The Future of Political Theory: The Normative Science of Politics

Alexandra Oprea Duke University
Abstract: This essay examines the contrasts between political theory and political science, with a view toward the future trends of the two in relation to one another in the twenty-first century. The author ventures to make three predictions and invites readers to offer their own insights in response.

Political theory is written by perfectionists.

Think of Rousseau’s Confessions describing the painful process through which sentence after sentence arranges itself in his head before he can transfer them to paper. And even then, he says, “[m]y scratched out, blotted, mixed up, indecipherable manuscripts attest to the trouble they have cost me” (Rousseau [1782] 1995, 95).

Political theory is written by perfectionists.

More recently, think of the hundreds of footnotes that adorn the book manuscripts of John Rawls as he continued to address criticisms, qualifications, and extensions to his arguments. His 1971 magnum opus, A Theory of Justice, can be described as a 560-page refinement of his 1950s paper, “Justice as Fairness.” And more personally, think of the written, rewritten, and endlessly edited papers that are currently sitting in either your physical or your electronic folders.

But if political theory admits of the endless tinkering of perfectionists in search of justice, truth, and reason, the practice of politics offers no such hope of precision, nuance, and certainty. Whether we describe politics, following Weber ([1919] 2004), as “a slow, powerful drilling through hard boards” (93), or whether we follow Foucault ([1976] 2003) in describing it as “a continuation of war by other means” (16), political life is populated by imperfect people with imperfect information working out imperfect agreements.

So how can our breed of perfectionists understand the imperfect world of politics? Or, more precisely, what might a normative science of politics look like given the nature of our subject matter?

So how can our breed of perfectionists understand the imperfect world of politics? Or, more precisely, what might a normative science of politics look like given the nature of our subject matter?

The question I ask is by no means original. In one guise or another, it has preoccupied political thinkers from Aristotle to Arendt and from Wollstonecraft to Wolin. As a political theorist of the perfectionist bent myself, I dare not proceed to speculations about the future before (at least briefly) consulting the wisdom of the past.

In the Nicomachean Ethics, Aristotle described the topic of inquiry for political science as “noble things and just things” ([Aristotle] Reeve 2014, 3). This subject, Aristotle claimed, showed such variation among individuals and societies that we can only expect to describe the truth about it “roughly and in outline” ([Aristotle] Reeve 2014, 3). Lest this pronouncement disappoint, however, Aristotle reminded us that different standards apply to different areas of inquiry, since “accepting persuasive arguments from a mathematician is like demanding demonstrations from a rhetorician” ([Aristotle] Reeve 2014, 3). Political scientists therefore need to find the proper standard that their subject allows. And since our field, he claimed, seeks after noble and just actions in the political societies we inhabit, we might as well accept that our knowledge will never amount to the perfect mathematical demonstration our perfectionism aspires to.

Despite Aristotle’s warnings, many of the political theorists we study and teach sought precisely such mathematical proofs in their study of politics. Arguably the most famous of them is Thomas Hobbes. Looking admiringly toward geometry, Hobbes faulted his contemporaries for not being rigorous enough in their own studies of politics: “For there is not one of them that begins his ratiocination from the Definitions, or Explications of the names they are to use; which is a method that hath been used onely in Geometry; whose Conclusions have thereby been made indisputable” (Hobbes [1651] 1996, 34). The Leviathan shows just how far deductive political theory can arrive with just a few simple axioms of human behavior (and possibly a few creative definitions). Assuming neither religion nor morality to begin with, he builds an entire system of politics on the twin edifices of fear and calculating self-interest. This imposing masterpiece continues to stare us down every time our perfectionist aspirations carry us too far into the contentious territory of democratic politics and distributive justice.

In the search for a method for the normative study of politics, Hobbes was hardly a solitary voice. Descartes and Bacon, Locke and Rousseau, Mill and Tocqueville, all aspired to discover or create their own “new political science," whether or not for “a world altogether new” (Tocqueville [1835] 2000, 7). While some expected this political science to function deductively on the model of mathematics, others expected their conclusions to have more of the probabilistic precision of rhetoric. What they shared was a focus on the collective life that human beings had lived, could live, and should live.

I make my predictions for the future as an invitation for those more skilled at observing and more experienced in living in such societies to offer their own insights in response or in refutation.

I take the opportunity of this brief historical investigation to reintroduce a figure who historians of political thought have generally forgotten, but who can be read as bringing together the scientific aspirations of both Aristotle and Hobbes. This is François Guizot. Living through the turbulent aftermath of the French Revolution, Guizot was both a scholar and a politician, the Chair of Modern History at the Sorbonne, and the Prime Minister of France. His political thought dramatizes the difficulty of developing a normative science of politics. On the one hand, Guizot acknowledged truth, reason, and justice as the standards for political life. These standards are given “from above” and their existence is fundamentally independent of the will of human beings, individually or collectively (Guizot [1861] 2002, 165). On the other hand, Guizot never forgets that all political societies are imperfect, transient, and historically contingent. Laws, institutions, and leadership are always and necessarily prone to falsehood, irrationality, and injustice. It is in the nature of human communities to perpetually arrange and rearrange their institutions to approximate truth, reason, and justice while perpetually failing to actualize them. The best one can hope for in the world of politics is finding more rather than less of truth, reason, and justice in the communities we inhabit and shape. As a political theorist, Guizot’s method combines moral reasoning with what we might call comparative politics in a historical perspective, aiming at justice while studying the different degrees of injustice in the available institutions of France, England, and the rest of Europe.

This brief historical detour finally brings us to the more mundane world of departmental politics. I choose these political societies explicitly committed to the study of politics as a microcosm of the issues confronting political theorists who continue to search for the appropriate degree of rigor and the appropriate methods for understanding the problems of the twenty-first century. I make my predictions for the future as both a junior member of such a political community and a participant in the comparative study of political science departments that takes place at professional conferences and through the social practice of gossip. But I also make these predictions as an invitation for those more skilled at observing and more experienced in living in such societies to offer their own insights in response or in refutation.

Prediction 1: Political theorists are here to stay.

By this prediction I do not only mean the general prediction that the world will still include individuals who will think and write about politics, whether they do so in the columns of flagship newspapers and political magazines or in the margins of mainstream politics. I intend to make the specific prediction that political theorists will continue to find their home in large numbers in political science departments, where they will continue to have mutually beneficial (if occasionally challenging) interactions with more empirically oriented political scientists.

While the placement of political theorists within political science departments is a contingent phenomenon of developing disciplinary boundaries within American universities during the late nineteenth and early twentieth centuries, I want to argue that this accident could be justified on principle. Given the methods applied by canonical political theorists in their search for a normative science of politics, our subfield has always relied on both philosophical reasoning and empirical evidence about the comparative development of political institutions and societies. In our striving for both theories of justice, democracy, and liberalism as well as action-guiding principles for the imperfect world of politics, we deploy a variety of methods appropriate for the merely probabilistic precision allowed by political matters.

For the political societies of political scientists and political theorists to continue to flourish in the twenty-first century, I see two complementary trends that I expect will lead to a more peaceful cohabitation (at least a modus vivendi if not an overlapping consensus). I intend these predictions as both empirical claims about current and future developments in the field and as normative claims about how empiricists and theorists should respond to these changes.

Prediction 2: Empirical political scientists will become more transparent about their normative goals.

The twenty-first century has already witnessed significant developments in three areas of empirical political science: (1) data collection; (2) technology for analyzing large data; and (3) methods for analyzing data (the identification revolution). All three of these developments, which I briefly address in turn, have dramatically improved the technical possibilities for empirical research in political science. As a result, normative debates about which aspects of politics are important to measure and analyze will become more important aspects of the empirical work done by our colleagues.

The past decade alone has witnessed a proliferation of new data sources from a variety of governmental and private sources. At the macrolevel, the available measures of GDP and political regime types (Polity IV, Freedom House, and others) have been supplemented by a variety of macrolevel indicators about literacy, education, health, inequality, well-being, and other quality of life measures (many of these are included in the Human Development Index). These are supplemented by increasingly precise measures of quality of political institutions, including measures of corruption, transparency, clientelism, electoral competition, political polarization, and economic, political, and civil liberties. At the state level, more governments have improved their data collection and reporting, and nongovernmental organizations have become more adept at independently verifying the data released by governments, especially through representative surveys of the population and other tools for the independent verification of local data collection agencies. As the survey and reporting technology improves, political scientists will have more and more detailed information available to study politics at the local, national, and international level.

As their empirical tools are becoming increasingly sophisticated, political scientists are able to refine the concepts they are aiming to empirically measure, such as democracy, education, liberalism, polarization, or equality of opportunity.

As better information becomes available in larger quantities, the increase in data analysis capacity has proceeded apace. While political science is not generally dealing with the petabytes of data currently used in other fields, increasingly fine-grained, individual-level information that fluctuates through time is starting to be available to survey voter behavior online. The hardware for storing these data and the software for processing them are allowing political scientists to investigate previously unknown features of voting behavior in a mass democracy, as well as features of citizens’ social and economic behavior that affect the quality of political institutions.

Finally, the identification revolution combined with methodological refinements coming out of statistics, economics, and other social and natural sciences has allowed political scientists to become increasingly precise in the type of causal claims and predictions that they can make. Among the methodological innovations that can allow for fruitful collaboration between empirical and normative political scientists are historical discontinuity research designs through which political scientists can use unexpected historical junctions as natural experiments to disentangle a variety of causal mechanisms.

As their empirical tools are becoming increasingly sophisticated, political scientists are able to refine the concepts they are aiming to empirically measure, such as democracy, education, liberalism, polarization, or equality of opportunity. With their unique focus on conceptual analysis, theory building, and investigation of fundamentals, political theorists will be increasingly necessary to the field as it tries to understand topics of growing global importance: authoritarian politics, illiberal democracies, the future of capitalism, and environmental politics.

Prediction 3: Normative political scientists will become more conversant with the methods and findings of empirical political science

In one of his lesser read pedagogical essays called Some Thoughts Concerning Reading and Study for a Gentleman, Locke divided politics into two distinct parts. The first part, he claimed, was concerned with “the original of societies, and the rise and extent of political power” (Locke [1689] 1824, 240). For this study, Locke recommends the treatises of Hooker, Sydney, and Pufendorf, as well as his own (anonymous) Two Treatises of Government. The second part he calls “the art of governing men in society” (Locke [1689] 1824, 240). This aspect of political science is primarily concerned with the particular politics of one’s political society as distilled from experience and the study of history, geography, and law. We might be tempted to read Locke as identifying a natural division of labor between political theorists and political scientists, between those of us concerned with philosophically deducing how the world should be and those concerned with empirically explaining and describing how the world actually is. I would argue against this interpretative choice. Even if political theory is primarily about the type of work Locke undertakes in his Two Treatises, his analysis of what governments should do depended on his extensive research into legal and political history, anthropology, and comparative politics. To derive his conclusions about what should be, he closely studied what has been and what can be.

I prefer to read Locke’s division instead as a psychic division of labor within the mind of individual political theorists, a division of labor between the normative philosopher and the political scientist.

The development of academic political science has offered political theorists a wealth of new studies of political institutions and political behavior, not only in the American context, but cross-culturally and across different historical periods. 

Political theorists have always constructed their theories of justice and their ideal political institutions upon what Immanuel Kant called “such warped wood as that which man is made of” ([1784] 1970, 42). Their unique concern with both the noumenal world of reason, truth, and justice and the phenomenal world of politics has always posed unique challenges. Whether they were concerned with the search for the very best political institutions, such as Plato, or with the bare minimum to avoid political conflict, such as Hobbes, political theorists have always built upon the research undertaken in other fields. These have included fields as varied as economics (Marx) and musicology (Plato), mathematics (Descartes) and anthropology (Locke), psychology (Freud) and botany (Rousseau), history (Guizot) and law (Grotius). The development of academic political science has offered political theorists a wealth of new studies of political institutions and political behavior, not only in the American context, but cross-culturally and across different historical periods. Unfortunately, the presentation of these findings in terms of statistical results and mathematical proofs has led to reduced cross-subfield interactions of political scientists. Along with the identification revolution and the increasing methodological sophistication of political scientists, however, I predict that the presentation of empirical analysis will also become more visually, analytically, and conceptually sophisticated. The growth of visualization technology speaks to the potential for presenting empirical research to nonempirical scientists. The normative political theorists can drive this process by increasingly demanding, citing, and incorporating the findings of political scientists. Over time, I predict that specialized training for theorists interested in interpreting empirical data will become more easily available.

One of the advantages of political theorists engaging more closely with political science is the potential it holds for research in comparative political theory. Our colleagues in comparative politics have produced extensive studies of political societies in Asia, the Middle East, and Latin America that political theorists promoting cross-cultural normative inquiry in Chinese, Indian, Islamic, or Latin American political thought can benefit from. Another normative advantage would be the return of normative political theory to the issues being addressed by scholars working in political economy, particularly economic and political inequality, income redistribution, and the provision of public services. A more empirically grounded normative science of politics would be able to complement the work of these scholars to consider how much inequality political societies should tolerate and which public services should be provided as a matter of justice.

At the end of these three predictions, I return to the point of departure:

Political theory is written by perfectionists. But human beings, even if perfectible, are not capable of perfection.

Political theory is written by perfectionists. But human beings, even if perfectible, are not capable of perfection. At the end of one his last and most empirically informed studies in normative political theory, called Considerations on the Government of Poland and on Its Proposed Reformation, Rousseau reminded his readers pursuing either a normative or an empirical science of politics that all the works of men are as imperfect, transitory and perishable as man himself” ([1772] 1985, 115).


Aristotle. 2014. Nicomachean Ethics. Translated by C. D. C. Reeve. Indianapolis, IN: Hackett Publishing Company.

Foucault, Michel. (1976) 2003. “7 January 1976.” In “Society Must be Defended”: Lectures at the College de France 1975–1976, edited by Mauro Bertani and Alessandro Fontana, 1-23. Translated by David Macey. Picador, NY: St. Martin’s Press.

Guizot, François. (1861) 2002. The History of the Origins of Representative Government in Europe. Translated by Andrew R. Scoble, introduction by Aurelian Craiutu. Indianapolis, IN: Liberty Fund.

Hobbes, Thomas. (1651) 1996. Leviathan. Edited by Richard Tuck. Cambridge, UK: Cambridge University Press.

Kant, Immanuel. (1784) 1970. “Idea for a General History with a Cosmopolitan Purpose.” In Kant Political Writings, edited by H. S. Reiss, 41–54. Cambridge, UK: Cambridge University Press.

Locke, John. (1689) 1824. “Some Thoughts Concerning Reading and Study for a Gentleman.” In The Works of John Locke in Nine Volumes. Vol. 2: An Essay concerning Human Understanding Part 2 and Other Writings. London: Rivington.

Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press.

Rousseau, Jean-Jacques. (1772) 1985. The Government of Poland. Translated by Willmoore Kendall. Indianapolis, IN: Hackett Publishing Company.

Rousseau, Jean-Jacques. (1782) 1995. “The Confessions.” In The Confessions and Correspondence, Including the Letters to Malesherbes, translated by Christopher Kelly, edited by Christopher Kelly, Roger D. Masters, and Peter G. Stillman. Hanover, NH: University Press of New England.

Tocqueville, Alexis de. (1835) 2000. Democracy in America. Translated and edited by Harvey C. Mansfield and Delba Winthrop. Chicago, IL: University of Chicago Press.

Weber, Max. (1919) 2004. “Politics as a Vocation.” In The Vocation Lectures, edited and with an introduction by David Owen and Tracy B. Strong. Translated by Rodney Livingstone. Indianapolis, IN: Hackett Publishing Company.

See also the brief essays on the future of political theory by Nora Hanagan, Michael Gillespie, Chris Kennedy, and Sam Bagg, which come out of the same event and are intended to be in conversation with this paper.

This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. The images in this work are not subject to this license.