On Social Media,Solidarity,and the Catastrophe of Climate Change

The Roots of the Problem

In the past decade, there has been increasingly intense public debate about whether there is something wrong with social media, something problematic for both politics and society. Most of the debate has been focused on particular scandals, or particular solutions. As a result, the larger context and roots of the problem are usually missed. It is not, however—this is important to emphasize—that social media platforms are themselves the root cause of social and political polarization: to say that would be historically ignorant (Banaji & Bhat, 2022) as well as technologically determinist.

A deep problem with social media platforms, as they currently exist in the West and probably also China, emerged with the Cambridge Analytica scandal of 2018 and the 2021 testimonies of Meta whistleblowers Frances Haugen and Sophie Zhang. That problem is the business models of large platforms, which are designed to drive profits by attracting advertisers through the mechanism of fueling “engagement” with platform content in ways that maximally exploit the platform’s data assets (Ghosh, 2020).

Even if you are relaxed, in general, about funding our social spaces through advertising, you should not be at all relaxed about platforms generating their advertising income through mechanisms that maximize user “engagement” with the platform (i.e., activity countable as evidence of user attention from which advertisers can be persuaded they might profit).

The scandals just mentioned gave a sense of why this business model might be problematic. The details are well known, so let me highlight just one point: that the model encourages users into competitive behavior that, while counting as “engagement” from the point of view of Meta’s accountants, might also be psychologically harmful (for those who are young and/or otherwise vulnerable) and potentially harmful for wider communities. Haugen’s and Zhang’s testimony brought out clearly that Meta itself was for some time aware of those “real-world” consequences of its business model, but chose not to alter it (Hao, 2021; Silverman et al., 2020). Whatever the details, it is the business models of social media platforms that must be challenged, and yet they are so often left out of the proposals for reforming social media by governments and civil society organizations. While the examples highlighted by those whistleblowers concerned, for example, the amplification of negative body image for young women or the fueling of inter-ethnic hatred on Instagram, examples closer to climate change are not hard to find. Consider the regular vilification on social media of climate scientists such as Michael Mann and George Monbiot, and the circulation of disinformation memes about climate science and potential climate policy. Both make building solidarity around combating climate change more difficult.

The second root case of today’s problems with social media is bittersweet, since it lies not in social media platforms directly, but in the vast human interconnectedness that the Internet has enabled. Who doesn’t celebrate the fact that they can point to a website containing any sort of digital content and make it immediately accessible to someone on the other side of the planet? Who doesn’t celebrate the fact that, through some means or other, we can be present to each other online, though physically apart (Couldry, 2012, chapter 2)? But the wonderful thing about the Internet and the protocol-based version of it that we call the worldwideweb—the fact that it connects in principle any computer and any content with a url—has a social consequence so profound that we have long since stopped noticing it: to create an effectively infinite information space that is accessible potentially from every point in physical and social space on the planet. This was at most a theoretical consequence of the information space that Tim Berners-Lee built, but it became a reality as every computer-based device went online and a high percentage of the world’s population acquired constant access to such connected devices, now of course including mobile phones.

This vast new hybrid space (over-layering information, physical, and social space) brought a significant and irreducible risk: that, through it, anyone can get access to people who are and content that is morally repugnant or psychologically dangerous and to which they otherwise would have had no direct access. Since evil is an unavoidable aspect of humanity (Dews, 2008), connecting up humanity necessarily created a non-trivial risk that people’s exposure to distant evil would be increased. But that risk was only effectively manageable before two other factors intervened, neither of them anticipated by Berners-Lee. One was the emergence of focusing devices that massively reduce the effective size of the Internet down to a space that is more individually manageable: I mean search engines and platforms. The other factor was the decision—made initially in the United States but with consequences everywhere—that those focusing devices should be run by private organizations to make profit, rather than to advance positive civic or social ends (Van Dijck et al., 2019). I have already touched on the problems with platform business models. This transformation of the Internet’s infinite space of information circulation into a space powerfully focused by commercial forces has made the optimistic reading that pervaded a decade ago of the worldwideweb as a beneficial space of information circulation (Weinberger, 2011) seem sadly irrelevant today.

The third root cause of today’s problems must be briefly indicated here, since it only emerges clearly from comparing the particular way social media platforms have been designed and what, long before, we knew about the causes of political and social polarization from the social sciences. There is now broad consensus among general commentators and political communications researchers that the growing dominance of social media platforms—as the spaces where we live socially and politically—is associated with growing polarization in many countries, not least the United States. Once again, it is important to emphasize that there is no question of claiming that the roots of polarization lie in technology itself, or in the particular technologies that make social media platforms possible. But a leading article on the problem of affective polarization (i.e., polarization of felt identities between groups, rather than polarization of opinions and factual beliefs) provides an important clue to where we should look to explain the problem with social media, a problem that would persist even if their business models were changed.

Shanto Iyengar and colleagues (Iyengar et al., 2012, p. 406) ground their work on affective polarization in 1980s social psychology, in particular the “social identity” theory of Tajfel and Turner about the conditions under which polarization inevitably grows. The core point of that work, as explained by Turner and Oakes (1986), is that the more in any interaction space individuals have incentives to mark their identity as X, and therefore implicitly not as Y, the greater the pressures, over time, for Xs and Ys to perceive themselves in terms not just of different, but polarized, group identities. ¹

Yet no one noticed when building social media platforms that, for entirely commercial reasons, they were installing exactly the properties social psychology had shown are likely to cause polarization, spaces where individuals are incentivized to regularly mark their identity and affinity with particular groups (by signaling “like,” tagging, commenting, etc.). Social media’s goal of tracking engagement, in effect, created a polarization machine. We should not, therefore, be surprised when a recent report in Nature found growing polarization of opinions on climate change as expressed on social media (Falkenberg et al., 2022).

The result was a fundamental design error in building social media platforms, at least as seen from the point of view of their social and political outcomes, an error that business models designed to incentivize engagement only amplified, along with the inevitable drive of platforms as capitalist enterprises to reach an ever larger global user base.

The outcome is a fundamental problem for anyone who wants today to create greater solidarity between different groups or nations, as we surely must if climate change is to be combated. The point of solidarity is to build connections between those who might otherwise perceive themselves as different. But today’s platformed space of politics is not a space that any version of political or social theory would have proposed for enhancing solidarity: the spaces on which social and political theory modeled were massively smaller than the planetary space that the Internet and its platforms bring together. But then our inherited political and social theory has yet to integrate the basic fact of the Internet’s existence into their formulations.

How Did We Miss This, and What Next?

How could humanity have made this basic error, of designing a space of interconnection that was likely to undermine rather than build solidarity? The main reason, I suggest, is that no one ever imagined, until very recently, that human beings, let alone particular capitalist enterprises, could do what Big Tech over the course of two decades ended up doing: redesigning and reconstructing the larger space in which human interactions principally take place. In a forthcoming book, I call this space “the space of the world” (Couldry, forthcoming). The space of the world has in fact gone on changing through history, but slowly and without anyone planning it—until the past three decades when it has not only changed rapidly, but become a target for commercial exploitation. Not only was this unanticipated by the Internet’s architects, but so too was the devastating side effect for human solidarity, that is, for the conditions under which humanity has a chance of addressing climate change.

But we are where we are, and we need a way forward. The only way forward that I can see is to learn the lessons of this category error and find practical ways to work together to dismantle the current platform-based space of the world and build a better one. This will be work of a generation, and right now no one can predict its outlines. But it is just possible that recent moves toward federated models of online social platforms such as Mastodon point the way toward a better and more sustainable way that human beings can be connected.

If so, the choices many people are making today about where they spend their online time are not choices that we can any longer avoid. They are choices that must be made as part of the wider struggle to avert the self-made catastrophe of climate change.