"Democracy" at Harvard and America

By Jason Morganbesser

Source: Harvard Government Department

Throughout the past few decades, American politics has shifted focus from traditional issues like the economy or social policy to an emphasis on “defending democracy.” In the 2024 election, a third of voters said their vote was primarily motivated by fears that American democracy was under threat, with a plurality of both Democrat and Republican voters saying that “democracy in the U.S.” was “very threatened” by contemporary politics. Within Harvard’s Government department, defending democracy and “democratic institutions” has similarly become the central research question for intellectual inquiry. Unfortunately, this trend has undermined our ability to have serious discussions about politics, both in America and at Harvard.

The emergence of “defending democracy” as the central issue of American political life has been as confusing as it has been bipartisan. On the right, following the 2020 election, concerns about the “deep state” and illegal voting have led both politicians and voters to prioritize fairness of elections above nearly any other political issue. As of March this year, the only major Republican domestic policy proposal was to nationalize the management of elections and require voter ID nationwide. This policy has been declared by President Donald Trump to be the “most important & consequential piece of legislation” in American politics, necessary to stop Democrats from, in his words, “rigg[ing] elections.” Indeed, Trump promised to halt all other congressional policy (in a Republican Congress) if they cannot pass this singular bill, and in support of this goal he has advocated for overturning traditional institutions of the American Republic like the filibuster.

Meanwhile, since the 2020 election, Democrats have also made “defending democracy” central to their political project. Many Democrats such as former President Joe Biden have declared that “MAGA Republicans” are “determined to destroy democracy.” Thus, Democrats say, America must nationalize election management and overturn stringent voter ID requirements in conservative states. During the Biden Administration, one of the flagship policies put forth by Congress was the For the People Act, a bill that sought to nationalize election management which Senate Majority Leader Chuck Schumer argued was necessary even if passing it required overturning institutions of the American republic like the filibuster. This shared political objective has become popular at the same time among both Democrat and Republican lawmakers without the slightest self-consciousness on either side. There must be some reason why rhetoric focused on “democracy” is naturally attractive to politicians.

As philosopher Herman Cappelen argues, the attractiveness of this rhetoric comes from the fact that the actual meaning of “democracy” is nebulous. Some (usually populist) politicians say that anything which limits majoritarian decision-making is undemocratic. Other (usually institutionalist) politicians say that anything which limits the influence of civil society is undemocratic.¹ Organizations like The Economist and Freedom House provide entire lists of factors to determine whether a country is a democracy, including everything from the presence of “civil liberties” to the “effectiveness” of the government.² Because there are so many different definitions of democracy, any political partisan can choose to define “democracy” however they wish without worrying about whether they are infringing on the actual definition of the term.³ As a particularly extreme example, one Chinese government document argues that China is not just a democracy — it’s the only functioning democracy in the world!⁴ “Democracy” can thus be — and often is — defined however a politician wishes, and thus, all that a politician needs to connect their goals to the “defense of democracy” is a creative enough redefinition of democracy.

Making this sort of rhetoric even more attractive, the term “democracy” is highly emotionally charged. When we read or hear “democracy,” we naturally associate it with positive emotions. These emotional impacts are what Cappelen calls the “lexical effects” of the term, and they strongly influence how we feel about the content of what is defined as “democratic.”⁵ When a politician says that checks on executive power, or voter ID laws, or any other political policy “defends democracy,” that makes us much more likely to support those policies, because hearing that something is associated with democracy emotionally influences us to support it. Thus, because of our cultural and emotional understanding of democracy, “defending democracy” often functions as a way for political partisans to use the lexical effects of “democracy” to justify unrelated political policies.⁶

As a result, as Cappelen argues, politicians can very effectively manipulate voters. Even worse, our disagreements become purely verbal and our discussion degrades. When Joe Biden and Donald Trump can both convincingly say to their supporters that they are defending democracy from the other, this is not because of a mere disagreement on facts. Instead, it is a disagreement based on two completely different understandings of what “democracy” even means. In this case, their disagreement is “purely linguistic,” a disagreement which results exclusively from different definitions of terms with strong lexical effects. Thus, cross-ideological discussion becomes intractable, and politics becomes polarized.⁷

This use of the lexically powerful term “democracy” as a method to justify political preferences has become equally significant at Harvard. In Harvard’s “Government” department, 37 out of Harvard’s 64 courses (around 58%) over the 2025-6 school year include readings or course content analyzing “democracy” or “democratic” institutions. Not including 10 purely quantitative courses in which students are exclusively taught skills (like how to conduct a survey), nearly 70% of courses include some analysis in terms of “democracy.” In contrast, just 7 courses include any analysis whatsoever of what it means to be a “republic.”

This is in spite of the fact that, just like in ordinary political life, it is unclear what “democracy” refers to in political science and political theory. Within academia, definitions abound, often referring to entirely different things.⁸ We can categorize definitions into five basic ones: “minimalistic” definitions, “pluralist” definitions, “substantive” definitions, “epistemic” definitions, and “agonistic” definitions.⁹ The significant disagreement over the criteria used to evaluate whether a system is a “democracy” is reflected in the pedagogy of Harvard’s Government department. Among the 37 courses with readings or course content that base analysis on a particular conception of “democracy,” no more than 12 use any one definition of democracy (32%).¹⁰ Thus, different students receive wildly different understandings of democracy in their classes — and with those wildly different understandings, wildly different prejudices to influence whatever policies they are naturally predisposed to support.

Without a clear definition to ground them, these courses often invoke “democracy” just as rhetorically as do ordinary political partisans. Take, for instance, GOV 94OA, “Inequality and American Democracy.” In the description of the course, the syllabus makes it very clear that “responsive democracy may have stalled, and in some cases reversed” because of “disparities of income, wealth, and access to opportunity.” The course readings include yet more lexically charged terminology justifying its political ideology. One book is about how the “rural versus urban” divide “threatens democracy.” Another reading is about how “invisible government policies” like tax relief for the wealthy also “undermine American democracy.” Here, the lexical effects of “democracy,” redefined to include certain normative ends, are used to emotionally influence students to accept a worldview. Perhaps contemporary America is too unequal. However, the right way to discuss that in an educational environment is to directly describe the exact problems resulting from that inequality, avoiding language with strong lexical effects such as rhetoric about the “defense of democracy.” Courses like GOV 94OA fail to discuss these problems neutrally.

However, even less overtly ideological courses have this same problem. GOV 1719, a course on “Political Communication Breakdown,” argues that contemporary partisanship and the resulting “epistemic problems” have led to a situation in which “American democracy is in crisis.” By defining these epistemic problems as displays that our system is not democratic enough and as contradicting our democratic principles, this course description implies that these epistemic problems are “undemocratic.” This course, too, uses the lexical effects of “democracy” to influence student perspectives, albeit in a less overt way than GOV 94OA. Again, the conclusions — in this case that our political system often fails to communicate the truth to voters — may in fact be right. However, by defining “democracy” to include professors’ preferred normative ends, even if those normative ends are right, these courses have moved away from dispassionate analysis and towards political advocacy.

With such a wide diversity of concepts of “democracy,” it is hardly surprising that students increasingly find it difficult to discuss politics across political lines. Whether measured by high levels of political self-censorship, fears of political disagreement, or attempts to shut down visiting speakers, Harvard students are unusually avoidant of political discourse. But who can blame them? The main form of “political discourse” taught to politically interested students at Harvard is not the disinterested discussion of political principles, but rather the emotional cajoling of students into adopting ideologies because of their affinity to artificially constructed conceptions of “democracy.” Thus, students’ disagreements with one another arise emotionally, merely as a result of however democracy has been redefined to a particular student.

Therefore, upon meeting another student with an alternate view of “democracy,” how can two Harvard students with different opinions based upon these redefinitions of democracy possibly hope to communicate and adjudicate their different conclusions? These differences of opinion, after all, are not based in rational disagreements but rather in the students’ different definitions of democracy. Those definitions, just like in the ordinary political case, will inevitably render any further disagreement “purely linguistic” and thus unproductive. As a consequence, it is only rational for Harvard students to avoid impossible-to-resolve political discussions. Just like how our politics polarizes and stagnates, so does our campus discourse similarly polarize and stagnate.

Of course, most professors do not use “democracy” terminology purposely to manipulate. They are, after all, just trying to accurately depict political debate. And most politics today is debated in terms of the “defense of democracy.” As a result, much political-academic literature is written in terms of that defense. Thus, as Cappelen notes, our pedagogical problems are likely less a problem of individual professors and more a problem with the structure of political theory sneaking normativity into the discussion through the use of the term “democracy.”¹¹ But it is unnecessary to define political discussion according to those terms. As Cappelen argues, the term “democracy” does not add much of anything to political-academic life. Who votes, what institutions structure their voting, and what results they reach can be easily described and debated directly, without mediation through the term “democracy.”¹²

Pedagogically speaking, this problem is even more easily resolved. There is no need for discussion with students to be held in terms of “democracy.” Instead, professors can teach directly about what they want to discuss. Debates about the “impact of political polarization on our democratic crisis” in GOV 1719 can be easily re-defined as “debates about the impact of political polarization on our ability to arrive at policies based on truth.” Debates about whether inequality undermines democracy in GOV 94OA can easily be reformulated as discussions about whether inequality undermines the political power of the poor and how that loss in political power harms America. These are completely ordinary and valuable ways of discussing these issues — and they involve none of the emotional sleights of hand which our current debates entail.

What makes resolving this problem difficult, however, is the underlying collective action problem that is incentivizing the use of “democracy”-laden rhetoric in classes. Politically interested students will be more interested in taking classes about topics on which they already have strong opinions. And, of course, with contemporary politics focused on the “defense of democracy,” students will naturally be interested in classes which use such lexically powerful words.

While this collective action problem is difficult to solve, similar collective action problems are resolved by universities all the time. A recent example is the Harvard Administration decision to cap the assignment of A’s at 20% of students in any particular course. As Amanda Claybaugh’s report itself recognized, the proposed policies were aggressive. Nevertheless, they were, in the eyes of the administration, necessary because “collective action problem[s]” such as grade inflation necessitate aggressive action. As President Alan Garber has recognized, Harvard’s lack of genuine campus discourse has also become a crisis. Similarly aggressive action is thus necessary to respond to Harvard’s crisis of self-censorship. If Harvard is serious about bringing back campus discourse, it should cap the number of courses on “democracy” offered in the Government department each year.

The policy would be controversial and assertive. But viewpoint diversity cannot be solved just by task forces and reports. As Harvard has recognized, decades-long collective action problems require aggressive engineering to solve them. To stop the continued polarization of political discourse at Harvard, we need strong action. Refocusing the purpose of Harvard’s Government department on what we think rather than what we feel would be a strong first step. Only through this decisive action can we hope to engender political discourse on campus that does not devolve into the mudslinging of partisan politics or the silence that follows collective incomprehension.

1. Cappelen, Herman. The Concept of Democracy: An Essay on Conceptual Amelioration and Abandonment. Oxford University Press, 2023, pp. 128-30.

2. Ibid., pp. 171-172.

3. Ibid., pp. 197-199.

4. Ibid., p. 11.

5. Ibid., pp. 90-91.

6. Ibid., pp. 92-3.

7. Ibid., pp. 130-132.

8. Ibid., p. 13, pp. 130-2.

9. To give exact descriptions of these definitions: “Minimalist” definitions say a system is “democratic” if it has certain institutions such as elections, “pluralist” definitions say that a system is a “democracy” if it has a set of methods such as interest groups, political parties, and legislative debate that ensure public participation, “substantive” definitions say that a system is democratic if its institutions result in certain desirable ends such as equality or liberty, “epistemic” definitions say that for a system to be a democracy, there must be constant communication between political decision-makers and the general population in the determination of policy, and “agonistic” definitions say that democracy must include popular mobilization and struggle between interest groups. Joseph Schumpeter is an exemplar of a minimalist theorist, Robert Dahl of a pluralist theorist, C.B. MacPherson of a substantive theorist, Hélène Landemore of an epistemic theorist, and Chantal Mouffe of an agonistic theorist.

10. To be precise, an estimated 10 courses use minimalist definitions (GOV 20, 30, 94OA, 94OF, 94FO, 1135, 1171, 1248, 1270, 1295), 12 use pluralist definitions (GOV 30, 40, 94EU, 94GD, 94JO, 1003, 1061, 1062, 1074, 1102, 1270, 1730), 12 use substantive definitions (GOV 10, 63, 94AU, 94CT, 94EM, 94JE, 94PE, 1003, 1004, 1041, 1102, 1203), 10 use agonistic definitions (GOV 20, 94AA, 94AU, 94FA, 1060, 1171, 1192, 1248, 1295, 1719), and 9 use epistemic definitions (GOV 10, 94DC, 1003, 1004, 1061, 1074, 1146, 1203, 1719). Some courses are double counted because readings in the syllabus adhere to multiple definitions.

11. Cappelen, The Concept of Democracy, pp. 190-1.

12. Ibid., Ch. 8-9.