I think I need to eat crow. As my friends at Pomona will probably recognize, I spent a lot of the last few years being (at least outwardly) skeptical of political theory and philosophy. And, as you can probably tell from reading this blog, I’ve been thinking a lot while being in the UK about political theory. So I thought I ought to write a bit about why I used to be skeptical of the enterprise and why I’ve come to change my mind.
It’s worth starting, I think, by clarifying what I mean by “political theory” and what I used to believe more in. American political science is typically divided into a few different areas: American politics, comparative politics (literally the domestic politics of any non-American country), international relations (traditionally, the interactions between countries), public law (the law that concerns the relationship between a government and people, such as constitutional law and criminal law), public administration/public policy (concerning questions of implementing governmental policy), and political theory (basically, everything else concerning questions of government and politics!). As I understand it, political theory is distinguishable from these other subfields primarily by its concern with broader questions (e.g., in political theory, we’d ask questions such as, “what is a good government?”; “what is law for?”; “what is representation?”) and its broader methodological scope. As far as I can tell, in other areas of political science (except for public law), we focus on empirical, observable behaviors, and the purpose of the discipline is to explain and study those behaviors. In contrast, political theory engages with other kinds of questions, even those that are not operationalizable as empirical questions.
I used to believe a lot more in what we might lump as “quantitative, empirical political science” — efforts to understand politics by measuring phenomena and then applying the tools of statistics. Examples of this kind of work that I’m familiar with are the literature on presidential vetoes (when does the President veto legislation and why?), political polarization (what causes Congress to vote in increasingly partisan ways?), international treaty compliance (when do states parties follow human rights treaties?), and so on. This kind of work, I used to — and still do, to a certain extent — believe, can help us understand how politics works by building causal theories that are “falsifiable,” that is, theories that are testable and disprovable.
Three reasons led me to believe in the value of this kind of political science. First, methodological rigor helps us temper our ideological biases. We are all surrounded by and implicated in the work of politics. Political scientists all the more so — when you spend your entire life studying something, you tend to care a lot about it. And because politics concerns the big questions in life, your ideological biases can matter a lot. There are works of political science that try to argue, for example, that Democratic presidents tend to grow the economy while Republican presidents don’t (and vice versa). That kind of eminently partisan argument is very liable to ideological bias. Methodological transparency can temper those biases because they expose your work to criticism: How do you measure economic growth? What factors are you controlling for? And so on.
Second, quantitative analysis, especially, can temper other kinds of personal biases. We all have gut inclinations about how politics operates, and those are built off our individual life experiences. They’re subject to all sorts of cognitive biases that aren’t going to represent the complex, dynamic systems political scientists study. Aggregating data can help the analyst deal with those individual biases. Nate Silver, for example, loves to make fun of the traditional method of covering presidential elections by sending a journalist to a diner in the Midwest, interviewing a few random people, and then forecasting the results of an election. Obviously most qualitative political science tends to be more methodologically robust, but the strawperson example does help draw our attention to the dangers of non-systematic analyses.
Third, and most fundamentally, science is meant to be constructive — to develop over time, resulting in the accretion of greater and greater knowledge so that we can, in the end, say that we’ve actually learned something about how the world works. What made me skeptical of political theory, I think, is that it doesn’t feel like it grows; it seems like the endless negotiation of competing worldviews and values that lack any methodological or subject-matter constraint to enable the building of knowledge.
In my very first reading for my college introductory physics class — my beloved Physics 70 — our textbook argued that a scientific discipline requires four criteria: (1) a community of scholars; (2) commitment to logical consistency of models; (3) use of reproducible experiments to test those models; and (4) theory that enables research.
The vast majority of empirical, quantitative political science, we might say, barely (if at all) meets these criteria. Only a small proportion of political science is done by experiment, typically (as I understand it) in the field of political psychology. But what does distinguish empirical political science is that it attempts to build models to explain the behavior of political actors; it is grounded in, and tested against, observable reality. Critically, then, empirical political science is falsifiable, even if it is not experimental.
Take, for example, the “credible commitment” hypothesis, which is an important theory in regulatory studies. The basic idea is that democratic countries have legislatures that cannot bind themselves to a future action — a legislature that establishes, say, a carbon-trading mechanism could always, after an election, vote to abolish that system. The result is that corporations are skittish and are less likely to invest in that country because they think the country’s regulatory scheme lacks a credible commitment. (Note that this basic idea has normative breadth; arguments in development studies about the rule of law being crucial to economic growth, for example, are premised on this same underlying faith in property guarantees and regulatory stability.) The credible commitment hypothesis argues that countries will therefore create independent regulatory agencies with power to create stable, expert-driven regulation that can survive political changes. The probability of the creation of an independent regulatory agency — and the extent to which the agency is actually independent — will vary, according to this model, based on the number of veto points in the country’s legislature (i.e., is it easy for a subsequent legislature to create new law?). Regardless of whether you buy this logic — I’m skeptical, for reasons that aren’t relevant here — it leads to an empirically verifiable hypothesis; we can test it and disprove it. There’s something comforting, I think, in that possibility. For it suggests that we can learn about politics — that we can become better analysts, and perhaps even practitioners, of politics.
There are those for whom politics is simply chaos and arbitrary circumstance. For such fatalists, political success or failure is a function of luck rather than ability or knowledge. What a dismal view of politics. It is one that I simply can’t agree with, even as many of my professors seem to believe it.
In the opening of The Federalist, Hamilton wrote: “It has been frequently remarked that it seems to have been reserved to the people of this country, by their conduct and example, to decide the important question, whether societies of men are really capable or not of establishing good government from reflection and choice, or whether they are forever destined to depend for their political constitutions on accident and force.” I think the debate between empirical political science and political theory is mirrored by Hamilton’s central question — whether government can, in fact, be created and act based on “reflection and choice” rather than “accident and force.” Hamilton, one of America’s first political scientists, thought it could. Indeed, in a different essay, he relied on this essential logic. He acknowledged that previous republics had failed, and dismally so: “It is impossible to read the history of the petty republics of Greece and Italy without feeling sensations of horror and disgust at the distractions with which they were continually agitated, and at the rapid succession of revolutions by which they were kept in a state of perpetual vibration between the extremes of tyranny and anarchy.” But the “science of politics,” he thought, “has received great improvement.” Armed with a more perfect political science, Hamilton suggested, the American experiment in republican self-rule might endure.
The Federalist, it is sometimes remarked, was a piece of political propaganda. Madison, Hamilton, and Jay wrote these essays supporting the 1787 Constitution; their work might be better understood as disingenuous advocacy than sincere belief. But I think concentrating on Hamilton’s use of political science is helpful for us. For one, it helps us understand how political science is inherently political — it is invariably tied to, and thus shapes, the phenomena it studies. The political science of Montesqieu was embedded into the American Constitution in its separated powers and checks-and-balances. Even the scholar who studies political science solely for intellectual stimulation ought to recognize that political science can and will be used for normative ends.
Hamilton’s essay, we should note, also reveals the limit of empirical political science. It is hard to read the writings of the Framing generation without feeling that all of their thought was fundamentally, irreconcilably, absurdly wrong. How, we wonder, can one write about liberty, and equality, and human rights — of government based on “reflection and choice” rather than “force” — when one partakes in a system built on chattel slavery? The political theorist Judith Shklar wrote of the Framing generation’s intellectual giants (Hamilton, Madison, and Jefferson): “Not one could even suggest a plausible solution to America's greatest single failure, black chattel slavery. History, institutional analysis, and economic calculation were all equally useless. For all their confidence in political knowledge and in their own ability to determine the fate of a nation, they were unable to imagine a multiracial citizenry. . . . The narrowness of the politics of rational compromise confront us starkly (as do those of merely accurate knowledge) when we recall the drift to the Civil War.”
Hamilton, Madison, and Jefferson were America’s early empirical political scientists, and they failed so utterly in responding to the profound moral dilemma of their time. Their failure, I think, points to the first reason why I’ve become more disillusioned with empirical, quantitative political science. It is a subfield that, I think, can’t answer — and doesn’t even engage with — many of the important questions of politics. There are fundamental ideas in politics that we just can’t operationalize for quantitative, empirical analysis. How do we talk about, quantitatively and systematically, the pull of rhetoric or the visceral effects of a story? Yet there's no doubt that both of those things matter a lot. Think about the canons of American political rhetoric: the better angels of our nature; life, liberty, and the pursuit of happiness; government of laws, and not of men; a republic, if you can keep it; government of the people, by the people, and for the people; all men are created equal. Political rhetoric in America often reads like a poorly-cited judicial opinion, sprinkled as it is with echoes of the past. Think, too, of the power of individual stories: the killing of George Floyd or the federalization of the Arkansas National Guard in response to Orval Faubus’s refusal to desegregate Little Rock Central High School. These stories profoundly shape the American political self-image in ways that we might peripherally measure but cannot ultimately test.
A second argument is that political theory is necessary for widening our imagination and situating empirical analysis. We need a name and causal mechanism before we can study a phenomenon, as I have suggested previously. Moreover, theory-driven empirical research is important as a methodological constraint: it helps us justify that we are finding more than mere correlations, but actual causal mechanisms. And theory enables us to interpret quantitative results. Consider electoral analysis. What does it mean to say that rural voters are more Republican than urban voters? To a certain extent, that empirical finding alone means the exact same as “Republican voters are more Republican than non-Republicans.” For neither offers any causal, explanatory value. Where you live, on its own, doesn’t explain why you behave in a certain way. We need a theoretical reason to think that where you live ought to matter; otherwise, we might not be finding anything helpful, and the rural-urban divide might be better explained by other factors, such as education, attitudes toward race, and so on.
A third and most fundamental reason for caring about political theory is that it clarifies our values. Empirical political science can help us trace why someone votes a certain way. And there is insight there. But it is a limited insight. Hanna Pitkin wrote about the “democratic” answer to political questions, where we simply say, “they do X because of A, B, and C.” But it doesn’t give us a reason that they should act in that way. In her analysis of representation, she explained, “I will not ask what causes people to have a psychological feeling of being represented, but what reasons can be given for supposing someone or something is being represented.” There’s a value in theoretical abstraction because it requires you to confront the fundamental questions at hand, rather than attempting to explain them in terms of empirical issues. No wonder, then, that political scientists love to dodge questions by saying, “that’s an empirical question.”
It is this third reason that has felt most compelling to me recently. From time to time, I worry that I lack concrete political convictions. My personal sense is that I tend to judge too quickly, and so I try to force myself to learn more before rendering judgment. But I’ve noticed myself recently feeling very contrarian — I feel that I’m finding ways to disagree with just about anyone I’m talking with. Everyone, I feel, seems quite certain, and it’s making me feel more and more uncertain even as I wish I were more certain.
The last time I really felt this way was when I spent eight months in DC. Two people in particular — one of the line attorneys at DOJ and my supervisor at CCE — really inspired me because they seemed to have figured it out: they were both extremely certain of the righteousness of their causes and also had extremely firm groundings for those beliefs. As always, there is a good Taylor Swift lyric about this feeling (although, as is often the case, it comes from a romantic song whose ideas are capable of much broader application): “You're so much older and wiser, and I / I wait by the door like I'm just a kid.”
I just finished a book about Louis Brandeis, and his method for dealing with this concern about righteous certainty was to focus on facts. He abandoned the search for a grand theory in favor of contingent judgment. It’s really hard to imagine in the abstract the meaning of “justice,” but it’s easy for us to point to examples of injustice that we witness. And we can act based on our shared agreement about such injustices. In a speech called “The Employer and Trades Unions,” he explained, “Nine-tenths of the serious controversies which arise in life result from misunderstanding, result from one man not knowing the facts which to the other man seem important, or otherwise failing to appreciate his point of view.” When talking about the development of law, he chastised Lochner-era jurisprudence for being overly focused on abstract legal concepts rather than the lived realities of people’s experiences; “no law,” he contended, “can be understood without a full knowledge of the facts out of which it arises, and to which it is to be applied.” Full understanding of the context surrounding the novel legislation — in these cases, dealing with labor conditions, such as minimum-wage or maximum-hour laws — would help judges realize the rationality of laws that, facially, abridge the freedom of contract. Conflicting absolute principles of law — the police power of states weighed against the due-process right to liberty of contract — could give way to reasoned compromise by focusing on facts and the particular circumstances surrounding each piece of legislation.
That kind of common-law-like method of thinking about politics — successively making contingent judgments that are highly fact-bound — can be valuable, but I worry that it can lead one to get lost in the weeds. For those kinds of judgments can only really be made with regard to specific questions, and they’re not always the important ones. Here’s an example. I met up with another Marshall yesterday, and she posed a very different kind of question, one that we don’t typically ask. She noted that Western countries whose companies have developed vaccines were able to do so because, for the last few centuries, they have plundered and looted the world’s riches, enabling them to concentrate wealth and thereby facilitate scientific development. Do these countries not, therefore, owe an obligation to other countries to share the fruits of that wealth (perhaps narrowly the coronavirus vaccine, but also more broadly)? And if they do, how can that obligation be fulfilled?
To put a finer point on it: there is value in asking questions that do not normally arise in our political discourse, in imagining future worlds that do not seem possible today. The vast majority of people, a year ago, would not have thought the abolition of policing was a possible policy. Many today, it seems, still hold that belief. But that policy has become the subject of mainstream political debate; it has entered the realm of the possible because people chose to imagine a different world free of fundamental injustices. We can critique — and I have critiqued — political theory for being abstract, and academic, and impractical. Try approaching a politician or policymaker and asking them for their thoughts on the ideal Gandhian versus Weberian politician; I don’t imagine you’ll get a meaningful response. But is it wrong for us to ask what we ought to expect of our politicians, to wonder whether a principled, religious politician versus a deal-making machine-raised politician is better? Of course not. But when we’re sucked into the messy political disputes of the day — how large a check should people get in a COVID stimulus package? how much money should go to local and state governments? — we lose sight of those animating, broader questions that undergird political practice.
I’ve written previously about the necessity of judgment in thinking about politics. Politics can’t, in the end, be reduced to the mere application of absolute principles. There are simply too many conflicting principles that are each individually desirable. When we have those conflicting principles, we are driven to narrower, fact-bound assessments, as Brandeis would have it. But we need to balance that perspective with that of the theorist who seeks to build a constructive political morality — who tries to answer the broader questions of what is good, and right, and just, as opposed to merely what is better.
There will be moments in my life, I think, where I feel a stronger pull toward the former perspective — the Brandeisian idea of emphasizing facts and narrow, negotiated compromises through the muddied process of democratic politics. But right now, I feel that what is missing in my political thought is much more the latter perspective — that clarity of values and of guiding ideas that act as a lodestar for rendering those fact-bound political judgments.
* * * * *
Roses: Two things this week. First, I’ve continued to go for bike rides with my LSE friend! It has been quite fun, although I’m still pretty bad at approaching a red light with cars stopped near the sidewalk; the combination of going slowly and not having much space for error means I tend to lose my balance. I think we’re going to try doing longer rides soon to build up to the Runnymede trip. Here’s a picture of us.
Second, as I mentioned earlier, I met up with another Marshall in the town of Ely — a cute little town in Cambridgeshire. The folk story (disputed, according to Wikipedia) is that the town is named after eels, because there was a lot of eel-fishing in the area. (Apparently the Domesday Book recorded the number of eels being fished in the area!) My friend and I decided to take a selfie with all of the eel-themed art we found as we walked along the eel trail (yes it is a real thing, and yes, I did pay £1 to buy their eel trail pamphlet). Also, the town is also famous because Oliver Cromwell’s house is there, and his house serves as the visitor center for the town. I should note, also, that Ely’s cathedral is the first British cathedral I’ve been in. And it was absolutely breath-taking. The size and grandeur of the space was astonishing, and I could really understand the sense of reverence that it would’ve induced in a medieval church-goer. Anyhow, here are some pictures.
Bud: I’ve started two books that are really exciting and stimulating. First is Hanna Pitkin’s The Concept of Representation, which I quoted from earlier; it is about, as one might guess, representation. I decided to read it because I realized that I hear and use the language of representation a lot, but I have no meaningful way of thinking about representation. Second is Eric Foner’s The Fiery Trial, about Abraham Lincoln and slavery. As I’ve hinted at in this post (and my very first post), I want to think about how one ought to respond to systemic injustice. History, just like political theory, can provide guidance.
Thorn: Since the last lockdown ended, I’ve been able to restart going to the gym, which has been nice, although I’ve hit a point in deadlifting where the constraining factor is my forearm grip strength, which is unfortunate. That was going to be the thorn, but as I just learned in the last few hours, London is entering a new tier of lockdown. So I guess the bigger thorn now is that I won’t be able to go to the gym for the rest of the year!