The old Democratic party doesn't fit new media

Melding new modes of policy making with new modes of public making

[“One Third of a Nation,” by O. Louis Gugliemi, from the Met Collection]

A few months back, I argued that political scientists should read Chris Hayes’ fantastic book, The Siren’s Call, and Chris’s subsequent conversation with Ezra Klein, to understand the relationship between the attention economy and politics. Some weeks back, Ezra and Chris had a follow-up conversation about Mamdani’s primary victory in New York, which I’ve been thinking about ever since.

Below is an attempt to remix that conversation above the beats of other, more academic discussions with Hahrie Han, Nate Matias, Marion Fourcade and Cosma Shalizi amongst others. None of the above are to blame for any stupidities, and I’m not sure that this proposed synthesis will work. But even if it doesn’t, the mistakes might be useful. So here goes.

The background to the conversation is that (a) there is something obviously not working in how Democrats communicate with ordinary people, and (b) Mamdani seems to be connecting better than other Democrats, at least to New York primary voters. But the most interesting parts of the discussion are those that go well beyond the particulars. Ezra and Chris both believe that this problem won’t be fixed by One Weird Trick or Discovering the Liberal Joe Rogan. They see the problem as symptomatic of a more systematic failure. Rather than just figuring out how to get voters to tick the right box, they are both convinced that there is a fundamental mismatch between the way we do politics and the environment that we do politics in.

I think that what they are looking for is an understanding of democracy that is better at generating and organizing useful variety in Brian Eno’s sense, while also being more politically robust. Put slightly differently, their problem is the following. How can you build a new mode of politics that is both (a) usefully experimental, and (b) capable of getting and sustaining sufficient public attention to be politically viable?

Of course, the difficulty in the US context, as in many other countries,* is that we are starting from a very bad place. In the U.S., both the Republicans and Democrats are what Daniel Schlozman and Sam Rosenfeld call “hollow parties,” political organizations that are largely incapable of doing the job they are supposed to. Fixing the Republican party looks impossible right now, but the Democrats aren’t much better. As Schlozman and Rosenfeld put it in a recent essay, the Democratic party is best described as a series of ineffectual gestures towards a political environment that it obviously don’t understand:

from the geriatric TikTok videos to the little bingo paddles House members waved at Trump’s address to Congress to the feckless inside-baseball legislative maneuvering. …less a political party than a vehicle for institutional self-consciousness

Chris is kinder in emphasizing that Democratic communications tend to be more inwardly than outwardly focused.

Democratic Party politics are really complicated politics of multiracial, multiethnic, multilingual coalitions. I think often the things that success in Democratic politics selects for is skill at managing these coalitional tensions, which is a really difficult thing to do. Hakeem Jeffries is very good at that. Nancy Pelosi is the best at it.

There is much more that could be said about the reasons why this coalition is so fragile, but the more immediate problem is that this style of communication is very badly suited to the actually existing attention economy.

No one — and I think including Nancy Pelosi — would be like: I want to listen to a Nancy Pelosi podcast. Nancy Pelosi is not a great public communicator. She is a legendary, all-time great manager of coalitional tension.

I think the coalitional politics of Democratic politics often select for people who are very skilled at managing these very different, difficult coalitional issues. That is a different skill than public communication to the normies.

As Ezra says, the fundamental problem is:

you need to understand attention not as something other people gift to you but as something you earn yourself.

Working in the new media environment is actually not about communicating existing priorities better. It is about building new and different connections between politics, policies and the public in a world where the machineries of the attention economy have fundamentally transformed those relationships.

One way of bridging between Chris and Ezra’s ideas and academic arguments, might be this. We talk a lot, both in broader conversation and specialied academic discussions about what the “public” believes and wants. The problem is that we cannot actually see the public directly, or even make out the beliefs and wants of the various micropublics (interest groups, parties, local organizations and the like) that are its constitutive parts. All of them are just too big, too complex and too inchoate to be visible as coherent entities. Hence, we rely on a variety of technologies to translate publics into simplified pictures that can be grasped more readily (what computer scientists might call ‘lossy representations,’ or physicists might call “coarse-grainings”).

Take, for example opinion polls, which even at the best of times are dramatic simplifications of what the public believes on a particular issue, filtered through the polling questions, weighting decisions and statistical techniques. To some extent, it is reasonable to say that opinion polls don’t measure publics so much as they make them. Much the same can be said of the eighteenth century coffee shops that were idealized in Jurgen Habermas’s discussion of the public sphere, or twenty-first century social media platforms. All of these provide images of the public that flatten, distort, and to some extent actively create the public’s beliefs and desires.

These technologies shape publics, not just in the eyes of outside observers, but in the eyes of the putative members of the public themselves. If you are on social media, your understanding of what the public is - in the U.S, for example, your understanding of Democrats, Republicans and what they agree and disagree about - is shaped by the media that you participate in, the algorithms (whether they simply just present a flow in reverse chronology like Bluesky, or select some posts as Twitter/X does) that direct those media flows, and the social practices that users semi-spontaneously create within the broader affordances of the technology.

Democratic publics, in short, are a partial by-product of the machineries of attention that Ezra and Chris are talking about. Different machineries will represent public opinion, will in part make publics in different ways, emphasizing this aspect of what they collectively believe and want, and discounting that other aspect.** Ezra and Chris worry that new media tend to flatten out the aspects of democratic discussion that make for a workable relationship between politics and policy.*** Ezra:

The fundamental reality of the Twitter text box is that it’s a compression mechanism.

Chris:

I do worry that the structural nature of public opinion now is negative in a way that makes even good governance not resonate with people. Or the structural limitations on governing — one of the two. That it’s just very hard because of how many things contribute to a working-class person who lives on Fordham Road feeling the squeeze in every direction.

That, then, is why Mamdani’s primary victory is interesting to them. Not simply because he’s more skilled at using TikTok than the modal Democratic politician, but because he’s apparently trying to figure out how to meld a new mode of policy making with a new mode of public-making.

As Ezra makes clear, it is hard to connect policy making to publics:

Translating this kind of communication from campaign to governance — not that many people have had to do it. But Obama had to, and I think I would say he failed to do that. I think the sense is that he was an amazing campaigner. And then given the reality of incremental victory, he was never sort of able to narrativize that. … I think that’s in some ways why the liberalism he represented, after him, for at least some time, had a hard time because he had raised hopes so high for a lot of people. … how do you narrativize the difference between people’s hopes for your campaign and what they got? Donald Trump is interesting because he comes after Obama. He also makes huge, sweeping, wild promises. … They don’t build the wall. But Donald Trump has his way of communicating throughout his entire presidency.

Simplifying crudely: Obama’s approach works for a while, because he builds on the audacity of hope, but he delivers disappointment rather than transformative change, because that is what the system allows. Trump also makes big promises, but they are performative. It isn’t clear that his public really expects him to deliver on them, and his partial political success partly rests on a radical disconnection between purported proposals and actual outcomes.

So the question then is:

Can you use this kind of communication mastery, this capability over attention, to generate force for the things you want to pass? Can you use it as a form of power? But then can you use it if you’re not able to get it done?

More specifically, Chris asks whether the new technologies meld with some new version of Franklin Delano Roosevelt’s experimentalist approach to policy making. As FDR famously said in 1932:

The country needs and, unless I mistake its temper, the country demands bold, persistent experimentation. It is common sense to take a method and try it: If it fails, admit it frankly and try another. But above all, try something. The millions who are in want will not stand by silently forever while the things to satisfy their needs are within easy reach.

Chris sees Mamdani’s approach as similar in some important ways.

One thing I would say is, I like experimentation and new ideas. When [Mamdani] was asked about [publicly funded grocery stores] — I think it’s in “The Bulwark Podcast” — he says: We’ll try it, and if it doesn’t work, c’est la vie. I love that answer. Politicians never give that answer. The person who really most embodied that spirit is F.D.R. If you go back and read about the first hundred days, they’re just trying a lot. We now think about F.D.R. as this colossus who remade the relationship between the citizen and the federal government. A lot of that stuff did not work — like, fully failed. A lot of the interventions failed. They did a lot of clunky stuff. It was a totally different time. He had these enormous mandates, and it was a crisis. I like the idea of experimentation and these ideas coming from outside of the consensus around sensible policy. The test for it is: Can you deliver?

As the last question suggests, it might not work now! The possible failure modes don’t just involve Mamdani’s particular strengths and weaknesses, but the fact that he is operating in a very different environment than FDR, and under different institutional constraints. But asking the question of how to do this seems to me to be exactly the right way to frame the problem. Given current circumstances, how can you build an experimentalist democracy, which can keep on trying things out to discover how to get out of the multiple intersecting crises that we find ourselves in? Can experimenting over new forms of communicating with the public reinforce experimenting with new ways of delivering policy, or do they cut against each other?

Those are the big questions that we face. The Democratic party’s crisis is one particular manifestation of it - a congeries of different groups that don’t particularly trust each other and are more focused on their internal problems and rivalries than the external challenges that they face. It’s enormously difficult to undertake bold, persistent experimentation under these circumstances, because every experiment is likely to piss off some key group or another.**** But if you look at the U.S. as a whole, the question reappears on a bigger scale. U.S. democracy is not built so much to solve problems so much as to dissipate political disagreements. Other countries face related dilemmas. So too for global politics.

I keep on coming back to Brian Eno’s description of what an adaptive organization ought look like as a practical description for how democracy ought work.

An organism operating in this way must have something more than a centralized control structure. It must have a responsive network of subsystems capable of autonomous behaviour, and it must regard the irregularities of the environment as a set of opportunities around which it will shape and adjust its own identity.

In practice, this implies regular and repeated experimentation, investigating the irregularities of the environment and figuring out possible responses to them. This is an apt framework to understand both parties trying to appeal to the public, and policy makers trying to get things done. Reconciling the two aspects of experimentation - generating new variant relationships with the public that are compatible with new ways of solving the public’s problems - is the most fundamental challenge of democracy.

So how can you build political parties, broader publics, democratic systems and global institutions that will do this? In particular, can you make this system robust given the particular compression mechanisms and machineries of attention that we have right now? Can you engage in piecemeal democratic engineering (itself experimental) to make these systems more responsive to our actual problems?

These are enormous challenges, which need to be addressed practically, rather than from abstract theoretical first principles. Even so, you still need to get the questions right to get started. I found Ezra and Chris’s conversation enormously helpful in thinking about those questions. Possibly, you will too.

Update: When writing, I hadn’t seen this really interesting Greg Sargent article on the intricacies of the Mamdani team’s approach to social media.

* Schlozman and Rosenfeld’s diagnosis, discussed right below, shares much common ground with Peter Mair’s posthumous book, Ruling the Void, on the failure of party politics in the UK and Europe.

** Here, I’m simplifying. Hahrie and I have a piece coming out very soon, where we define a democratic public as “an evolving, living relationship between the people who collectively constitute it, and the feedback loops (technologically mediated or otherwise) that allow them to understand themselves as a collective and act accordingly.” More soon on the evolving, living part of publics, as opposed to the technological channeling thereof.

*** As an aside - one of the reasons that the Democrat mainstream may be increasingly out of touch is that the filters that it deploys to understand the public may be increasingly out of sync with the mechanisms through which the public understands itself. Elected politicians rely on opinion polls, and are mostly members of the demographics that watch cable and network TV, rather than following influencer beefs, or diving headfirst into the endless feed of YouTube etc. I have no idea of how to investigate whether this hypothesis is true, but it seems at least superficially plausible.

**** This includes centrists and moderates. People who deplore the “groups” as the fundamental problem of the Democratic party are themselves often attached to a faction that reacts splenetically whenever its interests are threatened by some possible political compromise.