Politics with Proper Names

On Przeworski and Teune, Gove and "Rove"

Cover Art from Przeworski and Teune, 1970

Last week I committed sins in the eyes of both “X” and Bluesky. Different sins, of course. In our new Manichean social media environment, there is no middle path. Centrist dads lie like battered corpses in the no-mans land of the 2025 culture wars.

My sins could be attributed to hosting an episode of Rethink on liberalism and post-liberalism. It was tricky territory, I agree. To take seriously the critics of liberalism, without necessarily agreeing with their underlying assumptions. To underline the core merits of liberal thought, without dismissing the alternatives tout court.

My producer, Ben Cooper, and I thought the best way would be to start with defining liberalism - here we asked Edmund Fawcett to go over his four core principles from his brilliant Liberalism: The Life of an Idea. Here is a great review of that book by Jeremy Waldron.

The four principles are: (a) an acknowledgement that social conflict and disagreement are unavoidable; (b) a mistrust of concentrated power; (c) a belief in human progress; and (d) the importance of ‘civic respect’.

Those might sound obvious but conservatives, socialists, and populists disagree. Many conservatives believe that tradition trumps perfectibility; that there is a natural and harmonious order; that not all groups or cultures deserve equal respect; and that humans are frail and sinful. Socialists believe that (class) conflict is inevitable before but not after socialism and that concentrated power is needed to overcome the inequalities of capitalism. Populists believe that conflict only arises because of a corrupt elite; that concentrated power is needed to overcome the establishment blob; and that only some groups (citizens, perhaps co-ethnic ones) deserve equal respect. And so on.

After setting out the tenets of liberalism, the show then turned to the post-liberal challenge to these principles. In the last third of the show, I upset the denizens of X by casting the (self-professedly) illiberal parties of the European right as opposed to these principles. Perhaps they were upset by my tone of implied critique or the general absence of uncritical support of illiberal parties. Who is to say?

But it was the middle third of the show that upset some respondents on Bluesky. And the reason was because I had interviewed two British politicians who came from non-liberal traditions but had some sympathy for liberals. On the left, I interviewed Jon Cruddas, until recently the MP for Dagenham and Rainham, and whose political views might be placed on the communitarian side of the Labour Party - somewhat but not entirely ‘Blue Labour’. Jon’s points about the experience of Dagenham during the era of liberal triumphalism in the first decade of the millennium are well worth listening to.

It wasn’t interviewing Jon, though, that set the cat among the pigeons. It was my other interview - with Michael Gove, former Cabinet Minister and now editor of the Spectator. Gove was a minister during a more (right-leaning) liberal era under the Cameron government and also during the conservative/post-liberal governments of May, Johnson, and Sunak. I won’t claim to have great insights into Gove’s true beliefs but I think there is a mix of liberal ones - support for gay marriage, free markets, robust debate - and post-liberal ones - Brexit (as happened), tradition and custom, and suspicion of ‘identity politics’ (or at least some people’s identity politics). I will say this, though: as many others have found, he was a pleasant and thoughtful interviewee, whether one agrees or disagrees with his politics.

For some people, the mere presence of Michael Gove was reason to avoid the show or be upset about it. That’s fine - no-one owes me a listen. I hope in the future we can have a world in which voices from the left and right are on the same show and people are fine with it but I recognise this is a polarised time.

From my perspective, I was able to ask two questions I had rather wanted to ask Gove over the years. One was whether conservatives and liberals had switched roles - with the former now dismissive of establishment institutions and wanting reforms and the latter protective and cautionary. He appeared to agree, for whatever that’s worth.

The second question - the one I am sure many of you have thought about a lot - is that when Gove famously claimed people had ‘had enough of experts’, was this really a critique of contemporary liberalism? Here is his response, as heard on the show:

“Inviting people to defer to experts in every area and suggesting that the the common sense of the people should be superseded by the consensus of those at the centre who have the credentials and the qualifications to claim the title of expert, I do think that that was liberalism's fatal conceit. And I also think that the post liberal view is… that you cannot rule the country on a technocratic basis”

Gove’s famous initial exchange about experts, with Faisal Islam, in 2016 was perhaps slightly misconstrued in popular memory. Here it is in full:

Michael Gove: I think the people in this country have had enough of experts, with organisations from acronyms, saying—
Faisal Islam: They've had enough of experts? The people have had enough of experts? What do you mean by that?
Michael Gove: People from organisations with acronyms saying that they know what is best and getting it consistently wrong.
Faisal Islam: The people of this country have had enough of experts?
Michael Gove: Because these people are the same ones who got consistently wrong what was happening.
Faisal Islam: This is proper Trump politics this, isn't it?
Michael Gove: No it's actually a faith in the—
Faisal Islam: It's Oxbridge Trump.
Michael Gove: It's a faith, Faisal, in the British people to make the right decision.

The line that everyone remembers is just the first part, before Faisal’s flabbergasted interruption. But it’s fairly clearly a claim that technocrats should not have the sole responsibility for making political decisions, which should be left to the people. We can argue about whether the British public made a wise decision in this case - in my view, it is increasingly apparent that they did not - but I don’t think this is an illegitimate view.

Indeed, it’s one with a grand tradition dating back at least to Rousseau and including thinkers from Andrew Jackson to Maximilien Robespierre. Now there’s an episode of Come Dine with Me. The argument is that expertise, past practice, and elite consensus should not override popular will. It is, I suppose, leaning on the democracy part of ‘liberal democracy’, rather than the liberal part.

It also reminded me of another much-derided quote from a decade or so earlier. A quote that the journalist Ron Suskind received in 2004 from a high-level Bush administration official, rumoured but never confirmed to be Karl Rove. Hence I have placed Rove in scare quotes in this piece’s subtitle… Here’s the quote from Suskind:

“The aide said that guys like me were 'in what we call the reality-based community,' which he defined as people who 'believe that solutions emerge from your judicious study of discernible reality.' [...] 'That's not the way the world really works anymore,' he continued. 'We're an empire now, and when we act, we create our own reality. And while you're studying that reality—judiciously, as you will—we'll act again, creating other new realities, which you can study too, and that's how things will sort out. We're history's actors...and you, all of you, will be left to just study what we do'.”

Liberals, of course, seized on the dismissal of the ‘reality-based community’. Who would denounce people who cared about reality? Why, the same kinds of monsters who would dismiss “experts”, I’d imagine.

But there is something in common with both this quote and that from Gove (be it to me or to Faisal Islam). A critique of a certain type of (liberal) thinking. A critique of a logic that goes something like this:

There is a social reality out there that has laws and regularities and that we can analyse objectively and, employing robust tools derived from social science, can use to infer from past experiences the ‘right thing to do’.

In other words, listen to social scientists.

Now, as may have occurred to you, I am both a social scientist and, broadly speaking, a liberal (inasmuch as I am neither a conservative nor a socialist). So, I feel the temptation of this logic. My job could be described as trying to look past my subjective opinions and the fluctuations of contingent historical events to derive deeper ‘truths’ about how people behave, in order that we might make the world a better place.

But there’s a lot going on in there. Can one ever truly look past their subjective opinions? Are there really deeper truths that are out there? Should we actually use those ‘truths’ to make the world better place? Who gets to decide on ‘better’? And are historical events just jetsam on a sea of underlying structural social forces?

It would be comforting for us social scientists to be able to look at an uncontested set of past facts, use statistics or the logic of comparison to derive underlying patterns, and then make recommendations to policymakers about how to set policies to improve the world, and for those policymakers to faithfully follow those prescriptions.

I mean that’s basically how my colleagues in economics think the world should work.

Ho ho ho… <Frank Lampard voice>… but seriously, this is a common refrain from the academy - especially the applied parts of social science and medicine. We saw it with public health scholarship during COVID; we saw it with the economics profession following the Great Recession; we see it with political scientists like me following polls around election time.

We have carefully adjudicated what’s really going on. We have recommendations. Follow us.

But what if that doesn’t work? Politically. Or even on its own terms.

Let’s start with the political part. This is the essence of the Gove critique - that technocratic advice should not outweigh democratic choice. Even if the latter is ‘wrong’. Because the ‘wrong’ is usually defined by the technocrats, who themselves do not have unimpeachable track records. I find myself ever more convinced by the argument that experts - and I guess I’m one too - have been too blithe about ignoring mass opinion when it conflicts with their advice - on the merits of free trade, on the costs of shutting schools during COVID, on the costs of net zero.

The experts might be 'right’, in a narrow sense of that word - but if the advice is ignored, indeed if the public start to do the opposite of the advice in a backlash, then we are all worse off. So in part, this is a communications challenge for experts and importantly for the liberal politicians attracted to their recommendations. How to not condescend, how to listen, but how to still hold fast to some set of truths and advice beyond the immediate vagaries of public opinion.

One might say, how to lead.

This should not be viewed as ‘giving in to populism’ but as accepting that some of the strategies and skills populists have are worth mimicking - in particular, learning how to communicate effectively, simply, and engagingly. If we want to critique Gove’s claim about the people having had enough of experts, ‘we’ experts need to up our game.

But let me go now to the second, more challenging critique. That the social scientist view of the world doesn’t work even on its own terms. This is the ‘reality based community’ argument. Think back to that quote. Karl Rove, or someone very much like Rove, critiqued politicians and elites who engaged in the “judicious study of discernible reality” all while he and the Bush administration were creating their own new reality as “history’s actors”.

Set aside the sneering arrogance of the claim and there is an important insight here. There is something very limiting in thinking that all politics is foretold. That today’s politics is simply an extension of previous events that can be studied. That today’s world is a function of yesterday’s world. That the lagged dependent variable explains everything.

For those of us from mainstream social science traditions this is a challenging critique.

Here is how I usually do analysis. I collect existing, observational data about the world - perhaps on inequality levels and measures of democracy, or survey data about party ID and attitudes towards government redistribution. I then look at the correlations across these variables. Or maybe I seek to make more causal claims and I run an experiment in a survey or in the lab, changing some variable randomly to see how it impacts some other outcome variable. Or I read a bunch of primary sources - I spent a week one summer sitting in the archives in Saskatchewan reading annual reports of mental asylums and sanatoria, trying to figure out how decentralised funding was or whether religious authorities were involved, all so I could code the province on a one to three score. Fun.

In these cases - including the one where I was reading about Weyburn Mental Hospital (once the second largest building in North America after the Pentagon!) - I am interested in (a) things that have already happened, or (b) how one social variable relates to another, or (c) both.

In other words, I am engaged in the “judicious study of discernible reality”, and of things that have already happened. From which I draw conclusions about politics and then say, this is what you need to know to understand politics. It’s all backward-looking analysis not forward looking action. Because that is how academics create knowledge in the social sciences.

I’m not saying there is anything wrong with that. That’s my job. But that is a different job to either being a politician or advising one. Because political actors, especially those at the top of the tree, do create their own reality. Indeed, they create all of our realities. They do not do so solely relying on interpreting the past and adjusting it slightly. They are not Bayesian algorithms shunting us ever close to some optimum. They are not looking for marginal effects under some constraints. They are changing the very constraints themselves.

That means a fully technocratic liberal form of government misunderstands politics. It acts as if it is solving an equation within a set of given institutions. But the institutions can be changed by the very actors they are supposed to constrain. It is all, and here’s a term, endogenous.

My last essay was on the perils and promise of disruption. In that essay I noted that disruption could often be damaging - breaking eggs in the promise of an omelette; forgetting that the eggs have jobs, families, lives.

But there is a reverse risk that I hope is coming through in this post - assuming that politics is about fixed rules and institutions, following the ‘evidence’, being bound by the social science laws of the past.

As the news reminds us daily - were it only so. I am no great fan of the Great Man theory of history. Structural forces do matter, do constrain to some degree, do shape incentives that leaders face. But leaders and democratic publics can act in ways that oppose, ignore, or even disgust expert consensus.

And that takes me to the image at the top of the post, which comes from a seminal work in political science - The Logic of Comparative Social Inquiry by Adam Przeworski and Henry Teune. Przeworski remains perhaps the greatest contemporary thinker on democracy and I strongly advise reading his new Substack. But I am going to push back on one of his claims from over half a century ago.

Przeworski and Teune used this book to set out the modern ‘comparative method’ in political science and along the way they did so by advocating a ‘variable-oriented’ approach, as contrasted to a ‘case-oriented’ approach. That means that understanding politics is about understanding underlying social objects in general, not the real people and places that are instances of them. Let me quote from their 1966-67 article in Public Opinion Quarterly, which preceded the book:

“In formal terms, cross-national analysis is an operation by which a relationship between two or more variables is stated for a defined population of countries. In analysis, no proper names of societies or cultures are mentioned. The goal is not to "understand" Ghana or Cuba, not to describe Hitler, Stalin, Roosevelt, or Churchill, but to see to what extent external crises and internal control, military prowess and economic frustration, nationalism and persecutions, are related, and to know the generality of each relationship. Whether variables are related depends on the observations of Ghana, or Hitler. But these are the observations that are means to an end-the end of testing relationships between variables, even at the cost of obscuring some differences between specific units.” (p.554)

I have italicised the words “no proper names” because that is the advice that generations of social scientists took from Przeworski and Teune - real understanding on politics could only come from analysing social objects in their general, not their specific forms. The names of leaders, the names of countries, the names of regions were not relevant. What mattered was the type of leader, the wealth of a country, the size of a region, and so forth. David Laitin later termed this abandoning ‘uppercase’ letters - that is, the G of Ghana and the R of Roosevelt. Or to refer to the title of this post again - this is politics without proper names.

That has been the tradition of social scientific thought since that day, not just in economics, but in sociology, political science, international relations, even anthropology. We have learned a great deal through abstraction.

But I want to push back a little - that abstraction may help us understand past politics better, it may even help us predict regular, normal future politics. But it is extremely unhelpful in times of great tumult, where past practices fail to predict current or future events. Where leaders, with very recognisable ‘proper names’ - let’s call them Trump or Netanyahu or Xi or Zelenskyy - can break our expectations from past practice. Where variables in normal times are sundered by contingent history with capital letters.

And this is where the advice of Michael Gove and (possibly) Karl Rove is important for those of us who are liberals to take seriously. Returning again and again to previous observations, to ‘what works’, to impregnable political institutions, will blind us to the very real political change going on around us. We will find ourselves permanently shocked and surprised by old patterns dissolving, social norms decaying, institutions collapsing. We might tie our hands to the institutional mast only to find it has been eaten through with woodworm.

The Cornell political scientist Tom Pepinsky recently wrote a series of posts on Bluesky that sum up the dilemma facing the liberal left in the US. I will quote them below:

A fundamental problem is the Democrats are utterly committed to the fallacy that public opinion polls are politics. This has destroyed their ability to actually do politics. 1/

Polls tell you what people like and dislike. This is important. But Democrats' obsession with poll-testing everything has led Dem leaders and pundits to completely misunderstand how politics works in a representative democracy. Winning in public opinion polls is not winning in politics. 2/

In short: for democracy to survive, our representatives must represent us. They cannot just declare that our views are more popular and wait for the world to turn. Democrats are losing because they are coming with spreadsheets and figures, when the opponent brings theater and muscle 3/

Democrats must abandon the belief that just standing for the more popular thing will create the right outcome. Public opinion will move wherever the leader directs it (watch what will happen on Russia in the coming days; negative polarization will dominate *any* GOP principles) 4/

Democrats will continue to lose until they realize that the rules have changed, that facts do not win elections, and that they must make a costly stand for what they believe in. Until then, we are ruled by a mad king and his billionaires, and the Democrats are but useful idiots /end

Set aside your views about how close democracy is to collapse in America - I remain on the more optimistic side of this debate, like Tom’s old colleague Andrew Little - though I must admit to feeling increasingly like my bet won’t come in.

What Tom is saying here is a helpful critique of the quasi-technocratic form of liberal politics besetting the Democrats. They are trying to analyse existing data - polling data on Americans’ attitudes - in the hope it will provide evidence for what arguments they should make. They are tied to the past and are following in its wake. But they, like us all, live in the present! And yet they are strategising in the rearview mirror.

Donald Trump and Elon Musk, whatever else one can say about them, are not governing by polls. They are making the waves themselves, to continue my nautical analogy. They clearly do not care for the judgment of the “reality-based community” and certainly not for the edicts of experts. They are destroying the constraints that liberals thought bound all politics. They are changing the rules of the game, rather than optimising within them. Because they can.

And yes much of this behaviour is not democratic - but some of it is. Trump has every right to change tariffs or to withdraw military forces. And he is doing it because he can and because he wants to. That too, is politics.

Pretending that everything will just go back to normal - that politics will go back to being predictable, marginal shifts from past practice - that attitude will fail us today. That is not the politics of our time. And those of us who are social scientists will harm the prospects of politicians if we pretend otherwise for our own convenience.

This is not to say that all bets are off, that constraints will never bind, that nothing matters LOL. I suspect DOGE will collapse ignominiously in the next year because its form of disruption is so vast and so arbitrary that it will slice through the legs of the US economy. NATO is under threat, quite clearly, but I still think the US would respond to an invasion of Poland or some other classic Article 5 issue.

But politics today is fluid. And let me give you a rather unusual example of that fluidity - Sir Keir Starmer. Over the past very challenging week, Starmer and his team have shown flexibility and a willingness to understand that the ground is shifting and that following standard practice won’t work. The apparently successful trip to the court of the MAGA king, with the surprise offering of a state visit, was one such move. But Friday’s debacle at the White House - a setup of such pointless venom it is hard to fathom - showed the sheer unpredictability, sheer nonlinearity of global politics right now.

And yet this weekend Starmer shifted again, this time to leading a hastily organised summit of what can only be called NATO without the US. It is hard to imagine predicting such a summit ever occurring if we only followed past practice. And yet here it was, a manifestation of making, rather than following, reality.

So fellow liberals, this is the message to hear, even from those you might rather not listen to. One cannot simply hope judicious study of the past provides our salvation. Unpredictability is not a friend to liberalism. But it is not its doom. Arguably, liberals have thrived most in times of change - in the early Industrial Revolution, in the Second World War, in the fall of the Soviet Union. Those were moments when liberal leaders did what comes more naturally to followers of other philosophies that emphasise action and force, rather than evidence and rules - they chose to make history themselves for liberal ends. That road is still open. We can create, not merely study, the reality we live in.