Civil society: good thing or bad thing? A lot of people (including myself) have spent years arguing that Americans’ weakening associational attachments helped birth the Trump era. But over the past few weeks, a few smart observers have challenged that thesis, or at least qualified it.
In a conversation with John Ganz, the social theorist Dylan O’Riley drew on Gramsci to argue that civil society is not inherently a bulwark against fascism but instead a site of struggle between fascist and anti-fascist forces. Building on this argument, David Sessions has suggested that the United States, rather than suffering from extreme social atomization, may be a victim of its opposite. We actually have an overdeveloped civil society, Sessions argues, which online communications platforms have driven into hypertrophy. Meanwhile, Henry Farrell has mounted a defense — or at least an articulation — of the liberal theory that an independent civil society is both one of the linchpins of a pluralist democracy and a critical line of defense against tyranny.
I’m sympathetic to both the liberal and Gramscian analyses of civil society, and I haven’t quite figured out how to resolve the tension between the two of them in my own mind. But I suspect that a greater synthesis begins with a more finely grained definition of civil society—or, rather, of civil societies.
Sessions writes:
What has become a general moral panic in the American media about social media, phones, fragmentation, and the “loneliness epidemic” perhaps leads us to overlook the real associative power of those technologies, the way they have created public spheres that extend beyond the internet and produce their own forms of organization. Just because their outbursts seem ephemeral and haven’t given rise to new political parties doesn’t mean we can take their weakness for granted. Personally, and only semi-jokingly, I would call armies of citizens getting people fired for kissing at a Coldplay concert an overdevelopment of civil society.
This is of course correct. But the character of “digital-first” associations — think Discord servers where most of the regular contributors live in different parts of the country — differs pretty substantially from the character of and “analog-first” ones, such as labor unions, the Elks, intramural sports teams, and church congregations. There’s a certain richness to routine face-to-face engagement that you can’t get on X, Zoom, or the metaverse. We all understand this intuitively; it’s why COVID-era social distancing was so ruinous for the mental health of so many, including a lot of people who were in regular online contact with friends and loved ones throughout the ordeal.
I’m convinced the richness of face-to-face sociality lends itself to a different type of political engagement than you find on social media or other remote communications platforms. The precise difference between those two modes of political engagement is something I can’t fully articulate just yet. But the work of social scientists who study in-person organizing can help us start to fumble toward a working definition. In another post, Henry Farrell summarizes some of the scholarship produced by his colleague (and recently certified MacArthur Genius) Hahrie Hahn:
Her second book, How Organizations Develop Activists: Civic Associations and Leadership in the 21st Century stressed the difference between "transactional mobilizing,” and “transformational organizing.” Lots of organizations focus on lowering the transaction costs of doing things - pressing a button to send an email expressing outrage to a politician. However, it is the ones that get people involved in deeper ways and build relationships that can forge enduring communities. Organizations do this less often than they might, because building such relationships is messy, difficult and hard.
[…]
Furthermore, as should already be clear, evangelical churches like Crossroads are more apt to transformational organizing than the transactional mobilizing that more traditional political organizations prioritize on both the left and right. People who get deeply involved in church life are transformed by their relationships. They are also likely to be able to apply the organizational lessons they have acquired in other contexts too. Again, large swathes of American liberalism and the left are structurally bad at offering those kinds of opportunities, because they have doubled down on shallower transactional forms of organizing.
It seems at least plausible to me that in-person gatherings lend themselves to transformational organizing much more than online interactions or one-way communications channels.
There are two other differences I think worth calling out, even though it’s something of a cliché at this point to talk about them. Briefly, then: online engagement lends itself to both epistemic filtering and context collapse in a way that in-person interaction does not. Obviously, every form of association can be an epistemic bubble of sorts — think of a homogeneously white suburb populated by people who really have social intercourse with the nonwhite residents of the neighboring town — but in extended face-to-face interactions you’re more likely to be confronted by cross-cutting identity markers which force you to manage and accommodate difference. Similarly, in-person interactions, especially when thy take place over long periods of time within an established community, have a contextual thickness to them that makes them the virtual opposite of the social media feed.
This does not make IRL civil society intrinsically anti-authoritarian. In fact, Dylan O’Riley and likeminded scholars like Sheri Berman are right to insist, contra the traditional Arendtian analysis, that Nazism infected a highly developed civil society instead of filling the vacuum created by its absence. In the United States, right-wing evangelical churches are probably among the strongest redoubts of classic IRL civil society, and they constitute a significant part of the MAGA movement’s social basis.
But there aren’t enough right-wing evangelicals to elect a president on their own. To better understand the Trumpism phenomenon, we need to understand the disaffected and unorganized voters who pushed Trump over the top. Many of them reside in battleground regions — especially in the Rust Belt — that used to have much thicker civil society ecosystems. I would argue that the Arendtian analysis actually does carry some force in these areas. And it is in these areas where I suspect that mass media and online civil society is a particularly potent vector for Trumpist thought.
All of which is to suggest that anti-fascists should be attentive not merely to civil society as a general category, but to different kinds of civil society. The Internet and mass media are a form of civil society, but in critical ways they are not a substitute for in-person association. (The same applies to professionally managed nonprofits.) I would also argue that, in critical respects, the Internet and mass media as currently constituted are better at transmitting some ideas than others. Fascism as a social contagion is more at home than liberal democracy on many of these platforms.
That’s just a hunch, though. And it shouldn’t be taken as a recommendation for anti-fascists to simply cede these channels to liberal democracy’s enemies. It’s just a reminder that you can’t build a real mass democratic politics through posting alone. You also need to organize some real-world hangs.