his concerns privately with Zuckerberg and Facebook’s chief operating officer, Sheryl Sandberg, in the fall of 2016. But as the months passed, it became clear that they were not going to offer a constructive response to Facebook’s rampant abuse of its platform and users. That’s when McNamee decided to “go public.” On March 6, just a month after the release of Zucked, Zuckerberg unveiled a new “privacy-focused vision for Facebook”? One can only speculate
about whether the two events were related. But if Zuckerberg’s change of posture and (perhaps) policy leads to real, constructive change, McNamee may well feel entitled to take a victory lap.
Same Old Capitalism
One of the strengths of Zucked is its tight focus on McNamee’s own journey from analysis to action. Zuboff also begins with a personal anecdote, recounting fieldwork that she conducted almost 40 years ago for her first book, In the Age of the Smart Machine: The Future of Work and Power. But in fewer than five pages of The Age of Surveillance Capitalism, she shifts into high theory, offering a definition of “surveillance capitalism” that is so abstract as to have almost no connection to observed human experience.
Over the course of 520 pages, Zuboff wants the reader to know that the evils she decries are not the inevitable result of digital technology, but rather a consequence of deliberate choices made in pursuit of profit. Her surveillance capitalists are users of technology; they are not its predetermined agents. On this sound point, she unwittingly echoes the great French historian of pre-modern capitalism Fernand Braudel, who observed that, “Capitalism does not … invent the market or production or consumption, it merely uses them.”
With similar insight, in the spirit of Karl Polanyi’s The Great Transformation, Zuboff observes that Google and Facebook have both commodified activities that already existed outside the market. Whereas Polanyi’s focus was on the “fictitious commodities” of land and labor, Zuboff’s is on what she calls “behavioral surplus” – that is, the “digital exhaust” generated whenever users “Search” and “Connect” online.
Unfortunately, such thoughtful observations are too often smothered with jargon. Zuboff is driven to offer a grand theory of capitalism and the pursuit of profit in the digital age. So, rather than simply explaining the phenomena she observes, she obscures her arguments behind portentous talk of “the third modernity” and “instrumentarianism,” a concept that she borrows from Alex Pentland of MIT’s Media Lab, whose own theoretical aspirations are a match for Zuboff’s own.
Zuboff approvingly cites Pentland’s assertion that “social phenomena are really just aggregations of small transactions between individuals,” which goes beyond even former British Prime Minister Margaret Thatcher’s notorious claim that “there is no such thing as society.” At least Thatcher made room for “families” in addition to “individual men and women.” Zuboff takes as pretentious a leap when arguing that the improved precision of machine tools, thanks to locally deployed digital feedback, represents a transition from the era of top-down, hierarchical control over mass production to the distributed “hive mind” of her “third modernity.”
Confidence Traps
To be sure, Zuboff makes an important distinction between the type of behavioral control pursued by totalitarian states and the predictive certainty that motivates the “instrumentarian” purveyors and purchasers of behavioral data. But, as Diane Coyle of Cambridge University notes in her own critique of the book, a historical account of surveillance as a persistent social
phenomenon from Jeremy Bentham through Michel Foucault is missing entirely. It is a lacuna that becomes more obvious with each repeated reference to the “unique,” “unprecedented” nature of surveillance capitalism.
Worse still, Zuboff shows no appreciation of the fundamental uncertainty surrounding human actions and their future economic and social implications, even though that has been a central theme explored by economists as varied as Frank Knight, John Maynard Keynes, and Friederich Hayek. As George Soros has observed from his long engagement with financial markets, one constant form of uncertainty is the “reflexivity” implied when human actions become variable inputs in decisions made by others. The irony, of course, is that while Zuboff denies the inevitability of Google and Facebook’s triumph, she simply assumes that they will achieve near-perfect predictive certainty with respect to people’s future behavior.
As Zuboff’s argument becomes more speculative, she becomes more assertive. She calls to this reader’s mind the “psychohistory” of Hari Seldon, a fictional character from Isaac Asimov’s Foundation Trilogy who could calculate future events algorithmically. Yet Seldon offered only probabilistic predictions, not the certainty that Zuboff ascribes to the “instrumentarian” villains at the center of her “third modernity.”
Indeed, Zuboff’s own certainty about current and future technological developments undermines her own argument, for she seems to accept without question the marketing claims made by corporations currently invested in machine learning and what is now the third wave of hype over artificial intelligence (AI). One would never know from reading her account of IBM’s Watson program that its highly publicized applications in health care have failed repeatedly in clinical settings at some of the most prestigious medical institutions in the world.
“Ah,” the poet George Meredith wrote more than a century ago, “what a dusty answer gets the soul, when hot for certainty in this our life.” And yet, even if one accepts that the only certainty in human behavior is its uncertainty, one still needs to respond to the real-world evils spawned by today’s digital giants. And here, what McNamee and Zuboff offer differs substantially.
Restoring Antitrust
Zuboff is hopeful that the European Commission’s General Data Protection Regulation (GDPR) will address some of the public’s concerns about privacy. But her principal appeal is for individuals “to be the friction” against the “Big Other,” which is her opaque label (a play on George Orwell’s Big Brother) for all of the digital forces she decries. But that is where she leaves it.
For his part, McNamee also calls for “user pressure” as a first, best choice for inducing reform at Facebook, but he goes further than Zuboff in upbraiding Google and Facebook for not “embracing GDPR” themselves. And, in contrast to Zuboff, he dives deep into the regulatory weeds in search of a solution, concluding that a reinvigoration of antitrust law is necessary.
McNamee takes the reader back to 1956, when the US Department of Justice’s Antitrust Division leveled the playing field of information technology through its settlement with AT&T.
The company was allowed to keep its franchise in long-distance telephony, its dominant position in local services, and its (vertically integrated) manufacturing arm, Western Electric. But it was denied entry into the nascent commercial computing market, and it was required to license Bell Labs’ portfolio of patents for free to all comers. When all was said and done, the DOJ had furnished the public with an endowment of enormous, if unquantifiable, value, and set the stage for rapid growth in the information-technology sector, which was now free from the threat of AT&T’s dominance.
Antitrust law was already becoming a hot topic in discussions about the digital age, but McNamee has turned up the heat. Over a long generation, antitrust law has been transformed by the doctrine known as “Law and Economics,” derived from the market fundamentalism associated with the University of Chicago. This holds that the only legitimate purpose of antitrust law is to protect consumers from abuse: competitive conditions on the supply side of the market should be ignored unless they are directly responsible for such abuse. Given the zero cost to consumers of digital services funded by advertising, strategic acquisitions by the giants that pre-empt potential competition – Google’s purchase of YouTube, Facebook’s purchase of Instagram and WhatsApp – have gone unexamined.
McNamee, citing Barry Lynn of the Open Markets Institute, adds an imaginative dimension to the discussion of antitrust when he highlights the value of the user data that is being monetized on digital platforms. Insofar as that data is lucrative as a source of advertising revenues, consumers have been surrendering more and more value in exchange for the same old search and connect functions. And as McNamee has pointed out to the Federal Trade Commission, the skewed ratio of value surrendered to value received on today’s digital platforms may well qualify as customer abuse.
Data of the World Unite
The issue of data is central to the rise of the digital giants, and both McNamee and Zuboff pay due attention to the questions of who should own it, who should have access to it, and to what end. Data constitute a fourth source of market power that is separate and distinct from network effects and economies of scale and scope. Network effects drove market concentration in the age of telephony, but no one other than intelligence or security services really cared about the content of the messages exchanged.
Now, it is all about content, and not just the substance of exchanged information, but the metadata that defines the context in which information is generated. The more data digital platforms have, the better their algorithms will be; the better the algorithms, the better the service to users and the real customers (advertisers); the better the service, the larger the market share; the larger the market share, the more data. That is the new feedback loop driving market concentration.
One genuinely creative approach to addressing this problem can be found in Radical Markets: Uprooting Capitalism and Democracy for a Just Society, by Eric A. Posner of the University of Chicago Law School and economist E. Glen Weyl. They suggest that we think about data in the same way that we think of capital and labor: as a productive resource for hire. No doubt, granting
individuals ownership of their data and requiring those who capture it to pay for its use would transform the dynamics of the digital economy. In their very different ways, McNamee and Zuboff both show why such radical interventions are now entering the policy agenda.