A Liberal Polity: Ideological Homogeneity in Political Science

President Trump “is a spectacularly unqualified and catastrophically unfit egomaniac who poses an overt threat to the Republic” (Mayer 2019, 1). Those words, from the syllabus of a University of Wisconsin course on the American Presidency, intimidated a student, provoked a lawmaker, and outraged a media personality. Wisconsin State Representative Dave Murphy (R-56) was “appalled by [Mayer’s] politically polarized characterization of the Trump presidency” (Murphy 2019). Tucker Carlson sputtered about the syllabus’s “ludicrousness…and its just sort of doctrinaire Democratic talking points” (Campus Craziness 2019). This is only the latest episode in the long-running claim that universities are dominated by liberal faculty and therefore have become “leftist echo chambers” promoting liberal ideology at the expense of conservative thought.

I had three reactions to this episode, which was covered in detail by Moynihan (2019).

As a political scientist, I had no opinion at all. The behaviors observed are utterly predictable. Political actors act strategically to advance their personal, partisan, and policy interests. The student scored an appearance on Fox. The legislator fed his partisan base while signaling his policy goals. The commentator inflamed his cable-leading audience. Ordinary stuff.

As a professional, I was troubled. Republicans have become increasingly suspicious of higher education, and public support for it has declined. To the extent that universities are seen as merely another arena for political struggle, our research and teaching missions are threatened.

As a person, I am disgusted—and, as we now know, liberals are difficult to disgust (Aarøe, Petersen, and Arceneaux 2017; McAuliffe 2019)—especially by Tucker’s signature combination of ignorance and invective (as Tucker stated, “[a]ll the dumb kids end up teaching at the University of Wisconsin”). In Mayer’s defense, the College Republicans issued a statement declaring that he is “an intellectually engaging professor [who] treats conservatives fairly” (College Republicans of UW–Madison 2019). Yet, had Mayer not disappeared from social media, I know his inbox would have overflowed with the ugliest vitriol. I am relieved it was not my syllabus.

Although I had all three reactions, my personal one was the most deeply felt and the scientific one the least important. Our strictly scientific roles typically play the smallest role in our professorial duties of scholarship, teaching, and service. Professors do not design their courses, advise their students, and interact with their colleagues on the basis of scientific principles; they are guided largely by their professional norms and personal perspectives. Although we might strive for neutrality, it is unlikely that we achieve it.

It might be tempting to dismiss the idea that political science has a distinctly liberal bias because the most partisan attacks are also the most implausible ones. However, I also believe that the core charge of liberal (or Democratic) bias is true; that the cause of the bias is the relative lack of ideological diversity; and that the greater the bias, the more problematic it is for our discipline.

As a professor, I feel uneasy about the state of ideological diversity within political science and the broader academic community. Perhaps that is because I live in a liberal bubble: my zip code supported Clinton over Trump by a 78–15 margin in 2016. The chances are good that you also live in one: the closer you live to a university, the higher your chances of living in a blue neighborhood. (To explore your neighborhood, see Bloch et al. 2018; see also Al-Gharbi 2019 and Sachs 2019, but also Abrams n.d.) At political science gatherings, it seems to me that the jokes about conservatives are pointed; about liberals, ironic. Conservatives might reasonably assume that their views face heightened criticism and suspicion, with liberal viewpoints less scrutinized. I have never heard liberal political scientists confide that they feel isolated on the basis of their ideology; I have heard these concerns from conservatives.

Let me return, for a moment, to the “controversial” American Presidency syllabus. The structure is neutral, with parallel sections beginning “To his supporters, [Trump]...gleefully flouts the norms of governing” and “To others, [Trump] is spectacularly unqualified” (Mayer 2019). In Trumpian fashion, these opinions are attributed to the unverifiable “some people say.” This is not the whole story, however. Trump receives two generic sentences of praise, whereas the President is given a full, detailed paragraph of specific, discrediting statements. I recognize the extraordinary challenges that the Trump administration poses for political scientists seeking to convey neutral messages, and that opposition to his presidency is hardly limited to liberals or Democrats. Still, had the praise-to-blame ratio of Mayer’s description been reversed, I would have judged it as projecting a highly pro-Trump message—yes, to be biased.

Perhaps you also have followed various controversies regarding ideology in the academy and have reflected on their validity and meaning. The PS editorial board has; hence, the idea for this symposium. The articles in this issue examine the extent of ideological diversity in political science (Atkeson and Taylor 2019; Marineau and Williams 2019) and consider the problems that the lack of diversity may create for our discipline (Campbell 2019; Gray 2019; Wilson 2019; Zigerell 2019). Atkeson and Taylor use voter data to show the large disparity between Democrat and Republican faculty registrations; in contrast, Marineau and Williams demonstrate that religious, liberal arts colleges are among the most politically diverse institutions in higher education. Wilson argues that political science lacks, in particular, representation from the cultural and religious right. Along with Gray, he contends that our political liberalism leads to blind spots, such as a tendency to view conservativism as a pathology to be diagnosed rather than an ideology to be understood. Campbell’s article concludes that political science’s liberalism “pose[s] a threat to the soundness of our research and teaching, and can lead to the erosion of trust, respect, and support for the discipline.” Zigerell furthers these concerns and offers suggestions to mitigate the problems of ideological bias.

Contributions for this symposium were solicited through word-of-mouth, my own networks, and venues such as the Political Science group on Facebook and #WomenAlsoKnowStuff. The original goal was to broadly consider ideological diversity; however, this symposium now is focused squarely on the study of US politics and ideology (i.e., liberalism and conservativism) as understood in the American context. We understand that different issues will arise among international political science and scientists, and we hope that these matters will receive more attention in the future. A second PS symposium, forthcoming later this year, will center on ideological concerns in the classroom.

The large literature on confirmation, availability, and intergroup bias suggests that we are unlikely to be as neutral as we might think. Imagining ourselves unbiased is perhaps a “tribute to the human capacity to take our ideological biases and convince ourselves that they’re not biases at all but are instead inescapable rationality” (Waldman 2019).

Some readers might object to the notion that political science lacks ideological diversity, despite the fact that a substantial majority of political scientists are liberal. That reaction is understandable. When they are accused of having a liberal bias, we might expect political scientists individually and collectively to respond in the ways that any interest group (e.g., the NRA and its members) would respond to criticism. One predictable response is to disregard the critics as outraged, aggrieved extremists (here’s to you, Dinesh D’Souza). Another is to avoid communal responsibility and to attribute any obvious liberal bias to the rot of a few bad apples—or simply to remain silent and hope the controversy soon fades away. Finally, interest groups—theirs, yours, and mine—claim the moral high ground: “We stand here not as a special interest but as the defender of universal values” (i.e., rights, liberty, harmony, and so forth). It is fairly rare for an interest group to look in the mirror and say, “Yes, your claims against us are valid, and so we pledge to reflect on our errors and to change our ways.” It is even rarer if those changes would reduce the interest-group’s autonomy, prerogatives, or influence.

It would not surprise me if our liberal readers do not believe that political science has a liberal bias or that, if it does, that the bias is problematic for our discipline. We might think, as I am tempted to, that I do not have a liberal bias in my role as a professor and, if that is true for each I, the sum of all Is are unbiased. Resist! The large literature on confirmation, availability, and intergroup bias suggests that we are unlikely to be as neutral as we might think. Imagining ourselves unbiased is perhaps a “tribute to the human capacity to take our ideological biases and convince ourselves that they’re not biases at all but are instead inescapable rationality” (Waldman 2019).

That conservatives believe that political science has a liberal bias would be consistent with research showing that members of minority groups are more likely to perceive bias than those within the dominant group. Blacks are more likely to perceive racial bias than whites (Pew Research Center 2016); women are more likely to perceive sexual bias than men (Pew Research Center 2015). White Christians are among those least likely to believe that gays and lesbians face substantial discrimination (Cox, Lienesch, and Jones 2017). That blacks, women, and LGBTQs, in fact, have faced more discrimination has not been persuasive to majority groups who believe they do not.

If political scientists are like other humans, we cannot be confident that liberal and conservative political scientists will see eye-to-eye on matters of bias within our discipline. As Mitchell et al. stated: “Partisan and ideological differences can even lead individuals to reach different interpretations of the same objective policy reality” (Mitchell et al. 2014, 342; citing Gaines et al. 2007). Of special relevance to us as scholars: “Perversely, the better informed more effectively used interpretations to buttress their existing partisan views” (Gaines et al. 2007, 957).

Political bias is difficult to prove (see, e.g., Musgrave and Rom 2015). Attempting to demonstrate that our discipline has a liberal bias, however unintentional, and that the bias is harmful is beyond the scope of this brief article. I can hear the doubters say: “Show me the bias. Without specific, compelling evidence, I will not reject the null hypothesis.” Yet, some reviewers of this article argued that the liberal bias of our discipline is so pronounced that to deny it is to be willfully blind. Pleasing neither side, I instead argue that we should conclude that such bias exists if there are disparate impacts or if bias is more probable than not. These standards have been promoted as reasonable when intention is difficult to prove and when civil claims are made. Liberals generally, but not always, have favored these standards in examining racial discrimination and sexual misconduct.

Whereas there are differing estimates as to what the ratio of Democrats to Republicans and liberals/progressives to conservatives is within political science, there can be no doubt that (1) the ratio is generally much higher than one; and (2) elite schools tend to have more liberal faculties, with—of course—wide geographic and institutional variation. It has been suggested (Al-Gharbi 2018) that ideological conservatives are more underrepresented in social science, relative to their proportion in the population, than women, blacks, and Latinos. Political conservatives are a distinct minority within our discipline.

Does the political ideology of those in our discipline matter in our work as scholars, teachers, and administrators? Does it matter that there are many more liberal political scientists than conservative ones? In thinking about this question, I reflected about what political scientists know about other political actors. As professors, we too work in a political environment and, through our universities and professional activities, we authoritatively allocate all sorts of values. Within this environment, is it possible to imagine that we, as political scientists, behave in ways that resemble those whom we often study: judges, politicians, bureaucrats, and citizens? We act as judges when we resolve the various disputes that invariably come across our desk. We act as politicians when we advocate for, deliberate on, and at times determine department and university policies, as well as when we distribute patronage. We act as bureaucrats when we implement these policies, often with broad discretion in doing so. As citizens, we act in that capacity in our daily interactions with others.

Within this environment, is it possible to imagine that we, as political scientists, behave in ways that resemble those whom we often study: judges, politicians, bureaucrats, and citizens?

All judges ever confirmed claimed that they would decide the law neutrally, without favor or bias, and I do not doubt the sincerity of their affirmations. Justices Clarence Thomas and Ruth Bader Ginsberg both stressed the importance of judicial impartiality in their confirmation hearings. Their impartiality nonetheless allows them to disagree in ideologically predictable ways: on virtually every 5–4 decision, they vote against each other. Although 5–4 decisions account for only 20% of SCOTUS decisions—most decisions do not invoke ideological differences—they often are those in which the stakes are highest. Indeed, the view that judges’ decisions are influenced by extralegal factors (including ideology, strategy, and political environment) is “[t]he dominant theory of judicial behavior in the field of political science” (Yates, Cann, and Boyea 2013, 849; cited in Segal and Champlain 2017).

When political scientists act as judges, is there reason to believe that they are exempt from the attitudinal model? Should we be concerned if a majority of our judges shared the same ideological predispositions, especially if those dispositions did not match ours?

Politicians, unlike judges, make no claims to impartiality. They believe that they are seeking to advance the public interest in ways that, not coincidentally, also benefit their partisan and personal interests. Given that political scientists are overwhelmingly Democratic—and have been since the 1950s, when the APSA was labeled “largely a one-party organization” (Turner and Hetrick 1972, 374; Turner, McClintock, and Spaulding 1963; Roettger and Winebrenner 1983)—it is reasonable to conclude that to the extent that partisanship bleeds into faculty behavior, the blood is much more likely to be blue than red. Moreover, whether or not political science favors Democrats, it is clear that whenever there has been a political tussle, Democrats have favored political scientists (see, e.g., Mulhere 2015).

We may not see ourselves in this way, but perhaps in most of our ordinary academic work, we are more like bureaucrats than judges or politicians: we implement policies in the classroom, department, and school. What lessons might the study of administration teach us? One lesson is that because we humans are prone to seek benefits for ourselves and our allies, administrators should strive to be neutral servants to avoid the political abuse of office (Weber 1978). Although the idea that bureaucrats can be purely neutral has been discredited, the core concept of avoiding partisan favor remains. Because we do not trust bureaucrats to be neutral of their own volition, we establish rules (e.g., the Hatch Act) to enforce political neutrality. Fortunately, the partisan composition of the federal workforce is more balanced than that of the academy, with roughly equal numbers of civil servants identifying as Republican and Democrat (Katz 2015). Although giving lectures and awarding grades are not as important as giving fines or awarding contracts, should we be concerned if the bureaucracy were dominated by civil servants from a single party?

Finally, we are citizens of our university communities. In this role, does our partisanship matter? Yes: citizens are more dubious of the motivations and ulterior motives of politicians from opposing parties than co-partisans (Munro, Weih, and Tsai 2010). Increasingly yes: since 1980, those who identify with a political party have developed more negative views of the other party and its supporters: “[b]oth Republicans and Democrats increasingly dislike, even loathe, their opponents” (Iyengar, Sood, and Lelkes 2012, 405). Resoundingly yes: “[H]ostile feelings for the opposing party are ingrained or automatic in voters’ minds…Today, the sense of partisan identification is all encompassing and affects behavior in both political and nonpolitical contexts” (Iyengar and Westwood 2015). Partisan hostility might not be part of our typical day, however. Citizens who are better educated, more affluent, and more privileged—like professors—are less likely than others to have political opposites in their social circles (Mutz 2006).

In a polity in which the judiciary, legislature, bureaucracy, and citizenry were dominated by a single party, those in the minority might understandably believe that their interests are neither heard nor included in deliberations and decisions. Or, as one reviewer of this article put it, “Unless political scientists are superhuman, social science laws apply to us like they apply to everyone else” (Anonymous 2019).

The APSA demonstrates the liberal value of inclusiveness regarding minority voices based on race, class, gender, and sexual orientation. As APSA’s Political Science in the 21st Century report states (APSA 2011, 13):

[W]e contend that who does the research matters and that political science still has a long way to go in diversifying the profession. We are not, in this instance, arguing for “diversity for diversity’s sake,” as an abstract progressive value, but rather for an understanding of how differently individuals are situated within society…The presumption that a group of individuals of mostly the same background…can comprehensively study the politics…is deeply flawed and can limit the accuracy and relevance of the resulting work.

Of course, conservatives—an ideological minority in political science—have experienced neither the pervasiveness nor the severity of bias felt by ethnic minorities, women, and LGBTQs. Conservatives have positions of power and prestige in our society. Within our discipline, however, their minority position disadvantages them, and it does not disadvantage them alone. Broadening our discipline’s ideological diversity will benefit the scholarship, teaching, and service of the liberal majority. The irony here is clear. Conservatives tend to oppose group-based identifications and diversity measures; liberals generally favor them. To be consistent, liberals should call for increasing representation from a politically relevant minority and conservatives should reject that call. I hope this symposium will lead to a productive discussion of the liberal polity that is the political science discipline.